List of Contributors
R. Anand, Center for Brain Health, The University of Texas at Dallas, 1966 Inwood Road, Dallas, TX 75235, USA W.S. Anderson, Department of Neurosurgery, Meyer Building 7-113, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21287-7713, USA P.C. Bickford, Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA S. Birnbaum, Department of Neuroscience, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA C.V. Borlongan, Department of Neurology, Medical College of Georgia, 15th Street, Augusta, GA 30912, USA A. Bronstone, Posit Science Corporation, 225 Bush Street, Seventh Floor, San Francisco, CA 94104, USA J.A. Brown, Department of Neurological Surgery, Wayne State University, 600 Northern Boulevard No. 118, Great Neck, NY 11021, USA S.B. Chapman, Center for Brain Health, The University of Texas at Dallas, 1966 Inwood Road, Dallas, TX 75235, USA A.J.-W. Chen, Division of Geriatrics, University of California at San Francisco, San Francisco, CA, USA N. Chen, Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA S.B. Chiu Wong, Center for Brain Health, The University of Texas at Dallas, 1966 Inwood Road, Dallas, TX 75235, USA L.G. Cook, Center for Brain Health, The University of Texas at Dallas, 1966 Inwood Road, Dallas, TX 75235, USA M. D’Esposito, Helen Wills Neuroscience Institute, University of California, 132 Barker Hall, Berkeley, CA 94720-1650, USA M.D. Devous Sr., Nuclear Medicine Center, The University of Texas Southwestern Medical Center, 6000 Harry Hines Boulevard, Dallas, TX 75390, USA H.R. Dinse, Department of Theoretical Biology, Experimental Biology Laboratory, Institute for Neuroinformatics, Ruhr-University Bochum, Building ND04, 44780 Bochum, Germany R. Egli, Department of Neuroscience, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA A. Frigon, CRSN Department of Physiology, Faculty of Medicine, Universite´ de Montreal, Pav. Paul-G.Desmarais room 4139, 2960 Chemin de la Tour, Montreal, QC H3 T 1J4, Canada J.F. Gamino, Center for Brain Health, The University of Texas at Dallas, 1966 Inwood Road, Dallas, TX 75235, USA S. Garbuzova-Davis, Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA ix
x
C. Gemma, Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA J. Grafman, Cognitive Neuroscience Section, National Institute of Neurological Disorders and Stroke, NIH, Building 10, Room 5C205, MSC 1440, 10 Center Drive, Bethesda, MD 20892-1440, USA R. Hartmann, Institute of Sensory Physiology and Neurophysiology, J.W. Goethe University School of Medicine, Frankfurt am Main, Germany S. Heid, Institute of Sensory Physiology and Neurophysiology, J.W. Goethe University School of Medicine, Frankfurt am Main, Germany A.E. Hillis, Department of Neurology, Johns Hopkins Hospital, Phipps 126, 600 N. Wolfe Street, Baltimore, MD 21287, USA S.H. Horng, Department of Brain and Cognitive Sciences, Picower Institute for Learning and Memory, Massachusetts Institute of Technology, 46-6237, 43 Vassar Street, Cambridge, MA 02139, USA S.G. Kernie, Department of Pediatrics and Center for Developmental Biology, The University of Texas Southwestern Medical Center, 6000 Harry Hines Boulevard, Dallas, TX 75390-9133, USA M.P. Kilgard, Neuroscience Program, School of Brain and Behavioral Sciences, GR 41, University of Texas at Dallas, 2601 North Floyd Road, Richardson, TX 75080, USA S.K. Klasko, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA R. Klinke, Institute of Sensory Physiology and Neurophysiology, J.W. Goethe University School of Medicine, Frankfurt am Main, Germany A. Kral, Institute of Neurophysiology and Pathophysiology, UKE, Martinistr. 52, D-20246 Hamburg, Germany H.C. Lawson, Department of Neurosurgery, Meyer Building 7-113, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21287-7713, USA F.A. Lenz, Department of Neurosurgery, Meyer Building 7-113, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21287-7713, USA J. Levenson, Department of Neuroscience, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA S.G. Lomber, University of Western Ontario, Centre for Brain and Mind, Robarts Research Institute, London, ON N6A 5K8, Canada B.W. Luikart, The Vollum Institute, Oregon Health and Science University, L474 3181 SW Sam Jackson Park Road, Portland, OR 97239-3098, USA H.W. Mahncke, Posit Science Corporation, 225 Bush Street, Seventh Floor, San Francisco, CA 94143, USA E.B. Marsh, Department of Neurology, Johns Hopkins Hospital, Phipps 126, 600 N. Wolfe Street, Baltimore, MD 21287, USA M.M. Merzenich, Keck Center for Integrative Neuroscience, University of California at San Francisco, Box 0472, 513 Parnassus Avenue, Room HSE-836, San Francisco, CA 94143, USA D.K. Miles, Department of Pediatrics and Center for Developmental Biology, The University of Texas Southwestern Medical Center, 6000 Harry Hines Boulevard, Dallas, TX 75390-9133, USA A.R. Møller, School of Behavioral and Brain Sciences, GR 41, University of Texas at Dallas,2601 North Floyd Road, Richardson, TX 75080, USA R. Moucha, Neuroscience Program, School of Brain and Behavioral Sciences, GR 41, University of Texas at Dallas, 2601 North Floyd Road, Richardson, TX 75080, USA L. Nyberg, Departments of Integrative Medical Biology (Physiology) and Radiation Sciences (Diagnostic Radiology), Umea˚ University, Umea˚, Sweden S. O’aHara, Department of Neurosurgery, Meyer Building 7-113, Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, MD 21287-7713, USA
xi
L.F. Parada, Center for Developmental Biology and Kent Waldrep Foundation Center for Basic Neuroscience Research on Nerve Growth and Regeneration, University of Texas Southwestern Medical Center NB-Room 5.208, 6000 Harry Hines, Dallas, TX 75390-9133, USA A. Pascual-Leone, Center for Noninvasive Brain Stimulation, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue, KS-452, Boston, MA 02215, USA J. Persson, Department of Psychology, University of Michigan, East Hall, 530 Church Street, Ann Arbor, MI 48109, USA D.D. Price, Oral and Maxillofacial Surgery, Health Science Center, University of Florida, 1600 Archer Road, Gainesville, FL 32610-0416, USA V. Raymont, Vietnam Head Injury Study, National Naval Medical Center, Bethesda, MD, USA S. Rossignol, CRSN Department of Physiology, Faculty of Medicine, Universite´ de Montreal, Pav. PaulG.-Desmarais room 4139, 2960 Chemin de la Tour, Montreal, QC H3 T 1J4, Canada T. Roth, Department of Neuroscience, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA C.D. Sanberg, Saneron CCEL Therapeutics Inc., 3802 Spectrum Boulevard, Suite 145, Tampa, FL 33612, USA P.R. Sanberg, Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA S. Saporta, Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA J.M. Schwartz, College of Medicine, University of California at Los Angeles, Neuropsychiatry Institute, 12304 Santa Monica Blvd, Suite 210, Los Angeles, CA 90025, USA S.C. Shalin, Department of Neuroscience, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA M. Sur, Picower Institute for Learning and Memory, Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, 46-6237, 43 Vassar Street, Cambridge, MA 02139, USA J.D. Sweatt, Department of Neuroscience, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA J. Tillein, Institute of Sensory Physiology and Neurophysiology, J.W. Goethe University School of Medicine, Frankfurt am Main, Germany R.-D. Treede, Institut fu¨r Physiologie und Pathophysiologie, Johannes Gutenberg-Universita¨t, Mainz, Germany G.N. Verne, Department of Gastroenterology, College of Dentistry, University of Florida, Gainesville, FL, USA A.E. Willing, Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA E.M. Woller, School of Behavioral and Brain Sciences, The University of Texas at Dallas, 2601 North Floyd Road, GR 41, Richardson, TX 75080, USA J.R. Wolpaw, Wadsworth Center, Laboratory of Nervous System Disorders, New York Department of Health, Empire State Plaza, P1 South Dock J3, Albany, NY 12237, USA S.K. Yi, School of Behavioral and Brain Sciences, The University of Texas at Dallas, 2601 North Floyd Road, GR 41, Richardson, TX 75080, USA
Foreword It has been known for a long time that the central nervous system is plastic and that the plasticity is based on the ability of synapses to alter their efficacy, and that axons and dendrites can be eliminated or that sprouting can occur. These changes can cause reprogramming and reorganization of the brain causing many different forms of changes in functions of motor and sensory systems and it can affect cognitive functions as well. Expression of neural plasticity can cause signs and symptoms of disease and it can adapt the nervous system to changing demands and compensate for lost functions. After injury, tasks can be redirected to other parts of the CNS through expression of neural plasticity. While it was earlier assumed that neural plasticity only could become expressed at young age, it is now known that plasticity of different parts of the nervous system can become expressed throughout life and under many different circumstances. The ability of the central nervous system to change its function has become clinically important such as for treatment of neural deficits from injuries to the nervous system, from aging and in connection with prostheses for the auditory system, from aging and in connection with prostheses for the auditory system. The use of deep brain stimulation to treat movement disorders, pain and tinnitus also depends on the possibility to activate neural plasticity. It has been known for a long time that training can ameliorate the symptoms and signs from brain damage such as from traumatic head injuries and stroke. It is now recognized that the benefit from such training is achieved through expression of neural plasticity. More recently it has been shown that specific training can ameliorate cognitive deficits and what was earlier regarded as normal age-related changes can also be reversed by activating plastic changes in the nervous system. The book discusses the basic properties of neural plasticity, illustrates how expression of neural plasticity can rewire the brain and how that can be used clinically to ameliorate symptoms and signs of disease. The first three chapters discuss the basis for neural plasticity and signaling in the CNS and memory. The following five chapters are devoted to neural plasticity in aging and clinical ways to reverse age-related changes using special training. The following seven chapters concern injury to the CNS and how its effect can be ameliorated through methods that activate neural plasticity. The two chapters that follow concern the spinal cord and how plastic changes can reverse the deficits from injury. The success of prostheses of hearing such as cochlear implants and auditory brainstem implants depends on neural plasticity and this is the topic of the following two chapters. The last three chapters concern the role of neural plasticity in pain and tinnitus and how activation of neural plasticity through deep brain stimulation can alleviate neuropathic pain. The book is based on presentation at two symposia organized by Dr. Sandra Chapman at the Center for Brain Health at the University of Texas at Dallas, held April 10–11, 2003 and April 7–8 2005, both with the title ‘‘Reprogramming the Human Brain: Translating Brain Plasticity Research into Clinical Practice’’. Aage R. Møller Dallas, May 2006
xiii
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 1
Visual activity and cortical rewiring: activitydependent plasticity of cortical networks Sam H. Horng and Mriganka Sur Picower Institute for Learning and Memory, Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
Abstract: The mammalian cortex is organized anatomically into discrete areas, which receive, process, and transmit neural signals along functional pathways. These pathways form a system of complex networks that wire up through development and refine their connections into adulthood. Understanding the processes of cortical-pathway formation, maintenance, and experience-dependent plasticity has been among the major goals of contemporary neurobiology. In this chapter, we will discuss an experimental model used to investigate the role of activity in the patterning of cortical networks during development. This model involves the ‘‘rewiring’’ of visual inputs into the auditory thalamus and subsequent remodeling of the auditory cortex to process visual information. We review the molecular, cellular, and physiological mechanisms of visual ‘‘rewiring’’ and activity-dependent shaping of cortical networks. Keywords: vision; auditory cortex; development; plasticity; connections; sensory processing developmental program that involves patterns of gene expression together with both spontaneous and environmentally derived patterns of neural activity (Fig. 1(A); O’Leary, 1989; Rakic, 1988; Job and Tan, 2003; Sur and Rubenstein, 2005). During embryonic development of the anterior neural tube, signaling centers induce regional and graded expression patterns of transcription factors (Figdor and Stern, 1993; Rubenstein et al., 1994, 1998; Ragsdale and Grove, 2001; Nakagawa and O’Leary, 2002; Grove and Fukuchi-Shimogori, 2003; Shimogori et al., 2004). In the early cortex, localized sources of fibroblast growth factor 8 (FGF8), sonic hedgehog (Shh), and bone morphogenetic protein 4 (Bmp4) contribute to regional gradients of Emx2, Foxg1, COUPTF1, Pax6, and other transcription factors (Chalepakis et al., 1993; Toresson et al., 2000; Bishop et al., 2000, Crossley et al., 2001; Monuki and Walsh, 2001; Muzio and Mallamaci, 2003). These patterns confer positional information leading to the formation of discrete
Introduction Two critical features of brain development include: (1) the formation of specific pathways that connect brain regions, and (2) the assembly of appropriate processing networks in each brain region. Each of these features relies on a combination of intrinsic and environmental influences. In general, pathways form early in brain development whereas processing networks are not shaped until late, with changes continuing often into adulthood. Experience-induced change, or plasticity, represents an adaptive response of networks to patterns of stimuli. In the cortex, for example, such plasticity then serves to match processing in a cortical area to its specific inputs. The specification of a cortical area, including the architecture of its neural networks, relies on a Corresponding author. Tel.: +1 (617) 253-8784; Fax: +1 (617) 253-9829; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57001-3
3
4
A
positional cues
B
targeting cues
activity dependent cues
Normal
Rewired
Auditory cortex
Visually responsive auditory cortex
Visual cortex Inferior colliculus
LGN
Visual cortex
LGN MGN
MGN
Cochlea Retina
Retina
Fig. 1. Development of brain areas and connectivity. The role of novel activity on areal specification. (A) Early signaling centers secrete diffusible molecules, or patterning cues, which dictate positional information and establish regional patterns of transcription factor expression. Targeting cues along the pathways and at the targets of outgrowing axons instruct the connectivity between areas. Finally, activity-dependent cues further refine the networks. During normal development, areal specificity is dictated by a combination of patterning, targeting, and activity-dependent cues. In rewired animals, normal targeting of retinal afferents to the thalamus is disrupted, while thalamic patterning remains intact. Furthermore, activity-dependent patterning of auditory cortex is altered despite normal cortical patterning and targeting by thalamocortical inputs. (B) The visual pathway in normal ferrets and mice consists of retinal afferents innervating the lateral geniculate nucleus (LGN) and superior colliculus (SC), with LGN projecting to the primary visual cortex (V1). The auditory pathway in normal ferrets and mice begins with inputs from the cochlear nucleus (not shown) to the inferior colliculus (IC), continuing to the medial geniculate nucleus (MGN) and the amygdala, and then on to the primary auditory cortex (A1). Ablating the IC in neonatal animals induces retinal afferents to innervate the MGN and repattern the auditory cortex to process visual information. (A) Adapted from Sur and Rubenstein (2005); (B) adapted from Sur and Leamey (2001). See Plate 1.1 in Colour Plate Section.
cortical regions. Altering the expression levels or location of certain signaling molecules, such as FGF8, or transcription factors, such as Emx2, it has been shown that the cortical map will shift in size or location in response to atypical expression of critical patterning molecules (Mallamaci et al., 2000; Fukuchi-Shimogori and Grove, 2001, 2003; Hamasaki et al., 2004). ‘‘Proto-areas’’ of the cortical map acquire structurally distinctive features and begin to send and receive connections. Molecular cues enable migrating axons to locate their
targets (Bolz et al., 2004). For thalamocortical axons, a number of sources, including subcortical regions of the basal ganglia, or prethalamus, the subplate, the cortex, as well as corticothalamic axons, have been implicated in the provision of necessary targeting cues (Garel et al., 2002; Hevner et al., 2002; Uziel et al., 2002; Garel and Rubenstein, 2004; Shimogori and Grove, 2005). Finally, activity provided by input pathways can refine and even alter connections within cortical areas (Cline, 2003; Sur and Rubenstein, 2005).
5
This chapter will focus on the potential of novel patterns of neural activity to alter the functional specificity and anatomical connectivity of cortical areas. We will review a model in which rerouting of retinal afferents to the auditory thalamus, or medial geniculate nucleus (MGN), after auditory deafferentation (Fig. 1(B)) directs visual information to the primary auditory cortex (A1), where neurons acquire novel retinotopic and feature-selective response properties (Sur and Leamey, 2001). Characterizing visually ‘‘rewired’’ A1 has allowed us to measure the extent to which cortical networks are modified to process information with novel spatial and temporal properties. While other models, such as the ability of eye-specific pathways in the primary visual cortex (V1) to change their spatial territory in response to imbalances of input, a paradigm referred to as ‘‘ocular dominance plasticity,’’ (Wiesel and Hubel, 1965; Shatz and Stryker, 1978; Antonini and Stryker, 1993; Katz and Shatz, 1996) are used to investigate the role of a relative reduction in the amount of activity, together with inter-pathway competition, in weakening and strengthening synaptic connections, the rewiring model probes the role of novel patterns of activity in shaping synaptic connections and function in existing cortical structures.
Developmental plasticity of sensory pathways: rewiring retinal inputs into the auditory thalamus During development, sensory axons target unique regions, or nuclei, of the thalamus, which in turn send projections to specific areas of the cortex. Typically, these pathways process and transmit information of primarily one sensory modality (Wallace et al., 2004), while convergence at later stages represents the bulk of multisensory processing. In the visual pathway, retinal ganglion cells project to the dorsal and ventral subdivisions (LGd; LGv) of the lateral geniculate nucleus (LGN) in the thalamus and to the superior colliculus (SC) in the brainstem (Fig. 2(A)). From the LGd, thalamocortical axons project to V1. In the auditory pathway, cochlear afferents synapse first in the inferior colliculus (IC), which sends fibers along the brachium of the IC (BIC) to the MGN,
which then projects to A1 (Fig. 2(A)). Using hamsters, Schneider discovered that retinal afferents could be induced to innervate the ventral MGN (MGv) when both the IC and SC were ablated after birth (Fig. 2(A); Schneider, 1973; Kalil and Schneider, 1975; see also Frost, 1982; Frost and Metin, 1985). IC loss deprives the MGN of auditory afferents and ipsilateral fibers of the BIC, while SC loss deprives both the MGN of contralateral fibers of the BIC and retinal axons of a normal target (Angelucci et al., 1998). ‘‘Rewiring’’ retinal afferents to MGN has subsequently been demonstrated and studied in the ferret and mouse models (Sur et al., 1988; Roe et al., 1990, 1992; Lyckman et al., 2001; Newton et al., 2004; Ellsworth et al., 2005). ‘‘Rewiring’’ alters anatomic and physiologic features of MGN to more closely resemble those of the normal LGN. In the ferret, rewired MGN neurons exhibit center-surround visual receptive fields during extracellular recording (Roe et al., 1993), as well as eye-specific segregation (Angelucci et al., 1997). Retinal axons in the MGN are topographically ordered with the central and peripheral fields located from medial to lateral and the ventral and dorsal fields represented from dorsal to ventral (Roe et al., 1991). Nonetheless, certain morphological aspects of rewired MGN are retained from normal MGN. Retinal axon terminations are elongated along the typical isofrequency axis, or lamellae, of the MGN as opposed to more focal, isotropic distributions in the LGN (Pallas et al., 1994). Furthermore, the eye-specific clusters of retinal inputs are smaller and cruder than the eye-specific layers of LGN (Angelucci et al., 1997).
Novel inputs alter functional specificity in the cortex: anatomical, physiological and behavioral consequences of rewiring Similar degrees of functional and limited structural remodeling occur in rewired A1. Like their inputs, rewired A1 cells respond to visual-field stimulation and comprise a functional retinotopic map of visual space (Roe et al., 1990). However, the thalamocortical axons transmitting this information
6
A. Thalamic Projections Normal LGN
Normal MGN
Rewired MGN D
D
A
M
M
MGd
M
MGd
A C
A1
MGv
MGv from IC
from retina
from retina
IC to MGN
retina to LGN contralateral fibers
retina to MGN
ipsilateral fibers
B. Orientation Maps Rewired A1
Normal V1
C. Horizontal Connectivity of Superficial Cortex Normal V1
Normal A1
Rewired A1
140
Fig. 2. Anatomical and physiological consequences of rewiring. (A) In normal ferrets, retinogeniculate axons terminate in eye-specific regions of the LGN (horizontal plane). Auditory afferents from the IC project to the ventral subdivision (MGv) of the MGN (coronal plane) and terminate along lamellae running parallel to the lateral-medial axis. Rewired auditory afferents innervate the MGv along adjacent, nonoverlapping eye-specific terminals within the MGv lamellae. (B) Orientation maps are observed in normal V1 and rewired A1 of ferrets using optical imaging of intrinsic signals. Hemodynamic changes in reflectance due to oxygen consumption are recorded from the cortex while gratings of different orientations are presented to the animal. A composite map of orientation preference is calculated by computing a vector average of the response signal at each pixel. Color bar: color coding representing different orientations. Scale bar: 0.5 mm.(C) Retrograde label reveals the distribution of horizontal connections in the superficial layers of the cortex in normal V1, normal A1 and rewired A1 of ferrets. The pattern of horizontal connectivity in rewired A1 more closely resembles that of normal V1 than normal A1, and potentially subserves the refinement of orientation mapping within rewired A1. Scale bars: 500 mm. (A) Adapted from Sur and Leamey (2001); (B) adapted from Sharma et al. (2000). See Plate 1.2 in Colour Plate Section.
7
retain their pattern of elongated projections along the anteroposterior axis of A1, which typically correspond to isofrequency bands (Pallas et al., 1990). In order to create a functional map of focal retinotopic representations, there must be either a refinement of the inputs from these elongated projections by an intracortical inhibitory network or a difference in drive along the projection itself (Sur et al., 1990). Thus, despite persistent structural features of A1 and thalamocortical input, functional representation is shaped by the novel patterns of activity transmitted to the cortex. Rewired A1 also acquires computational response features, such as orientation and direction selectivity, that develop in normal V1 cells (Sharma et al., 2000). In ferrets, maps of orientation selectivity can be visualized by using optical imaging of cortical hemodynamic signals (Rao et al., 1997). In both normal V1 and rewired A1, but not in normal A1, orientation domains are present and converge at pinwheel centers representing the transition point between multiple domains (Fig. 2(B); Sharma et al., 2000). In rewired A1, orientation maps are smaller and less organized, though intracortical injections of retrograde tracer reveal that inhibitory horizontal connections bridge distantly located domains of the same orientation preference, as in V1 (Fig. 2(B); Sharma et al., 2000). In normal A1, horizontal connections are limited to isofrequency domains of the tonotopic map and stretch along these bands. Furthermore, calbindin-immunoreactive GABAergic neurons of rewired A1 display markedly more elongated axonal arbors in contrast to those in normal A1 (Gao et al., 2000). Orientation and direction selectivity arise from the summed inputs of thalamocortical projections representing aligned receptive fields (Somers et al., 1995). Feed-forward inputs onto the cortical cells create levels of activity in preferential response to contrast edges of different orientations. Intracortical inhibitory inputs further refine the selectivity of these responses. Changes in the inhibitory microcircuitry of rewired A1 suggest that similar mechanisms organize the response preferences of V1 and rewired A1 cells. The rewired auditory pathway also mediates functional changes in behavior. Experiments using
ferrets with a unilaterally rewired left hemisphere demonstrate that after training to distinguish a left visual-hemifield stimulus from an auditory stimulus, the animals accurately perceive a right visualhemifield stimulus as visual even after left LGN ablation (von Melchner et al., 2000). After left LGN ablation, the ferrets exhibit diminished yet intact spatial acuity in the right hemifield. Subsequent ablation of rewired A1 abolishes the animals’ ability to distinguish a right-hemifield stimulus presented as visual. These experiments demonstrate that rewired A1 is sufficient and necessary in the absence of the ipsilateral visual pathway to detect a visual percept in trained ferrets. In mice, direct subcortical projections from the MGN to the amygdala have been implicated in mediating the rapid acquisition of a fear-conditioned response to an auditory cue (Rogan and LeDoux, 1995; Doran and LeDoux, 1999; Newton et al., 2004). Because of an indirect pathway from LGN through V1 and the perirhinal cortex to the amygdala, a fear-conditioned response to a visual cue requires many more training sessions (Heldt et al., 2000). In rewired mice, the acquisition time of a fear-conditioned response to a visual cue is accelerated and resembles that of a normal mouse in response to an auditory cue (Newton et al., 2004). In sum, a number of functional properties of the rewired auditory pathway have been characterized: (1) structural reorganization of horizontal connections within the auditory cortex, (2) functional acquisition of visual receptive-field properties in rewired MGN and A1, (3) novel organization of retinotopic and orientation selective maps in rewired A1, (4) use of the rewired auditory pathway to mediate visual percepts, and (5) use of rewired subcortical connections to the MGN to rapidly entrain fear-conditioned behaviors in response to a visual cue.
Gene expression and the molecular mechanisms of rewiring Although the response properties of the rewired auditory pathway have been characterized, the mechanisms by which retinal afferents target the MGN remain unclear. Experiments in ephrin A2/
8
A5 double knock-out mice reveal that surgically induced rewiring is enhanced in the absence of these repulsive ligands for which retinal axons bear receptors (Lyckman et al., 2001). Ipsilateral projections are increased, as they originate from the temporal retina and express the highest levels of ephrinA receptor (Ellsworth et al., 2005). Although the loss of ephrin A2 and A5 ligands is not sufficient to induce rewiring without surgery, the ability to enhance rewiring in their absence suggests that molecular cues play a role in the targeting of sensory axons to their respective thalamic compartments. We hypothesize that surgical ablation of the IC induces changes in gene expression within the thalamus that mediate the aberrant ingrowth of retinal axons to the MGN. In order to discover molecular substrates for the targeting of retinal axons to the MGN, one strategy is to use gene microarrays to analyze gene expression in normal LGN and MGN and compare patterns of gene expression to rewired MGN. Differential patterns of gene expression between regions of neonatal (P0) mouse thalamus have been identified previously (Nakagawa and O’Leary, 2001; Jones and Rubenstein, 2004). We expect that molecules in the MGN that play a role in aberrant retinal targeting will either alter their levels of expression to resemble those of normal LGN or acquire novel levels of expression. In other words, targeting cues responsible for rewiring potentially correspond to those responsible for normal retinal ingrowth in the LGN or consist of novel factors uniquely expressed in the rewired MGN. Similar experiments could be applied to compare gene expression in normal A1, normal V1, and rewired A1 in order to discover changes in gene expression with a potential role in the structural and functional repatterning of A1 in response to novel input. A number of developmental questions pertinent to the rewiring paradigm remain to be addressed. Presumably, the surgical ablation of IC (and SC) must be performed during a period in which retinal and BIC fibers target the LGN and MGN. In the mouse, both retinal and BIC fibers innervate the thalamus from E15 through birth (Tuttle et al., 1998; Gurung and Fritzsch, 2004). However, the precise developmental time window for effective
rewiring has not been well-characterized. In the hamster, evidence for exuberant projections from the retina to the MGN early in normal development suggests that the rewiring surgery might simply stabilize these projections (Frost, 1982). In contrast, in the ferret, normal fibers from the retina do not reach the MGN and thus, rewiring must induce these fibers to innervate its novel target (Sur et al., 1988). In the mouse, retinal axons have not been observed to target MGN in normal development (Ellsworth and Sur, unpublished observations).
Implications of the rewiring model for human disability and comparative neurology Processes of visual ‘‘rewiring’’ in the hamster, ferret, and mouse models could inform our understanding of functional compensation after early sensory loss in humans. Functional magnetic-resonance imaging has demonstrated that cortical areas can acquire novel processing properties in the absence of typical sensory input (Weeks et al., 2000). In congenitally blind individuals, auditory and somatosensory activation of the visual cortex can be seen (Cohen et al., 1997), while the auditory cortex may be activated by visual stimuli in congenitally deaf individuals (Bavelier and Neville, 2002). The phenomenon of ‘‘phantom limb’’ sensation suggests that somatosensory networks previously devoted to processing absent areas of the body, acquire new somatotopic receptive fields while still retaining residual percepts (Jones and Pons, 1998). Despite these examples of cross-modal plasticity in humans, the novel patterns of activation do not necessarily reflect actual changes in sensory input, in contrast to activation of existing intracortical, multisensory networks. Single-unit recording in the mouse demonstrate the presence of multimodal cells in primary sensory regions (Wallace et al., 2004). Activity in these areas can be modulated by stimuli of other modalities (Komura et al., 2005), and activation of primary sensory areas in the cortex can be elicited by stimuli of other modalities in humans (Schroeder and Foxe, 2005). Nonetheless, understanding the potential of primary sensory cortical areas for processing novel
9
stimuli could lead to important clinical strategies of optimizing sensory compensation. Finally, rewiring of the auditory pathway may contribute to understanding general principles of brain evolution. Two strains of congenitally deaf mice (NKCC1, PMCA KO) exhibit visual rewiring in the absence of surgical IC ablation (Hunt et al., 2005). Furthermore, animal models of cross-modal plasticity are not limited to visual rewiring to the auditory cortex. Mice with congenital retinal defects, such as the ZRDCT/An mutant, exhibit auditory innervation of the LGN (Piche et al., 2004). Other strains, such as the oj/oj mutant, and mice enucleated at birth acquire somatosensory innervation of the LGN (Asanuma and Stanfield, 1990). Surgical ablation of IC and SC can also lead to retinal ingrowth to the ventrobasal nucleus (VB), or somatosensory thalamus (Frost and Metin, 1985). The general principle of sensory inputs compensating the target of a deprived modality suggests that divergence in the structure and function of brain networks could have developed after genetic mutations altering the strength or wholesale presence of certain sensory inputs. It has been hypothesized that the blind mole rat acquired expanded areas of the cortex devoted to somatosensory processing and audition after the loss of vision (Heil et al., 1991; Bronchti et al., 1991, 2002; Hunt et al., 2005). Thus, developmental plasticity in the targeting of sensory afferents to the thalamus may represent a general mechanism of evolving novel functional specificities for brain areas and networks.
Conclusions ‘‘Rewiring’’ retinal afferents to the auditory pathway has enabled us to investigate the role of novel patterns of activity in shaping the response properties of cortical and thalamic areas. Areas within the cortex and thalamus appear to have a structural blueprint determined by gene-expression programs of the developing brain. This framework can nonetheless be instructed to process novel information and reorganize its microcircuitry to some degree in order to mediate functional behaviors appropriate to the stimulus. Mechanisms
of rewiring are now being investigated to better understand the extent to which patterning, targeting, and activity-dependent programs of development can be dissociated and altered in order for a cortical area to acquire new functional response properties appropriate to novel stimuli. References Angelucci, A., Clasca, F., Bricolo, E., Cramer, K.S. and Sur, M. (1997) Experimentally induced retinal projections to the ferret auditory thalamus: development of clustered eye-specific patterns in a novel target. J. Neurosci., 17: 2040–2055. Angelucci, A., Clasca, F. and Sur, M. (1998) Brainstem inputs to the ferret medial geniculate nucleus and the effect of early deafferentation on novel retinal projections to the auditory thalamus. J. Comp. Neurol., 400: 417–439. Antonini, A. and Stryker, M.P. (1993) Rapid remodeling of axonal arbors in the visual cortex. Science, 260: 1819–1821. Asanuma, C. and Stanfield, B.B. (1990) Induction of somatic sensory inputs to the lateral geniculate nucleus in congenitally blind mice and in phenotypically normal mice. Neuroscience, 39: 533–545. Bavelier, D. and Neville, H.J. (2002) Cross-modal plasticity: where and how? Nat. Rev. Neurosci., 3: 443–452. Bishop, K.M., Goudreau, G. and O’Leary, D.D. (2000) Regulation of area identity in the mammalian neocortex by Emx2 and Pax6. Science, 288: 344–349. Bronchti, G., Rado, R., Terkel, J. and Wollberg, Z. (1991) Retinal projections in the blind mole rat: a WGA-HRP tracing study of a natural degeneration. Brain Res. Dev. Brain Res., 58: 159–170. Bronchti, G., Heil, P., Sadka, R., Hess, A., Scheich, H. and Wollberg, Z. (2002) Auditory activation of ‘‘visual’’ cortical areas in the blind mole rat (Spalaxehrenbergi). Eur. J. Neurosci., 16: 311–329. Bolz, J., Uziel, D., Muhlfriedel, S., Gullmar, A., Peuckert, C., Zarbalis, K., Wurst, W., Torii, M. and Levitt, P. (2004) Multiple roles of ephrins during the formation of thalamocortical projections: maps and more. J. Neurobiol., 59: 82–94. Chalepakis, G., Stoykova, A., Wijnholds, J., Tremblay, P. and Gruss, P. (1993) Pax: gene regulators in the developing nervous system. J. Neurobiol., 24: 1367–1384. Cline, H. (2003) Sperry and Hebb: oil and vinegar? Trends Neurosci., 26: 655–661. Cohen, L.G., Celnik, P., Pascual-Leone, A., Corwell, B., Falz, L., Dambrosia, J., Honda, M., Sadato, N., Gerloff, C., Catala, M.D. and Hallett, M. (1997) Functional relevance of cross-modal plasticity in blind humans. Nature, 389: 180–183. Crossley, P.H., Martinez, S., Ohkubo, Y. and Rubenstein, J.L. (2001) Coordinate expression of Fgf8, Otx2, Bmp4, and Shh in the rostral prosencephalon during development of the telencephalic and optic vesicles. Neuroscience, 108: 183–206.
10 Doran, N.N. and LeDoux, J.E. (1999) Organization of projections to the lateral amygdale from auditory and visual areas of the thalamus in the rat. J. Comp. Neurol., 430: 235–249. Ellsworth, C.A., Lyckman, A.W., Feldheim, D.A., Flanagan, J.G. and Sur, M. (2005) Ephrin-A2 and -A5 influence patterning of normal and novel retinal projections to the thalamus: conserved mapping mechanisms in visual and auditory thalamic targets. J. Comp. Neurol., 488: 140–151. Figdor, M.C. and Stern, C.D. (1993) Segmental organization of embryonic diencephalon. Nature, 363: 630–634. Frost, D.O. (1982) Anomalous visual connections to somatosensory and auditory systems following brain lesions in early life. Brain Res., 255: 627–635. Frost, D.O. and Metin, C. (1985) Induction of functional retinal projections to the somatosensory system. Nature, 317: 162–164. Fukuchi-Shimogori, T. and Grove, E.A. (2003) Emx2 patterns the neocortex by regulating FGF positional signaling. Nat. Neurosci., 6: 823–825. Fukuchi-Shimogori, T. and Grove, E.A. (2001) Neocortex patterning by the secreted signaling molecule FGF8. Science, 294: 1071–1074. Gao, W.J., Wormington, A.B., Newman, D.E. and Pallas, S.L. (2000) Development of inhibitory circuitry in visual and auditory cortex of postnatal ferrets: immunocytochemical localization of calbindin- and parvalbumin-containing neurons. J. Comp. Neurol., 422: 140–157. Garel, S., Yun, K., Grosschedl, R. and Rubenstein, J.L. (2002) The early topography of thalamocortical projections is shifted in Ebf1 and Dlx1/2 mutant mice. Development, 129: 5621–5634. Garel, S. and Rubenstein, J.L. (2004) Intermediate targets in formation of topographic projections: inputs from the thalamocortical system. Trends Neurosci., 27: 533–539. Grove, E.A. and Fukuchi-Shimogori, T. (2003) Generating the cerebral cortical area map. Annu. Rev. Neurosci., 26: 355–380. Gurung, B. and Fritzsch, B. (2004) Time course of embryonic midbrain and thalamic auditory connection development in mice as revealed by carbocyanine dye tracing. J. Comp. Neurol., 479: 309–327. Hamasaki, T., Leingartner, A., Ringstedt, T. and O’Leary, D.D. (2004) EMX2 regulates sizes and positioning of the primary sensory and motor areas in neocortex by direct specification of cortical progenitors. Neuron, 43: 359–372. Heil, P., Bronchti, G., Wollberg, Z. and Scheich, H. (1991) Invasion of visual cortex by the auditory system in the naturally blind mole rat. Neuroreport, 2: 735–738. Heldt, S., Sudin, V., Willott, J.F. and Falls, W.A. (2000) Posttraining lesions of the amygdale interfere with fear-potentiated startle to both visual and auditory conditioned stimuli in C57BL/6J mice. Behav. Neurosci., 114: 749–759. Hevner, R.F., Miyashita-Lin, E. and Rubenstein, J.L. (2002) Cortical and thalamic axon pathfinding defects in Tbr1, Gbx2, and Pax6 mutant mice: evidence that cortical and thalamic axons interact and guide each other. J. Comp. Neurol., 447: 8–17.
Hunt, D.L., King, B., Kahn, D.M., Yamoah, E.N., Shull, G.E. and Krubitzer, L. (2005) Aberrant retinal projections in congenitally deaf mice: how are phenotypic characteristics specified in development and evolution? Anat. Rec. A Discov. Mol. Cell Evol. Biol., 287: 1051–1066. Job, C. and Tan, S. (2003) Constructing the mammalian neocortex: the role of intrinsic factors. Dev. Biol., 257: 221–232. Jones, E.G. and Pons, T.P. (1998) Thalamic and brainstem contributions to large-scale plasticity of primate somatosensory cortex. Science, 282: 1121–1125. Jones, E.G. and Rubenstein, J.L.R. (2004) Expression of regulatory genes during differentiation of thalamic nuclei in mouse and monkey. J. Comp. Neurol., 47: 55–80. Kalil, R.E. and Schneider, G.E. (1975) Abnormal synaptic connections of the optic tract in the thalamus after midbrain lesions in newborn hamsters. Brain Res., 100: 690–698. Katz, L.C. and Shatz, C.J. (1996) Synaptic activity and the construction of cortical circuits. Science, 274: 1133–1138. Komura, Y., Tamura, R., Uwano, T., Nishijo, H. and Ono, T. (2005) Auditory thalamus integrates visual inputs into behavioral gains. Nat. Neurosci., 8: 1203–1209. Lyckman, A.W., Jhaveri, S., Feldheim, D.A., Vanderhaeghen, P., Flanagan, J.G. and Sur, M. (2001) Enhanced plasticity of retinothalamic projections in an ephrin-A2/A5 double mutant. J. Neurosci., 21: 7684–7690. Mallamaci, A., Muzio, L., Chan, C.H., Parnavelas, J. and Boncinelli, E. (2000) Area identity shifts in the early cerebral cortex of Emx2 / mutant mice. Nat. Neurosci., 3: 679–686. Monuki, E.S. and Walsh, C.A. (2001) Mechanisms of cerebral cortical patterning in mice and humans. Nat. Neurosci., 4(Suppl): 1199–1206. Muzio, L. and Mallamaci, A. (2003) Emx1, emx2 and pax6 in specification, regionalization and arealization of the cerebral cortex. Cereb. Cortex, 13: 641–647. Nakagawa, Y. and O’Leary, D.D. (2001) Combinatorial expression patterns of LIM-homeodomain and other regulatory genes parcellate developing thalamus. J. Neurosci., 21: 2711–2725. Nakagawa, Y. and O’Leary, D.D. (2002) Patterning centers, regulatory genes and extrinsic mechanisms controlling arealization of the neocortex. Curr. Opin. Neurobiol., 12: 14–25. Newton, J.R., Ellsworth, C., Miyakawa, T., Tonegawa, S. and Sur, M. (2004) Acceleration of visually cued conditioned fear through the auditory pathway. Nat. Neurosci., 7: 968–973. O’Leary, D.D. (1989) Do cortical areas emerge from a protocortex? Trends Neurosci., 12: 400–406. Pallas, S.L., Roe, A.W. and Sur, M. (1990) Visual projections induced into the auditory pathway of ferrets. I. Novel inputs to primary auditory cortex (AI) from the LP/pulvinar complex and the topography of the MGB-AI projection. J. Comp. Neurol., 298: 50–68. Pallas, S.L., Hahm, J. and Sur, M. (1994) Morphology of retinal axons induced to arborize in a novel target, the medial geniculate nucleus. I. Comparison with arbors in normal targets. J. Comp. Neurol., 349: 343–362. Piche, M., Robert, S., Miceli, D. and Bronchti, G. (2004) Environmental enrichment enhances auditory takeover of the
11 occipital cortex in anophthalmic mice. Eur. J. Neurosci., 20: 3463–3472. Ragsdale, C.W. and Grove, E.A. (2001) Patterning the mammalian cerebral cortex. Curr Opin Neurobiol., 11: 50–58. Rakic, P. (1988) Specification of cerebral cortical areas. Science, 242: 170–176. Rao, S.C., Toth, L.J. and Sur, M. (1997) Optically imaged maps of orientation preference in primary visual cortex of cats and ferrets. J. Comp. Neurol., 387: 358–370. Roe, A.W., Pallas, S.L., Hahm, J.O. and Sur, M. (1990) A map of visual space induced in primary auditory cortex. Science, 250: 818–820. Roe, A.W., Hahm, J.O. and Sur, M. (1991) Experimentally induced establishment of visual topography in auditory thalamus. Soc. Neurosci. Abs., 17: 898. Roe, A.W., Pallas, S.L., Kwon, Y.H. and Sur, M. (1992) Visual projections routed to the auditory pathway in ferrets: receptive fields of visual neurons in primary auditory cortex. J. Neurosci., 12: 3651–3664. Roe, A.W., Garraghty, P.E., Esguerra, M. and Sur, M. (1993) Experimentally induced visual projections to the auditory thalamus in ferrets: evidence for a W cell pathway. J. Comp. Neurol., 334: 263–280. Rogan, M.T. and LeDoux, J.E. (1995) LTP is accompanied by commensurate enhancement of auditor-evoked responses in a fear conditioning circuit. Neuron, 15: 127–136. Rubenstein, J.L., Martinez, S., Shimamura, K. and Puelles, L. (1994) The embryonic vertebrate forebrain: the prosomeric model. Science, 266: 578–580. Rubenstein, J.L., Shimamura, K., Martinez, S. and Puelles, L. (1998) Regionalization of the prosencephalic neural plate. Annu. Rev. Neurosci., 21: 445–477. Schneider, G.E. (1973) Early lesions of superior colliculus: factors affecting the formation of abnormal retinal projections. Brain Behav. Evol., 8: 73–109. Schroeder, C.E. and Foxe, J. (2005) Multisensory contributions to low-level, ‘unisensory’ processing. Curr. Opin. Neurobiol., 15: 454–458. Sharma, J., Agelucci, A. and Sur, M. (2000) Induction of visual orientation modules in auditory cortex. Nature, 404: 841–847. Shatz, C.J. and Stryker, M.P. (1978) Ocular dominance in layer IV of the cat’s visual cortex and the effects of monocular deprivation. J. Physiol., 281: 267–283.
Shimogori, T., Banuchi, V., Ng, H.Y., Strauss, J.B. and Grove, E.A. (2004) Embryonic signaling centers expressing BMP, WNT and FGF proteins interact to pattern the cerebral cortex. Development, 131: 5639–5647. Shimogori, T. and Grove, E.A. (2005) Fibroblast growth factor 8 regulates neocortical guidance of area-specific thalamic innervation. J. Neurosci., 25: 6550–6560. Somers, D.C., Nelson, S.B. and Sur, M. (1995) An emergent model of orientation selectivity in cat visual cortical simple cells. J. Neurosci., 15: 5448–5465. Sur, M., Garraghty, P.E. and Roe, A.W. (1988) Experimentally induced visual projections into auditory thalamus and cortex. Science, 242: 1437–1441. Sur, M., Pallas, S.L. and Roe, A.W. (1990) Cross-modal plasticity in cortical development: differentiation and specification of sensory neocortex. Trends Neurosci., 13: 227–233. Sur, M. and Leamey, C. (2001) Development and plasticity of cortical areas and networks. Nat. Rev. Neurosci., 2: 251–262. Sur, M. and Rubenstein, J.L.R. (2005) Patterning and plasticity of the cerebral cortex. Science, 310: 805–810. Toresson, H., Potter, S.S. and Campbell, K. (2000) Genetic control of dorsal–ventral identity in the telencephalon: opposing roles for Pax6 and Gsh2. Development., 127: 4361–4371. Tuttle, R., Braisted, J.E., Richards, L.J. and O’Leary, D.D. (1998) Retinal axon guidance by region-specific cues in diencephalon. Development, 125: 791–801. Uziel, D., Muhlfriedel, S., Zarbalis, K., Wurst, W., Levitt, P. and Bolz, J. (2002) Miswiring of limbic thalamocortical projections in the absence of ephrin-A5. J. Neurosci., 22: 9352–9357. Von Melchner, L., Pallas, S.L. and Sur, M. (2000) Visual behaviour mediated by retinal projections directed to the auditory pathway. Nature., 404: 871–876. Wallace, M.T., Ramachandran, R. and Stein, B.E. (2004) A revised view of sensory cortical parcellation. Proc. Natl. Acad. Sci. USA, 101: 2167–2172. Weeks, R., Horwitz, B., Aziz-Sultan, A., Tian, B., Wessinger, C.M., Cohen, L.G., Hallett, M. and Rauschecker, J.P. (2000) A positron emission tomographic study of auditory localization in the congenitally blind. J. Neurosci., 20: 2664–2672. Wiesel, T.N. and Hubel, D.H. (1965) Comparison of the effects of unilateral and bilateral eye closure on cortical unit responses in kittens. J. Neurophysiol., 28: 1029–1040.
A
positional cues
B
targeting cues
activity dependent cues
Normal
Rewired
Auditory cortex
Visually responsive auditory cortex
Visual cortex Inferior colliculus
LGN
Visual cortex
LGN MGN
MGN
Cochlea Retina
Retina
Plate 1.1. Development of brain areas and connectivity. The role of novel activity on areal specification. (A) Early signaling centers secrete diffusible molecules, or patterning cues, which dictate positional information and establish regional patterns of transcription factor expression. Targeting cues along the pathways and at the targets of outgrowing axons instruct the connectivity between areas. Finally, activity-dependent cues further refine the networks. During normal development, areal specificity is dictated by a combination of patterning, targeting, and activity-dependent cues. In rewired animals, normal targeting of retinal afferents to the thalamus is disrupted, while thalamic patterning remains intact. Furthermore, activity-dependent patterning of auditory cortex is altered despite normal cortical patterning and targeting by thalamocortical inputs. (B) The visual pathway in normal ferrets and mice consists of retinal afferents innervating the lateral geniculate nucleus (LGN) and superior colliculus (SC), with LGN projecting to the primary visual cortex (V1). The auditory pathway in normal ferrets and mice begins with inputs from the cochlear nucleus (not shown) to the inferior colliculus (IC), continuing to the medial geniculate nucleus (MGN) and the amygdala, and then on to the primary auditory cortex (A1). Ablating the IC in neonatal animals induces retinal afferents to innervate the MGN and repattern the auditory cortex to process visual information. (A) Adapted from Sur and Rubenstein (2005); (B) adapted from Sur and Leamey (2001).
A. Thalamic Projections Normal LGN
Normal MGN
Rewired MGN D
D
A
M
M
MGd
M
MGd
A C
A1
MGv
MGv from IC
from retina
from retina
IC to MGN
retina to LGN contralateral fibers
retina to MGN
ipsilateral fibers
B. Orientation Maps Rewired A1
Normal V1
C. Horizontal Connectivity of Superficial Cortex Normal V1
Normal A1
Rewired A1
140
Plate 1.2. Anatomical and physiological consequences of rewiring. (A) In normal ferrets, retinogeniculate axons terminate in eyespecific regions of the LGN (horizontal plane). Auditory afferents from the IC project to the ventral subdivision (MGv) of the MGN (coronal plane) and terminate along lamellae running parallel to the lateral-medial axis. Rewired auditory afferents innervate the MGv along adjacent, nonoverlapping eye-specific terminals within the MGv lamellae. (B) Orientation maps are observed in normal V1 and rewired A1 of ferrets using optical imaging of intrinsic signals. Hemodynamic changes in reflectance due to oxygen consumption are recorded from the cortex while gratings of different orientations are presented to the animal. A composite map of orientation preference is calculated by computing a vector average of the response signal at each pixel. Color bar: color coding representing different orientations. Scale bar: 0.5 mm.(C) Retrograde label reveals the distribution of horizontal connections in the superficial layers of the cortex in normal V1, normal A1 and rewired A1 of ferrets. The pattern of horizontal connectivity in rewired A1 more closely resembles that of normal V1 than normal A1, and potentially subserves the refinement of orientation mapping within rewired A1. Scale bars: 500 mm. (A) Adapted from Sur and Leamey (2001); (B) adapted from Sharma et al. (2000).
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 2
Receptor tyrosine kinase B-mediated excitatory synaptogenesis Bryan W. Luikart and Luis F. Parada Center for Developmental Biology & Kent Waldrep Foundation Center for Basic Neuroscience Research on Nerve Growth and Regeneration, University of Texas Southwestern Medical Center, Dallas, TX 75390-9133, USA
Abstract: The receptor tyrosine kinase B, TrkB, is the high-affinity receptor for brain-derived neurotrophic factor (BDNF). Much evidence supports a role for TrkB signaling in excitatory synapse formation. There have been a number of recent advances in understanding the cell biology of TrkB-mediated excitatory synaptogenesis. The predominant mechanism by which TrkB supports excitatory synaptogenesis appears to be due to cell-autonomous signaling in both pre- and postsynaptic cells. This signaling appears to contribute to the growth and stabilization processes necessary for the net formation of synapses during development. Further, the molecular mechanisms by which TrkB contributes to these growth and stabilization processes are beginning to be elucidated. Keywords: neurotrophins; TrkB; synaptogenesis; BDNF; conditional knockouts we discuss how brain-derived neurotrophic factor (BDNF) contributes to both growth and stabilizing processes through the receptor tyrosine kinase B (TrkB). Further, we discuss evidence that the effects of TrkB on synapse formation are cell autonomous in both pre- and postsynaptic cells.
Introduction Synapse formation is a fundamentally dynamic process. This is due to the necessity of balancing two apposing forces. For synaptogenesis to occur, growth processes must allow for interneuronal contact. These growth processes, however, must be tempered by the adhesive stabilization of appropriate contacts. Both presynaptic and postsynaptic neurons participate in the growth and stabilization processes that underlie synapse formation. For example, live imaging studies have revealed that growth-cone and filopodial dynamics of axons and dendrites contribute to synapse formation (Niell and Smith, 2004). Furthermore, adhesion molecules, necessary for the stabilization of synaptogenic contacts, are both homophilically and heterophilically expressed in pre- and postsynaptic compartments (Washbourne et al., 2004). Here,
History When TrkB was first cloned, it was noted that it was highly expressed in embryos as well as adults (Klein et al., 1989), and that its expression pattern was more pronounced in central nervous system (CNS) tissues than was that of its previously cloned relative, TrkA (Klein et al., 1990; MartinZanca et al., 1990). Subsequently, it was found that the preferred ligand for TrkB is BDNF and, to a lesser extent, neurotrophin-3 (NT-3) (Soppet et al., 1991). At this time, the growth- and survival-promoting properties of BDNF for peripheral nervous system neurons had been established
Corresponding author. Tel.: +1 (503) 494–5436; Fax: +1 (503) 494–1249; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57002-5
15
16
(Lindsay et al., 1985) and the trophic functions of BDNF in the CNS were being discovered (Hyman et al., 1991). As the roles for BDNF and TrkB in the differentiation, survival, and growth of neurons were being elucidated, another important discovery was made — BDNF and TrkB mRNA levels are increased by pathological levels of neuronal activity (Gall and Isackson, 1989; Zafra et al., 1990; Ernfors et al., 1991). Further, it appeared that BDNF transcription was regulated by levels of stimulation thought to be important for nonpathological activity-dependent plasticity (Castren et al., 1992; Patterson et al., 1992). Those labs interested in plasticity made the pioneering discovery that BDNF’s activation of TrkB could potentiate synaptic transmission (Lohof et al., 1993; Kang and Schuman, 1995). It was also discovered that TrkB activation is necessary for the late phase of long-term potentiation (Korte et al., 1995, 1996; Levine et al., 1995; Kang and Schuman, 1996; Patterson et al., 1996). Later, it became apparent that BDNF and TrkB are important for the activity-dependent synaptic remodeling of the early postnatal visual cortex (Cabelli et al., 1996, 1997). Thus, it is clear that TrkB activation is important for plasticity in both developing and mature synapses. The relative contribution of synaptogenic versus synaptic strengthening processes in developing versus adult systems is unknown.
TrkB mediates synaptogenesis BDNF was shown to be sufficient for the de novo formation of both excitatory and inhibitory synapses through application to E18.5 dissociated hippocampal cultures (Vicario-Abejon et al., 1998). Further, the necessity of TrkB for hippocampal excitatory synapse formation in vivo was demonstrated by decreased numbers of morphological synapses from electron micrographs of knockout animals (Martinez et al., 1998). These data remained subject to the caveat that the effects might not be direct given that TrkB mutant mice had complex defects and generally poor health. Later, it was shown that the effect of TrkB loss on synapse formation was direct — when healthy conditional knockout animals were shown to have
decreased numbers of excitatory and inhibitory synapses (Rico et al., 2002; Luikart et al., 2005). Along with decreased synapse number at the ultrastructural level, knockout animals have decreased expression of the presynaptic proteins synaptobrevin, synaptotagmin, synaptophysin, syntaxin-1, and SNAP-25 (Martinez et al., 1998; Pozzo-Miller et al., 1999; Carmona et al., 2003; Luikart et al., 2005). Decreases in postsynaptic receptor components of both excitatory and inhibitory synapses have also been reported (Carmona et al., 2003; Jourdi et al., 2003; Luikart et al., 2005). While the ultrastructural, immunohistochemical, and biochemical results supporting a role for TrkB in synapse formation have been relatively consistent, reported studies on the role for TrkB in dendritic spine formation and maintenance have been confusing. Live imaging experiments in slices from P25-28 ferrets show no change in spine dynamics in response to BDNF overexpressing neurons (Horch and Katz, 2002) but do show destabilization and decreased density in response to exogenous application (Horch et al., 1999). Application of BDNF to 2- to 3-week-old dissociated cultures from E1920 rat embryos has been shown to decrease spine density as well (Murphy et al., 1998). Conversely, in co-cultures of Purkinje cells and granule cells from P0 mouse pups, long-term application of BDNF elicits an increase in Purkinje spine density (Shimada et al., 1998). Further, long-term application of BDNF in hippocampal slices from P7 rats has also been found to increase spine density (Tyler and Pozzo-Miller, 2001; Alonso et al., 2004). Consistent with the latter results, conditional knockout of TrkB results in a decrease in spine density in vivo (Luikart et al., 2005). Several factors may explain the inconsistent results on spine density. It is clear that focal versus global application of BDNF results in spatially restricted growth (Horch et al., 1999; Horch and Katz, 2002). This indicates that less physiological methods and levels of BDNF application may elicit exaggerated responses. This has been further evidenced by experiments examining the effects of different doses of BDNF. Low doses of BDNF (5 ng/mL) result in increased spine densities, whereas higher doses (25 ng/mL) result in increased
17
filopodial density and dendritic growth as well as spine density (Ji et al., 2005). In the presence of cAMP, the effects at low BDNF doses are more similar to those at high doses due to cAMP-controlled phosphorylation of TrkB (Ji et al., 2005). Differing studies may also examine acute versus chronic or developmental requirements for TrkB/ BDNF, each of which may have impact on the readout. The low-affinity neurotrophin receptor p75 also mediates physiological responses to BDNF. Recently it has been demonstrated that p75 activation acts to decrease spine formation and dendritic growth (Zagrebelsky et al., 2005). Thus, the relative signaling through TrkB or p75 would affect the results of BDNF application. Studies using the knockout mice for TrkB and p75 appear to support the notion that TrkB activation leads to increased spine density, whereas p75 activation apposes that increase (Luikart et al., 2005; Zagrebelsky et al., 2005).
Cell-autonomous function of TrkB in excitatory synaptogenesis Although it is clear that TrkB is important for both excitatory and inhibitory synaptogenesis, the mechanisms by which this occurs are still unclear. It has been suggested that TrkB may indirectly regulate synapse formation by globally altering activity; however, several results support the conclusion that TrkB can signal in a cell-autonomous fashion to support excitatory synapse formation. The in vivo deletion of TrkB in only presynaptic cells of the Schaffer collateral synapse results in a specific reduction in the number of presynaptic varicosities as assessed by neuronal tracing, immunohistochemistry, and electron microscopy. When TrkB is knocked out in both the pre- and postsynaptic cells of the Schaffer collateral synapse, this reduction extends to postsynaptic densities. In vitro, the postsynaptic deletion of TrkB yields a cell with relatively normal presynaptic input, as assessed by synaptophysin staining, but a deficiency in the ability to generate postsynaptic structures apposed to that input (Luikart et al., 2005).
TrkB-mediated excitatory synaptogenesis — growth and guidance TrkB activation increases axonal and dendritic growth and thus may promote synaptogenesis by directing synaptogenic contacts. A number of experiments have demonstrated that application or overexpression of BDNF is sufficient to alter axonal arborization. Using the Xenopus retinaltectal system in vivo, imaging experiments demonstrated that BDNF application elicits increased axonal arborization in the tectum (CohenCory, 1999; Lom and Cohen-Cory, 1999). By expressing green fluorescent protein (GFP)-tagged synaptobrevin, it was found that the BDNF-induced axonal elaboration is accompanied by increased synapse formation and that this synapse formation may precede the effects on axonal branching (Alsina et al., 2001). It appears that the effects of BDNF on axons may be not only to elicit growth and branching but also to act as chemo-attractive agents. Dissociated Xenopus spinal neurons appear to grow toward BDNF in the presence of cAMP signaling (Song et al., 1997). Imaging of sensory neuron axons of mice in vivo also demonstrated that implanted BDNF-coated beads are both trophic and chemo-attractive (Tucker et al., 2001). It was also demonstrated that function-blocking antibodies could reduce growth in this system. Such trophic actions have been demonstrated in dendrites as well. In the ferret visual cortex, the application of BDNF was found to elicit increased dendritic growth of layer IV neurons (McAllister et al., 1995). Live imaging of dendritic responses to single neurons overexpressing BDNF in slices revealed very localized effects on dendritic branching (Horch and Katz, 2002). Thus, the trophic actions of TrkB have been demonstrated and spatial restriction of these actions proposed. In vivo, it is likely that the range of BDNF activity is spatially restricted. The expression of TrkB and BDNF is localized to both axonal and dendritic compartments and to synaptic structures within those compartments (Connor et al., 1997; Fawcett et al., 1997; Yan et al., 1997). Time-lapse imaging of BDNF tagged with GFP demonstrated activity-dependent release of BDNF from both
18
axons and dendrites (Haubensak et al., 1998; Hartmann et al., 2001; Kohara et al., 2001). Further, BDNF is a highly basic protein (pI ~ 11.8) and thus has limited diffusion characteristics. In vivo, one may predict that the effects of TrkB activation on dendritic and axonal growth may be spatially limited, leading to very local synaptogenic growth of axons and dendrites at the level of filopodia, varicosities, and spines. Dendritic filopodia and growth cones appear to be important for synapse formation in vitro as well as in vivo (Dailey and Smith, 1996; Ziv and Smith, 1996; Fiala et al., 1998; Niell et al., 2004). To a lesser degree, axonal filopodia also contribute to synapse formation (Jontes et al., 2000). It has been suggested that dendritic filopodia may be precursors of dendritic spines. However during development, a large number of excitatory synapses appear on the shafts of cells with highly dynamic filopodia. Thus, filopodia may participate in the generation of shaft synapses that subsequently develop into spine synapses. BDNF has been shown to regulate filopodial length and motility in a variety of systems. BDNF- and nerve growth factor (NGF)-coated beads were reported to elicit filopodial sprouting of dorsal root ganglion (DRG) axons via a TrkA-dependent mechanism, regardless of p75 activation (Gallo and Letourneau, 1998). However, BDNF appears to regulate axonal filopodial length via a p75-dependent mechanism (Gehler et al., 2004a, b). Thus, BDNF may influence filopodia differently depending on the receptor complement of the axon (Gallo and Letourneau, 2004). Exogenously applied BDNF to dissociated hippocampal cultures increased filopodial density contingent upon zipcode-binding protein-1 (ZBP1) expression (Eom et al., 2003). There has also been evidence that the truncated TrkB.T1 receptor may enhance filopodial motility, whereas p75 and the kinase-bearing TrkB appose this effect (Hartmann et al., 2004). Increased filopodial motility, however, was correlated to the cAMP-induced phosphorylation of TrkB (Ji et al., 2005). These results underscore the ability of neurotrophins to influence filopodia as well as the need for more intense investigation into the molecular mechanisms of this influence.
TrkB-mediated excitatory synaptogenesis — stabilization of contacts While a large body of evidence indicates TrkB may modulate growth and guidance, there is also mounting evidence that it may contribute to the stabilization of those contacts as well. The overexpression of the truncated TrkB.T1 receptor results in the disassembly of acetylcholine receptor clusters at neuromuscular junctions (Gonzalez et al., 1999). In 8–10-day-old hippocampal dissociated cultures, BDNF scavenging appeared to decrease the NMDA receptor clustering of immature synapses (Elmariah et al., 2004). In a similar culture paradigm, overexpression of TrkB.T1 at 10 days in vitro (DIV) led to fewer synapses (Klau et al., 2001). In Xenopus tadpoles, in vivo live-imaging experiments demonstrated that BDNFneutralizing antibodies result in the dismantling of recently formed synaptic clusters. It is important to note that in all of these systems the disruption of BDNF/TrkB signaling is taking place in developmental systems. In adult mice, it does not appear that mature excitatory synapses found on dendritic spines maintain their neurotrophin requirement for stabilization (Luikart et al., 2005). Thus, TrkB signaling appears important for the stabilization of those more labile synapses of developing systems. Over time, mature spine synapses recruit a number of scaffolding proteins likely to enhance stability and obviate the requirement for TrkB signaling.
Model of TrkB-dependent excitatory synaptic morphogenesis The above evidence corroborates the model of neurotrophins as synaptic morphogens (Poo, 2001). The increasing evidence that dendritic filopodia are prominent mediators of synaptogenic contacts (Dailey and Smith, 1996; Ziv and Smith, 1996; Fiala et al., 1998; Niell et al., 2004) and that BDNF influences dendritic filopodial motility (Eom et al., 2003; Ji et al., 2005) warrants their inclusion in this model. BDNF is released from active presynaptic terminals (Fig. 1(a)). Persistent release from these presynaptic terminals allows for the growth and
19
attraction of filopodia from depolarized neurons. When TrkB is deleted, both pre- and postsynaptically in conditional knockout animals, filopodial recruitment does not occur, resulting in a net decrease in synapse number (Luikart et al., 2005). In normal animals, this filopodialvaricosity interaction may stabilize, allowing the formation of two postsynaptic densities onto one varicosity (Fig. 1(b)). Interestingly, synaptogenic situations often result in the increase of this type of multiplesynapse bouton (Woolley et al., 1996; Fiala et al., 1998; Geinisman et al., 2001), suggesting that it may be common for new synapses to form by dendritic recruitment of existing axonal varicosities. Repeated activation of this new synapse results in the upregulation of the expression and release of BDNF from the postsynaptic dendritic structure (Fig. 1(b)). This retrogradely released BDNF, in turn, contributes to localized presynaptic growth processes resulting in the formation of an additional varicosity, thus maintaining the majority of synapses in the single synaptic-bouton configuration (Fig. 1(c)). When TrkB is conditionally deleted in the presynaptic cell, these local axonal growth processes (in Fig. 1(c)) do not occur, resulting in a decrease in the density of axonal varicosities and an increase in the number of multiple-synapse boutons (as in Fig. 1(b)) (Luikart et al., 2005). Over time, the persistence of the growth-promoting signals of BDNF at a limited number of spatially discrete sites can lead to larger modifications of dendritic and axonal arbors. The persistent modifications may be similar to those produced by pharmacological application, or overexpression, of BDNF. Fig. 1. Model of BDNF/TrkB function in activity-dependent refinement of synapse number. Initially, BDNF is released at high levels from active presynaptic boutons (red terminal in (a)). The presynaptically released BDNF then signals to TrkB receptors expressed on nearby filopodia (green). The TrkB activation allows for the additional growth and attraction of filopodia to the presynaptic terminal (b). The postsynaptic activation of TrkB as well as activity allows for the up-regulation and secretion of BDNF from the postsynaptic structures back onto the active presynaptic terminal (b). The TrkB activation in the presynaptic terminal then mediates local growth processes in the axon so that one presynaptic bouton interacts with a single postsynaptic density (c). See Plate 2.1 in Colour Plate Section.
Molecular mechanisms of TrkB-mediated excitatory synaptogenesis Understanding the molecular mechanisms by which TrkB supports synapse formation is complicated by its ability to activate diverse intracellular responses (Fig. 2). Further, there is redundancy of intracellular and extracellular mechanisms regulating all of TrkB’s downstream signaling. Thus, it is not likely for TrkB’s ability to activate MAPK, PI3 K, or PLC-g that allows for
20
Fig. 2. TrkB signaling mediates diverse cellular responses. See Plate 2.2 in Colour Plate Section.
the emergent impact of its signaling. It is more likely for its ability to act in concert with other signaling inroads to the MAPK, PI3 K, and PLC-g pathways to alter the amplitude of these molecular signals more subtly. This is exemplified by the ability of cAMP to gate the effects of TrkB on axonal and dendritic growth (Song et al., 1997; Ji et al., 2005). Further, the ability of Trk receptors to regulate the cellular localization of downstream signaling events may play a significant role in its cellular impact (Patterson et al., 2001; Delcroix et al., 2003; Kuruvilla et al., 2004). Synapse formation requires a number of cellular processes working together. Transcription and translation are necessary to produce the building blocks for all cellular growth processes, for the production of synaptic adhesion molecules, and for those proteins mediating synaptic transmission and reception. Both the MAPK pathways and PI3 K pathways have been demonstrated to contribute to cellular transcription and translation
(Tao et al., 1998; Tang et al., 2002; Kelleher et al., 2004). However, the nature of, and degree to which, TrkB-mediated transcription and translation contribute to synapse formation is unknown. It is intriguing that TrkB’s ability to spatially direct local protein synthesis may enhance the impact of translation on synapse formation and strengthening (Sutton and Schuman, 2005). The growth of dendritic and axonal arbors is fundamentally linked to synapse formation. In vivo imaging of dendritic growth reveals a process whereby filopodia are stochastically stabilized by synaptic contacts, resulting in dendritic arborization (Niell et al., 2004). Filopodia are also necessary for axonal growth cone guidance (Bentley and Toroian-Raymond, 1986; Chien et al., 1993; Zheng et al., 1996). Dendritic and axon arborization as well as axon guidance have all been demonstrated to be the direct result of TrkB signaling. TrkB-induced axon growth is dependent on both PI3 K and MEK signaling (Atwal et al., 2000).
21
Further, PLC-g and PI3 K signaling are necessary for BDNF-induced growth cone guidance (Ming et al., 1999; Song and Poo, 2001). Trk-dependent net dendritic growth has also been demonstrated to be dependent on MEK and CaMKII; however, the mechanism may depend on selective stabilization (Redmond et al., 2002; Vaillant et al., 2002). Pten knockdown and PI3 K activation appear to be more directly linked to the growth process (Kwon et al., 2001, 2003; Jaworski et al., 2005). Thus, it appears that TrkB may use distinct molecular signaling pathways to direct and elicit growth as well as stabilize new processes, and that synapse formation hangs in the balance of these processes (Fig. 2). Neuronal activity fundamentally mediates growth, synapse formation, and scaling of synaptic strength (Murthy et al., 2001; Burrone et al., 2002; Chen and Ghosh, 2005). TrkB has been proposed to regulate activity at the network level by a number of mechanisms. It may alter the excitatory versus inhibitory nature of GABAergic activity by altering KCC2 expression (Aguado et al., 2003), altering voltage-gated potassium channel activity (Tucker and Fadool, 2002) or affecting presynaptic vesicle pool size and release (Tyler and Pozzo-Miller, 2001). Thus, TrkB may affect synapse formation indirectly by altering the global level of activity. In summary, TrkB signaling has been shown to mediate neuronal growth, synapse stabilization, and activity. All of these processes synergize to produce the net effect of synaptogenesis. The relative importance of TrkB for growth, stabilization, and activity remains to be elucidated, as do the mechanistic relationships of growth, stabilization, and activity and their relative contributions to synaptogenesis. It is likely that the molecular signaling cascades (MAPK, PI3 K, PLC-g) downstream of TrkB are interrelated, but unique with regard to their relative impact on growth, stabilization, and activity. Thus, the exact nature or the molecular mechanisms by which TrkB influences synapse formation in vivo will be context dependent. Furthermore, it will be important to dissect the relative impact of all possible mechanisms on the complicated process of synapse formation. This will require highly sensitive
quantitative assays and precise molecular manipulations that only the most advanced techniques offer.
References Aguado, F., Carmona, M.A., Pozas, E., Aguilo, A., MartinezGuijarro, F.J., Alcantara, S., Borrell, V., Yuste, R., Ibanez, C.F. and Soriano, E. (2003) BDNF regulates spontaneous correlated activity at early developmental stages by increasing synaptogenesis and expression of the K+/Cl co-transporter KCC2. Development, 130: 1267–1280. Alonso, M., Medina, J.H. and Pozzo-Miller, L. (2004) ERK1/2 activation is necessary for BDNF to increase dendritic spine density in hippocampal CA1 pyramidal neurons. Learn Mem., 11: 172–178. Alsina, B., Vu, T. and Cohen-Cory, S. (2001) Visualizing synapse formation in arborizing optic axons in vivo: dynamics and modulation by BDNF. Nat. Neurosci., 4: 1093–1101. Atwal, J.K., Massie, B., Miller, F.D. and Kaplan, D.R. (2000) The TrkBShc site signals neuronal survival and local axon growth via MEK and P13-kinase. Neuron, 27: 265–277. Bentley, D. and Toroian-Raymond, A. (1986) Disoriented pathfinding by pioneer neurone growth cones deprived of filopodia by cytochalasin treatment. Nature, 323: 712–715. Burrone, J., O’Byrne, M. and Murthy, V.N. (2002) Multiple forms of synaptic plasticity triggered by selective suppression of activity in individual neurons. Nature, 420: 414–418. Cabelli, R.J., Allendoerfer, K.L., Radeke, M.J., Welcher, A.A., Feinstein, S.C. and Shatz, C.J. (1996) Changing patterns of expression and subcellular localization of TrkB in the developing visual system. J. Neurosci., 16: 7965–7980. Cabelli, R.J., Shelton, D.L., Segal, R.A. and Shatz, C.J. (1997) Blockade of endogenous ligands of trkB inhibits formation of ocular dominance columns. Neuron, 19: 63–76. Carmona, M.A., Martinez, A., Soler, A., Blasi, J., Soriano, E. and Aguado, F. (2003) Ca(2+)-evoked synaptic transmission and neurotransmitter receptor levels are impaired in the forebrain of trkb (/) mice. Mol. Cell Neurosci., 22: 210–226. Castren, E., Zafra, F., Thoenen, H. and Lindholm, D. (1992) Light regulates expression of brain-derived neurotrophic factor mRNA in rat visual cortex. Proc. Natl. Acad. Sci. USA, 89: 9444–9448. Chen, Y. and Ghosh, A. (2005) Regulation of dendritic development by neuronal activity. J. Neurobiol., 64: 4–10. Chien, C.B., Rosenthal, D.E., Harris, W.A. and Holt, C.E. (1993) Navigational errors made by growth cones without filopodia in the embryonic Xenopus brain. Neuron, 11: 237–251. Cohen-Cory, S. (1999) BDNF modulates, but does not mediate, activity-dependent branching and remodeling of optic axon arbors in vivo. J. Neurosci., 19: 9996–10003. Connor, B., Young, D., Yan, Q., Faull, R.L., Synek, B. and Dragunow, M. (1997) Brain-derived neurotrophic factor is
22 reduced in Alzheimer’s disease. Brain Res. Mol. Brain Res., 49: 71–81. Dailey, M.E. and Smith, S.J. (1996) The dynamics of dendritic structure in developing hippocampal slices. J. Neurosci., 16: 2983–2994. Delcroix, J.D., Valletta, J.S., Wu, C., Hunt, S.J., Kowal, A.S. and Mobley, W.C. (2003) NGF signaling in sensory neurons: evidence that early endosomes carry NGF retrograde signals. Neuron, 39: 69–84. Elmariah, S.B., Crumling, M.A., Parsons, T.D. and BaliceGordon, R.J. (2004) Postsynaptic TrkB-mediated signaling modulates excitatory and inhibitory neurotransmitter receptor clustering at hippocampal synapses. J. Neurosci., 24: 2380–2393. Eom, T., Antar, L.N., Singer, R.H. and Bassell, G.J. (2003) Localization of a beta-actin messenger ribonucleoprotein complex with zipcode-binding protein modulates the density of dendritic filopodia and filopodial synapses. J. Neurosci., 23: 10433–10444. Ernfors, P., Bengzon, J., Kokaia, Z., Persson, H. and Lindvall, O. (1991) Increased levels of messenger RNAs for neurotrophic factors in the brain during kindling epileptogenesis. Neuron, 7: 165–176. Fawcett, J.P., Aloyz, R., McLean, J.H., Pareek, S., Miller, F.D., McPherson, P.S. and Murphy, R.A. (1997) Detection of brain-derived neurotrophic factor in a vesicular fraction of brain synaptosomes. J. Biol. Chem., 272: 8837–8840. Fiala, J.C., Feinberg, M., Popov, V. and Harris, K.M. (1998) Synaptogenesis via dendritic filopodia in developing hippocampal area CA1. J. Neurosci., 18: 8900–8911. Gall, C.M. and Isackson, P.J. (1989) Limbic seizures increase neuronal production of messenger RNA for nerve growth factor. Science, 245: 758–761. Gallo, G. and Letourneau, P.C. (1998) Localized sources of neurotrophins initiate axon collateral sprouting. J. Neurosci., 18: 5403–5414. Gallo, G. and Letourneau, P.C. (2004) Regulation of growth cone actin filaments by guidance cues. J. Neurobiol., 58: 92–102. Gehler, S., Gallo, G., Veien, E. and Letourneau, P.C. (2004a) p75 neurotrophin receptor signaling regulates growth cone filopodial dynamics through modulating RhoA activity. J. Neurosci., 24: 4363–4372. Gehler, S., Shaw, A.E., Sarmiere, P.D., Bamburg, J.R. and Letourneau, P.C. (2004b) Brain-derived neurotrophic factor regulation of retinal growth cone filopodial dynamics is mediated through actin depolymerizing factor/cofilin. J. Neurosci., 24: 10741–10749. Geinisman, Y., Berry, R.W., Disterhoft, J.F., Power, J.M. and Van der Zee, E.A. (2001) Associative learning elicits the formation of multiple-synapse boutons. J. Neurosci., 21: 5568–5573. Gonzalez, M., Ruggiero, F.P., Chang, Q., Shi, Y.J., Rich, M.M., Kraner, S. and Balice-Gordon, R.J. (1999) Disruption of Trkb-mediated signaling induces disassembly of postsynaptic receptor clusters at neuromuscular junctions. Neuron, 24: 567–583.
Hartmann, M., Brigadski, T., Erdmann, K.S., Holtmann, B., Sendtner, M., Narz, F. and Lessmann, V. (2004) Truncated TrkB receptor-induced outgrowth of dendritic filopodia involves the p75 neurotrophin receptor. J. Cell Sci., 117: 5803–5814. Hartmann, M., Heumann, R. and Lessmann, V. (2001) Synaptic secretion of BDNF after high-frequency stimulation of glutamatergic synapses. Embo. J., 20: 5887–5897. Haubensak, W., Narz, F., Heumann, R. and Lessmann, V. (1998) BDNF-GFP containing secretory granules are localized in the vicinity of synaptic junctions of cultured cortical neurons. J. Cell Sci., 111(Pt 11): 1483–1493. Horch, H.W. and Katz, L.C. (2002) BDNF release from single cells elicits local dendritic growth in nearby neurons. Nat. Neurosci., 5: 1177–1184. Horch, H.W., Kruttgen, A., Portbury, S.D. and Katz, L.C. (1999) Destabilization of cortical dendrites and spines by BDNF. Neuron, 23: 353–364. Hyman, C., Hofer, M., Barde, Y.A., Juhasz, M., Yancopoulos, G.D., Squinto, S.P. and Lindsay, R.M. (1991) BDNF is a neurotrophic factor for dopaminergic neurons of the substantia nigra. Nature, 350: 230–232. Jaworski, J., Spangler, S., Seeburg, D.P., Hoogenraad, C.C. and Sheng, M. (2005) Control of dendritic arborization by the phosphoinositide-30 -kinase-Akt-mammalian target of rapamycin pathway. J. Neurosci., 25: 11300–11312. Ji, Y., Pang, P.T., Feng, L. and Lu, B. (2005) Cyclic AMP controls BDNF-induced TrkB phosphorylation and dendritic spine formation in mature hippocampal neurons. Nat. Neurosci., 8: 164–172. Jontes, J.D., Buchanan, J. and Smith, S.J. (2000) Growth cone and dendrite dynamics in zebrafish embryos: early events in synaptogenesis imaged in vivo. Nat. Neurosci., 3: 231–237. Jourdi, H., Iwakura, Y., Narisawa-Saito, M., Ibaraki, K., Xiong, H., Watanabe, M., Hayashi, Y., Takei, N. and Nawa, H. (2003) Brain-derived neurotrophic factor signal enhances and maintains the expression of AMPA receptor-associated PDZ proteins in developing cortical neurons. Dev. Biol., 263: 216–230. Kang, H. and Schuman, E.M. (1995) Long-lasting neurotrophin-induced enhancement of synaptic transmission in the adult hippocampus. Science, 267: 1658–1662. Kang, H. and Schuman, E.M. (1996) A requirement for local protein synthesis in neurotrophin-induced hippocampal synaptic plasticity. Science, 273: 1402–1406. Kelleher, R.J., 3rd, Govindarajan, A., Jung, H.Y., Kang, H. and Tonegawa, S. (2004). Translational control by MAPK signaling in long-term synaptic plasticity and memory. Cell, 116 467479. Klau, M., Hartmann, M., Erdmann, K.S., Heumann, R. and Lessmann, V. (2001) Reduced number of functional glutamatergic synapses in hippocampal neurons overexpressing full-length TrkB receptors. J. Neurosci. Res., 66: 327–336. Klein, R., Martin-Zanca, D., Barbacid, M. and Parada, L.F. (1990) Expression of the tyrosine kinase receptor gene trkB is confined to the murine embryonic and adult nervous system. Development, 109: 845–850.
23 Klein, R., Parada, L.F., Coulier, F. and Barbacid, M. (1989) TrkB, a novel tyrosine protein kinase receptor expressed during mouse neural development. EMBO J., 8: 3701–3709. Kohara, K., Kitamura, A., Morishima, M. and Tsumoto, T. (2001) Activity-dependent transfer of brain-derived neurotrophic factor to postsynaptic neurons. Science, 291: 2419–2423. Korte, M., Carroll, P., Wolf, E., Brem, G., Thoenen, H. and Bonhoeffer, T. (1995) Hippocampal long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor. Proc. Natl. Acad. Sci. USA, 92: 8856–8860. Korte, M., Griesbeck, O., Gravel, C., Carroll, P., Staiger, V., Thoenen, H. and Bonhoeffer, T. (1996) Virus-mediated gene transfer into hippocampal CA1 region restores long-term potentiation in brain-derived neurotrophic factor mutant mice. Proc. Natl. Acad. Sci. USA, 93: 12547–12552. Kuruvilla, R., Zweifel, L.S., Glebova, N.O., Lonze, B.E., Valdez, G., Ye, H. and Ginty, D.D. (2004) A neurotrophin signaling cascade coordinates sympathetic neuron development through differential control of TrkA trafficking and retrograde signaling. Cell, 118: 243–255. Kwon, C.H., Zhu, X., Zhang, J. and Baker, S.J. (2003) mTor is required for hypertrophy of Pten-deficient neuronal soma in vivo. Proc. Natl. Acad. Sci. USA, 100: 12923–12928. Kwon, C.H., Zhu, X., Zhang, J., Knoop, L.L., Tharp, R., Smeyne, R.J., Eberhart, C.G., Burger, P.C. and Baker, S.J. (2001) Pten regulates neuronal soma size: a mouse model of Lhermitte-Duclos disease. Nat. Genet., 29: 404–411. Levine, E.S., Dreyfus, C.F., Black, I.B. and Plummer, M.R. (1995) Brain-derived neurotrophic factor rapidly enhances synaptic transmission in hippocampal neurons via postsynaptic tyrosine kinase receptors. Proc. Natl. Acad. Sci. USA, 92: 8074–8077. Lindsay, R.M., Barde, Y.A., Davies, A.M. and Rohrer, H. (1985) Differences and similarities in the neurotrophic growth factor requirements of sensory neurons derived from neural crest and neural placode. J. Cell Sci. Suppl., 3: 115–129. Lohof, A.M., Ip, N.Y. and Poo, M.M. (1993) Potentiation of developing neuromuscular synapses by the neurotrophins NT-3 and BDNF. Nature, 363: 350–353. Lom, B. and Cohen-Cory, S. (1999) Brain-derived neurotrophic factor differentially regulates retinal ganglion cell dendritic and axonal arborization in vivo. J. Neurosci., 19: 9928–9938. Luikart, B.W., Nef, S., Virmani, T., Lush, M.E., Liu, Y., Kavalali, E.T. and Parada, L.F. (2005) TrkB has a cell-autonomous role in the establishment of hippocampal Schaffer collateral synapses. J. Neurosci., 25: 3774–3786. Martinez, A., Alcantara, S., Borrell, V., Del Rio, J.A., Blasi, J., Otal, R., Campos, N., Boronat, A., Barbacid, M., Silos-Santiago, I. and Soriano, E. (1998) TrkB and TrkC signaling are required for maturation and synaptogenesis of hippocampal connections. J. Neurosci., 18: 7336–7350. Martin-Zanca, D., Barbacid, M. and Parada, L.F. (1990) Expression of the trk proto-oncogene is restricted to the sensory cranial and spinal ganglia of neural crest origin in mouse development. Genes Dev., 4: 683–694.
McAllister, A.K., Lo, D.C. and Katz, L.C. (1995) Neurotrophins regulate dendritic growth in developing visual cortex. Neuron, 15: 791–803. Ming, G., Song, H., Berninger, B., Inagaki, N., TessierLavigne, M. and Poo, M. (1999) Phospholipase C-gamma and phosphoinositide 3-kinase mediate cytoplasmic signaling in nerve growth cone guidance. Neuron, 23: 139–148. Murthy, V.N., Schikorski, T., Stevens, C.F. and Zhu, Y. (2001) Inactivity produces increases in neurotransmitter release and synapse size. Neuron, 32: 673–682. Murphy, D.D., Cole, N.B. and Segal, M. (1998) Brain-derived neurotrophic factor mediates estradiol-induced dendritic spine formation in hippocampal neurons. Proc. Natl. Acad. Sci. USA, 95: 11412–11417. Niell, C.M., Meyer, M.P. and Smith, S.J. (2004) In vivo imaging of synapse formation on a growing dendritic arbor. Nat. Neurosci., 7: 254–260. Niell, C.M. and Smith, S.J. (2004) Live optical imaging of nervous system development. Annu. Rev. Physiol., 66: 771–798. Patterson, S.L., Abel, T., Deuel, T.A., Martin, K.C., Rose, J.C. and Kandel, E.R. (1996) Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal LTP in BDNF knockout mice. Neuron, 16: 1137–1145. Patterson, S.L., Grover, L.M., Schwartzkroin, P.A. and Bothwell, M. (1992) Neurotrophin expression in rat hippocampal slices: a stimulus paradigm inducing LTP in CA1 evokes increases in BDNF and NT-3 mRNAs. Neuron, 9: 1081–1088. Patterson, S.L., Pittenger, C., Morozov, A., Martin, K.C., Scanlin, H., Drake, C. and Kandel, E.R. (2001) Some forms of cAMP-mediated long-lasting potentiation are associated with release of BDNF and nuclear translocation of phosphoMAP kinase. Neuron, 32: 123–140. Poo, M.M. (2001) Neurotrophins as synaptic modulators. Nat. Rev. Neurosci., 2: 24–32. Pozzo-Miller, L.D., Gottschalk, W., Zhang, L., McDermott, K., Du, J., Gopalakrishnan, R., Oho, C., Sheng, Z.H. and Lu, B. (1999) Impairments in high-frequency transmission, synaptic vesicle docking, and synaptic protein distribution in the hippocampus of BDNF knockout mice. J. Neurosci., 19: 4972–4983. Redmond, L., Kashani, A.H. and Ghosh, A. (2002) Calcium regulation of dendritic growth via CaM kinase IV and CREB-mediated transcription. Neuron, 34: 999–1010. Rico, B., Xu, B. and Reichardt, L.F. (2002) TrkB receptor signaling is required for establishment of GABAergic synapses in the cerebellum. Nat. Neurosci., 5: 225–233. Shimada, A., Mason, C.A. and Morrison, M.E. (1998) TrkB signaling modulates spine density and morphology independent of dendrite structure in cultured neonatal Purkinje cells. J. Neurosci., 18: 8559–8570. Song, H.J., Ming, G.L. and Poo, M.M. (1997) cAMP-induced switching in turning direction of nerve growth cones. Nature, 388: 275–279. Song, H. and Poo, M. (2001) The cell biology of neuronal navigation. Nat. Cell Biol., 3: E81–E88.
24 Soppet, D., Escandon, E., Maragos, J., Middlemas, D.S., Reid, S.W., Blair, J., Burton, L.E., Stanton, B.R., Kaplan, D.R., Hunter, T., et al. (1991) The neurotrophic factors brain-derived neurotrophic factor and neurotrophin-3 are ligands for the trkB tyrosine kinase receptor. Cell, 65: 895–903. Sutton, M.A. and Schuman, E.M. (2005) Local translational control in dendrites and its role in long-term synaptic plasticity. J. Neurobiol., 64: 116–131. Tang, S.J., Reis, G., Kang, H., Gingras, A.C., Sonenberg, N. and Schuman, E.M. (2002) A rapamycin-sensitive signaling pathway contributes to long-term synaptic plasticity in the hippocampus. Proc. Natl. Acad. Sci. USA, 99: 467–472. Tao, X., Finkbeiner, S., Arnold, D.B., Shaywitz, A.J. and Greenberg, M.E. (1998) Ca2+ influx regulates BDNF transcription by a CREB family transcription factor-dependent mechanism. Neuron, 20: 709–726. Tucker, K. and Fadool, D.A. (2002) Neurotrophin modulation of voltage-gated potassium channels in rat through TrkB receptors is time and sensory experience dependent. J. Physiol., 542: 413–429. Tucker, K.L., Meyer, M. and Barde, Y.A. (2001) Neurotrophins are required for nerve growth during development. Nat. Neurosci., 4: 29–37. Tyler, W.J. and Pozzo-Miller, L.D. (2001) BDNF enhances quantal neurotransmitter release and increases the number of docked vesicles at the active zones of hippocampal excitatory synapses. J. Neurosci., 21: 4249–4258. Vaillant, A.R., Zanassi, P., Walsh, G.S., Aumont, A., Alonso, A. and Miller, F.D. (2002) Signaling mechanisms underlying reversible, activity-dependent dendrite formation. Neuron, 34: 985–998.
Vicario-Abejon, C., Collin, C., McKay, R.D. and Segal, M. (1998) Neurotrophins induce formation of functional excitatory and inhibitory synapses between cultured hippocampal neurons. J. Neurosci., 18: 7256–7271. Washbourne, P., Dityatev, A., Scheiffele, P., Biederer, T., Weiner, J.A., Christopherson, K.S. and El-Husseini, A. (2004) Cell adhesion molecules in synapse formation. J. Neurosci., 24: 9244–9249. Woolley, C.S., Wenzel, H.J. and Schwartzkroin, P.A. (1996) Estradiol increases the frequency of multiple synapse boutons in the hippocampal CA1 region of the adult female rat. J. Comp. Neurol., 373: 108–117. Yan, Q., Radeke, M.J., Matheson, C.R., Talvenheimo, J., Welcher, A.A. and Feinstein, S.C. (1997) Immunocytochemical localization of TrkB in the central nervous system of the adult rat. J. Comp. Neurol., 378: 135–157. Zafra, F., Hengerer, B., Leibrock, J., Thoenen, H. and Lindholm, D. (1990) Activity dependent regulation of BDNF and NGF mRNAs in the rat hippocampus is mediated by nonNMDA glutamate receptors. EMBO J., 9: 3545–3550. Zagrebelsky, M., Holz, A., Dechant, G., Barde, Y.A., Bonhoeffer, T. and Korte, M. (2005) The p75 neurotrophin receptor negatively modulates dendrite complexity and spine density in hippocampal neurons. J. Neurosci., 25: 9989–9999. Zheng, J.Q., Wan, J.J. and Poo, M.M. (1996) Essential role of filopodia in chemotropic turning of nerve growth cone induced by a glutamate gradient. J. Neurosci., 16: 1140–1149. Ziv, N.E. and Smith, S.J. (1996) Evidence for a role of dendritic filopodia in synaptogenesis and spine formation. Neuron, 17: 91–102.
Plate 2.1. Model of BDNF/TrkB function in activity-dependent refinement of synapse number. Initially, BDNF is released at high levels from active presynaptic boutons (red terminal in (a)). The presynaptically released BDNF then signals to TrkB receptors expressed on nearby filopodia (green). The TrkB activation allows for the additional growth and attraction of filopodia to the presynaptic terminal (b). The postsynaptic activation of TrkB as well as activity allows for the up-regulation and secretion of BDNF from the postsynaptic structures back onto the active presynaptic terminal (b). The TrkB activation in the presynaptic terminal then mediates local growth processes in the axon so that one presynaptic bouton interacts with a single postsynaptic density (c).
Plate 2.2. TrkB signaling mediates diverse cellular responses.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 3
Signal transduction mechanisms in memory disorders Sara C. Shalin, Regula Egli, Shari G. Birnbaum, Tania L. Roth, Jonathan M. Levenson and J. David Sweatt Department of Neuroscience, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA
Abstract: This chapter explores some of the molecular events contributing to memory formation and how, when these events malfunction, disturbances in memory occur. After a brief discussion of signaling in the hippocampus, we will explore the topics of human mental retardation syndromes that involve disruption of these processes, including Angelman syndrome (AS), Neurofibromatosis 1 (NF1)-associated learning disorders, Coffin–Lowry syndrome (CLS), Rubinstein–Taybi syndrome (RTS), and Rett syndrome (RTT). Keywords: CaMKII; ERK; NF1; RSK; CBP; MeCP2; MAPK; learning; epigenetics; mental retardation instances, they are also providing opportunities for possible therapeutic options. Our discussion will focus on genetic syndromes in which cognitive impairments play a major role, including Angelman syndrome (AS), Neurofibromatosis 1 (NF1), Coffin–Lowry syndrome (CLS), Rubinstein–Taybi syndrome (RTS), and Rett syndrome (RTT).
Introduction Signaling within the nervous system is remarkably complex, but a vital process necessary for an organism’s survival. It is difficult on the surface to conceive that the phosphorylation state of a protein could be important for the creation of a memory, or that a small imbalance of certain signaling molecules could trigger deficits in learning and memory processing. Yet it is becoming increasingly clear that these types of signaling events must be precisely regulated to allow for normal learning. Moreover, evidence is accumulating that not only mutations of genes directly involved in certain signaling cascades can contribute to disorders of memory, but genetic mutations, which indirectly affect widespread processes such as epigenetic chromatin remodeling and genomic imprinting, can also play a role in the pathogenesis of cognitive disorders. Mouse models of some of these diseases have brought us tremendous insight into what molecular events go awry and contribute to the learning disorders in particular. In some
The importance of signaling cascades within the hippocampus Excitatory neurotransmission within the central nervous system is mediated primarily by the neurotransmitter glutamate. Following glutamate binding to AMPA and NMDA receptors in the postsynaptic density, a wave of activity occurs in the postsynaptic dendritic spine. Much of this activity is initiated by the influx of calcium. Calcium, in addition to its essential role in the presynaptic neuron in regulating neurotransmitter release, plays an equally important role in postsynaptic activity, primarily through its activation of kinases. Calcium-calmodulin kinase II (CaMKII), protein kinase C (PKC), protein kinase A (PKA), and mitogen-activated protein kinase (MAPK) can all be activated, either directly or
Corresponding author. Tel.: +1 205 975-5196; Fax: +1 205 975-5097; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57003-7
25
26
indirectly, through the actions of calcium. Initiation of these kinase cascades leads ultimately to the generation of long-term potentiation (LTP), what is believed to be a cellular correlate to processes of learning and memory formation. In overview, in this chapter we will first describe the four major protein kinase signal transduction cascades that have been implicated in synaptic plasticity and memory formation. We then will describe five separate human mental retardation syndromes that involve disruption of these signaling cascades. The unifying theme for the chapter is the remarkable convergence of basic studies regarding the molecular basis of plasticity and memory formation with more disease-oriented investigations originating from the identification of human mental retardation-associated genes. In our view, a unifying model is beginning to emerge as the normal processes underlying memory formation intersect with the derangement of these same processes in learning and memory disorders.
CaMKII Calcium binds to calmodulin (CaM) and activated CaM leads to activation of CaMKII, a key player in the regulation of LTP and memory formation. Once activated, CaMKII has the unique ability to phosphorylate itself at Thr286, which renders the molecule no longer completely dependent on calcium for its activity. For this reason, autophosphorylatable CaMKII has often been cited as a memory molecule. Pharmacologic and genetic manipulations have made it clear that CaMKII activation is necessary and sufficient for the induction and maintenance of early forms of LTP (for review, see Lisman et al., 2002). In addition, CaMKII can interact with NMDA receptors, phosphorylate AMPA receptors, and through this phosphorylation may contribute to their increased incorporation into the plasma membrane following LTP induction. CaMKII can also inhibit Synaptic GTPase activating protein (Syn-GAP), which then enhances the guanosine triphosphate (GTP) bound form (and thus, active) form of Ras, thereby contributing to indirect regulation of the
extracellular signal regulated (ERK)/MAPK cascade (Waltereit and Weller, 2003). PKA Protein kinase A (PKA) becomes activated following elevations of cAMP, secondary to activation of adenylyl cyclase. Adenylyl cyclase activation can be triggered through elevations in calcium or through coupling to Gas-protein-linked receptors. Binding of cAMP to the regulatory subunits of PKA leads to their dissociation from the active, catalytic subunits (Kandel et al., 2000). In mammalian systems, PKA is particularly implicated in gene regulation following L-LTP-inducing stimuli. In hippocampal cell culture, inhibition of PKA decreases cAMP response element (CRE)-mediated transcription and phosphorylation of the transcription factor cAMP response element binding protein (CREB) (Impey et al., 1998). This observation is likely due to PKAs cross talk with the ERK/MAPK cascade, as PKA inhibition also blocks ERKs translocation to the nucleus and phosphorylation of nuclear RSK, the kinase that phosphorylates CREB (Impey et al., 1998; Roberson et al., 1999). PKA has also been linked to phosphorylation of the transcription factor Elk-1, again through its interactions with the ERK pathway (Vossler et al., 1997). PKA and/or cAMP activity have been reported to be required for L-LTP and the Morris water maze (Roberson and Sweatt, 1996; Waltereit and Weller, 2003), and inhibition of PKA impairs both short and long-term, cued and contextual memories in a fear conditioning paradigm (Ahi et al., 2004). PKC The PKC family includes calcium-responsive isoforms (a, b, g), as well as atypical isoforms (d, e, Z, y, m, l, z) that are regulated independently of calcium concentrations. Best characterized are the calcium-responsive isoforms, which are present in the hippocampus and are activated following the phospholipid hydrolysis by phospholipase C (PLC) that generates diacylglycerol (DAG) and
27
inositol-triphsophate (IP3) (for review, see (Angenstein and Staak, 1997). Like PKA and CaMKII, PKC has been shown to be important in the induction and maintenance of LTP as well as in behavioral assessments of memory. Pharmacological and genetic manipulations have independently ascertained a role for the various PKC isoforms in both LTP and learning paradigms (Abeliovich et al., 1993; Hvalby et al., 1994; Wang and Kelly, 1996; Vianna et al., 2000; Weeber et al., 2000). Like CaMKII, PKC can become autonomously active through oxidation of its side chains by reactive oxygen species or by autophosphorylation events in its C-terminus, and synthesis of the constitutively active fragment PKCz is increased following LTP induction (Sweatt, 2003). These modifications make PKC an attractive candidate for a molecular correlate of potentiation. In addition, although generally attributed to CaMKII, PKC can also phosphorylate AMPA receptors at Ser831, thereby increasing their conductance. One of the targets of PKC phosphorylation that has implicated in LTP induction is the small protein neurogranin. Phosphorylation of neurogranin releases its association with CaM, thereby enhancing the availability of CaM (Gerendasy and Sutcliffe, 1997). This of course allows a general amplification of calcium signaling, as CaM activates CaMKII and adenylyl cyclase. PKC activation through group I mGluRs has also been implicated in the control of dendritic protein synthesis (Antar et al., 2004) and transcription factor phosphorylation (Meberg et al., 1996; Roberson et al., 1999), possibly through its ability to trigger activation of the MAPK cascade. All in all, PKC represents yet another diverse modulator of synaptic plasticity.
ERK/MAPK signaling in neurons The ERK/MAPK cascade has also been implicated in normal memory formation as well as in several human disorders of memory formation. The ERK/MAPK cascade is a ubiquitously expressed series of kinases, which, when activated sequentially, go on to exert many downstream effects. The ERKs are part of a superfamily of
MAPKs including the JNK and p38 SAP kinases. Each cascade is characterized by the presence of a MAPK kinase kinase, which phosphorylates and activates MAPK kinase, which phosphorylates and activates MAPK. The ERK cascade consists of Raf kinase, which phosphorylates MEK, which then phosphorylates and activates ERK. This basic premise of sequential phosphorylation belies the remarkable complexity with which this cascade is regulated. Not only are there myriad signals that feed forward into ERK activation that must be precisely regulated, but the many diverse effectors of ERK are specifically controlled as well. A large body of research has established ERK’s involvement in rodent behavior and memory, including associative fear conditioning (Atkins et al., 1998; Schafe et al., 2000), spatial learning (Blum et al., 1999; Selcher et al., 1999), and conditioned place preference (Gerdjikov et al., 2004; Miller and Marshall, 2005). ERK has also been implicated as critical for normal synaptic plasticity in several LTP induction paradigms (English and Sweatt, 1997; Watabe et al., 2000; Mazzucchelli et al., 2002; Selcher et al., 2003; Kelleher et al., 2004). When activated, many different neuronal receptors can lead to activation of the ERK cascade (Roberson et al., 1999; Watabe et al., 2000; Krapivinsky et al., 2003; Morozov et al., 2003; Waltereit and Weller, 2003,). These diverse membrane receptors feed into MAPK activation primarily through either a PKC/Ras-dependent pathway or a PKA-dependent pathway. Early characterizations of ERK involved its activation by various growth factors. Binding of a growth factor to its respective receptor leads to tyrosine kinase activation of the receptor, the recruitment of adaptor proteins, and subsequent activation of Ras. Ras can then trigger the membrane translocation and activation of Raf-1, the best characterized Raf isoform. G-protein coupled receptors can also link to the ERK cascade. Receptors that couple to Gas stimulate adenylyl cyclase and PKA activation. PKA can lead to ERK phosphorylation by the activation of the small, Ras-like, GTP-binding Rap1 which then can activate the B-raf isoform (Dhillon and Kolch, 2002). Additionally, activation of Gaq-coupled receptors can generate phospholipid
28
hydrolysis and DAG production. IP3 contributes to calcium entry, which, among other effects, can elicit PKC activation, and DAG can lead to PKC activation directly. PKC-mediated ERK activation can be Ras-dependent or Ras-independent but proceeds through activation of the Raf-1 isoform (Dhillon and Kolch, 2002). Calcium entry through voltage-gated calcium channels or NMDA receptors also leads to the activation of PKC; of course, this elevation in intracellular calcium can also generate cAMP and PKA activation. Taken together, it is increasingly clear that the ERK activation necessary for synaptic plasticity and memory formation is not a linear signaling module but a complex network of tightly regulated events. ERK phosphorylates numerous targets in many different subcellular compartments. ERK phosphorylation of these targets has been linked to numerous processes including changes in cellular excitability (Adams et al., 2000a; Yuan et al., 2002), protein transcription (Vossler et al., 1997; Impey et al., 1998; Adams et al., 2000b), protein translation (Kelleher et al., 2004), remodeling of the dendritic cytoskeleton (Wu et al., 2001; Goldin and Segal, 2003), and alterations in chromatin structure (Levenson et al., 2004). ERK activity leads to the phosphorylation of transcription factors either through the activation of intermediary kinases or through direct phosphorylation events. ERK phosphorylates the kinase p90RSK, which subsequently phosphorylates the well-studied transcription factor CREB (Impey et al., 1998; Adams et al., 2000b). ERK can also translocate into the nucleus to directly phosphorylate such transcription factors as Elk-1, c-fos, and c-Jun (Pearson et al., 2001). More recently, ERK has even been shown to affect chromatin structure through its ability to acetylate and phosphorylate histone proteins, which affects the ability of transcriptional machinery to interact with DNA (Levenson et al., 2004). In addition, ERK has other targets outside of the nucleus. For instance, Kv4.2, the likely molecular identity of the A-type potassium channels that are observed in neuronal dendrites, is a membrane substrate of ERK (Adams et al., 2000a). When phosphorylated, the channel becomes less
active, allowing the neuron to become more excitable and neuronal signals to be better propagated. Both PKC and PKA can modulate the channel’s properties through ERK, causing an increase in the amplitude of back-propagating action potentials and increasing neuronal excitability (Yuan et al., 2002). In summary, a multitude of pathways are activated postsynaptically, and our discussion is certainly not exhaustive. The processes we have discussed are summarized in Fig. 1. Human mental retardation syndromes We will now turn our discussion to human diseases of which learning disorders or mental retardation are a prominent component. The associated retardation in AS may result from disruptions in normal CaMKII signaling, while the learning disorders associated with NF1 and CLS are likely due to dysregulation of the MAPK cascade. Rubinstein–Taybi and Rett syndromes reflect how imprecise regulation of the transcription machinery and the epigenome may ultimately lead to significant cognitive impairments. Angelman syndrome Angelman syndrome (AS), described in 1965 by Harry Angelman, is a relatively rare retardation disorder (incidence of 1 out of 10–20,000) characterized by severe learning disabilities as well as several prominent physical features. Among other symptoms, AS patients present with ataxic movements, an inability to speak (or very limited speech), epileptic seizures with a characteristic EEG, and dysmorphic facial features. Behaviorally, AS patients exhibit a happy demeanor — they are prone to sudden bouts of laughter, are hyperactive and very social, which, coupled with their jerky movements, gave rise to the unfortunate initial name of ‘‘Happy Puppet Syndrome,’’ which has since been replaced with AS (Clayton-Smith and Laan, 2003). A genetic etiology was suspected for AS, but this was not confirmed until Magenis et al. (1987) described two unrelated girls displaying characteristics consistent with AS who had
29
=Transcription Factor = G-protein (Gαs or Gαq) AMPA R
= Ribosome
β -AR AC
PKA P
NMDA CaM
R
CaMKII
Ca++
Glutamate
MAPK
VGCC
AMPA R
P
PKC PLC
mGluR Ras RTKs
Fig. 1. Signaling at the postsynaptic neuron: receptors, channels, and kinases. Signaling at the postsynaptic neuron is complex. Some of the key players are depicted here, although this is an oversimplification of the diverse sets of receptors, channels, and kinases that are activated following glutamate release from the presynaptic terminal and subsequent calcium release in the postsynaptic neuron. CaMKII, PKA, PKC, and MAPK are all activated by various mechanisms, and events downstream of these kinases contribute to the generation of long-term potentiation through the phosphorylation of targets such as AMPA and NMDA receptors, the insertion of AMPA receptors into the postsynaptic membrane, and the induction of transcription and translation. See text for details. See Plate 3.1 in Colour Plate Section.
deletions in the 15q11-13 chromosome. Since then, much work has been devoted to further elucidating the molecular mechanisms underlying AS, most of which has focused on how disruption of the 15q1113 chromosome impacts normal development. There are four primary types of disruption of 15q11-13 locus which result in AS, and all of them are unique because they affect the maternal expression of the UBE3A gene (Clayton-Smith and Laan, 2003). Class I, which accounts for 70–75% of AS patients, is the result of the interstitial deletion of chromosome 15q11-13. Class II is the result of uniparental disomy for chromosome 15, which results in the loss of inheritance of a
maternal copy of UBE3A. Class III mutations involve the abnormal methylation of chromosome 15 (indicating defective imprinting processes), while Class IV cases are caused by mutations in the UBE3A gene itself. All of the classes of defects result in the abnormal maternal expression of UBE3A and result in the deletion of the protein product of UBE3A, the E6-associated protein (E6AP). E6-AP is a cellular ubiquitin ligase enzyme which participates in the ubiquitin degradation pathway by forming a covalent link between ubiquitin and its target protein, thus marking the protein for degradation. There are very few known substrates of E6-AP — to date, the three identified
30
targets of E6-AP are Src (a tyrosine kinase), HHR23A,B (involved in DNA repair), and p53 (a cell-cycle control protein). Unfortunately, how the disruption of any of the known targets of E6-AP results in the AS phenotype is not immediately clear. It is for this reason that the production of an AS mouse was such an important development in the study of the mechanism of this disorder. The AS mouse was made by the targeted disruption of the UBE3A gene, which resulted in an ube3a null mouse (Jiang et al., 1998). Mice with a maternal deficiency in ube3a (m /p+) showed a phenotype similar to that seen in human AS patients: a reduction in brain weight, deficiency in motor skills, inducible seizures with a characteristic EEG, and poor performance on a hippocampal-based learning task (context-dependent fear conditioning). In addition, the mice had elevated levels of p53, one of the target proteins of E6-AP. Interestingly, initial physiological characterization of these animals indicated that basal hippocampal synaptic physiology (input/output curve and paired pulse facilitation) was normal. However, using a relatively mild plasticity protocol of two trains of 1 s 100 Hz stimulation separated by 20 s at 25 1C, Jiang et al. (1998) found that LTP was severely blunted. This reduction in synaptic plasticity suggested that further examination of the mechanism behind the disruption of LTP may lead to a potential explanation of the disruption of normal learning and development seen in AS patients. Weeber et al. (2003) performed a more in-depth characterization of the ube3a-deficient AS mice, examining the expression of various types of plasticity, as well as examining potential changes in signaling molecules known to underlie normal plasticity. This group tested a stronger LTP protocol (three sets of high-frequency stimulation — each HFS consisting of two trains of 100 Hz separated by 20 s at 32 1C) and found that while the ube3a-null mice could still support plasticity, they showed an altered threshold for the induction of this plasticity. These data suggested a potential role of altered Ca2+ signaling in the ube3a-null mice, since postsynaptic Ca2+ influx and signaling is crucial for the induction of LTP (Malenka et al., 1989; Malinow et al., 1989). To determine whether
the mechanism of LTP alteration was upstream (via NMDAR activation) or downstream of Ca2+ influx, Weeber et al. tested whether these mice could support a form of LTP that is NMDARindependent, but still Ca2+-dependent. Interestingly, the ube3a-null mice did not support this form of LTP, indicating that the alteration in Ca2+ signaling leading to the disruption of normal LTP induction is downstream of Ca2+ influx. This observation led to the examination of signaling molecules that are downstream of Ca2+ entry into the postsynaptic cell. There are four primary signaling molecules known to be involved in normal LTP that could be affected by altered Ca2+ signaling in these mice — PKC, PKA, ERK, and CaMKII. Weeber et al. tested all four of these and found that while there were no changes in overall protein levels of any of the molecules, there was a significant increase in the level of autophosphorylated CaMKII (Thr-286 phosphorylated CaMKII, Thr-286-P-CaMKII). Since the phosphorylation state of CaMKII is known to be a point of regulation for normal LTP, this finding suggests that the dysregulation of normal LTP and normal learning behavior seen in the AS mice (and potentially in AS patients) may be due to altered CaMKII signaling. To further examine a potential mechanism mediated by altered CaMKII phosphorylation, in vitro CaMKII activity experiments were performed. Interestingly, despite the increase in levels of autophosphorylated CaMKII, the kinase showed an overall reduction in responsiveness to activation with Ca2+ and CaM. This prompted the authors to examine levels of Thr-305 phosphorylated CaMKII, which proved also to be elevated. Phosphorylation of CaMKII at Thr305 results in an inhibitory effect on the kinase activity because it blocks the binding of the kinase to Ca2+ and CaM (Colbran and Soderling, 1990) and thus, an increase in Thr305-P-CaMKII could account for the reduction in overall CaMKII activity seen in the in vitro assays. It is also known that the phosphorylation state of CaMKII can affect the association of the kinase with the postsynaptic density (PSD), with Thr-286-P-CaMKII showing a higher association with the PSD than Thr305-P-CaMKII (Elgersma et al., 2002). Consistent with this, AS mice show an
31
overall reduction in levels of CaMKII associated with the PSD, which has previously been associated with a decrease in LTP and learning (Elgersma et al., 2002). The finding that the phosphorylation state of CaMKII is altered in AS mice is interesting, but the mechanism by which the deletion of E6-AP can manifest this is unclear. To probe further into the mechanism mediating the phosphorylation changes in CaMKII, Weeber et al. (2003) examined the activity of the phosphatases that modulate CaMKII phosphorylation, PP1 and PP2A. Overall levels of these phosphatases were unaltered, but their activity was found to be reduced, thus providing a potential mechanism for the increase in both Thr-286-P-CaMKII and Thr-305-P-CaMKII seen in the AS mice. The effect seen on PP1 and PP2A activity in these mice is suggestive of a mechanism mediated by the ube3a pathway, because recent work has indicated that there is a link between E6-AP and these phosphatases. As stated earlier, AS mice have increased levels of the E6-AP target protein, p53. p53 has an associated binding protein, p53BP2, which is also a binding partner for PP1 and inhibits its phosphatase activity (Phelps and Ledoux, 2005). Thus, it is possible, that the misregulation of p53 in these AS mice may result in downstream effects involving alterations of the interaction between p53BP2 and PP1, resulting in the dysfunction of the normal CaMKII signaling that is required for normal learning and mental development.
Neurofibromatosis type 1 and associated learning disabilities Neurofibromatosis type 1 (NF1) is an autosomal dominant genetic disease, and, at a prevalence of 1 in 3500–4000, represents a fairly common disorder. NF1 is also called von Recklinghausen’s disease and results from a mutation on chromosome 17 in the gene NF1, which encodes the protein neurofibromin. Approximately half of the mutations are inherited, while half represent spontaneous, new mutations (Kayl and Moore, 2000). Evidence suggests that the NF1 gene product functions as a tumor suppressor, and disruptions
in the gene, through translocations, deletions, point mutations, and de novo insertions, lead to an inactivation of the protein (Andersen et al., 1993). As such, patients carrying the mutation in the neurofibromin gene develop many benign tumors (often optic gliomas and Lisch nodules). Other hallmarks of the disease include cafe´-au-lait spots — large, birthmark-like dark patches on the skin, and freckling in the axillary or inguinal region (Kayl and Moore, 2000). An additional feature in a subset of patients with NF1 is associated learning disabilities. Only between 4% and 8% of NF1 patients actually classify as mentally retarded (defined as an IQ below 70), which is approximately twice that found in the general population (Ozonoff, 1999). However, a larger proportion (30–45%) of individuals with NF1 are classified as having learning disabilities (defined as scoring significantly below on academic achievement what one would predict based on intelligence scores) and slightly lower than average IQs (Ozonoff, 1999; Kayl and Moore, 2000). As the physical features associated with NF1 are variable in their manifestations, so are the cognitive impairments. The most commonly reported learning deficits in patients with NF1 are related to visual-spatial perception. These defects are such that one study found that performance on a combination of visual-spatial/motor ability tests was able to accurately identify about 90% of patients with NF1 (Schrimsher et al., 2003). In addition to visual-spatial impairments, NF1 patients often exhibit problems with both fine and gross motor skills, reading disabilities, and executive dysfunction, with some patients displaying features of autism and ADHD (Ozonoff, 1999; Kayl and Moore, 2000). How does the absence of the NF1 gene product contribute specifically to the learning impairments observed in patients with NF1? The protein product, neurofibromin, has several known functions, including Ras-GTPase activity, modulation of adenylyl cyclase activity, and interaction with microtubules (Das, 2003), and, in theory, any of these functions might contribute to a learning phenotype. Alcino Silva and his colleagues investigated the role of neurofibromin within the central nervous
32
system in a series of elegant and sophisticated experiments using genetic techniques. First, a mouse model of NF1 was developed by knocking out NF1. Mice that were homozygous for the mutation of NF1 died in utero at approximately embryonic day 13 (Brannan et al., 1994; Tong et al., 2002), but animals heterozygotic for the mutation were viable and normal in terms of many baseline behaviors (Silva et al., 1997). These NF1+/ mice were found to have impairments in learning as well as a predisposition to develop tumors. Thus, it was clear that normal NF1 activity was necessary for development and for the suppression of tumors, as well as for normal learning. Specifically, the NF1+/ mice showed deficits in spatial learning of the Morris water maze, a phenotype that seemed to recreate the deficits seen in human patients with NF1. Additional training of the mice led to normal performance, another aspect that seemed to mimic the human phenotype (Silva et al., 1997). To tease out how exactly the absence of neurofibromin was contributing to the learning impairments, Silva and colleagues took advantage of the existence of two splice variants of the NF1 gene. It had been previously reported that the NF1 gene exhibited an alternative splice site within the GAP-related domain of the gene that resulted in the insertion of exon 23a, an additional 21 amino acids (Andersen et al., 1993). The GAP-domain refers to a region of the protein with GTP-ase activating protein function, meaning that the protein will accelerate the conversion of GTP-bound Ras to the GDP-bound state. Both splice variants, type I (exclusion of exon 23a) and type II (inclusion of exon 23a), displayed GAP activity (Andersen et al., 1993). Although type II had slightly weaker GAP activity, type II actually displayed higher affinity for GTP-Ras (Zhu and Parada, 2001). A mouse strain specifically lacking exon 23a was developed to probe the roles of the different splice variants. Interestingly, mice with homozygous mutations of NF123a were viable and lacked a disposition to tumor formation, thus ruling out a role for exon 23a in development and tumor suppression. In fact, these mice specifically displayed learning impairments in visual-spatial tasks, which implicated the GAP activity of exon 23a as critical for normal memory formation (Costa et al., 2001).
NF123a / mice were found to perform poorly in the Morris water maze task, a spatial behavior task, and in contextual discrimination, two tasks that rely on normal hippocampal function (Costa et al., 2001). Again, as with the NF1+/ mice, extended training in both the Morris water maze and the contextual discrimination task was able to reverse the deficits seen (Costa et al., 2001). How does this genetic defect translate into signaling defects and a resultant behavioral phenotype? Because of its GAP activity, the presence of neurofibromin would limit the amount of GTPbound (and thus, activated) Ras. Absence of GAP-activity, as in NF1, would upset the balance between GTP-bound and GDP-bound Ras, enhancing the amount of active Ras. This hyperactivation of Ras would lead to overactive pathways downstream of Ras, including the ERK/MAPK pathway. As we have discussed, this pathway is critical for normal learning and memory functions as a result of the downstream effectors that impinge on transcription, translation, and cellular excitability. In order to test the hypothesis that the upregulation of activated Ras was responsible for the learning deficits seen in the NF1+/ and NF123a / mice (and possibly to explain the similar deficits in human NF1 patients), the NF1+/ mice were crossed with a strain of mice heterozygous for a mutation in the K-Ras gene (Costa et al., 2002). K-Ras+/ mice also displayed deficits in the Morris water maze, just like the NF1+/ mice. However, when the NF1+/ mice were crossed with the K-Ras+/ mice, the resultant strain was rescued with respect to the spatial learning deficits, that is, the double mutant mice performed similarly to wild type mice (Costa et al., 2002). Presumably, the decreased GAP activity contributed by the NF1+/ mice, coupled with the decreased levels of Ras contributed by the K-Ras+/ mice, served to recreate the balance of Ras-signaling. To show that this rescue was specific to the act of learning rather than a developmental effect, a farnesyl-transferase inhibitor was injected acutely to NF1+/ mice before training. Farnesylation is a post-translational modification necessary for normal Ras function, so inhibition of farnesylation
33
would decrease Ras activity. Blocking farneslyation was also successful at reversing the spatial learning deficits in the NF1+/ mice (Costa et al., 2002). More recent work has focused on a possible therapeutic option for patients with neurofibromatosis. The commonly prescribed cholesterol-synthesis inhibitor lovastatin has the additional effect of decreasing Ras activity. Lovastatin treatment in NF1+/ mice resulted in normalization of ERK/MAPK levels, and a rescue of the LTP and spatial learning deficits (Li et al., 2005). As lovastatin is an approved and welltolerated drug, it represents a candidate drug treatment for the cognitive impairments found in human patients with Neurofibromatosis 1. Taken together, this set of studies illustrates how a specific component of a disease, namely the learning deficits observed in patients with NF1, was mapped to a specific region of the mutated gene. Moreover, the learning deficits were then explained by the molecular events that followed as a consequence of reduced GAP-activity contributed by exon 23a — hyperactivation of Ras and downstream hyperactivation of the ERK cascade — all leading to impaired synaptic plasticity and behavioral learning deficits. Restoring the balance of Ras activity pharmacologically and genetically rescued the observed deficits and has suggested a means for treatment in clinical populations.
Coffin– Lowry syndrome Coffin–Lowry syndrome (CLS), first reported by Coffin et al. (1966) and Lowry et al. (1971), is a rare X-linked disorder (estimated to affect 1 in 40,000–50,000 births) in which the patients suffer from mental retardation as well as characteristic facial and skeletal abnormalities. Although this disorder affects both males and females in equal numbers, typical of X-linked disorders, the symptoms in females are commonly milder than those in males. The physical symptoms are variable, but generally include short stature (average 4 ft 8 in. in males), skeletal abnormalities, broad hands with short tapering fingers, and facial features including a protruding forehead, broad nose, prominent ears, a large mouth with ‘‘turned out’’ lips and dental
anomalies. Behavioral features are also quite variable and include mental retardation (generally severe in males only), language delays, perseverating behaviors and sometimes anxiety. Many patients also display a cataplexy or a ‘‘drop episode’’ in which they suddenly loose muscle tone in response to an unexpected tactile or loud auditory stimulus. Most of these symptoms are readily apparent by the age of 2 years. However, due to the large variability and the nonspecific nature of the symptoms, clinical diagnosis may not distinguish CLS from other disorders. Several early genetic studies implicated X-linkage at Xp22.2-p22.1 in CLS (Biancalana et al., 1992; Hanauer et al., 1988; Partington et al., 1988). Finally, in 1996, mutations in the gene encoding p90 ribosomal S6 kinase-2 (RPS6KA3; RSK-2) were associated with the disease (Trivier et al., 1996). It is now known that a large variety of mutations in the RSK-2 gene are associated with CLS including nonsense mutations, missense mutations, frameshift mutations, splicing errors, and short deletion or insertion events (Jacquot et al., 1998; Delaunoy et al., 2001). The mutations have been found to be distributed throughout the gene and there is no correlation between the specific location of these mutations and the severity of the symptoms. Most of these mutations are predicted to result in loss of function of the protein. Although CLS is inherited as an X-linked semi-dominant disorder, many patients have no prior history of the disorder in their families. These cases are thought to be due to spontaneous mutations in the RSK-2 gene. However, there are some patients with CLS symptoms that do not appear to have any mutations of RSK-2. The cause of the disorder in these patients is currently unknown. Furthermore, four isoforms of the p90 (RSK) family are known to arise from distinct genes but little is known about their individual functions, and only RSK-2 has been associated with CLS. It is important to note, however, that mutation of RSK4 has also been documented in association with X-linked mental retardation (Yntema et al., 1999). RSK-2 is activated when it is phosphorylated (see the earlier discussion of pathways) and, in turn, phosphorylates a variety of other proteins. Only a few downstream substrates of RSK-2 have yet been identified, including CREB (Xing et al., 1996), activating transcription factor 4 (ATF4)
34
(Daw et al., 2004), the tumor suppressor p53 protein (Collins et al., 2006), and histone H3 (SassoneCorsi et al., 1999). As expected from the signaling pathways discussed earlier, cells lines established from patients with CLS show attenuated phosphorylation of CREB in response to EGF or PKC stimulation (De Cesare et al., 1998; Harum et al., 2001), and a concurrent decrease in c-fos expression (De Cesare et al., 1998). Furthermore, data from Harum et al (2001) suggest that the ability of RSK-2 to phosphorylate CREB in CLS patients is correlated with their cognitive ability as measured by a standardize test of intelligence. In contrast, data from mutant mice lacking a functional RSK2 gene (Dufresne et al., 2001) do not show similar changes in CREB phosphorylation or c-fos levels. These mice showed similar characteristics as patients with CSL: they were shorter, weighed less, and showed impairments in both coordination and learning (although interpretation of their data is problematic as the mice have poor motor coordination which may have interfered with their measure of learning). However there was no difference between the wildtype and mutant mice in either insulin- or exercise-induced c-fos mRNA expression or CREB phosphorylation levels (neither group showed any increase in CREB phosphorylation). Differences in tissue (fibroblasts vs. skeletal muscle) as well as methods used to activate RSK-2 (EGF, PKC activator, insulin, or exercise) may well explain the different results observed in these studies. Although we currently have a limited understanding of how deficits in RSK-2 signaling produces the symptoms of CLS, further investigations with the RSK-2 knockout mice will hopefully yield more insight into the changes and underlying mechanism that lead to CLS. In addition, changes in chromatin remodeling may contribute to the deficits observed in CLS. A study by Sassone-Corsi et al. (1999) demonstrated that RSK-2 can directly phosphorylate histone H3 in response to EGF stimulation, and that this phosphorylation was lacking in fibroblasts derived from a patient with CLS. However, these data were not replicated by another group (Soloaga et al., 2003). The story is likely more complicated as RSK-2 has recently been shown to activate p53, which appears to be essential for RSK-2
phosphorylation of histone H3 (Collins et al., 2006). Furthermore, p53 may also interact with CREB and alter acetylation of histone H3 through recruitment of CREB-binding protein (CBP). Mitogenic stimulation can induce both phosphorylation and acetylation of H3 through an interaction between RSK-2 and CBP (Merienne et al., 2001). These changes in chromatin remodeling ultimately affect the ability of transcriptional machinery to interact with DNA. As the dynamic associations between RSK-2, CREB, CBP, and p53 may be altered in CLS, gene expression may also be affected, resulting in the observed cognitive impairments. Interestingly, disruption of p53 or CREB/CBP have been implicated in other mental retardation disorders as well (see the discussion on Angelman’s syndrome and RTS). Moreover, the direct mechanism that links the loss of RSK-2 function with the physical characteristics of CLS has been very difficult to demonstrate. One study suggested that activation of the transcription factor ATF4, a substrate of RSK-2, is critical for osteoblast differentiation as well as regulation of proteins essential for bone formation (Daw et al., 2004). Thus, mutations of RSK-2 which reduce ATF4 activity may directly lead to many of the skeletal abnormalities observed in CLS. A direct role of RSK-2 in regulating similar pathways in neurons may be found to underlie the mental retardation and other symptoms of CLS. Until more is understood about the direct role of RSK-2 in regulating development, there is no effective cure or treatment for patients with CLS. Treatments generally are supportive and include physical and speech therapy as well as educational services.
Rubinstein– Taybi syndrome Rubinstein–Taybi syndrome (RTS) was first described by Rubinstein and Taybi (1963). RTS occurs 1 in every 125,000 live births, and accounts for 1 in 300 patients with mental retardation. RTS is an inherited, autosomal dominant disease that has been mapped to chromosome 16p13.3, which contains the gene for the transcriptional coactivator CREB-binding protein (CBP) (Petrij et al., 1995). Several studies have shown that RTS
35
patients have a variety of mutations in CBP, including point mutations and 5’- or 3’-deletions (Blough et al., 2000; Coupry et al., 2004). It should be noted that patients are typically heterozygous for a mutation in CBP, suggesting that RTS is due to either haploinsufficiency or a dominant negative effect (Petrij et al., 1995). CBP is a transcriptional coactivator that facilitates gene transcription coupled to activation of the transcription factor CREB (cAMP response element binding protein). CBP facilitates transcription via endogenous histone acetyltransferase activity (Kalkhoven, 2004), and, in the context of CREB-mediated transcription, is thought to acetylate histones associated with cAMP response elements (CRE) (Ogryzko et al., 1996). Acetylation of histones alters chromatin structure and can ultimately enhance accessibility of chromatin to transcription factors (Norton et al., 1989; Vettese-Dadey et al., 1996). Indeed, an increase in CREB-mediated transcription is associated with an increase in acetylation of histones located near CREs (Ogryzko et al., 1996). To better understand the molecular and cellular ramifications of a loss of CBP function on cognition, several mouse models have been generated that have loss of CBP function. Mice heterozygous for a dominant negative form of truncated CBP (CBP+/ DN ) (Oike et al., 1999) have significant deficits in step-through passive avoidance, novel object recognition, and cued fear conditioning (Oike et al., 1999; Bourtchouladze et al., 2003). Unfortunately, CREB and CBP play major roles in development (Andrisani, 1999; Mannervik et al., 1999). Not surprisingly, CBP+/ DN mice have several developmental abnormalities making straightforward interpretation of memory formation in these animals difficult (Oike et al., 1999). To avoid the effects of CBP on development, three laboratories independently developed CBPdeficient mice that lack the problems of the classic CBP+/ DN animals. The genetic models of CBP derangement include coupling a CBPDN allele to an inducible promoter (CBP+/ I DN), knocking out one allele of CBP (CBP+/ ), and expression of a transgene where the CaMKIIa promoter was coupled to a truncated form of CBP (CBPa1), restricting expression of the dominant negative form of CBP
to forebrain neurons (Alarcon et al., 2004; Korzus et al., 2004; Wood et al., 2005). All of the newly developed CBP mouse models exhibit significant deficits in the spatial water maze task and contextual fear conditioning (Alarcon et al., 2004; Korzus et al., 2004;Wood et al., 2005). Both the +/ mice also exhibited deficits CBP+/ I DN and CBP in novel object recognition (Alarcon et al., 2004; Korzus et al., 2004). Considered together, these results suggest that impairment of CBP function has serious consequences for formation of longterm memory in the hippocampus, and further support the hypothesis that regulation of the epigenome is an important component to memory formation in mammals. Histone acetylation is an enzymatic process, governed by histone acetyltransferases (HATs) and histone deacetylases (HDACs). CBP displays HAT activity, and thus RTS may be due to a deficiency in CBP-mediated HAT activity. This suggests the intriguing hypothesis that inhibitors of HDAC activity, which would increase overall levels of histone acetylation, may ameliorate some or all the effects of RTS on cognitive ability. Indeed, several studies have suggested that HDAC inhibition might be a viable treatment for such diseases as Huntington’s disease, Alzheimer’s disease, Schizophrenia, and RTS (Kimberly et al., 2001; Steffan et al., 2001; Ferrante et al., 2003; Hockly et al., 2003; Kim et al., 2004; Numachi et al., 2004; Rouaux et al., 2004; Von Rotz et al., 2004; Grayson et al., 2005; Tremolizzo et al., 2005). Inhibition of HDAC activity with various compounds restored normal hippocampus-dependent long-term +/ memory formation in CBP+/ I DN and CBP mice, suggesting that HDAC inhibitors represent a viable treatment for patients with RTS (Alarcon et al., 2004; Korzus et al., 2004). These results further support the idea that RTS is a disease of the epigenome, as drugs that specifically affect the state of the epigenome can restore normal cognitive ability in CBP-deficient animals.
Rett syndrome First described in 1966 by Austrian pediatrician Andreas Rett, RTT is a childhood neurodevelopmental
36
disorder that affects primarily females and results in profound mental retardation. Indeed, RTT is estimated to be the second leading cause of mental retardation in females (Ellaway and Christodoulou, 2001). The onset and severity of the disorder is variable, with the onset of symptoms typically occurring between 6 and 18 months of age. After a period of normal development, the progression of the neurodevelopmental disorder entails the loss of previously acquired motor, social, and cognitive skills. Prominent symptoms of RTT include microcephaly, aphasia, ataxia, the loss of purposeful hand movements, and autistic behavior (Hagberg et al., 1983; Ellaway and Christodoulou, 2001). The cause of RTT is attributed to mutations in the X chromosome-linked gene encoding for methyl CpG binding protein 2 (MeCP2) (Amir et al., 1999). MeCP2 is a transcriptional repressor, and functionally, MeCP2 links synaptic activity to nuclear gene transcription, a pivotal process subserving learning and memory. DNA methyl transferases (DNMTs) catalyze the methylation of cytosines at the 5-position of the pyrimidine ring. Once methylated, 5mCpG is bound by MeCP2, which then recruits a complex of proteins including HDACs and transcriptional corepressors such as Sin3A (Jones et al., 1998; Roopra et al., 2000). Ultimately, the histones associated with the 5mCpG become hypoacetylated, promoting tight association between DNA and histones. Overall, this results in a transcriptionally repressive heterochromatin complex which silences gene transcription. As discussed above, RTT is a progressive disease that does not result in symptoms until early childhood. It has recently been suggested that MeCP2 does not play a critical role in the early stages of brain development itself, meaning that early postnatal development is normal and that the neural dysfunction manifest in RTT develops after the prolonged deficiency of MeCP2 (Luikenhuis et al., 2004). Indeed, MeCP2 is predominately expressed in mature neurons, and its expression peaks postnatally, preceding synaptogenesis (Cohen et al., 2003). Another link between MeCP2 and neural function is found in the observation that mutations in MeCP2 produce deficits in synaptic proliferation, neuronal excitability, and
cortical activity (Harum et al., 2001; Dani et al., 2005). To date, we have a limited understanding of how deficits in MeCP2 produce cognitive impairments. Deficiency of MeCP2 in mice yields many of the same Rett phenotypes found in human clinical populations, including motor impairment, abnormal social interaction, and stereotypic limb movements (Chen et al., 2001; Nguyen et al., 2002; Shahbazian et al., 2002). More recently, in MeCP2308/y mice, another RTT rodent model in which the last one-third of MeCP2 was removed, significant deficits in hippocampus-dependent long-term memory formation and in the induction of synaptic plasticity in both the sensory-motor cortex and hippocampus were demonstrated (Moretti et al., 2006). Conditional MeCP2 knockout mice have also been generated to investigate specifically whether postnatal loss of the protein is responsible for the behavioral facets of RTT (Gemelli et al., 2005). Results demonstrate that selective postnatal loss of MeCP2 also produces many of the behavioral deficits associated with RTT, such as impaired motor coordination and abnormal social interactions. And though these mice exhibit normal context-dependent fear conditioning, they do display impaired cue-dependent fear conditioning. Furthermore, overexpression of MeCP2 can rescue the RTT phenotype (Luikenhuis et al., 2004) and can enhance both long-term memory formation and hippocampal LTP induction (Collins et al., 2004). Specifically, these results relating to MeCP2 function support the hypothesis that the loss of functional MeCP2 protein contributes to the cognitive deficits associated with RTT. Overall, such studies demonstrate that alterations in transcription states of the genome within the CNS have consequences on plasticity and memory formation in the adult animal.
Conclusion In summary, as our basic understanding of molecular and epigenetic mechanisms increases, so does our knowledge about the pathogenesis of cognitive impairments. While the learning
37
disabilities found in some disorders, such as NF1 or CLS, can be traced to the mutation of a gene whose protein function directly affects a signaling pathway integral for memory formation, other disorders are increasingly complex. These genetic mutations can lead to more widespread effects, such as interference with the overall transcription state of the genome (RTS, RTT) or, as in AS and possibly CLS, dysregulation of a number of proteins that eventually indirectly affects mediators of synaptic plasticity and memory formation. References Abeliovich, A., Chen, C., Goda, Y., Silva, A.J., Stevens, C.F. and Tonegawa, S. (1993) Modified hippocampal long-term potentiation in PKC gamma-mutant mice. Cell, 75: 1253–1262. Adams, J.P., Anderson, A.E., Varga, A.W., Dineley, K.T., Cook, R.G., Pfaffinger, P.J. and Sweatt, J.D. (2000a) The Atype potassium channel Kv4.2 is a substrate for the mitogenactivated protein kinase ERK. J. Neurochem., 75: 2277–2287. Adams, J.P., Roberson, E.D., English, J.D., Selcher, J.C. and Sweatt, J.D. (2000b) MAPK regulation of gene expression in the central nervous system. Acta Neurobiol. Exp. (Warsz), 60: 377–394. Ahi, J., Radulovic, J. and Spiess, J. (2004) The role of hippocampal signaling cascades in consolidation of fear memory. Behav. Brain Res., 149: 17–31. Alarcon, J.M., Malleret, G., Touzani, K., Vronskaya, S., Ishii, S., Kandel, E.R. and Barco, A. (2004) Chromatin acetylation, memory, and LTP are impaired in CBP+/- mice: a model for the cognitive deficit in Rubinstein–Taybi syndrome and its amelioration. Neuron, 42: 947–959. Amir, R.E., Van Den Veyver, I.B., Wan, M., Tran, C.Q., Francke, U. and Zoghbi, H.Y. (1999) Rett syndrome is caused by mutations in X-linked MECP2, encoding methylCpG-binding protein 2. Nat. Genet., 23: 185–188. Andersen, L.B., Ballester, R., Marchuk, D.A., Chang, E., Gutmann, D.H., Saulino, A.M., Camonis, J., Wigler, M. and Collins, F.S. (1993) A conserved alternative splice in the von Recklinghausen neurofibromatosis (NF1) gene produces two neurofibromin isoforms, both of which have GTPase-activating protein activity. Mol. Cell. Biol., 13: 487–495. Andrisani, O.M. (1999) CREB-mediated transcriptional control. Crit. Rev. Eukaryot. Gene Expr., 9: 19–32. Angenstein, F. and Staak, S. (1997) Receptor-mediated activation of protein kinase C in hippocampal long-term potentiation: facts, problems and implications. Prog. Neuropsychopharmacol. Biol. Psychiatry, 21: 427–454. Antar, L.N., Afroz, R., Dictenberg, J.B., Carroll, R.C. and Bassell, G.J. (2004) Metabotropic glutamate receptor activation regulates fragile mental retardation protein and
FMR1 mRNA localization differentially in dendrites and at synapses. J. Neurosci., 24: 2648–2655. Atkins, C.M., Selcher, J.C., Petraitis, J.J., Trzaskos, J.M. and Sweatt, J.D. (1998) The MAPK cascade is required for mammalian associative learning. Nat. Neurosci., 1: 602–609. Biancalana, V., Briard, M.L., David, A., Gilgenkrantz, S., Kaplan, J., Mathieu, M., Piussan, C., Poncin, J., Schinzel, A., Oudet, C. and Hanauer, A. (1992) Confirmation and refinement of the genetic localization of the Coffin–Lowry syndrome locus in Xp22.1-p22.2. Am. J. Hum. Genet., 50: 981–987. Blough, R.I., Petrij, F., Dauwerse, J.G., Milatovich-Cherry, A., Weiss, L., Saal, H.M. and Rubinstein, J.H. (2000) Variation in microdeletions of the cyclic AMP-responsive elementbinding protein gene at chromosome band 16p13.3 in the Rubinstein–Taybi syndrome. Am. J. Med. Genet., 90: 29–34. Blum, S., Moore, A.N., Adams, F. and Dash, P.K. (1999) A mitogen-activated protein kinase cascade in the CA1/CA2 subfield of the dorsal hippocampus is essential for long-term spatial memory. J. Neurosci., 19: 3535–3544. Bourtchouladze, R., Lidge, R., Catapano, R., Stanley, J., Gossweiler, S., Romashko, D., Scott, R. and Tully, T. (2003) A mouse model of Rubinstein–Taybi syndrome: defective longterm memory is ameliorated by inhibitors of phosphodiesterase 4. Proc. Natl. Acad. Sci. USA, 100: 10518–10522. Brannan, C.I., Perkins, A.S., Vogel, K.S., Ratner, N., Nordlund, M.L., Reid, S.W., Buchberg, A.M., Jenkins, N.A., Parada, L.F. and Copeland, N.G. (1994) Targeted disruption of the neurofibromatosis type-1 gene leads to developmental abnormalities in heart and various neural crest-derived tissues. Genes Dev., 8: 1019–1029. Chen, R.Z., Akbarian, S., Tudor, M. and Jaenisch, R. (2001) Deficiency of methyl-CpG binding protein-2 in CNS neurons results in a Rett-like phenotype in mice. Nat. Genet., 27: 327–331. Clayton-Smith, J. and Laan, L. (2003) Angelman syndrome: a review of the clinical and genetic aspects. J. Med. Genet., 40: 87–95. Coffin, G.S., Siris, E. and Wegienka, L.C. (1966) Mental retardation with osteocartilaginous anomalies. Am. J. Dis. Child., 112: 205–213. Cohen, D.R., Matarazzo, V., Palmer, A.M., Tu, Y., Jeon, O.H., Pevsner, J. and Ronnett, G.V. (2003) Expression of MeCP2 in olfactory receptor neurons is developmentally regulated and occurs before synaptogenesis. Mol. Cell. Neurosci., 22: 417–429. Colbran, R.J. and Soderling, T.R. (1990) Calcium/calmodulinindependent autophosphorylation sites of calcium/calmodulin-dependent protein kinase II. Studies on the effect of phosphorylation of threonine 305/306 and serine 314 on calmodulin binding using synthetic peptides. J. Biol. Chem., 265: 11213–11219. Collins, A.L., Levenson, J.M., Vilaythong, A.P., Richman, R., Armstrong, D.L., Noebels, J.L., Sweatt, J.D. and Zoghbi, H.Y. (2004) Mild overexpression of MeCP2 causes a progressive neurological disorder in mice. Hum. Mol. Genet., 13: 2679–2689.
38 Collins, M.O., Husi, H., Yu, L., Brandon, J.M., Anderson, C.N., Blackstock, W.P., Choudhary, J.S. and Grant, S.G. (2006) Molecular characterization and comparison of the components and multiprotein complexes in the postsynaptic proteome. J. Neurochem., 97(suppl. 1): 16–23. Costa, R.M., Federov, N.B., Kogan, J.H., Murphy, G.G., Stern, J., Ohno, M., Kucherlapati, R., Jacks, T. and Silva, A.J. (2002) Mechanism for the learning deficits in a mouse model of neurofibromatosis type 1. Nature, 415: 526–530. Costa, R.M., Yang, T., Huynh, D.P., Pulst, S.M., Viskochil, D.H., Silva, A.J. and Brannan, C.I. (2001) Learning deficits, but normal development and tumor predisposition, in mice lacking exon 23a of Nf1. Nat. Genet., 27: 399–405. Coupry, I., Monnet, L., Attia, A.A., Taine, L., Lacombe, D. and Arveiler, B. (2004) Analysis of CBP (CREBBP) gene deletions in Rubinstein–Taybi syndrome patients using realtime quantitative PCR. Hum. Mutat., 23: 278–284. Dani, V.S., Chang, Q., Maffei, A., Turrigiano, G.G., Jaenisch, R. and Nelson, S.B. (2005) Reduced cortical activity due to a shift in the balance between excitation and inhibition in a mouse model of Rett syndrome. Proc. Natl. Acad. Sci. USA, 102: 12560–12565. Das, U.N. (2003) Can memory be improved? A discussion on the role of ras, GABA, acetylcholine, NO, insulin, TNF-alpha, and long-chain polyunsaturated fatty acids in memory formation and consolidation. Brain Dev., 25: 251–261. Daw, N., Rao, Y., Wang, X.F., Fischer, Q. and Yang, Y. (2004) LTP and LTD vary with layer in rodent visual cortex. Vision Res., 44: 3377–3380. De Cesare, D., Jacquot, S., Hanauer, A. and Sassone-Corsi, P. (1998) RSK-2 activity is necessary for epidermal growth factor-induced phosphorylation of CREB protein and transcription of c-fos gene. Proc. Natl. Acad. Sci. USA, 95: 12202–12207. Delaunoy, J., Abidi, F., Zeniou, M., Jacquot, S., Merienne, K., Pannetier, S., Schmitt, M., Schwartz, C. and Hanauer, A. (2001) Mutations in the X-linked RSK2 gene (RPS6KA3) in patients with Coffin–Lowry syndrome. Hum. Mutat., 17: 103–116. Dhillon, A.S. and Kolch, W. (2002) Untying the regulation of the Raf-1 kinase. Arch. Biochem. Biophys., 404: 3–9. Dufresne, S.D., Bjorbaek, C., El-Haschimi, K., Zhao, Y., Aschenbach, W.G., Moller, D.E. and Goodyear, L.J. (2001) Altered extracellular signal-regulated kinase signaling and glycogen metabolism in skeletal muscle from p90 ribosomal S6 kinase 2 knockout mice. Mol. Cell. Biol., 21: 81–87. Elgersma, Y., Fedorov, N.B., Ikonen, S., Choi, E.S., Elgersma, M., Carvalho, O.M., Giese, K.P. and Silva, A.J. (2002) Inhibitory autophosphorylation of CaMKII controls PSD association, plasticity, and learning. Neuron, 36: 493–505. Ellaway, C. and Christodoulou, J. (2001) Rett syndrome: clinical characteristics and recent genetic advances. Disabil. Rehabil., 23: 98–106. English, J.D. and Sweatt, J.D. (1997) A requirement for the mitogen-activated protein kinase cascade in hippocampal long term potentiation. J. Biol. Chem., 272: 19103–19106.
Ferrante, R.J., Kubilus, J.K., Lee, J., Ryu, H., Beesen, A., Zucker, B., Smith, K., Kowall, N.W., Ratan, R.R., LuthiCarter, R. and Hersch, S.M. (2003) Histone deacetylase inhibition by sodium butyrate chemotherapy ameliorates the neurodegenerative phenotype in Huntington’s disease mice. J. Neurosci., 23: 9418–9427. Gemelli, T., Berton, O., Nelson, E.D., Perrotti, L.I., Jaenisch, R. and Monteggia, L.M. (2005) Postnatal loss of methylCpG binding protein 2 in the forebrain is sufficient to mediate behavioral aspects of Rett syndrome in mice. Biol. Psychiatry., 59: 468–476. Gerdjikov, T.V., Ross, G.M. and Beninger, R.J. (2004) Place preference induced by nucleus accumbens amphetamine is impaired by antagonists of ERK or p38 MAP kinases in rats. Behav. Neurosci., 118: 740–750. Gerendasy, D.D. and Sutcliffe, J.G. (1997) RC3/neurogranin, a postsynaptic calpacitin for setting the response threshold to calcium influxes. Mol. Neurobiol., 15: 131–163. Goldin, M. and Segal, M. (2003) Protein kinase C and ERK involvement in dendritic spine plasticity in cultured rodent hippocampal neurons. Eur. J. Neurosci., 17: 2529–2539. Grayson, D.R., Jia, X., Chen, Y., Sharma, R.P., Mitchell, C.P., Guidotti, A. and Costa, E. (2005) Reelin promoter hypermethylation in schizophrenia. Proc. Natl. Acad. Sci. USA, 102: 9341–9346. Hagberg, B., Aicardi, J., Dias, K. and Ramos, O. (1983) A progressive syndrome of autism, dementia, ataxia, and loss of purposeful hand use in girls: Rett’s syndrome: report of 35 cases. Ann. Neurol., 14: 471–479. Hanauer, A., Alembik, Y., Gilgenkrantz, S., Mujica, P., Nivelon-Chevallier, A., Pembrey, M.E., Young, I.D. and Mandel, J.L. (1988) Probable localisation of the Coffin–Lowry locus in Xp22.2-p22.1 by multipoint linkage analysis. Am. J. Med. Genet., 30: 523–530. Harum, K.H., Alemi, L. and Johnston, M.V. (2001) Cognitive impairment in Coffin–Lowry syndrome correlates with reduced RSK2 activation. Neurology, 56: 207–214. Hockly, E., Richon, V.M., Woodman, B., Smith, D.L., Zhou, X., Rosa, E., Sathasivam, K., Ghazi-Noori, S., Mahal, A., Lowden, P.A., Steffan, J.S., Marsh, J.L., Thompson, L.M., Lewis, C.M., Marks, P.A. and Bates, G.P. (2003) Suberoylanilide hydroxamic acid, a histone deacetylase inhibitor, ameliorates motor deficits in a mouse model of Huntington’s disease. Proc. Natl. Acad. Sci. USA, 100: 2041–2046. Hvalby, O., Hemmings Jr., H.C., Paulsen, O., Czernik, A.J., Nairn, A.C., Godfraind, J.M., Jensen, V., Raastad, M., Storm, J.F., Andersen, P. and Greengard, P. (1994) Specificity of protein kinase inhibitor peptides and induction of long-term potentiation. Proc. Natl. Acad. Sci. USA, 91: 4761–4765. Impey, S., Obrietan, K., Wong, S.T., Poser, S., Yano, S., Wayman, G., Deloulme, J.C., Chan, G. and Storm, D.R. (1998) Cross talk between ERK and PKA is required for Ca2+ stimulation of CREB-dependent transcription and ERK nuclear translocation. Neuron, 21: 869–883. Jacquot, S., Merienne, K., De Cesare, D., Pannetier, S., Mandel, J.L., Sassone-Corsi, P. and Hanauer, A. (1998) Mutation
39 analysis of the RSK2 gene in Coffin–Lowry patients: extensive allelic heterogeneity and a high rate of de novo mutations. Am. J. Hum. Genet., 63: 1631–1640. Jiang, Y.H., Armstrong, D., Albrecht, U., Atkins, C.M., Noebels, J.L., Eichele, G., Sweatt, J.D. and Beaudet, A.L. (1998) Mutation of the Angelman ubiquitin ligase in mice causes increased cytoplasmic p53 and deficits of contextual learning and long-term potentiation. Neuron, 21: 799–811. Jones, P.L., Veenstra, G.J., Wade, P.A., Vermaak, D., Kass, S.U., Landsberger, N., Strouboulis, J. and Wolffe, A.P. (1998) Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat. Genet., 19: 187–191. Kalkhoven, E. (2004) CBP and p300: HATs for different occasions. Biochem. Pharmacol., 68: 1145–1155. Kandel, E.R., Schwartz, J.H. and Jessel, T.M. (2000) Principles of Neural Science. The McGraw-Hill Companies, Inc, New York. Kayl, A.E. and Moore III, B.D. (2000) Behavioral phenotype of neurofibromatosis, type 1. Ment. Retard. Dev. Disabil. Res. Rev., 6: 117–124. Kelleher III, R.J., Govindarajan, A., Jung, H.Y., Kang, H. and Tonegawa, S. (2004) Translational control by MAPK signaling in long-term synaptic plasticity and memory. Cell, 116: 467–479. Kim, H.S., Kim, E.M., Kim, N.J., Chang, K.A., Choi, Y., Ahn, K.W., Lee, J.H., Kim, S., Park, C.H. and Suh, Y.H. (2004) Inhibition of histone deacetylation enhances the neurotoxicity induced by the C-terminal fragments of amyloid precursor protein. J. Neurosci. Res., 75: 117–124. Kimberly, W.T., Zheng, J.B., Guenette, S.Y. and Selkoe, D.J. (2001) The intracellular domain of the beta-amyloid precursor protein is stabilized by Fe65 and translocates to the nucleus in a notch-like manner. J. Biol. Chem., 276: 40288–40292. Korzus, E., Rosenfeld, M.G. and Mayford, M. (2004) CBP histone acetyltransferase activity is a critical component of memory consolidation. Neuron, 42: 961–972. Krapivinsky, G., Krapivinsky, L., Manasian, Y., Ivanov, A., Tyzio, R., Pellegrino, C., Ben-Ari, Y., Clapham, D.E. and Medina, I. (2003) The NMDA receptor is coupled to the ERK pathway by a direct interaction between NR2B and RasGRF1. Neuron, 40: 775–784. Levenson, J.M., O’riordan, K.J., Brown, K.D., Trinh, M.A., Molfese, D.L. and Sweatt, J.D. (2004) Regulation of histone acetylation during memory formation in the hippocampus. J. Biol. Chem., 279: 40545–40559. Li, W., Cui, Y., Kushner, S.A., Brown, R.A., Jentsch, J.D., Frankland, P.W., Cannon, T.D. and Silva, A.J. (2005) The HMG-CoA reductase inhibitor lovastatin reverses the learning and attention deficits in a mouse model of neurofibromatosis type 1. Curr. Biol., 15: 1961–1967. Lisman, J., Schulman, H. and Cline, H. (2002) The molecular basis of CaMKII function in synaptic and behavioural memory. Nat. Rev. Neurosci., 3: 175–190. Lowry, B., Miller, J.R. and Fraser, F.C. (1971) A new dominant gene mental retardation syndrome. Association
with small stature, tapering fingers, characteristic facies, and possible hydrocephalus. Am. J. Dis. Child., 121: 496–500. Luikenhuis, S., Giacometti, E., Beard, C.F. and Jaenisch, R. (2004) Expression of MeCP2 in postmitotic neurons rescues Rett syndrome in mice. Proc. Natl. Acad. Sci. USA, 101: 6033–6038. Magenis, R.E., Brown, M.G., Lacy, D.A., Budden, S. and Lafranchi, S. (1987) Is Angelman syndrome an alternate result of del(15)(q11q13)? Am. J. Med. Genet., 28: 829–838. Malenka, R.C., Kauer, J.A., Perkel, D.J., Mauk, M.D., Kelly, P.T., Nicoll, R.A. and Waxham, M.N. (1989) An essential role for postsynaptic calmodulin and protein kinase activity in long-term potentiation. Nature, 340: 554–557. Malinow, R., Schulman, H. and Tsien, R.W. (1989) Inhibition of postsynaptic PKC or CaMKII blocks induction but not expression of LTP. Science, 245: 862–866. Mannervik, M., Nibu, Y., Zhang, H. and Levine, M. (1999) Transcriptional coregulators in development. Science, 284: 606–609. Mazzucchelli, C., Vantaggiato, C., Ciamei, A., Fasano, S., Pakhotin, P., Krezel, W., Welzl, H., Wolfer, D.P., Pages, G., Valverde, O., Marowsky, A., Porrazzo, A., Orban, P.C., Maldonado, R., Ehrengruber, M.U., Cestari, V., Lipp, H.P., Chapman, P.F., Pouyssegur, J. and Brambilla, R. (2002) Knockout of ERK1 MAP kinase enhances synaptic plasticity in the striatum and facilitates striatal-mediated learning and memory. Neuron, 34: 807–820. Meberg, P.J., Kinney, W.R., Valcourt, E.G. and Routtenberg, A. (1996) Gene expression of the transcription factor NFkappa B in hippocampus: regulation by synaptic activity. Brain Res. Mol. Brain Res., 38: 179–190. Merienne, K., Pannetier, S., Harel-Bellan, A. and Sassone-Corsi, P. (2001) Mitogen-regulated RSK2-CBP interaction controls their kinase and acetylase activities. Mol. Cell. Biol., 21: 7089–7096. Miller, C.A. and Marshall, J.F. (2005) Molecular substrates for retrieval and reconsolidation of cocaine-associated contextual memory. Neuron, 47: 873–884. Moretti, P., Levenson, J.M., Battaglia, F., Teague, R., Antalffy, B., Armstrong, D., Arancio, O., Sweatt, J.D. and Zoghbi, H. (2006) Learning and memory and synaptic plasticity are impaired in a mouse model of Rett Syndrome. J. Neurosci., 26: 319–327. Morozov, A., Muzzio, I.A., Bourtchouladze, R., Van-Strien, N., Lapidus, K., Yin, D., Winder, D.G., Adams, J.P., Sweatt, J.D. and Kandel, E.R. (2003) Rap1 couples cAMP signaling to a distinct pool of p42/44MAPK regulating excitability, synaptic plasticity, learning, and memory. Neuron, 39: 309–325. Nguyen, A., Burack, W.R., Stock, J.L., Kortum, R., Chaika, O.V., Afkarian, M., Muller, W.J., Murphy, K.M., Morrison, D.K., Lewis, R.E., Mcneish, J. and Shaw, A.S. (2002) Kinase suppressor of Ras (KSR) is a scaffold which facilitates mitogen-activated protein kinase activation in vivo. Mol. Cell. Biol., 22: 3035–3045.
40 Norton, V.G., Imai, B.S., Yau, P. and Bradbury, E.M. (1989) Histone acetylation reduces nucleosome core particle linking number change. Cell, 57: 449–457. Numachi, Y., Yoshida, S., Yamashita, M., Fujiyama, K., Naka, M., Matsuoka, H., Sato, M. and Sora, I. (2004) Psychostimulant alters expression of DNA methyltransferase mRNA in the rat brain. Ann. NY Acad. Sci., 1025: 102–109. Ogryzko, V.V., Schiltz, R.L., Russanova, V., Howard, B.H. and Nakatani, Y. (1996) The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell, 87: 953–959. Oike, Y., Hata, A., Mamiya, T., Kaname, T., Noda, Y., Suzuki, M., Yasue, H., Nabeshima, T., Araki, K. and Yamamura, K. (1999) Truncated CBP protein leads to classical Rubinstein–Taybi syndrome phenotypes in mice: implications for a dominant-negative mechanism. Hum. Mol. Genet., 8: 387–396. Ozonoff, S. (1999) Cognitive impairment in neurofibromatosis type 1. Am. J. Med. Genet., 89: 45–52. Partington, M.W., Mulley, J.C., Sutherland, G.R., Thode, A. and Turner, G. (1988) A family with the Coffin–Lowry syndrome revisited: localization of CLS to Xp21-pter. Am. J. Med. Genet., 30: 509–521. Pearson, G., Robinson, F., Beers Gibson, T., Xu, B.E., Karandikar, M., Berman, K. and Cobb, M.H. (2001) Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocr. Rev., 22: 153–183. Petrij, F., Giles, R.H., Dauwerse, H.G., Saris, J.J., Hennekam, R.C., Masuno, M., Tommerup, N., Van Ommen, G.J., Goodman, R.H., Peters, D.J. and Breuning, M.H. (1995) Rubinstein–Taybi syndrome caused by mutations in the transcriptional co-activator CBP. Nature, 376: 348–351. Phelps, E.A. and Ledoux, J.E. (2005) Contributions of the amygdala to emotion processing: from animal models to human behavior. Neuron, 48: 175–187. Roberson, E.D., English, J.D., Adams, J.P., Selcher, J.C., Kondratick, C. and Sweatt, J.D. (1999) The mitogen-activated protein kinase cascade couples PKA and PKC to cAMP response element binding protein phosphorylation in area CA1 of hippocampus. J. Neurosci., 19: 4337–4348. Roberson, E.D. and Sweatt, J.D. (1996) Transient activation of cyclic AMP-dependent protein kinase during hippocampal long-term potentiation. J. Biol. Chem., 271: 30436–30441. Roopra, A., Sharling, L., Wood, I.C., Briggs, T., Bachfischer, U., Paquette, A.J. and Buckley, N.J. (2000) Transcriptional repression by neuron-restrictive silencer factor is mediated via the Sin3-histone deacetylase complex. Mol. Cell. Biol., 20: 2147–2157. Rouaux, C., Loeffler, J.P. and Boutillier, A.L. (2004) Targeting CREB-binding protein (CBP) loss of function as a therapeutic strategy in neurological disorders. Biochem. Pharmacol., 68: 1157–1164. Rubinstein, J.H. and Taybi, H. (1963) Broad thumbs and toes and facial abnormalities. A possible mental retardation syndrome. Am. J. Dis. Child., 105: 588–608. Sassone-Corsi, P., Mizzen, C.A., Cheung, P., Crosio, C., Monaco, L., Jacquot, S., Hanauer, A. and Allis, C.D. (1999)
Requirement of RSK-2 for epidermal growth factor-activated phosphorylation of histone H3. Science, 285: 886–891. Schafe, G.E., Atkins, C.M., Swank, M.W., Bauer, E.P., Sweatt, J.D. and Ledoux, J.E. (2000) Activation of ERK/MAP kinase in the amygdala is required for memory consolidation of pavlovian fear conditioning. J. Neurosci., 20: 8177–8187. Schrimsher, G.W., Billingsley, R.L., Slopis, J.M. and Moore III, B.D. (2003) Visual-spatial performance deficits in children with neurofibromatosis type-1. Am. J. Med. Genet. A, 120: 326–330. Selcher, J.C., Atkins, C.M., Trzaskos, J.M., Paylor, R. and Sweatt, J.D. (1999) A necessity for MAP kinase activation in mammalian spatial learning. Learn. Mem., 6: 478–490. Selcher, J.C., Weeber, E.J., Christian, J., Nekrasova, T., Landreth, G.E. and Sweatt, J.D. (2003) A role for ERK MAP kinase in physiologic temporal integration in hippocampal area CA1. Learn. Mem., 10: 26–39. Shahbazian, M., Young, J., Yuva-Paylor, L., Spencer, C., Antalffy, B., Noebels, J., Armstrong, D., Paylor, R. and Zoghbi, H. (2002) Mice with truncated MeCP2 recapitulate many Rett syndrome features and display hyperacetylation of histone H3. Neuron, 35: 243. Silva, A.J., Frankland, P.W., Marowitz, Z., Friedman, E., Laszlo, G.S., Cioffi, D., Jacks, T. and Bourtchuladze, R. (1997) A mouse model for the learning and memory deficits associated with neurofibromatosis type I. Nat. Genet., 15: 281–284. Soloaga, A., Thomson, S., Wiggin, G.R., Rampersaud, N., Dyson, M.H., Hazzalin, C.A., Mahadevan, L.C. and Arthur, J.S. (2003) MSK2 and MSK1 mediate the mitogen- and stress-induced phosphorylation of histone H3 and HMG-14. EMBO J., 22: 2788–2797. Steffan, J.S., Bodai, L., Pallos, J., Poelman, M., Mccampbell, A., Apostol, B.L., Kazantsev, A., Schmidt, E., Zhu, Y.Z., Greenwald, M., Kurokawa, R., Housman, D.E., Jackson, G.R., Marsh, J.L. and Thompson, L.M. (2001) Histone deacetylase inhibitors arrest polyglutamine-dependent neurodegeneration in Drosophila. Nature, 413: 739–743. Sweatt, J.D. (2003) Mechanisms of Memory. Elsevier, San Diego, CA. Tong, J., Hannan, F., Zhu, Y., Bernards, A. and Zhong, Y. (2002) Neurofibromin regulates G protein-stimulated adenylyl cyclase activity. Nat. Neurosci., 5: 95–96. Tremolizzo, L., Doueiri, M.S., Dong, E., Grayson, D.R., Davis, J., Pinna, G., Tueting, P., Rodriguez-Menendez, V., Costa, E. and Guidotti, A. (2005) Valproate corrects the schizophrenia-like epigenetic behavioral modifications induced by methionine in mice. Biol. Psychiatry, 57: 500–509. Trivier, E., De Cesare, D., Jacquot, S., Pannetier, S., Zackai, E., Young, I., Mandel, J.L., Sassone-Corsi, P. and Hanauer, A. (1996) Mutations in the kinase RSK-2 associated with Coffin–Lowry syndrome. Nature, 384: 567–570. Vettese-Dadey, M., Grant, P.A., Hebbes, T.R., Crane- Robinson, C., Allis, C.D. and Workman, J.L. (1996) Acetylation of histone H4 plays a primary role in enhancing transcription factor binding to nucleosomal DNA in vitro. EMBO J., 15: 2508–2518.
41 Vianna, M.R., Barros, D.M., Silva, T., Choi, H., Madche, C., Rodrigues, C., Medina, J.H. and Izquierdo, I. (2000) Pharmacological demonstration of the differential involvement of protein kinase C isoforms in short- and long-term memory formation and retrieval of one-trial avoidance in rats. Psychopharmacology (Berl.), 150: 77–84. Von Rotz, R.C., Kohli, B.M., Bosset, J., Meier, M., Suzuki, T., Nitsch, R.M. and Konietzko, U. (2004) The APP intracellular domain forms nuclear multiprotein complexes and regulates the transcription of its own precursor. J. Cell. Sci., 117: 4435–4448. Vossler, M.R., Yao, H., York, R.D., Pan, M.G., Rim, C.S. and Stork, P.J. (1997) cAMP activates MAP kinase and Elk-1 through a B-Raf- and Rap1-dependent pathway. Cell, 89: 73–82. Waltereit, R. and Weller, M. (2003) Signaling from cAMP/ PKA to MAPK and synaptic plasticity. Mol. Neurobiol., 27: 99–106. Wang, J.H. and Kelly, P.T. (1996) The balance between postsynaptic Ca(2+)-dependent protein kinase and phosphatase activities controlling synaptic strength. Learn. Mem., 3: 170–181. Watabe, A.M., Zaki, P.A. and O’dell, T.J. (2000) Coactivation of beta-adrenergic and cholinergic receptors enhances the induction of long-term potentiation and synergistically activates mitogen-activated protein kinase in the hippocampal CA1 region. J. Neurosci., 20: 5924–5931. Weeber, E.J., Atkins, C.M., Selcher, J.C., Varga, A.W., Mirnikjoo, B., Paylor, R., Leitges, M. and Sweatt, J.D. (2000) A role for the beta isoform of protein kinase C in fear conditioning. J. Neurosci., 20: 5906–5914.
Weeber, E.J., Jiang, Y.H., Elgersma, Y., Varga, A.W., Carrasquillo, Y., Brown, S.E., Christian, J.M., Mirnikjoo, B., Silva, A., Beaudet, A.L. and Sweatt, J.D. (2003) Derangements of hippocampal calcium/calmodulin-dependent protein kinase II in a mouse model for Angelman mental retardation syndrome. J. Neurosci., 23: 2634–2644. Wood, M.A., Kaplan, M.P., Park, A., Blanchard, E.J., Oliveira, A.M., Lombardi, T.L. and Abel, T. (2005) Transgenic mice expressing a truncated form of CREB-binding protein (CBP) exhibit deficits in hippocampal synaptic plasticity and memory storage. Learn. Mem., 12: 111–119. Wu, G.Y., Deisseroth, K. and Tsien, R.W. (2001) Spaced stimuli stabilize MAPK pathway activation and its effects on dendritic morphology. Nat. Neurosci., 4: 151–158. Xing, J., Ginty, D.D. and Greenberg, M.E. (1996) Coupling of the RAS-MAPK pathway to gene activation by RSK2, a growth factor-regulated CREB kinase. Science, 273: 959–963. Yntema, H.G., Van Den Helm, B., Kissing, J., Van Duijnhoven, G., Poppelaars, F., Chelly, J., Moraine, C., Fryns, J.P., Hamel, B.C., Heilbronner, H., Pander, H.J., Brunner, H.G., Ropers, H.H., Cremers, F.P. and Van Bokhoven, H. (1999) A novel ribosomal S6-kinase (RSK4; RPS6KA6) is commonly deleted in patients with complex X-linked mental retardation. Genomics, 62: 332–343. Yuan, L.L., Adams, J.P., Swank, M., Sweatt, J.D. and Johnston, D. (2002) Protein kinase modulation of dendritic K+ channels in hippocampus involves a mitogen-activated protein kinase pathway. J. Neurosci., 22: 4860–4868. Zhu, Y. and Parada, L.F. (2001) A particular GAP in mind. Nat. Genet., 27: 354–355.
=Transcription Factor = G-protein (Gαs or Gαq) AMPA R
= Ribosome
β -AR AC
PKA P
NMDA CaM
R
CaMKII
Ca++
Glutamate
MAPK
VGCC
AMPA R
P
PKC PLC
mGluR Ras RTKs
Plate 3.1. Signaling at the postsynaptic neuron: receptors, channels, and kinases. Signaling at the postsynaptic neuron is complex. Some of the key players are depicted here, although this is an oversimplification of the diverse sets of receptors, channels, and kinases that are activated following glutamate release from the presynaptic terminal and subsequent calcium release in the postsynaptic neuron. CaMKII, PKA, PKC, and MAPK are all activated by various mechanisms, and events downstream of these kinases contribute to the generation of long-term potentiation through the phosphorylation of targets such as AMPA and NMDA receptors, the insertion of AMPA receptors into the postsynaptic membrane, and the induction of transcription and translation. See text for details.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 4
Altered brain activity in healthy seniors: what does it mean? Jonas Persson1, and Lars Nyberg2 2
1 Department of Psychology, University of Michigan, Ann Arbor, MI, USA Departments of Integrative Medical Biology (Physiology) and Radiation Sciences (Diagnostic Radiology), Umea˚ University, Umea˚, Sweden
Abstract: Age-related performance decreases are frequently observed on various memory tasks. Recent brain imaging studies using positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) suggest a link between altered patterns of brain activity in older adults and memory performance. Convergent neuroimaging evidence shows that older adults have decreased activity in multiple regions important for memory tasks. Such relative under-activation in older adults is likely related to agerelated reductions in cognitive performance. Age-comparative neuroimaging studies have also provided convincing support for regional over-activation by older adults. Such findings indicate that the older brain can re-organize to better cope with cognitive and other challenges. Although over-activation may play a compensatory role when cognitive decline is limited, under-activation seems to be the typical pattern when cognitive impairment is in a more progressed state. This pattern of age-related changes suggests that compensation through over-activation is restricted to the early stages of cognitive impairment in aging. Keywords: aging; compensation; fMRI; memory; prefrontal; cognitive the brain constraints that typically accompany advancing age. Age-related reductions in task-induced brain activity may be the most common empirical observation, but some functional brain imaging studies have also found that older adults on occasion may activate certain regions to a greater extent than younger adults (e.g., Cabeza et al., 1997; Madden et al., 1999; Reuter-Lorenz et al., 2000; Logan et al., 2002; Rosen et al., 2002). Relative age-related increases, rather than age-related reductions, in brain activity may be seen as a form of re-organization of functional networks in the brain through expression of neural plasticity, which occurs as a compensatory response initiated by age-related changes. That the brain can show structural as well as functional changes in response to various conditions has been known for a long
Introduction The cognitive neuroscience of aging is currently a topic of intense investigation (Cabeza et al., 2005). Much of the research in this area is focused on the neural basis for impairment of sensory/perceptual, cognitive, and other functions in older age. For example, functional and structural MRI studies have demonstrated that older adults show less hippocampal activity during episodic tasks and also a reduction in hippocampal volume as compared with younger adults (e.g., Grady et al., 1999; Daselaar et al., 2003a, b; Raz et al., 2004). Such and other findings are beginning to define some of
Corresponding author. Tel.: +1 734 764 6399; Fax: +1 734 763 7480; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57004-9
45
46
time (see Ward and Cohen, 2004; Pascual-Leone et al., 2005; Burke and Barnes, 2006 for recent reviews), and it is discussed in many chapters of the current volume. While there is substantive evidence that such changes can signify positive adaptations to a new situation, considerably less is known about the functional relevance of relative increases in older compared with younger adults. In this chapter we will therefore review the results of some recent studies of age-related alterations of functional brain activity, primarily during tests of episodic long-term memory and working memory, with the overall goal of trying to shed some light on the possible functional relevance of increased brain activity.
Overview of findings Age-related changes in activation have been reported in diverse prefrontal regions and across a variety of episodic and working memory tasks. It is important to note that these changes are not always easy to interpret, and that there are an immense number of activation and deactivation patterns possible to be associated with memory function. Older adults could activate less, more, or even different neural substrates to perform a memory task than young adults do. The relation between age-related differences in activation and behavioral performance is complicated further by volumetric decreases in neural tissue that occur with age. Volumetric changes with aging are especially noticeable in the prefrontal cortex (PFC) and the hippocampus, regions that are highly involved in memory functions (Raz, 2000; Raz et al., 2004). Although neuroimaging studies attribute age-related changes in regional blood-flow or blood oxygen level-dependent (BOLD) signal to neural activity, this coupling may be altered by changes in the neurovascular system that occurs with age. It is important to point out, however, that studies on simple motor or sensory tasks suggest that although the signal may be smaller in older adults, the overall characteristics of the signal are similar for older and younger adults. Despite these issues, neuroimaging studies have provided valuable information about the neural
mechanisms of aging and memory. This may be especially true for certain issues for which behavioral methods are inadequate, such as assessing regions underlying specific components of memory, such as encoding and retrieval.
Working memory and aging Working memory — the capacity to hold information actively in mind for a brief (several seconds) period of time — has been a major focus of behavioral research on aging because this capacity is key to higher order cognitive abilities, including language comprehension, reasoning, and problem solving. Working memory tasks differ in the amount of ‘‘work’’ they require, with some emphasizing the maintenance of several items (i.e., 7 or less) and others requiring additional manipulation or complex processing of the stored contents (e.g., reordering the items, successive updating, and resolving interference from previous targets). Tasks that require more work place greater burden on cognitive control, and are more likely to show age dependence in performance (see Reuter-Lorenz and Sylvester, 2005, for a review). Neuroimaging evidence suggests that PFC regions sustain working memory processes, and that it may be functionally segregated as follows: ventral PFC mediates maintenance processes, and dorsal PFC is recruited when additional cognitive control of information is needed to perform the task (Cabeza and Nyberg, 2000; e.g., Smith and Jonides, 1999). In the past 10 years there have been several studies investigating working memory, which included a sample of older adults (see Table 1). A majority of these have used verbal material in the delayed match-to-sample task with varying target load (in general two to six items). In one of the early studies of working memory (Grady et al., 1998), age differences in working memory for faces were investigated using a delayed match-to-sample task. In this study a face was presented with a variable delay ranging from 1 to 21 s. Accuracy was comparable across age groups, but older adults were generally slower than young adults. Although both young and old adults activated the
Table 1. List of studies included in the review Study Working memory Grady et al. (1998) Reuter-Lorenz et al. (2000) Mitchell et al. (2000) Jonides et al. (2000) Rypma and D’Esposito (2000) Reuter-Lorenz et al. (2001) Rypma and D’Esposito (2001) Smith et al. (2001) Grossman et al. (2002) Cabeza et al. (2004) Rypma et al. (2005) Episodic memory encoding Grady et al. (1995) Cabeza et al. (1997) Grady et al. (1999) Madden et al. (1999) Anderson et al. (2000) Iidaka et al. (2001) Rosen et al. (2002) Grady et al. (2002) Logan et al. (2002) Stebbins et al. (2002) Schiavetto et al., 2002 Daselaar et al. (2003a, b) Morcom et al. (2003) Gutchess et al. (2005) Episodic memory retrieval Grady et al. (1995) Cabeza et al. (1997) Madden et al. (1999) Cabeza et al. (2000)
Behavioral performance
(letters) (locations)
(words) (pictures)
(item) (temporal)
Anderson et al. (2000) Grady et al. (2002) Schiavetto et al. (2002) Daselaar et al. (2003a, b) Cabeza et al. (2004)
Y faster Y faster; Y more accurate Y faster Y more accurate Y better than Oa Y faster Y faster; Y more accurate Y faster Y faster Y faster n.s. Y faster; Y more accurate Y more accurate n.s. Y more accurate Y faster Y more accurate Y more accurate Y more accurate Y more accurate Y more accurate Y more accurate n.s. Y more accurate Y faster Y faster; Y more accurate n.s. Y more accurate n.s. Y faster Y more accurate Y more accurate Y more accurate Y more accurate Y more accurate Y faster Y faster
Temporal activation
Parietal activation
Occipital activation
MTL activation
Prefrontal activation
L
L
R
L O+
L Y+
Y+
O+
L Y+ Y+ Y+
Y+
Y+
Y+
O+
Y+
O+ O+
Y+
R
Y+
O+
Y+
R
R
Y+ ¼ Y+ ¼ Y+ O+ ¼ ¼
Y+
Y+
Y+ Y+
Y+
Y+
Y+
Y+ Y+ Y+ O+ Y+ ¼ ¼ Y+ Y+ O+ Y+ O+ Y+ O+ O+
Y+ Y+
Y+
Y+
Y+
Y+
O+
O+ O+ ¼ O+ Y+ O+ ¼ ¼ ¼
Y+
Y+ Y+
Y+ Y+
O+ Y+
Y+ Y+
¼
O+ O+ Y+
O+ Y+
Y+
Y+
Y+
Y+ Y+ Y+ O+
Y+
Y+ O+
Y+ O+
Y+
Y+ O+
O+
R O+ O+ O+ Y+ ¼ O+ ¼ Y+ ¼ ¼ ¼
¼ Y+ O+ Y+ O+ Y+ ¼ O+ O+ ¼ Y+ O+ Y+ Y+ O+ ¼ O+ ¼
Note: MTL ¼ medial temporal lobe. The ‘‘Behavioral performance’’ column lists whether there was a significant between-group difference in reaction time or accuracy. a
Composite Z of accuracy and speed.
47
48
bilateral PFC, the peak activations varied with age. While younger adults activated ventrolateral PFC [VLPFC; Brodmann area (BA) 45] more than older adults, older adults activated the left dorsolateral PFC (DLPFC, BA9) more than younger adults. Age-related differences in maintenance of verbal material were examined in a positron emission tomography (PET) study by Reuter-Lorenz et al. (2000). In this study a four-item letter recognition task with a long delay was compared with a repeated single-item with a short delay. Both young and older adults performed the working memory task with high accuracy, although the older participants were significantly slower and less accurate than younger participants. Using a region-of-interest analysis, similar levels of activation were found for both age groups in left VLPFC (BA 44), left premotor cortex (BA 6), and left parietal areas (BA 40/7). Age differences were observed in anterior regions, with older individuals showing higher activation in right VLPFC and dorsolateral PFC (DLPFC) regions compared with their younger counterparts. In two event-related experiments using a similar paradigm, Rypma and colleagues (Rypma and D’Esposito, 2000, 2001) also found significant age differences in frontal regions. Both experiments included a manipulation of target load (six items vs. two items) with a 12 s retention interval. The event-related design made it possible to separately assess brain activation related to the different components of encoding, maintenance, and retrieval. In both these experiments it was found that younger individuals showed greater activation in right DLPFC compared with older individuals. This difference was restricted to the retrieval phase, and was more pronounced for high target load. Even though these results deviate considerably from the findings by Reuter-Lorenz et al. (2000, 2001), Rypma and D’Esposito (2000, 2001) also found a correlation between right DLPDC activation and performance. For older adults, greater activation in the right DLPFC was associated with shorter response times, which is similar to the findings by Reuter-Lorenz et al. (2001). Rypma and D’Esposito (2000) also reported an inverse relationship for younger adults. The relation between performance and activation suggests
age differences in activation–response time relationships, and is consistent with behavioral data suggesting that processing speed is an important variable in age-related performance variation (Salthouse, 1996). This finding has also been observed in different brain regions, and across task domains (Persson et al., 2004; Rypma et al., 2005). Taken together, despite the variability in imaging methods, task designs, and materials used, there are several consistent patterns across the studies on working memory and aging. First, prefrontal areas are predominantly, although by no means exclusively, the site of age differences. Second, observations of less activation in older adults are found in several studies, and this pattern of less activation is most consistent for left VLPFC (e.g., Grady et al., 1998; Jonides et al., 2000; Smith et al., 2001). Third, over-activation in older adults is found in several studies, although several exceptions to this pattern have been reported (e.g., Rypma and D’Esposito, 2000). Over-activation is most pronounced in the DLPFC, although some studies also find over-activation in VLPFC (see Rajah and D’Esposito, 2005, for a recent review).
Episodic encoding and aging In neuroimaging studies that examine memory for a particular learning episode (e.g., learning a list of items followed by subsequent memory retrieval) it is a typical finding that left prefrontal regions are activated during the memory encoding phase, and that right prefrontal regions become more activated during memory retrieval (Tulving et al., 1994; Nyberg et al., 1996). A pronounced departure from this pattern has been found in older adults. In one of the early reports on episodic memory, Grady et al. (1995) investigated age-related differences during intentional encoding of faces compared with passive viewing of a visual noise pattern or perceptual matching of faces. Younger adults showed more activation in left dorsal and inferior frontal cortex during face encoding than during passive viewing or perceptual matching, but older adults showed no encoding activation in the inferior frontal cortex. Reduced left frontal activation may reflect inadequate
49
encoding strategies and reduced processing resources, which, in turn, may lead to poor memory in older adults. This is partly supported by findings of age-invariant left prefrontal activation under conditions of specific encoding strategies. Logan et al. (2002) examined encoding of words under the following three conditions: intentional memory, deep incidental processing (abstract/concrete categorization), and shallow incidental encoding (temporal order of a letter). A similar gain in memory performance for young and old adults was found as a function of depth of encoding. In accordance with results from several other studies, older adults had overall less activation across prefrontal regions during shallow incidental encoding and also during intentional encoding. Regions included Brodmann areas 45 and 47, which are associated with semantic processing (Demb et al., 1995; Poldrack et al., 1999). A notable finding was that the observation of less activity in older adults disappeared during deep incidental encoding. One possibility is that age-related differences in memory processing might be reduced by providing older adults with effective encoding strategies, such as semantic categorization. Subsequent studies on age-related differences have found reduced activation for older adults in regions typically activated by younger adults, such as left PFC. Rather, a common finding is that older adults activate additional, often contralateral prefrontal regions, which are not generally activated by younger adults during episodic encoding (Cabeza et al., 1997; Logan et al., 2002; Stebbins et al., 2002; Morcom et al., 2003; Gutchess et al., 2005). In the seminal study by Cabeza et al. (1997), young and old adults were scanned using PET while intentionally encoding words. In addition to the decreased left prefrontal activation in older adults as reported previously, they found almost equivalent levels of activation in left and right PFC for older adults. In contrast, young adults showed specific engagement of left PFC. Based on these data, Cabeza and colleagues suggested that the finding of bilateral activation in old adults might indicate compensatory recruitment of right prefrontal regions as a function of insufficient engagement in the left hemisphere. It is important to note that a bilateral pattern of activation may be a
consequence of a reduction of activation in older adults, and not necessary activation in right frontal regions over and above activation in younger adults. For example, a bilateral pattern of activation can occur as a result of less activation in the dominant hemisphere by the older adults as compared to young adults, while activation in the nondominant hemisphere remains unchanged (e.g., Stebbins et al., 2002). Bilateral recruitment in this case may not represent compensation, since the level of activation in right frontal regions may be more or less equal for young and old adults. One important aspect is to examine how agerelated differences in activation at encoding relate to whether the encoding of a particular item was successful or not. Using event-related functional magnetic resonance imaging (fMRI) it is possible to measure the signal for a specific item, and then bin the items based on whether the item was remembered or not in a subsequent memory test. This kind of approach is particularly valuable as it can provide information about the relationship between greater activation, encoding success, and compensation in old age. To date, there are three studies that have examined changes in episodicencoding brain activation related to aging using this paradigm (Daselaar et al., 2003a, b; Morcom et al., 2003; Gutchess et al., 2005). Morcom et al. (2003) examined verbal encoding using an incidental semantic categorization task. When remembered items were compared with forgotten items, similar left inferior PFC activation was found for young and old adults. These results are in line with the findings by Logan et al. (2002) showing equivalent left frontal activation during deep encoding. Old individuals also showed more bilateral anterior PFC activation for remembered items compared with their younger counterparts. The functional significance of this additional recruitment is difficult to determine from these findings. Additional recruitment of the left PFC was found in the study by Gutchess et al. (2005) using a nonverbal (pictures) incidental deep-encoding task. Given the bilateral activation often found using pictures (Kelley et al., 1998), the increased engagement of left PFC above the level of young adults indicates that this selective activation may be compensatory. Although reduced hippocampal
50
recruitment in old adults was found in the study by Daselaar et al. (2003a, b), no age-related differences were found in prefrontal regions. Together, the findings of prefrontal activation using the subsequent memory paradigm have been quite variable, and clearly more studies are needed to specify the functional role of prefrontal activation in aging. To summarize, the literature on neural activation related to episodic encoding suggests that older adults show less recruitment of medial temporal regions, including the hippocampus, than younger adults. With regard to frontal regions, many studies show less left-sided activation in older adults compared with young adults. More pronounced activation in right- frontal regions for older adults is also found across studies, which suggests that older adults activate a different network of regions compared to younger adults.
Episodic retrieval and aging The behavioral literature suggests that episodic retrieval is generally less affected by aging than encoding, and disproportional impairments by interference during encoding compared with retrieval have been observed (Park et al., 1989; Anderson et al., 1998). Even so, older adults often report difficulties with episodic retrieval. Several previous neuroimaging studies have proposed a specific role for right frontal regions in episodic retrieval, which has been conceptualized in the HERA (hemispheric encoding/retrieval asymmetry) model (Tulving et al., 1994; Nyberg et al., 1996). In general, older adults seem to engage a somewhat different neural system compared with younger adults. Overall, age differences in brain activation at retrieval are most consistently seen in frontal regions, and less often in medial temporal regions. As noted earlier, Cabeza et al. (1997) found reduced right-PFC activation with age, resulting in a more bilateral pattern of frontal activation for older adults and unilateral activation for younger adults. Interestingly, the age differences were more accentuated during recall compared with recognition, suggesting less differentiation of neural activity in older adults as a function of retrieval condition. In addition to age-related re-
ductions in PFC activation, several studies have found increased activation in older adults. For example, Madden et al. (1999) found stronger activation in both right and left frontal regions for older adults compared with younger adults during retrieval of words using forced choice recognition. These results suggest that the neural systems mediating episodic retrieval is more widely distributed for older adults than for younger adults. Although the results from Madden et al. (1999) suggest a general increase in activation in older adults, most studies have found more pronounced recruitment of left frontal regions in older adults compared with younger adults. Recent studies suggest that reduced frontal activation may be linked to individual differences in performance. Daselaar et al. (2003a, b) divided old individuals on the basis of memory performance and found that low-performing older adults had more frontal activation than young adults, with high-performing older adults showing least activation. In general, brain patterns associated with episodic retrieval show less distinct age differences than encoding studies. In a majority of the studies, older adults showed left-PFC activation, which deviates from the pattern of activation typically found in younger adults. In many studies frontal decreases with age were observed, predominantly in the right PFC. Age-related hippocampal differences were less pronounced than for episodic encoding.
Structure– function correlates of longitudinal changes in memory performance In most studies to date, the critical comparison has been related to differences in brain activation in younger and older adults. However, psychological assessment of individuals across the life span makes it evident that not all older adults show an equal level of performance, or are similarly affected by advancing age. To understand the dynamics of functional and structural changes that occur with increasing age, it is important to take longitudinal cognitive performance into account, as well as predictors that may affect this
51
performance. For example, specific variables may represent normally occurring individual differences in the capacity to cope with detrimental effects associated with advancing age.
A central question is how changes in the structure and function of the aging brain relate to stable or declining cognitive performance over time. In a recent study this question was addressed by
Fig. 1. Overview of findings from Persson et al. (2006). Top — left: stable (N ¼ 20) and declining (N ¼ 20) longitudinal memory performance in older adults (composite score from three episodic tests). Top — right: fMRI results showing over-activation in right PFC during semantic categorization compared to fixation baseline for participants with stable (bottom) and declining (top) memory performance. Activations are displayed on transverse sections of an anatomical template brain. Bars show average signal change in the right ventral frontal cortex (BA 47). Bottom — left: mean height adjusted volume of the left (top) and right (bottom) hippocampus (in mm3). Error bars show SEM. Bottom — right: results from the DTI analyses show the anterior corpus callosum (genu) outlined on a transverse slice of a fractional anisotropy (FA) image. High signal intensity (brightness) reflects higher FA. Mean FA as a function of longitudinal memory performance is shown in the histogram on the left-hand side. See Plate 4.1 in Colour Plate Section.
52
identifying two groups of older adults that differed with regard to how their level of episodic memory performance changed over a decade (Persson et al., 2006). The groups were selected from an ongoing longitudinal study (Nilsson et al., 1997), based on composite scores from three episodic memory tests at three time points over 10 years. One group included participants with stable memory performance over time, and the other group included participants with declining memory performance over time (Fig. 1). The assessment of brain structural differences between the groups included manual tracing of the hippocampus, and measures of white-matter integrity by diffusion tensor imaging (DTI). We found that the volume of the hippocampus was significantly reduced in older adults with a declining memory performance (Fig. 1) (see also Rodrigue and Raz, 2004). Also, a reduction in white-matter integrity was found for DTI-measured fractional anisotropy in the anterior corpus callosum (Fig. 1). The DTI finding of group differences in the anterior part of the corpus callosum suggests that changes in white-matter integrity may contribute to memory dysfunction in old age. These results are also in agreement with the previous findings of negative correlations between white-matter integrity and behavior in anterior, but not posterior parts, of the corpus callosum in older adults (O’Sullivan et al., 2001; Madden et al., 2004). Using fMRI, brain activation was measured in stable and declining individuals during incidental episodic encoding. A central question was whether stable or declining longitudinal performance was associated with atypical neural activation. Given the previous findings of brain alterations in frontal regions in the brain, four prefrontal regions of interests were used for the functional assessment. Participants were scanned while performing a semantic (abstract/concrete) categorization task, which previously has been sensitive to age-related changes in activation. We found increased leftfrontal activation in both groups in this task, and additional right-PFC activation for elderly participants with the greatest decline in memory performance (Fig. 1). The combination of longitudinal behavioral scores and functional imaging data indicates that
increases in frontal activation are associated with age-related decline in cognitive function. In accordance with the HERA model discussed previously, strong right- frontal activation is atypical for this type of task in studies of younger adults (Cabeza and Nyberg, 2000). These results therefore suggest that older participants with stable memory performance recruit a similar pattern of brain regions as young adults, while older adults with declining performance recruit a different network of regions. Increased recruitment of frontal cortex with declining performance is consistent with the hypothesis that such differences in activity relate to age-associated disturbance in brain function (e.g., Li et al., 2001, 2002).
Discussion The main goal of the present chapter was to discuss the significance of age-related alterations in patterns of functional brain activity. One pattern to emerge from several studies is that younger adults activate certain brain regions to a greater extent than older adults. This is illustrated in Table 1, in which several findings of relative greater prefrontal activity of younger compared to older adults are summarized. A not-so-bold interpretation is that such relative under-activation of older adults is related to age-related reductions in cognitive performance. Support for this interpretation comes from Table 1, in which performance on cognitive tasks, when available, is summarized. As can be seen, the typical finding is that younger adults are more accurate and/or faster than older adults. This cognitive and brain-activity reduction in older age may, in part, be a reflection of agerelated changes in brain structure. There is much evidence for reductions in the volume of gray- as well as white-matter frontal volumes (Pfefferbaum et al., 1994; Raz et al., 1997; O’Sullivan et al., 2001). In addition, although some studies have not supported such a relation (e.g., Ylikoski et al., 2001), several studies have linked structural changes of the hippocampus to age-related cognitive changes (Golomb et al., 1994; Rodrigue and Raz, 2004). Results from recent analyses in our laboratory provide additional support for a
53
relationship (Forsberg, 2004). Manual tracing of the hippocampus (Lind et al., 2006) was used to define hippocampal volume for 58 adults ranging in age between 40 and 80 years. Left hippocampal volume was found to correlate significantly with chronological age, such that younger age was associated with larger hippocampi. It was also significantly correlated with episodic memory performance, and there was a typical negative age–memory relationship. Together, this correlative triad relates aging, cognition, and hippocampal volume. A second basis for age-related under-activation of functional brain activity may be process-based, rather than reflecting structural changes in the brain’s hardware. There are many findings from cognitive studies suggesting that older adults are less prone to spontaneously engage in elaborated mental activities that typically foster good cognitive performance (Craik, 1986). However, when given appropriate instructions and support, they are fully capable of using such more effective means to approach a task and thereby enhancing their performance levels. The above-mentioned study by Logan et al. (2002) adds neural data to this situation by showing that under-activation of frontal regions by older adults during intentional memory encoding can be reversed if they instead use instructor-guided means for incidental encoding. This elegant demonstration rules out a structural explanation for their under-activation in the first place, as it could be reversed by support. In addition to age-related under-activation, and of primary focus in the present chapter, age-comparative brain-imaging studies have also provided convincing evidence for regional over-activation by older adults (see Table 1). Such findings have opened up for several interesting speculations that the older brain can re-organize to better cope with cognitive and other challenges (Reuter-Lorenz et al., 2000; Cabeza et al., 2002; Gutchess et al., 2005). Our study of structural and functional correlates of cognitive decline in aging (Persson et al., 2005) provided evidence that additional regional activation was related to cognitive decline rather than stability, and showed that those elderly who over-activated had reductions of hippocampal volume and white-matter integrity in the
anterior corpus callosum. While these findings strongly suggest that the ‘‘best case’’ for older adults would be to show similar patterns of functional brain activity as younger adults (i.e., neither over-activation nor under-activation), they still leave room for a compensatory interpretation of over-activation by older adults. First, based on cross-sectional cognitive assessment, it has been shown that over-activation in a sample of older adults related to higher cognitive performance (Cabeza et al., 2002). Second, patients with Alzheimer’s disease (AD), who clearly are cognitively under-performing, have been observed to recruit additional regions compared with healthy elderly (Ba¨ckman et al., 1999), and the degree of recruitment of additional prefrontal regions by AD patients has been found to relate to task performance (Grady et al., 2003). Grady and colleagues argued that recruitment of additional cortical regions by AD patients is likely a response to degenerative disease processes that even may precede the onset of symptoms. This account is in good agreement with the conclusion by Persson et al. (2005) that additional brain activity goes with cognitive decline rather than stability. In short, in order to be viewed as a compensatory response, there has to be something to compensate for. Importantly, though, there may be limits to functional compensation as indexed by regional over-recruitment. The findings by Cabeza et al. (2002) that low-performing older adults did not display additional frontal activity may be one example of such constraints on compensation. Additional results that suggest limits to compensatory responses come from studies of patients who under-performed on neuropsychological performance and progressed to AD within 8 months after assessment (Elgh et al., 2003). Compared to age-matched controls, these AD at-risk persons showed under-activation of task-relevant frontal regions. Similar findings were reported by Kato et al. (2001). A possible reason for this pattern is that at a certain stage of cognitive impairment, individuals may no longer be able to come up with additional strategies and compensatory responses, but instead may approach the task in a more passive manner. At this stage, structural brain changes may also constrain the possibilities
54
Fig. 2. Tentative model of how increased activation (over-recruitment) in older adults might relate to age-related changes in cognition.
for compensatory responses. Findings of robust group differences in default mode of brain activity may be seen as providing additional evidence that cognitively under-performing individuals have reductions in spontaneous brain activity (Lustig et al., 2003). Taken together, the available evidence on overactivation in older age may be summarized as in the schematic model in Fig. 2. When cognitive performance is unaffected by the aging process, or only affected marginally, older adults display little or no over-recruitment of brain regions. When older adults become more cognitively impaired (be it as a course of normal aging, mild cognitive impairment, or dementia), over-recruitment is salient. Finally, when the degree of cognitive impairment has progressed even more, under-activation seems to be the typical pattern with no or little signs of compensation in terms of overrecruitment.
Conclusions and future directions In conclusion, functional brain imaging studies have converged on the interesting pattern of not only age-related decreases in functional brain activity but also age-related increases. The available evidence suggests that age-related increases reflect a compensatory response to various age-related cognitive and brain changes through expression of
neural plasticity. Still, although driven by negative age changes, it is quite possible that over-recruitment is a positive (compensatory) response to a detrimental process. As such, there are reasons in future studies to examine to what extent specific factors influence degree of over-recruitment. Such factors may include a cognitive reserve, or spared neurocognitive capacity, that can be drawn upon to meet mental challenges (e.g., Stern et al., 2005). The ability to compensate for negative changes that occur with increasing age may be influenced by numerous lifestyle factors including physical fitness, exercise, diet, education, and social and intellectual engagement throughout adulthood, and later life. Recent results from the cognitive neuroscience of aging concern the possibility for adaptive reorganization and benefits from physical (Colcombe et al., 2004), and cognitive (Nyberg et al., 2003) training, and reduced age-related decline in cognition through social participation (Lo¨vde´n et al., 2005).
References Anderson, N.D., Craik, F.I.M. and Naveh-Benjamin, M. (1998) The attentional demands of encoding and retrieval in younger and older adults. 1. Evidence from divided attention costs. Psychol. Aging, 13: 405–423. Ba¨ckman, L., Andersson, J.L.R., Nyberg, L., Winblad, B., Nordberg, A. and Almkvist, O. (1999) Brain regions associated with episodic retrieval in normal aging and Alzheimer’s disease. Neurology, 52: 1861–1870. Burke, S.N. and Barnes, C.A. (2006) Neural plasticity in the ageing brain. Nat. Rev. Neurosci., 7: 30–40. Cabeza, R., Anderson, N.D., Locantore, J.K. and McIntosh, A.R. (2002) Aging gracefully: compensatory brain activity in high-performing older adults. Neuroimage, 17: 1394–1402. Cabeza, R., Grady, C.L., Nyberg, L., McIntosh, A.R., Tulving, E., Kapur, S., et al. (1997) Age-related differences in neural activity during memory encoding and retrieval: a positron emission tomography study. J. Neurosci., 17: 391–400. Cabeza, R. and Nyberg, L. (2000) Imaging cognition II: empirical review of 275 PET and fMRI studies. J. Cogn. Neurosci., 12: 1–47. Cabeza, R., Nyberg, L. and Park, D. (2005) Cognitive Neuroscience of Aging. Oxford University Press, New York. Colcombe, S.J., Kramer, A.F., Erickson, K.I., Scalf, P., McAuley, E., Cohen, N.J., Webb, A., Jerome, G.J., Marquez, D.X. and Elavsky, S. (2004) Cardiovascular fitness, cortical plasticity, and aging. Proc. Nat. Acad. Sci. USA, 101: 3316–3321.
55 Craik, F.I.M. (1986) A functional account of age differences in memory. In: Klix, F. and Hagendorf, H. (Eds.), Human Memory and Cognitive Abilities, Mechanisms, and Performances. Elsevier, North-Holland, The Netherlands. Daselaar, S.M., Veltman, D.J., Rombouts, S.A., Lazeron, R.H., Raaijmakers, J.G. and Jonker, C. (2003a) Neuroanatomical correlates of episodic encoding and retrieval in young and elderly subjects. Brain, 126: 43–56. Daselaar, S.M., Veltman, D.J., Rombouts, S.A., Raaijmakers, J.G. and Jonker, C. (2003b) Deep processing activates the medial temporal lobe in young but not elderly adults. Neurobiol. Aging, 24: 1005–1011. Demb, J.B., Desmond, J.E., Wagner, A.D., Vaidya, C.J., Glover, G.H. and Gabrieli, J.D.E. (1995) Semantic encoding and retrieval in the left inferior prefrontal cortex: a functional MRI study of task difficulty and process specificity. J. Neurosci., 15: 5870–5878. Elgh, E., Larsson, A., Eriksson, S. and Nyberg, L. (2003) Altered prefrontal brain activity in persons at risk for Alzheimer’s disease: an fMRI study. Int. Psychogeriatr., 15: 121–133. Forsberg, K., (2004) An MRI-study of the hippocampus volume in normal aging: relation to explicit memory. Unpublished undergraduate thesis, Department of Molecular Biology, Umea˚ University, Sweden. Golomb, J., Kluger, A., de Leon, M.J., Ferris, S.H., Convit, A., Mittleman, M., et al. (1994) Hippocampal formation size in normal human aging: a correlate of delayed secondary memory performance. Learn. Memory, 1: 45–54. Grady, C.L., McIntosh, A.R., Beig, S., Keightley, M.L., Burian, H. and Black, S.E. (2003) Evidence from functional neuroimaging of a compensatory prefrontal network in Alzheimer’s disease. J. Neurosci., 23: 986–993. Grady, C.L., McIntosh, A.R., Bookstein, F., Horwitz, B., Rapoport, S.I. and Haxby, J.V. (1998) Age-related changes in regional cerebral blood flow during working memory for faces. Neuroimage, 8: 409–425. Grady, C.L., McIntosh, A.R., Horwitz, B., Maisog, J.M., Ungerleider, L.G., Mentis, M.J., et al. (1995) Age-related reductions in human recognition memory due to impaired encoding. Science, 269: 218–221. Grady, C.L., McIntosh, A.R., Rajah, M.N., Beig, S. and Craik, F.I.M. (1999) The effects of age on the neural correlates of episodic encoding. Cereb. Cortex, 9: 805–814. Gutchess, A.H., Welsh, R.C., Hedden, T., Bangert, A., Minear, M., Liu, L.L., et al. (2005) Aging and the neural correlates of successful picture encoding: frontal activations compensate for decreased medial-temporal activity. J. Cogn. Neurosci., 17(1): 84–96. Jonides, J., Marshuetz, C., Smith, E.E., Reuter-Lorenz, P.A., Koeppe, R.A. and Hartley, A. (2000) Age differences in behavior and PET activation reveal differences in interference resolution in verbal working memory. J. Cogn. Neurosci., 12: 188–196. Kato, T., Knopman, D. and Liu, H.Y. (2001) Dissociation of regional activation in mild AD during visual encoding — a functional MRI study. Neurology, 56: 1143–1153.
Kelley, W.M., Miezin, F.M., McDermott, K.B., Buckner, R.L., Raichle, M., Cohen, N.J., et al. (1998) Hemispheric specialization in human dorsal frontal cortex and medial temporal lobe for verbal and nonverbal memory encoding. Neuron, 20: 927–936. Li, S.C., Lindenberger, U. and Sikstro¨m, S. (2001) Aging cognition: from neuromodulation to representation. Trends Cogn. Sci., 5: 479–486. Lind, J., Larsson, A., Persson, J., Ingvar, M., Nilsson, L.G., Ba¨ckman, L., et al., (2006) Reduced hippocampal volume in non-demented carriers of the apolipoprotein E varepsilon4: relation to chronological age and recognition memory. Neurosci. Lett., Jan. 4 [Epub ahead of print]. Logan, J.M., Sanders, A.L., Snyder, A.Z., Morris, J.C. and Buckner, R.L. (2002) Under-recruitment and nonselective recruitment: dissociable neural mechanisms associated with aging. Neuron, 33: 827–840. Lo¨vde´n, M., Ghisletta, P. and Lindenberger, U. (2005) Social participation attenuates decline in perceptual speed in old and very old age. Psychol. Aging,, 20: 423–434. Lustig, C., Snyder, A.Z., Bhakta, M., O’Brien, K.C., McAvoy, M., Raichle, M.E., et al. (2003) Functional deactivations: change with age and dementia of the Alzheimer type. Proc. Nat. Acad. Sci. USA, 100: 14504–14509. Madden, D.J., Turkington, T.G., Provenzale, J.M., Denny, L.L., Hawk, T.C., Gottlob, L.R., et al. (1999) Adult age differences in the functional neuroanatomy of verbal recognition memory. Hum. Brain Mapp., 7: 115–135. Madden, D.J., Whiting, W.L., Huettel, S.A., White, L.E., MacFall, J.R. and Provenzale, J.M. (2004) Diffusion tensor imaging of adult age differences in cerebral white matter: relation to response time. Neuroimage, 21: 1174–1181. Morcom, A.M., Good, C.D., Frackowiak, R.S.J. and Rugg, M.D. (2003) Age effects on the neural correlates of successful memory encoding. Brain, 126: 213–229. Nilsson, L.-G., Ba¨ckman, L., Erngrund, K., Nyberg, L., Adolfsson, R., Bucht, G., et al. (1997) The Betula prospective cohort study: memory, health and aging. Aging Neuropsychol. Cogn., 4: 1–32. Nyberg, L., Cabeza, R. and Tulving, E. (1996) PET studies of encoding and retrieval: the HERA model. Psychonom. Bull. Rev., 3: 135–148. Nyberg, L., Sandblom, J., Jones, S., Stigsdotter Neely, A., Petersson, K.M., Ingvar, M., et al. (2003) Neural correlates of training-related memory improvement in adulthood and aging. Proc. Nat. Acad. Sci. USA, 100: 13728–13733. O’Sullivan, M., Jones, D.K., Summers, P.E., Morris, R.G., Williams, S.C.R. and Markus, H.S. (2001) Evidence for cortical ‘‘disconnection’’ as a mechanism of age-related cognitive decline. Neurology, 57: 632–638. Park, D.C., Smith, A.D., Dudley, W.N. and Lafronza, V.N. (1989) Effects of age and a divided attention task presented during encoding and retrieval on memory. J. Exp. Psychol.: Learning, Memory, Cogn., 15: 1185–1191. Pascual-Leone, A., Amedi, A., Fregni, F. and Merabet, L.B. (2005) The plastic human brain cortex. Ann. Rev. Neurosci., 28: 377–401.
56 Persson, J., Nyberg, L., Lind, J., Larsson, A., Nilsson, L.G., Ingvar, M., et al. (2006) Structure — function correlates of cognitive decline in aging. Cereb. Cortex, 16(7): 907–915. Persson, J., Sylvester, C.-Y.C., Nelson, J.K., Welsh, K.M., Jonides, J. and Reuter-Lorenz, P.A. (2004) Selection requirements during verb generation: differential recruitment in older and younger adults. NeuroImage, 23: 1382–1390. Pfefferbaum, A., Mathalon, D.H., Sullivan, E.V., Rawles, J.M., Zipursky, R.B. and Lim, K.O. (1994) A quantitative magnetic resonance imaging study of changes in brain morphology from infancy to late adulthood. Arch. Neurol., 51: 874–887. Poldrack, R.A., Wagner, A.D., Prull, M.W., Desmond, J.E., Glover, G.H. and Gabrieli, J.D.E. (1999) Functional specialization for semantic and phonological processing in the left inferior prefrontal cortex. Neuroimage, 10: 15–35. Rajah, M.N. and D’Esposito, M. (2005) Region-specific changes in prefrontal function with age: a review of PET and fMRI studies on working and episodic memory. Brain, 128: 1964–1983. Raz, N. (2000) Aging of the brain and its impact on cognitive performance: integration of structural and functional findings. In: Craik, F.I.M. and Salthouse, T.A. (Eds.), Handbook of Aging and Cognition. Erlbaum, Mahwah, NJ, pp. 1–90. Raz, N., Gunning, F.M., Head, D., Dupuis, J.H., McQuain, J., Briggs, S.D., et al. (1997) Selective aging of the human cerebral cortex observed in vivo: differential vulnerability of the prefrontal gray matter. Cereb. Cortex, 7(3): 268–282. Raz, N., Gunning-Dixon, F., Head, D., Rodrigue, K.M., Williamson, A. and Acker, J.D. (2004) Aging, sexual dimorphism, and hemispheric asymmetry of the cerebral cortex: replicability of regional differences in volume. Neurobiol. Aging, 25: 377–396. Reuter-Lorenz, P.A., Jonides, J., Smith, E.E., Hartley, A., Miller, A., Marschuetz, C., et al. (2000) Age differences in the frontal lateralization of verbal and spatial working memory revealed by PET. J. Cogn. Neurosci., 12: 174–187. Reuter-Lorenz, P.A., Marshuetz, C., Jonides, J., Smith, E.E., Hartley, A. and Koeppe, R. (2001) Neurocognitive ageing of storage and executive processes. Euro. J. Cogn. Psychol., 13: 257–278. Reuter-Lorenz, P.A. and Sylvester, C.-Y.C. (2005) The cognitive neuroscience of working memory and aging. In: Cabeza, R., Nyberg, L. and Park, A. (Eds.), Cognitive Neuroscience
of Aging. Oxford University Press, New York, NY, pp. 186–217. Rodrigue, K.M. and Raz, N. (2004) Shrinkage of the entorhinal cortex over five years predicts memory performace in healthy adults. J. Neurosci., 24: 956–963. Rosen, A.C., Prull, M.W., O’Hara, R., Race, E.A., Desmond, J.E., Glover, G.H., et al. (2002) Variable effects of aging on frontal lobe contributions to memory. Neuroreport, 13: 2425–2428. Rypma, B., Berger, J.S., Genova, H.M., Rebbechi, D. and D’Esposito, M. (2005) Dissociating age-related changes in cognitive strategy and neural efficiency using event-related fMRI. Cortex, 41: 582–594. Rypma, B. and D’Esposito, M. (2000) Isolating the neural mechanisms of age-related changes in human working memory. Nat. Neurosci., 3: 509–515. Rypma, B. and D’Esposito, M. (2001) Age-related changes in brain-behavior relationships: evidence from event-related functional MRI studies. Euro. J. Cogn. Psychol., 13: 235–256. Salthouse, T.A. (1996) The processing-speed theory of adult age differences in cognition. Psychol. Rev., 103(3): 403–428. Smith, E.E., Geva, A., Jonides, J., Miller, A., Reuter-Lorenz, P. and Koeppe, R.A. (2001) The neural basis of task-switching in working memory: effects of performance and aging. Proc. Nat. Acad. Sci. USA, 98: 2095–2100. Smith, E.E. and Jonides, J. (1999) Storage and executive processes in the frontal lobes. Science, 283: 1657–1661. Stebbins, G.T., Carrillo, M.C. and Dorman, J. (2002) Aging effects on memory encoding in the frontal lobes. Psychol. Aging, 17: 44–55. Stern, Y., Habeck, C., Moeller, J., Scarmeas, N., Anderson, K.E., Hilton, H.J., et al. (2005) Brain networks associated with cognitive reserve in healthy young and old adults. Cereb. Cortex, 15: 394–402. Tulving, E., Kapur, S., Craik, F.I.M., Moscovitch, M. and Houle, S. (1994) Hemispheric encoding/retrieval asymmetry in episodic memory: positron emission tomography findings. Proc. Nat. Acad. Sci. USA, 91: 2016–2020. Ward, N.S. and Cohen, L.G. (2004) Mechanisms underlying recovery of motor function after stroke. Arch. Neurol., 61: 1844–1848. Ylikoski, R., Salonen, O., Mantyla, R., Ylikoski, A., Keskivaara, P., Leskela, M., et al. (2001) Hippocampal and temporal lobe atrophy and age-related decline in memory. Acta Neurol. Scand., 101: 273–278.
Plate 4.1. Overview of findings from Persson et al. (2006). Top — left: stable (N ¼ 20) and declining (N ¼ 20) longitudinal memory performance in older adults (composite score from three episodic tests). Top — right: fMRI results showing over-activation in right PFC during semantic categorization compared to fixation baseline for participants with stable (bottom) and declining (top) memory performance. Activations are displayed on transverse sections of an anatomical template brain. Bars show average signal change in the right ventral frontal cortex (BA 47). Bottom — left: mean height adjusted volume of the left (top) and right (bottom) hippocampus (in mm3). Error bars show SEM. Bottom — right: results from the DTI analyses show the anterior corpus callosum (genu) outlined on a transverse slice of a fractional anisotropy (FA) image. High signal intensity (brightness) reflects higher FA. Mean FA as a function of longitudinal memory performance is shown in the histogram on the left-hand side.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 5
Cortical reorganization in the aging brain Hubert R. Dinse Institute for Neuroinformatics, Department of Theoretical Biology, Experimental Neurobiology Laboratory, Ruhr-University Bochum, Bochum, Germany
Abstract: Aging exerts major reorganization and remodeling at all levels of brain structure and function. Studies in aged animals and in human elderly individuals demonstrate that sensorimotor cortical representational maps undergo significant alterations. Because cortical reorganization is paralleled by a decline in perceptual and behavioral performance, this type of cortical remodeling differs from the plastic reorganization observed during learning processes in young individuals where map changes are associated with a gain in performance. It is now clear that brain plasticity is operational into old age; therefore, protocols for interventions such as training, exercising, practicing, and stimulation, which make use of neuroplasticity principles, are effective to ameliorate some forms of cortical and behavioral age-related changes, indicating that aging effects are not irreversible but treatable. However, old individuals cannot be rejuvenated, but restoration of function is possible through the emergence of new processing strategies. This implies that cortical reorganization in the aging brain occurs twice: during aging, and during treatment of age-related changes. Keywords: aging; plasticity; cortical maps; cortical processing; perception; behavior; enriched environment; amelioration of aging effects; intervention In fact, we witness a unique restructuring of the aging pattern in the societies of the industrial nations, characterized by an increasing probability to reach old age (Fig. 1). Concomitantly, the probability to suffer from age-related disorders has increased dramatically, indicating an urgent need for a more comprehensive understanding of the different facets of aging. Therefore, the investigation of the aging brain is not only fascinating from the standpoint of how aging affects neural structures, but also vital with respect to the many implications of aging on social disciplines, such as psychology, sociology, health care, and politics in general. However, surprisingly little research efforts are devoted to unravel the diverse aspects of the aging brain. Given the restructuring in the industrial civilizations, the preservation of every-day life competence of aged populations becomes increasingly important. In particular, the maintenance of cognitive
Introduction Aging societies From the very beginning of human civilization, the process of aging appeared to exert a peculiar attraction and fascination. The fear of aging and the associated inevitable vanishing of life quality find their expression in a desire for measures that provide longevity, a yearning frequently captured in fine arts. However, it is only for a few decades that we experience a dramatic increase of life span. Although aging is an old problem, the emergence of such longevity for a substantial portion of the population is a fairly new phenomenon.
Corresponding author. Tel.: +49 234 3225565; Fax: +49 234 3214209; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57005-0
57
58
Fig. 1. Development of the aging structure of the industrialized countries USA, Europe, and Japan (adopted from world population prospects. The 2002 Revision, UN, New York).
and sensorimotor abilities is crucial for maintaining an independent life style at old age. In this context it is of far-reaching consequences to understand whether age-related changes are due to the accumulation of degenerative processes and thus largely irreversible, or whether age-related changes can be understood in the framework of neuroplasticity and subsequent compensatory mechanisms, for which there are possibilities of effective treatment. Few other medical disciplines have to contest similar misconceptions and anecdotal and erroneous knowledge as gerontology. For example, when young and adult people were questioned about their expectancies about the social and psychological conditions at old age, their response pattern revealed a significant underestimation of the quality of life at older age, exemplifying the misconception of aging in the public opinion (Palmore, 1988). According to the ‘‘Berlin study,’’ there exists a clear negative correlation between age and measures of intelligence (Linderberger and Baltes, 1992). However, in spite of this overall correlation, the interindividual variability in the population of people aged between 70 and 103 years is enormous. As a result, despite the negative correlation, the highest rating in intelligence performance is reached by a woman aged 87. Aging theories In the search of understanding of aging processes, many ‘‘aging theories’’ have been developed. Some
of these theories are stochastic in nature and describe aging processes by a probabilistic accumulation of factors that progressively exert deteriorating effects onto the organism. Other theories are deterministic in nature and assume that aging is the consequence of endogenous and/or genetically programmed processes. Among the stochastic theories are ‘‘tear and wear’’ (Pearl, 1924), ‘‘free radicals’’ (Harman and Piette, 1966), ‘‘collagen/cross linkage’’ (Verzar, 1963), ‘‘error and fidelity theory’’ (Orgel, 1963) and the ‘‘immune theory’’ (Walford, 1967). Deterministic theories are known as the ‘‘absolute metabolic scope theory’’ (Rubner, 1908) and the ‘‘cell doubling theory’’ (Hayflick, 1968). While each of these theories can account for only some aspects observed during aging, there is now agreement that aging cannot be explained by a single theory, but instead, must almost be regarded as caused by a multitude of factors. Further evidence that aging must not automatically imply a general decline comes from studies of the so-called ‘‘oldest old,’’ i.e., participants 100 years and older, who characteristically display considerable mental and physical fitness. With few exceptions these individuals report a high degree of subjective wellness that include active participation in social and cultural life. Interestingly, no correlations have so far been established between specific events in their individual life-span history and the amount of vitality at very old age (Perls, 1995). It has been a main desire to be able to interfere with aging processes in order to delay or to ameliorate the impact of age-related changes. In rodents, it is well established that diet and caloric restrictions have significant life-extending effects. It has been discussed whether comparable effects exist in primates and humans (cf. Walford, 1985; Sohal and Weindruch, 1996). A longitudinal study at the University of Wisconsin of the effects of caloric restrictions on longevity and diseases in rhesus monkeys is expected to show results at or about year 2020 (Wanagat et al., 1999). Other evidence suggests that maintained physical and mental exercise are prerequisites for what has been called ‘‘successful aging,’’ although definite answers might not be revealed until the next decades
59
(Rowe and Kahn, 1987; Elward and Larson, 1992; Kramer et al., 1999). New approaches that are based on neuroplasticity and use specific training and stimulation protocols are currently being tested. These approaches provide new intervention techniques with the overall goal of ameliorating age-related decline in neural function (Dinse et al., 2005, 2006). The goal of these techniques is not to extend life expectancy, but to achieve healthy aging for as long as possible. Myths and facts An editorial with the title ‘‘How old is old?’’ addressed a most crucial question in aging research (Coleman, 1989) and discussed a survey of articles published in the journal Neurobiology of Aging in which rodents had been defined as ‘‘aged’’ or ‘‘old’’ at ages that varied from 18 to 53 months. One article describes changes at 7 months of age as an ‘‘early manifestation of aging.’’ Without further comments, this highlights the imperative need for multiple time points in aging studies (Coleman et al., 1990), not only in respect to rodents, but in all types of aging studies. The acknowledgment of a large heterogeneity and interindividual variability in brain structure was an important step in the development of gerontology. This variability seems to be a general characteristic and has been observed not only in elderly humans, but also in aged primates and rodents. A large degree of variability has been shown to be present at any possible level of description and for any possible variable and parameter, making studies of aging complex (Rapp and Amaral, 1992; Gallagher and Rapp, 1997). ‘‘The concept that cortical neurons are lost with age and that this is the basis for cognitive decline is so embedded in our culture that when someone elderly is a little forgetful it is often said that ‘He/ she is losing his/her neurons.’’ (cited from Peters et al., 1998a). In fact, the hypothesis that there is a significant loss of neurons during normal aging dates back to Brody (1955). However, recent studies revealed a remarkable degree of constancy of the number of neurons in the brain (Flood and Coleman, 1988; Morrison and Hof, 1997; Peters et al., 1998a). Estimations of the number of neurons
in the brain are hampered by technical problems and by the large individual variations in the number of neurons in the brain. For example, the size of area 17 in the human or primate brain can vary by a factor of three among individuals (Stensaas et al., 1974; Peters et al., 1998a, b). This enormous variability raises doubt about the significance of a loss of up to 10% of neurons, when individual variations can be more than 100%. Rapp and Gallagher (1996), in a study of neuron counts in representative samples of the entire hippocampus of behaviorally tested rats, showed no age-related loss of neurons, even in the animals that had the greatest age-related behavioral impairments. The question regarding the presence of detectable structural counterpart to the behavioral impairments that occurs during aging thus remains. Cellular and molecular changes during aging It had been suggested that only certain subpopulations of neurons might undergo losses and that there may be reductions of specific sets of dendritic spines demonstrated to occur in the prefrontal cortex during aging (Peters et al., 1998b). Molecular shifts in morphologically intact circuits have been described in the dentate gyrus in old monkeys (Gazzaley et al., 1996). Old rats with spatial learning deficits displayed significant reductions in synaptophysin immunoreactivity in CA3 of the hippocampus relative to either young controls or age-matched animals with preserved learning (Smith et al., 2000). More recently it was shown that behavioral performance correlates with N-methyl-D-aspartate (NMDA) receptor-dependent long-term depression (LTD) (NMDARLTD) in young animals and with non-NMDARLTD in old animals. NMDAR-LTD is reduced in old rats with learning deficits, but age-matched unimpaired animals show increased nonNMDAR-LTD. This suggests that high-functioning old rats maintain the ability to generate LTD, but do so by different mechanisms than those used by young adults (Lee et al., 2005). Thus, while hippocampal information processing can deteriorate during normal aging without detectable significant neuronal loss, these findings imply that a
60
circuit-specific pattern of variability in the molecular organization of the hippocampus is coupled to individual differences in cognitive function during normal aging. Aging of sensorimotor behavior and cortical map plasticity Decline of cognitive and sensorimotor abilities is clearly associated with aging but the magnitude and the time of onset of age-related changes differ considerably between individual animals. While there is a growing body of information about agerelated changes at the cellular and molecular levels, little is known about how aging affects the way in which neurons process and integrate sensory information, and therefore little is known about how aging affects the functional representation of sensory information in human cerebral cortices. While human aging processes influence all stages of sensorimotor processing, it is not known how cortical representations of somatosensory input are affected by normal, nonpathological aging, and how cortical changes affect tactile perception and sensorimotor functions. It is well documented that both the human and nonhuman somatosensory systems contains highly ordered maps of the body surface. These maps are not fixed, but subject to modification through expression of neural plasticity. Imaging studies have shown that continuous and long-lasting practice of specific sensorimotor functions resulted in expansions of the respective cortical areas, as described for blind Braille readers and musicians (PascualLeone et al., 1993; Elbert et al., 1995; Pantev et al., 1998; Sterr et al., 1998). These studies corroborated earlier animal studies, which demonstrated that training-induced enlargement of cortical maps as a result of expression of cortical plasticity (Recanzone et al., 1992, 1993; Dinse and Merzenich, 2002). There is often a direct proportionality between the amount of cortical reorganization and the individual improvement of performance and skills (Recanzone et al., 1992; Pascual-Leone et al., 1993; Xerri et al., 1994; Elbert et al., 1995; Buonomano and Merzenich, 1998; Dinse and Merzenich, 2002). The opposite effect has been demonstrated to occur from disuse such as after
immobilization through cast wearing (Liepert et al., 1995; Ragert et al., 2003). After days to weeks of wearing a cast the respective cortical representations shrink and that has been associated with decreased perceptual abilities. After removal of the cast these changes reversed and functions returned to the baseline conditions. These data thus demonstrate use-dependent neural plasticity applies to situations of severely reduced use. These findings regarding different forms of neural plasticity suggest that at least some forms of age-related changes of somatosensory processing and behavior can be explained in the framework of use-dependent plasticity. Taking advantage of new technologies such as magnetic resonance imaging (MRI) makes it possible to analyze gray matter density in healthy human individuals. In participants in an MRI study with age from 7 to 87 years, the density of gray matter decreased significantly with age. The decrease in dorsal frontal and parietal association cortices occurred in a nonlinear way, most rapidly between the age of 7 and 60 years (Sowell et al., 2003). The age effects were inverted in the left posterior temporal region, where the increase of the density of gray matter continued up to age 30 and then rapidly declined. Visual, auditory and limbic cortices where fibers are known to become myelinated early, showed a more linear pattern of aging. Similarly, data from longitudinal measures of five-year change in the regional brain volumes in healthy adults revealed substantial shrinkage of the caudate, the cerebellum, the hippocampus, and the association cortices, with minimal change in the entorhinal and none in the primary visual cortex (Raz et al., 2005). Both studies showed that age-related brain volume changes are very localized with large difference in time course and time of onset. Cortical reorganization in the aging brain occurs twice: during aging, and during treatment of agerelated changes In the following we provide a summary of results obtained from studies of a commonly used animal model of aging. The aim of the research was to explore the nature of age-related changes of tactile
61
perception and cortical reorganization in healthy elderly and the possibilities for ameliorating the age-related changes. It emerges from these studies that aging causes major restructuring of the brain and reorganization of its function. Moreover, it was demonstrated that intervention protocols using the principles of neural plasticity can ameliorate many forms of age-related deteriorations. These positive effects are based upon cortical remodeling, which leads to the emergence of what may be called the third brain: A reorganized aged brain that in many aspects differs from an untreated old brain (see p. 75).
1989a, b; Stoll et al., 1990; Spengler et al., 1995). Old rats use the more distal parts of the heels in addition to the digits and pads for locomotion used by younger rats. This leads to reduced sensory stimulation of the hindpaw (cf. Fig. 2). This sensorimotor impairment is mainly restricted to the hind-limb. We hypothesized that the use of the forelimbs is intact because they are engaged in cleaning, feeding, and looming behavior throughout life. However, despite the behavioral intactness of the forelimb, walking in old rats is slowed down and displays many compensatory changes such as increased step frequency and increased duty cycle (Schulze and Dinse, unpublished).
Rat aging Cortical reorganization in rats during normal aging Rats are convenient for studies of aging because they age within 2–3 years. We used male hybrid Fischer 344 Brown Norway (FBNF1) rats to study age-related changes of sensorimotor cortical representation (maps) and information processing using single- or multiunit recordings. The FBNF1-strain is specifically recommended for aging research by the National Institute of Aging (http://www.nia.nih.gov/ResearchInformation/ScientificResources). Animals were kept in standard housing environments to study age-related changes of somatosensory cortex. The 50% probability of survival in an aging colony is approximately 34.5 months for male FBNF1 (Sprott, 1997). If not otherwise stated, the FBNF1 animals were 29 months of age or older. A particular advantage of the FBNF1 animals is that they age in a rather healthy way, and that the gain in body weight during aging is small with 400–450 g in young vs. 500–530 g in old rats.
Age-related changes of rat sensorimotor behavior Deterioration of walking, particularly of the hindlimbs are the characteristic impairment of the sensorimotor state in old rats, which typically slide and drag their limbs due to insufficient elevation of the feet (Ingram, 1988; Schuurman and Traber,
Age-related changes of rat sensorimotor cortex organization The cortical receptive fields (RFs) in young adult animals (3–6 months of age) to cutaneous stimulation of the hindpaw usually are small comprising only single or neighboring digits and pads, while RFs on the proximal part of the paw represent larger skin areas. In contrast, cutaneous RFs of the hindpaw in old rats were significantly enlarged compared to adults (about 200% on average). RFs in old animals represented multiple digits and pads, and the RFs of the proximal parts of the paw were much larger than in younger adults (Fig. 3). Since the skin surface is about the same in young and old rats, the enlargement of the RF means that RFs overlap to a greater extent in old than in younger adult rats (Spengler et al., 1995). Mapping the hindpaw representation on the surface of the SI cortex of old rats revealed several differences from younger rats: First, because the size of the RFs increased, so did the cortical point-spread function (Fig. 4). The average cortical area excited by tactile stimulation in old rats was 0.051 mm2+/– 0.018 (s.d.) compared to 0.016 mm2 +/– 0.011 (s.d.) in young control rats (two-sided unpaired Student’s t-test, po0.001). Secondly, the overlap between zones of activity evoked by stimuli to different skin sites increased as well contributing to an overall loss of fine grained topography and the topographic order of the hindpaw map was severely deteriorated.
62
Fig. 2. Comparison of sensorimotor performance of young and of old rats. Footprints of the hindpaw as shown on the left are typically found in young rats serving as control group. This walking pattern is correlated with distinct and selective sensory inputs where single digits and pads are placed on the ground. Prints depicted in the middle and on the right are typical for old rats. The footprints shown in the middle are correlated with an intermediate state of sensorimotor performance; those on the right are correlated with multiple and diffuse inputs, sometimes even from the dorsal side of the paw; when the foot is twisted and dragged behind the body (reprinted from Spengler et al. (1995) with permission from Lippincott Williams & Wilkins).
To visualize the effects of aging on the topography of the underlying cortical representations, we reconstructed somatosensory maps using a computer interpolation algorithm based on a linear least square approximation of sampling coordinates of penetration sites and corresponding receptive field centers. (Reconstructions of a
cortical hindpaw representation are shown in Fig. 3 for a young and for an old rat where cortical topographies are represented as a regular lattice within somatosensory cortex.) Thirdly, the total amount of cortical territory devoted to the representation of the hindpaw was reduced by approximately 30% (Spengler et al., 1995;
63
Ju¨rgens and Dinse, 1997) in old rats. The changes in walking behavior of the hind-limbs in old rats occurred in parallel with a large decline of the functional organization of the somatosensory cortex. Studies of age-related maps of the motor cortex representing hind-limb muscles using intracortical microstimulation revealed profound age-related changes consisting of large reductions of the size of the maps, a loss in muscle topology, an increase in thresholds for evoking movements, and prolonged latencies of the electromyogram (EMG) responses (Dinse et al., 2001). The results indicate that the motor system of the hind-leg is affected by age in a similar way, as are sensory cortices.
Small age-related peripheral changes It is important to consider whether central changes are a mere consequence of age-related alterations developing in the periphery, and whether age-related changes of cortical sensory processing are a simple reflection of changes occurring already at the level of mechanoreceptors. We therefore investigated the effects of aging on rapidly (RA) and slowly adapting (SA) cutaneous mechanoreceptors by means of single fiber recordings and evoked sensory nerve action potentials (NAPs) of the hindpaw of the N. plantaris in adult and old rats. Recordings of NAPs revealed similar shapes and amplitudes in all animals of all age groups. In old rats, conduction velocities were reduced by approximately 15%, (differences were significant, p ¼ 0.001) and the number of SA units was reduced. However, there were no differences in RF size and in threshold between old and adult animals (Reinke and Dinse, 1996). Evidence for a similar lack of age-related effects at peripheral levels comes from studies of the monkey retina. Stereological procedures used to compare the densities, numbers, and soma sizes of retinal ganglion cells in young adult and old rhesus monkeys revealed no noticeable changes with age (Kim et al., 1996). Studies of the electroretinograms showed only modest age-related prolongation of the latencies of the retinal response and decrease of the amplitude of the electroretinogram (Trick et al., 1986; Porciatti et al., 1992).
Lack of comparable age-related changes in the rat forepaw system — evidence against global breakdown of function If age-related changes in cortical maps are due to degeneration, it would be expected that concomitant changes would occur in the representation of fore- and hindpaws. If the age-related cortical changes of the hindpaw system are in any way related to the behavioral state, the cortical forepaw system would be expected to have less age-related deterioration than that of the hindpaw. Studies have shown no similar increase in RF size in the cortical representation of the forepaw in old animals as seen for RF of the hindpaw (Fig. 3). Even in the oldest animal tested (43 months) there were no indications of changes in the size of the RF of the forepaw. The lack of changes in the RF for the forepaw indicates that there is no global breakdown of function at old age. Age-related prolongation of neuronal response latencies — evidence for global degeneration The conduction velocity in fiber tracts decreases during aging mainly due to demyelinization (Verdu et al., 2000; Peters, 2002). Given the robustness of the cortical forepaw representation in terms of RFs and map size during aging, it was therefore of interest to investigate whether latencies of cortical forepaw neurons remained similarly unaffected. Compared to adult controls, the latency of the response from neurons in the somatosensory cortex to stimulation of both the fore- and the hindpaw in old animals were significantly prolonged in the range of 30% (po0.001 for neurons recorded in the fore- and the hindpaw representation) (Ju¨rgens and Dinse, 1995, 1997). Degenerative vs. plastic-adaptive changes If the observed age-related changes were caused by degeneration, it might be expected that similar changes occurred in the cortical representation of both the fore- and hindpaws. However, the RFs of the cortical forepaw representation of old animals were not noticeably different from that of younger animals (Ju¨rgens and Dinse, 1997; Godde et al., 2002) nor were cutaneous RFs of the forepaw of rats 24–28 months of age different from those of younger adults (6.5–8 months) (Coq and Xerri,
64
a.
c.
d.
b.
65
2000, 2001). The sensorimotor behavior of the forelimbs were largely unaffected by age, presumably because the forepaws are used in feeding and cleaning behavior. These results imply that agerelated changes can be regionally specific, and the results implicate a link between age-related neural changes and specific behavioral alterations that occur during aging, extending the concept of usedependent plasticity to old age. Age-related changes in other cortical systems and modalities While reorganization of sensorimotor cortices during aging has been clearly demonstrated, the nature of age-related cortical alterations in other sensory modalities are less clear. For example, little age effects were reported in a series of papers for primary visual cortex of monkeys 24 years of age and older, neither in terms of cell loss, nor functionally (Kim et al., 1997; Peters et al., 1997, see also Spear, 1993). More recently, evidence for a significant degradation of orientation and direction selectivity in old macaque monkeys was described (Schmolesky et al., 2000). According to this study, the decreased direction selectivity of cells in old animals was accompanied by increased responsiveness to all orientations and directions as well as an increase in spontaneous activity (Schmolesky et al., 2000). The authors suggested that the decreased selectivity and increased excitability of cells in old animals might be attributable to an age-related degeneration of intracortical
inhibition. In contrast, the percentage of neurons in the auditory cortex of old rats that showed direction preference of FM sweep was not different from young animals (Mendelson and Ricketts, 2001). Age-related changes beyond cortical maps — temporal processing Repetition of stimuli can alter the behavior of cortical neurons as compared to presentation of a single stimulus isolated in time (Gardner and Costanzo, 1980; Lee and Whitsel, 1992; Tommerdahl et al., 1996, 1998; Buonomano, 1999). This means that response properties to repetitive stimulation are different from the response to solitary stimulation. This phenomenon is often referred to as short-term plasticity (Zucker, 1989; Varela et al., 1997; Buonomano, 1999) or paired-pulse suppression (Castro-Alamancos and Connors, 1996). Following repetitive stimulation with trains of tactile stimuli of variable interstimulus intervals (ISIs – 30–1000 ms) of the hindpaw (10 stimuli, intertrial pause of 5 s) large impairment of coding of such stimuli were observed in old rats as compared to young controls (Ju¨rgens and Dinse, 1995). While SI neurons of both young and aged animals can follow trains of tactile stimuli of ISIs of 200 ms, the response to stimuli of shorter ISIs decreased more in old than in young animals. The neurons of old animals could barely follow stimuli with ISIs of 30 ms (Fig. 5).
Fig. 3. Specific effects of age on receptive fields of the hindpaw recorded in somatosensory cortex of aged rats. In addition, representative examples of behavioral changes of walking pattern derived from footprint analysis are shown: (a) young, control animal; (b) old animal. On the left: hindpaw; on the right: forepaw. Note selectivity of walking impairment restricted to hind-leg. At bottom of (a) and (b) are examples of receptive fields recorded in the hindpaw representation ((a) and (b), left) and in the forepaw representation ((a) and (b), right) in a young (a) and an old animal (b). Age-related changes are limited to the behaviorally impaired extremity. To visualize the effects of aging on the topography of the underlying cortical maps, we reconstructed somatosensory maps using a computer-based interpolation-algorithm based on a linear least square approximation of sampling coordinates of penetration sites and corresponding receptive field centers. Reconstructions of a cortical hindpaw representation are shown for a control (c) and for an old rat (d). Left: examples of cortical topographies represented as a regular lattice within somatosensory cortex. Middle: the extrapolated cortical representation of a schematic and standardized drawing of the hindpaw. Dashed lines indicate horizontal, solid lines the vertical components of the lattice. One square of the lattice represents 1 mm2 skin area. Diamonds indicate penetration sites; squares give the interpolated RF centers. Dotted lines give the deviation between them. Right: back projection of the regular lattice of the cortical map onto the hindpaw. Squares give the interpolated, asterisks the measured RF centers. One square of the lattice represents the skin portion that is represented by 0.01 mm2 cortical areas. According to these reconstructions, maps of the hindpaw representation recorded in old animals, which are characterized by a selective impairment of the hind-limbs, show a dramatic distortion of their representational maps and a loss of topographic order (modified from Spengler et al., 1995; Ju¨rgens and Dinse, 1997b). See Plate 5.3 in Colour Plate Section.
66
Fig. 4. Local field potential (LFP) maps constructed on the basis of LFP recorded by multiple penetrations showing the spatial activity distributions evoked by tactile stimulation in the hindpaw representation of the primary somatosensory cortex of a young (left) and an old rat (right). Shown are activity distributions following stimulation of a digit. Normalized LFP-amplitudes are color-coded and projected onto the cortical area of the hindpaw representation reconstructed from [x, y] coordinates of penetration sites. Scale bar: 1 mm; each plot is oriented with rostral left and lateral up (reprinted from Spengler et al. (1995) with permission from Lippincott Williams & Wilkins). See Plate 5.4 in Colour Plate Section.
Treatability of age-related changes in the rat
improve cognitive function (Rosenzweig and Bennett, 1996), facilitate recovery from injury or stroke (Johansson, 2000), and prevent age-related decrease in synaptic density in the aged brain (Saito et al., 1994). Such enriched environments have also been shown to increase brain weight (Cummins et al., 1973), cortical thickness (Diamond and Connor, 1982), dendritic arborization (Connor et al., 1982), and neurotrophic factors (Mohammed et al., 1993). It is noteworthy that housing animals under enriched conditions can induce neurogenesis (Kempermann et al., 1997). More recently, beneficial effects have been described for the auditory system from housing rats in an acoustically enriched environment (Engineer et al., 2004).
Assuming that at least certain aspects of age-related cortical reorganizations are caused by expression of neural plasticity, it should be possible to reverse these changes by treatment using protocols activating appropriate forms of neuroplasticity. Housing animals in an enriched environment that exert behavioral challenges have beneficial effects on a wide range of morphological, molecular, and physiologic features of the brain. Enriched environments that target sensorimotor modalities (Dinse, 2004) have been shown to
Effects of enriched environment on aging rats — behavior and cortical organization Beneficial effects on cortical forepaw neurons had been reported for animals that were kept in enriched conditions for their entire life (Coq and Xerri, 2001). We have addressed the question whether age-related changes can be affected through enriched housing even after they developed. Rats at an age of 26–29 months needed only to be exposed to enriched environments for a few months to regain nearly normal walking and
Studies of the responses from single cells in the auditory cortex of young and old rats in response to frequency-modulated (FM) sweeps showed that the majority of cells in young rats responded most vigorously to fast and medium rates of changes in frequency of the sounds while most units in old animals responded best to sounds the frequency of which was changed slowly (Mendelson and Ricketts, 2001). These results demonstrate that aging has a pronounced effect on not only cortical maps and receptive fields but also processing temporal information.
67
Fig. 5. The effect of aging on temporal sequence representation was investigated using trains of 10 tactile stimuli of variable interstimulus intervals (ISIs). Shown are examples of poststimulus time histograms (PSTHs) recorded in a young rat (a) and an old rat (b). ISIs used were 200, 50 and 35 ms. Bin size is 1 ms, time on abscissa as indicated, and neural activity in spikes per bin on ordinate. Each PSTH gives the response accumulated over 32 trials; pause between each single trial was 5 s. For a slow repetition rate of tactile stimuli, neurons recorded in young and old rats follow truthfully each stimulus as indicated by about the same peak activity evoked by each stimulus. While neurons recorded in young animals are still able to represent the sequence of stimuli delivered at an ISI of 50 ms, there is a significant deterioration in the ability to follow this sequence in the neuron recorded in an old rat. This failure to represent fast sequences becomes even more dramatic at an ISI of 35 ms. In addition to massive changes of topography developing during aging (cf. Fig. 2), there is also a significant deterioration of temporal processing abilities, which are also correlated to the behavioral status of the hind-limb as the temporal deficits can be ameliorated by housing the animals in an enriched environment that ameliorates the hindlimb impairment (modified from Ju¨rgens and Dinse, 1995; Churs et al., 1996).
68
sensorimotor behavior of the hind-limbs. The typical age-related enlargement of RFs of the hindpaw was almost eliminated, and input sequence representation largely restored (Churs et al., 1996). Old rats exposed to enriched environment under the same schedule showed nearly complete recovery from age-related functional shrinkage of cortical motor territory typically found in animals housed under standard conditions (Dinse et al., 2001). The age-related prolongation of the latencies of cortical responses and of EMG responses remained unaffected. The decrease in the thresholds of the responses from motor cortex neurons, which normally occur with age, was unaffected by the enriched housing (see also p. 76).
Effects of enriched environment on aging rats — cellular changes Lipofuscin and gliosis increase with progressing age (12–36 months) in functionally characterized cortical areas, and area-specific loss of perineuronal nets occurs in the somatosensory cortical representation of the hind-limbs (Fig. 6). The accumulation of lipofuscin and increased gliosis, the loss of perineuronal nets, and the reduction of nonphosphorylated neurofilament H, which normally occur with age, were reduced or prevented by housing the animals under enriched environmental conditions between 33 and 36 months of age. Reduction of astrocytosis (by 20%) coincided with a reduction in the loss of extracellular matrix
Fig. 6. Schematic drawings of cortical sections +0.7 and 0.8 mm relative to Bregma illustrating the lipofuscin accumulation for animals of different age groups. FL, forelimb area; HL, hind-limb area; Fr1, Fr2 motor cortex according to Zilles (1985). Note the dramatic increase of lipofuscin accumulation with increasing age, particularly in the hind-limb area (modified according to Hilbig et al., 2002a).
69
components involved in forming the glia-neuroninterface demonstrates. These results indicate that the aging cortex can retain its potential for functional plasticity (Hilbig et al., 2002a), but the sensitivity to aging processes between the fore- and hind-limb is different regarding their SI representation. Changes in the expression of the nitric oxide synthase (NOS) isoforms have been implicated in age-related neurodegeneration. Immunohistochemical examination of the age-dependent NOS-I and NOS-II expression in rats at 3, 12, 24, and 36 months of age as well as in 36month-old rats, which were housed in an enriched environment for their last 3 months of life, revealed a significant decrease in NOS-I expression with aging while NOS-II was increased in rats of 36 months of age. NOS-II expression in old rats that were kept under enriched environmental conditions was much reduced. These results indicate that both NOS-I and II may contribute to agerelated degenerative processes, but in contrast to NOS-I, the age-dependent changes of NOS-II are reduced or even reversed by environmental stimulation (Hilbig et al., 2002b). Effects of pharmacological intervention on aging rats During aging, the cellular Ca2+ homeostasis is impaired due to an elevated influx through voltagegated Ca2+ channels, which leads to severe cytotoxic effects. Nimodipine blocks selectively the Ltype voltage-dependent Ca2+ channels (Godfraind and Govoni, 1995), which are important for maintenance of the neuronal Ca2+ homeostasis (Miller, 1987). Schuurman and Traber demonstrated that a long-term treatment with nimodipine can delay the typical deterioration of walking behavior in old rats. Animals that displayed an impairment of walking at 24 months of age had significant reduction of their sensorimotor deficits after receiving 6 weeks of nimodipine treatment (Schuurman et al., 1987; Schuurman and Traber, 1989a, b). We studied the effects of long-term administration of nimodipine on RFs in the fore- and hindpaw representations of primary somatosensory neurons of aged Wistar rats 23–31 months of age, starting the nimodipine treatment at 19 months of
age. These studies confirmed the beneficial effects of nimodipine on the overall state and walking pattern of the hind-limbs, and in addition, optical imaging revealed a normal layout of cortical hindpaw maps and reversal of the age-related increase of RF size to the size found in normal adult animals. This reversal was restricted to a period of 5.5–9 months of treatment corresponding to an age of 24.5–28 months. The RFs of the forepaw were not affected by administration of nimodipine. These results thus demonstrate that nimodipine can delay typical age-related changes of the hindpaw representation in a similar way as it affects the sensorimotor state of the hind-limbs. Nimodipine did not affect the age-related prolongation of the latencies of the responses from cortical cells representing the hindpaw or those representing the forepaw. These findings indicate a specific mode of action of nimodipine and an overall effectiveness of this kind of intervention in terms of decreasing age-related cortical changes (Berkefeldt et al., 1996; Ju¨rgens and Dinse, 1997). Recently, de Rivera et al. (2005) studied the effect of dietary supplements that were rich in antioxidants (blueberry phytochemicals) on temporal processing in primary auditory (AI) cortical neurons in old rats that had been placed on either a blueberry-supplemented or control diet 2 months prior to the physiological recordings. The results showed that most cells from the blueberry-fed rats responded vigorously to fast FM sweeps, similar to that of cells in young rats, but most cells recorded from the control rats showed a preference for slow FM sweep rates as described above (p. 66). These results indicate that age-related changes in temporal processing in AI may be reversed by dietary supplementation of blueberry phytochemicals (de Rivera et al., 2005), offering alternative options for slowing age-related changes. Hypothesis on rat aging Experimental data support the hypothesis that the primary cause of motor decline in old rats is mechanical in nature and consists of muscle atrophy resulting from reduced use. The beneficial effects of administration of nimodipine may be caused by unspecific factors, such as an elevation of the level of arousal, which in turn enhance motor activity.
70
This hypothesis is compatible with the concept of use-dependent plasticity whereby increased motor activity increases sensory stimulation that modifies cortical representations. The difference in the agerelated impairment of sensorimotor performance of the fore and hindpaws could be attributed to maintained intensive use. An alternative hypothesis may be that synaptic efficiency, and hence cortical information processing capacities, are progressively impaired during aging and that failure in maintaining Ca2+ homeostasis affects cellular and network properties such as inhibitory mechanisms that regulate RF size. In the case of nimodipine treatment, the Ca2+ overload developing during aging is prevented. This could explain why RFs of the hindpaw are not enlarged after nimodipine. Maintained use of the forepaws may counteract a Ca2+ overload, and therefore RF size of the forepaw remains unaffected by age. Combined, the results from the studies of the effect of an enriched environment and of administration of nimodipine show that age-related changes may be treatable. The results of these studies clearly imply that age-related changes can be reversed even if they have developed. This may also be taken as a sign that the properties of these changes are not degenerative by nature (Godde et al., 2002; Dinse, 2005).
Human aging Cortical reorganization in healthy humans during normal aging The studies in the rat data discussed above have inspired studies in humans. In these investigations, psychophysical studies were combined with noninvasive imaging studies (in cooperation with Dr. Martin Tegenthoff, Department of Neurology at the Ruhr-University). Assuming that age-related changes in humans develop in a similar way as in the rat, we wanted to study the implications of impaired cortical sensorimotor organization for tactile perception. Naturally, the different lifespan of rats and humans must be taken into account in such studies.
Earlier studies have found that the functional organization of the somatosensory cortex in humans is linearly related to tactile discrimination abilities (Pleger et al., 2001). We therefore assumed that ‘‘normal’’ cortical processing of sensory information is required for ‘‘normal’’ perceptual and motor performance, and that large changes in the cortical organization occurring in elderly humans lead to severe perceptual impairments. Studies on spatial (2-point) discrimination in elderly individuals revealed significantly impaired discrimination (Stevens, 1992; Woodward, 1993; Wohlert, 1996; Dinse et al., 2006). The impairment on the toe was much greater than that on the fingertip (400% deterioration of acuity on the foot as compared to 130% on the finger) (Stevens and Choo, 1996). These results are thus in agreement with results from the rat, and also in humans these differences may be caused by difference in the use of the foot and the hand.
Analysis of sensorimotor and cognitive functions in healthy elderly We have accumulated a large database on 2-point discrimination performance in young individuals in the context of tactile learning (Pleger et al., 2001, 2003; Dinse et al., 2003, 2005; Ragert et al.; 2004; Tegenthoff et al., 2005). Using that database as a reference we studied sensorimotor and cognitive functions in healthy elderly individuals including absolute touch threshold, localization performance, 2-point discrimination, and tactile object recognition. We also assessed cognitive performance using the Raven Progressive Matrices test, a nonreading, nonlanguage-based measure of fluid intelligence. Aged individuals were recruited by poster announcements in senior residences. All participants in the study were right-handed and underwent neurological examination, and were without neurological symptoms and in good physical condition. The participants were divided into three age groups (group 50: 45 55 years, group 70: 65–75 years, and group 80: 75–85 years). Young adults served as control (20–30 years). Eligibility criteria for participating in the study group were
71 lucidity, independence in activities of daily living, and absence of motor and sensory handicaps and of any impairment due to arthritis or other causes of joint immobility. Individuals with visual or hearing loss, former or actual diseases of the central or peripheral nervous system or individuals who took central nervous acting medication were excluded. Cognitive abilities were assessed using the ‘‘Mini Mental State Examination.’’ Only individuals scoring 27–30 out of 30 indicative of ‘‘no dementia’’ were included in the study.
Touch thresholds increased from 0.257 0,013 mN in young adults to 1.1671,71 mN in the intermediate age group, to 2.8272.65 mN in the group-70 and 2.0571.66 mN in the group 80 (Kleibel, Kalisch, Tegenthoff, Dinse, unpublished data). The results of 2-point discrimination tests confirmed previous findings showing a significant deterioration of performance with increasing age, beginning at the age of 50 years (Fig. 7). No significant correlation (Pearson) between the
Fig. 7. Tactile 2-point discrimination thresholds of the tip of the right index finger as a function of age (total of 120 participants). According to univariate ANOVA (with age F(2,118) ¼ 347.785; po0.001) the differences across the three age groups were significant. After coactivation (pink symbols), thresholds of the coactivated participants (young control group and group of elderly) were significantly reduced. Coactivationinduced improvement in the group aged 66–86 was several fold stronger in magnitude as compared to the young subject. As a result, after coactivation thresholds of the elderly resembled those found in the participants 47–55-years old (reprinted from Dinse et al. (2006) with permission from Wiley-Liss, Inc.). See Plate 5.7 in Colour Plate Section.
individual touch threshold and 2-point discrimination scores was present, indicating that both parameters change independently, and that touch threshold is not a predictor for tactile acuity. The effect of aging on tactile object recognition was studied by having the participants identify small cubic objects without visual support using a method modified from Newell et al. (2001). The results showed large effect of aging on object recognition. The number of errors and the time to fulfill the object-classification task increased significantly with age. Interestingly, the decline in object recognition was highly gender specific: Female participants had a much greater impairment than age-matched male participants. No gender specific trends were present within the young individuals (Kalisch and Dinse, submitted). Age-related changes of human SI cortical maps The studies discussed above showed that the representation of the hindpaw on the SI cortex undergoes considerable age-related changes. In humans the representation of the hands on the primary somatosensory cortex undergoes continuous adaptation through expression of neural plasticity arising from environmental input, with altered use or injury leading to substantial reorganization of body surface representation with regards to the size, extent, and position. Using SEP mapping in combination with electric source localization, we investigated the influence of healthy aging on the human cortical hand representation in relation to tactile performance. The results showed that the hand representations within S1 on both hemispheres in elderly between 60 and 85 years of age were substantially enlarged as indicated by an increase of the distance between the dipole of the index and the little finger by approximately 40%. Correlation analysis revealed a significant positive relation between perceptual performance (2-point discrimination) and cortical map size (difference between dipoles for index and little finger). These results indicate that normal aging affects the cortical organization of the hand representations within SI, and that the enlargement of the cortical hand representation is paralleled by impaired tactile acuity (Kalisch et al., submitted).
72
These results are of interest for two reasons: First, as discussed above (p. 60) map expansion observed in young individuals during learning is usually associated with a gain in performance. In contrast, the age-related changes in cortical maps and in tactile abilities changed in opposite directions (enlarged maps were associated with impaired performance). These changes may therefore be assumed to reflect different forms of cortical reorganization. Second, studies in rats showed that the cortical representation of the forepaw had little age-related changes, while the cortical territory representing the hindpaw decreased noticeably with age and the hindpaw motor map became extended. This means that animal and human data for cortical reorganization are contradictory. There are several possibilities explaining these differences such as differences in accumulated years or reorganization to compensate for slowing down of conduction and processing speed. Treatability of age-related changes in human elderly Given that aging in the rat is subject to various forms of interventions, it is of interest to explore if similar strategies are also effective in humans. It is well known that extensive training and repeated exercise improves perceptual and motor skills, and it is assumed that these improvements are at least to some extent a result of expression of neural plasticity (Recanzone et al., 1992; Pascual-Leone and Torres, 1993; Elbert et al., 1995; Pantev et al., 1998). Therefore, the typical approach to ameliorate age-related changes is to subject elderly to intense schedules of training and practicing, and there is no doubt about the effectiveness of such intervention even at old age (Bock and Schneider, 2002; Sawaki et al., 2003; Floel et al., 2005; Kornatz et al., 2005; Smith et al., 2005). However, since many elderly suffer from restricted mobility, development of additional and alternative approaches that could supplement, enhance, or even replace conventional training procedures would be advantageous. In search for enriched environments for elderly Given the high degree of efficacy of housing old rats under enriched environmental conditions, it is
natural to search for equivalent enriched environments for elderly humans. It was recently reported that dancing evokes many beneficial effects (Federici et al., 2005; Jacobson et al., 2005). Nordic walking or other forms of exercising might offer valuable alternatives (cf. Elward and Larson, 1992; Kramer et al., 1999; Jacobson et al., 2005). Both approaches have in common that they emphasize the role of physical exercise. They differ in that dancing involves a much broader scope of factors beyond pure exercising such as social interaction, divided attention capacity, navigation in highly populated spaces, and following musical rhythms. Significant improvement in balance in the dance group was demonstrated at the end of such a program (Federici et al., 2005). Tango dancing improves several balance measures related to moving in confined spaces, and for complex walking tasks (Jacobson et al., 2005). Adherence to tango was higher than for the walking group (1 vs. 4 dropouts), and out of the 25 individuals who participated in the study, 60% are still participating in tango classes, making this as a feasible alternative to other activity programs. Amelioration of the effects of aging through passive stimulation protocols As an alternative approach to training, we have recently introduced tactile coactivation to control and to improve tactile performance in humans on a time scale of only a few hours. Coactivation consists of ‘‘passive’’ and therefore unattended stimulation, which enforces localized activation patterns in the brain. A major advantage of coactivation is that it is applied passively and thus does not require active cooperation of the participants. Coactivation closely follows the principles of Hebbian learning, which states that synchronous neural activity drives plastic changes. The Hebbian nature of coactivation was demonstrated in a control experiment, in which only a very small skin area was stimulated (no coactivation). This protocol caused neither changes of thresholds nor changes in cortical activation, implying that ‘‘co’’activation is indeed crucial (Pleger et al., 2003; Ragert et al. submitted) Coactivation can induce improvement of tactile perceptual performance in parallel to cortical
73
reorganization in young individuals (Godde et al., 2000; Pleger et al., 2001, 2003; Dinse et al., 2003, 2005, 2006; Ragert et al., 2004). We used tactile coactivation as an alternative intervention to interfere with the aging-related impairment of tactile perception and demonstrated that plastic changes in neural organization are involved in the age-related decline in sensory performance, which is typical for elderly humans, and therefore these changes can be ameliorated through brief periods of this specific form of tactile learning. When the same coactivation protocol as used in previous studies in young individuals is applied to aged participants (Dinse et al., 2006), we found that thresholds were significantly lowered from 3.4270.50 mm to 2.8970.40 mm following this coactivation (Figs. 7 and 8). When all young participants were submitted to similar coactivation their 2-point skin discrimination improved from 1.5570.19 mm to 1.3370.19 mm (Figs. 7 and 8). In young individuals the gain in discrimination threshold was 0.2270.19 mm.The mean improvement in elderly participants was 0.5470.32 mm.These results demonstrate that the tactile coactivation protocol is also effective at old age improving discrimination thresholds in individuals of up to 89 years. Prior to coactivation there was a clear difference in the discrimination thresholds in the 50-year age group and in groups of older individuals. After coactivation this difference disappeared and the tactile acuity of the aged individuals after coactivation matched the average performance of participants aged 47–59 years. These results demonstrate that age-related decline of perception is not irreversible but can be improved by specific stimulation protocols. Typical improvement of acuity from coactivation is between 15% and 20%. It is not evident if this magnitude of improvement represents a major advantage for everyday life. Comparing traininginduced improvements of tactile acuity for pianists (Ragert et al., 2004) and violinists (unpublished data) to those evoked by coactivation revealed almost identical improvement of tactile discrimination when long-term training and short-time coactivation were compared (Dinse et al., 2005). This indicates that a short time of coactivation is
Fig. 8. Effects of coactivation on discrimination thresholds. Psychometric functions illustrating the discrimination performance obtained pre, post, and 24 h after coactivation for a young (a), and an elderly subject (b). Correct responses in percent (squares) are plotted as a function of separation distance together with the results of a logistic regression line (diamonds). 50% level of correct responses is indicated (dashed line) together with resulting thresholds (arrows). (top): precondition before coactivation; (middle): postcondition, immediately after coactivation; (bottom): recovery-condition, 24 h after termination of coactivation. In both the young (26 years) and the elderly subject (81 years), after coactivation there is a distinct shift in the psychometric functions toward lower separation distances, which recover to preconditions 24 h later. In the young subject, thresholds were reduced from 1.56 to 1.23 mm after coactivation, and recovered back to baseline (1.59 mm). In the elderly, thresholds were reduced from 4.2 to 2.8 mm thereby matching prethresholds typically found in a 50-year-old subject (mean threshold of 13 participants aged 47–55 was 2.6170.48 mm). 24 h later, threshold recovered back to baseline (4.1 mm) (reprinted from Dinse et al. (2006) with permission from Wiley-Liss, Inc.).
as effective in promoting perceptual improvement as long-term training. While coactivation improves tactile acuity in discrimination-impaired elderly, the lowest thresholds in elderly (2.34 mm) after coactivation were still above the thresholds typically observed in young individuals (approximately 1.5 mm). Linear correlation analysis (Pearson) revealed that the magnitude of coactivation-induced changes in the aged group was systematically dependent on their performance level before subjected to the coactivation treatment: The participants who had the
74
highest thresholds before the start of the study showed the largest improvement, while participants with low thresholds had only limited improvement, indicating the presence of ceiling effects. The performance observed after coactivation may represent the lowest limit in acuity that can be reached by elderly given the anatomical and morphological changes accumulating over age, or the discrimination thresholds may be further reduced by using different or more refined methods of intervention. The latter hypothesis is supported by the observation that tactile 2-point discrimination thresholds in elderly are much higher in sighted than in blind people of the same age (Stevens et al., 1996; Heinisch et al., 2006). Especially the members of a group of early blind individuals had significantly lower thresholds as young sighted and blind individuals (Heinisch et al., 2006). These results show that extraordinary discrimination can occur even in elderly individuals, which we take as an argument that physiological, particularly peripheral constraints are unlikely to limit acuity, and they also support the hypothesis that practicing is the major driving force for maintaining high acuity performance even into old age. Perspectives on passive unattended intervention methods In a previous study of young participants, it was found that significant improvements of spatial discrimination were present up to 6 h after coactivation (Godde et al., 2000). When coactivation was applied on three consecutive days, the magnitude of changes was not different but the effects lasted longer. Only on day 5 did the thresholds return to preconditions (Godde et al., 2000). Coactivating all fingertips of a hand instead of a single finger resulted in much stronger and longer-lasting effects (Kalisch et al., 2005). Application of highfrequency tactile coactivation mimicking longterm potentiation (LTP)-like stimulation for only 20 min evoked tactile acuity improvements comparable in magnitude to those of the standard coactivation protocol lasting 3 h , which recovered to baseline only after 48 h (Ragert et al., 2005, submitted). Conceivably, combining repeated applications with new forms of coactivation
protocols will lead to higher persistence of the evoked improvement. Alternative attempts to interfere with the agerelated decline of sensory capacities have been described (Dhruv et al., 2002; Priplata et al., 2003). Addition of noise to a signal that is to be transmitted can improve the ability to transfer reliably information, a phenomenon known as stochastic resonance (Collins et al., 1996). Electrical noise stimulation applied to the hand of elderly individuals lower touch thresholds (Dhruv et al., 2002), and noise stimulation to the foot can improve postural stability in young and elderly individuals (Priplata et al., 2003). While stochastic resonance affects thresholds by making inputs that would otherwise be subthreshold exceed the threshold, coactivation may alters the modes of neural processing due to specific changes of synaptic efficacy and synaptic connections (cf. Dinse et al., 2003; Pleger et al., 2003). The unique advantage of coactivation is its passive nature, i.e., it does not require the active cooperation and involvement of the individual person who is being treated. It can be applied even in parallel to other occupations and might therefore be substantially easier to implement. These properties, together with the effectiveness of coactivation to improve tactile discrimination, make coactivation-based principles prime candidates for therapeutic intervention programs that serve as training substitute in individuals with neurological impairments or it may be used in addition to conventional training. Preservation of sufficient tactile acuity to old age is an important prerequisite for the maintenance of independent and autonomous living. We therefore believe that the concept of coactivation might turn out to be beneficial in preserving everyday sensorimotor competence in the elderly through the use of unattended therapeutic interventions. Treatment of age-related cortical changes Assessment of tactile discrimination ability, recording of somatosensory evoked potentials and functional magnetic resonance imaging (fMRI) before and after coactivation in young human individuals showed that the coactivation-induced gain of perceptual performance was linearly
75
correlated with the amount of cortical reorganization of the finger representation in primary somatosensory cortex (Pleger et al., 2001, 2003; Dinse et al., 2003). fMRI studies confirmed that also in the elderly, learning processes localized in somatosensory cortex are likely to be involved in mediating the amelioration of age-related impairment of tactile acuity from coactivation (Pleger et al., unpublished). The third brain: interventions are no time-travel — insight from intervention studies At first glance, the old rats from the enriched housing were able to perform nearly as well as young animals. Only closer inspection of these animals revealed that old rats regained their ability by walking in a different way than young animals. Most notable, the behavioral changes were paralleled by significant effects on the cortical maps. In particular, the severe deteriorations of the maps as described above were largely restored. But similar to the behavioral findings, cortical maps in enriched animals were ‘‘new’’ in the sense that they combined features seen in young animals and those never been seen in old animals from standard housing, implying a use-enforced development of new strategies of cortical processing at old age. This observation led us to a more systematic evaluation of the effects of enriched housing (Table 1). To find out if housing animals in enriched environment restored the cortical organization to a status present earlier in life, we analyzed parameters that describe behavioral performance and cortical organization in animals of intermediate age (29 months) and in animals 34–36 months of age that were kept under enriched conditions (Ta-
ble 2). The animals were subjected to enriched housing at 29 months of age. Behavioral parameters were ground reaction forces as measured during walking, step length/frequency as obtained from videobased analysis of walking, duty factor (defined as the fraction of the stride period that a limb is in contact with the ground), footprints (various parameters such as print length and area), and beam walking (number of droppings, time to cross the beam). Cortical parameters were representational area (size of representational map), muscle complexity (indicates how many different muscles are represented at a single cortical location), muscle topography (topographic order of different muscle representations), location of motor maps (location of cortical map relative to absolute scull coordinates such as bregma), threshold for evoking EMG activity (current needed to evoke movements or EMG activity), and EMG latency (time between current application and onset of EMG activity).
As summarized in Table 1, housing animals under enriched conditions had beneficial effects on Table 2. Comparison of the 29-month status with that observed in animals aged 34–36 months housed under enriched environmental conditions Parameter
Compared to 29 months
Representational area Muscle complexity EMG threshold EMG latency Ground reaction forces Step length/frequency Duty factor Foot prints
Better Better ¼ 29 months ¼ 29 months Better ¼ 29 months Worse ¼ 29 months
Table 1. Effect of enriched housing on various behavioral and cortical parameters Restoration
‘‘New’’ properties
Representational area Muscle complexity Muscle topography Beam walking Ground reaction forces Step length/frequency Foot prints
Location of motor maps
Enhancement of aging
No effect EMG thresholds EMG latencies
Duty factor
76
many of the parameters studied, but clearly not every tested parameter was positively affected. Some parameters even showed enhanced aging effects, while others revealed behaviors never seen in young or aged individuals. Comparing the status of a 29-month-old with that of a 33-month-old that was kept under enriched conditions showed that enriched housing does not allow an individual rat to rejuvenate to the status of a younger animal (Table 2). While some parameters became similar to those of younger animals, other parameters improved, and some became worse than those of an untreated aged-matched control from standard housing. Even for that limited range of parameters investigated the results indicate that very complex remodeling of both behavior and cortical processing strategies was behind the observed regaining of sensorimotor abilities. Major constraints appear to be induced by the observed prolongation of the latencies and by elevation of the thresholds, which both impair control and regulation of walking, causing impaired coordination and stability. The restoration of the size of the cortical maps is a sign of recruitment of processing resources needed for compensation of impaired timing and excitability. Increasing the duty factor is one among other possibilities, which has a significant stabilizing effect and thereby counteracting impairment of walking. Sensory latencies were insensitive to enriched housing or nimodipine-treatment, suggesting that degenerative age changes are not affected by training and stimulation. Does this conclusion also apply to elderly humans? According to unpublished imaging data from our group (Pleger and coworkers), cortical maps of the fingers in SI cortex of elderly individuals clearly expand (cf. also Kalisch et al., submitted for EEG-based analysis). Coactivating these individuals leads to an improvement of tactile acuity as described above (Dinse et al., 2006), thereby ameliorating the age-related decline in discrimination performance. Similar to what has been observed in young individuals, coactivation in elderly induced a further enlargement of cortical maps indicating the coexistence of two forms of cortical map expansion: Enlargement of cortical representation after coactivation results in
perceptual improvement (in elderly in an amelioration of age-related perceptual decline), and the expansion that develops during aging occurs simultaneously with perceptual impairment. Conclusion Cognitive impairments during nonpathological aging have been suggested to reflect synaptic alterations in otherwise intact neural circuits rather than loss of neurons, thus an important prerequisite for being able to reverse age-related changes (Morrison and Hof, 1997; Hof and Morrison, 2004). Despite the accumulation of degenerative processes during aging, the findings presented here demonstrate that the typical age-related decline in tactile performance is not inevitable, but it is preventable and if it occurs it is subject to restoration by various forms of intervention. Based on experimental observations aging induces major cortical reorganization, and interventions aiming at ameliorating age-related changes lead to other forms of cortical reorganization, where the outcome is a ‘‘new’’ brain – the third brain – that differs significantly from brains seen in young or in old, but untreated individuals. Acknowledgments This research was supported by DFG (DI 334/101, 10-2, 10-3, 10-4, 15-1), the International Graduate School of Neuroscience (IGSN) of the RuhrUniversity Bochum, a grant of the Schering Stiftung, and a research grant of the Sandoz Foundation for Gerontological Research (EU-94-1029). We gratefully acknowledge the supply of FBNF1 rats by Bayer, Germany. References Berkefeldt, T., Godde, B. and Dinse, H.R. (1996) Optical imaging of age-related changes of rat somatosensory cortical representations and their sensitivity to the Ca2+-influxblocker nimodipine. Soc. Neurosci. Abstr., 22: 538.4. Bock, O. and Schneider, S. (2002) Sensorimotor adaptation in young and elderly humans. Neurosci. Biobehav. Rev., 26: 761–776.
77 Brody, H.D. (1955) Organization of the cerebral cortex III.. A study of aging in the human cerebral cortex. J. Comp. Neurol., 1023: 511–556. Buonomano, D.V. (1999) Distinct functional types of associative long-term potentiation in neocortical and hippocampal pyramidal neurons. J. Neurosci., 19: 6748–6754. Buonomano, D.V. and Merzenich, M.M. (1998) Cortical plasticity: from synapses to maps. Annu. Rev. Neurosci., 21: 149–186. Castro-Alamancos, M.A. and Connors, B.W. (1996) Cellular mechanisms of the augmenting response: short-term plasticity in a thalamocortical pathway. J. Neurosci., 16: 7742–7756. Churs, L., Spengler, F., Ju¨rgens, M. and Dinse, H.R. (1996) Environmental enrichment counteracts decline of sensorimotor performance and deterioration of cortical organization in aged rats. Soc. Neurosci. Abstr., 22: 102. Coleman, P. (Ed.). (1989) Editorial: how old is old? Neurobiol. Aging, 10: 115. Coleman, P., Finch, C. and Joseph, J. (1990) The need for multiple time points in aging studies. Neurobiol. Aging, 11: 1–2. Collins, J.J., Imhoff, T.T. and Grigg, P. (1996) Noise-enhanced tactile sensation. Nature, 383: 770. Connor, J.R., Wang, E.C. and Diamond, M.C. (1982) Increased length of terminal dendritic segments in old adult rats’ somatosensory cortex: an environmentally induced response. Exp. Neurol., 78: 466–470. Coq, J.O. and Xerri, C. (2000) Age-related alteration of the forepaw representation in the rat primary somatosensory cortex. Neurosci., 99: 403–411. Coq, J.O. and Xerri, C. (2001) Sensorimotor experience modulates age-dependent alterations of the forepaw representation in the rat primary somatosensory cortex. Neurosci., 104: 705–715. Cummins, R.A., Walsh, R.N., Budtz-Olsen, O.E., Konstantinos, T. and Horsfall, C.R. (1973) Environmentally-induced changes in the brains of elderly rats. Nature, 243: 516–518. de Rivera, C., Shukitt-Hale, B., Joseph, J.A. and Mendelson, J.R. (2005) The effects of antioxidants in the senescent auditory cortex. Neurobiol. Aging, (Epub ahead of print). Dhruv, N.T., Niemi, J.B., Harry, J.D., Lipsitz, L.A. and Collins, J.J. (2002) Enhancing tactile sensation in older adults with electrical noise stimulation. Neuroreport, 13: 597–600. Diamond, M.C., Connor, J.-R.J. and Plasticity of the aging cerebral cortex. (1982). Exp. Brain Res. Suppl., 5: 36–44. Dinse, H.R. (2004) Sound case for enrichment. J. Neurophysiol., 92: 36–37. Dinse, H.R. (2005) Treating the aging brain: cortical reorganization and behavior. Acta Neurochir. Suppl., 93: 79–84. Dinse, H.R. and Merzenich, M.M. (2002) Adaptations in the somatosensory system. In: M. Fahle, T. Poggio (Eds.), Perceptual Learning. MIT Press, Cambridge, pp. 19–42. Dinse, H.R., Ragert, P., Pleger, B., Schwenkreis, P. and Tegenthoff, M. (2003) Pharmacological modulation of perceptual learning and associated cortical reorganization. Science, 301: 91–94.
Dinse, H.R., Reinke, H., Schulz, M.H., Ju¨rgens, M. and Zepka, R. (2001) Altered walking behavior alters cortical motor representations — an animal study. 6th Congress of the European College of Sport Science, IPS42 — COL2011. Dinse, H.R., Kalisch, T., Ragert, P., Pleger, B., Schwenkreis, P. and Tegenthoff, M. (2005) Improving human haptic performance in normal and impaired human populations through unattended activation-based learning. Transaction Appl. Perc., 2: 71–88. Dinse, H.R., Kleibel, N., Kalisch, T., Ragert, P. and Tegenthoff, M. (2006) Tactile coactivation resets age-related decline of human tactile discrimination. Ann. Neurol., 60: 88–94. Elbert, T., Candia, V., Altenmuller, E., Rau, H., Sterr, A., Rockstroh, B., Pantev, C. and Taub, E. (1995) Increased cortical representation of the fingers of the left hand string players. Science, 270: 305–307. Elward, K. and Larson, E.B. (1992) Benefits of exercise for older adults. A review of existing evidence and current recommendations for the general population. Clin. Geriatr. Med., 8: 35–50. Engineer, N.D., Percaccio, C.R., Pandya, P.K., Moucha, R., Rathbun, D.L. and Kilgard, M.P. (2004) Environmental enrichment improves response strength, threshold, selectivity, and latency of auditory cortex neurons. J. Neurophysiol., 92: 73–82. Federici, A., Bellagamba, S. and Rocchi, M.B. (2005) Does dance-based training improve balance in adult and young old subjects? A pilot randomized controlled trial. Aging Clin. Exp. Res., 17: 385–389. Floel, A., Breitenstein, C., Hummel, F., Celnik, P., Gingert, C., Sawaki, L., Knecht, S. and Cohen, L.G. (2005) Dopaminergic influences on formation of a motor memory. Ann. Neurol., 58: 121–130. Flood, D.G. and Coleman, P.D. (1988) Neuron numbers and sizes in aging brain: comparisons of human, monkey, and rodent data. Neurobiol. Aging, 9: 453–463. Gallagher, M. and Rapp, P.R. (1997) The use of animal models to study the effects of aging on cognition. Annu. Rev. Psychol., 48: 339–370. Gardner, E.P. and Costanzo, R.M. (1980) Temporal integration of multiple-point stimuli in primary somatosensory cortical receptive fields of alert monkeys. J. Neurophysiol., 43: 444–468. Gazzaley, A.H., Siegel, S.J., Kordower, J.H., Mufson, E.J. and Morrison, J.H. (1996) Circuit-specific alterations of N-methyl-D-aspartate receptor subunit 1 in the dentate gyrus of aged monkeys. Proc. Natl. Acad. Sci., USA, 93: 3121–3125. Godde, B., Berkefeld, T., David-Ju¨rgens, M. and Dinse, H.R. (2002) Age-related changes in primary somatosensory cortex of rats: evidence for parallel degenerative and plastic-adaptive processes. Neurosci. Biobehav. Rev., 26: 743–752. Godde, B., Stauffenberg, B., Spengler, F. and Dinse, H.R. (2000) Tactile coactivation induced changes in spatial discrimination performance. J. Neurosci., 20: 1597–1604. Godfraind, T. and Govoni, S. (1995) Recent advances in the pharmocology of Ca2+ and K+ channels. Tr. Biochem. Sci., 16: 1–4.
78 Harman, D. and Piette, L.H. (1966) Free radical theory of aging: free radical reactions in serum. J. Gerontol., 21: 560–565. Hayflick, L. (1968) Human cells and aging. Sci. Am., 218: 32–37. Heinisch, C., Kalisch, T. and Dinse, H.R. (2006) Tactile and learning abilities in early and late-blind subjects. FENS Conference 2006, in press. Hilbig, H., Bidmon, H.J., Steingru¨ber, S., Reinke, H. and Dinse, H.R. (2002a) Enriched environmental conditions reverse age-dependent gliosis and losses of neurofilaments and extracellular matrix components but do not alter lipofuscin accumulation in the hind-limb area of the aging rat brain. J. Chem. Neuroanat., 23: 199–209. Hilbig, H., Hu¨ller, J., Dinse, H.R. and Bidmon, H.J. (2002b) In contrast to neuronal NOS-I, the inducible NOS-II expression in aging brains is modified by enriched environmental conditions. Exp. Toxic. Pathol., 53: 427–431. Hof, P.R. and Morrison, J.H. (2004) The aging brain: morphomolecular senescence of cortical circuits. Trends Neurosci., 27: 607–613. Ingram, D.K. (1988) Motor performance variability during aging in rodents. Assessment of reliability and validity of individual differences. Ann. N.Y. Acad. Sci., 515: 70–96. Jacobson, A.C., McKinley, P.A., Leroux, A. and Rainville, C. (2005) Argentine tango dancing as an effective means for improving cognition and complex task performance in at-risk elderly. Soc. Neurosci. Abstr., 757.7. Johansson, B.B. (2000) Brain plasticity and stroke rehabilitation. The Willis lecture. Stroke, 31: 223–230. Ju¨rgens, M. and Dinse, H.R. (1995) Spatial and temporal integration properties of cortical somatosensory neurons in aged rats. Soc. Neurosci. Abstr., 21: 197. Ju¨rgens, M. and Dinse, H.R. (1997) Differential effects of the Ca2+-influxblocker nimodipine on receptive field properties and response latencies of somatosensory cortical neurons in aged rats. Institut fu¨r Neuroinformatik, Ruhr-University Bochum. Internal Report IRINI, 10: 1–23. Kalisch, T. and Dinse, H.R. (submitted) Increasing genderspecific differences in haptic object recognition performance over the life-span. Kalisch, K., Ragert, P., Schwenkreis, P., Dinse, H.R. and Tegenthoff, M. (submitted) Human age-related decline in tactile perception is accompanied by enlargement of the hand representation in somatosensory cortex. Kalisch, T., Tegenthoff, M. and Dinse, H.R. (2005) Improvement of right hand haptic skills following 3 hours of tactile coactivation of all fingers. In: Zimmermann H. and Krieglstein K. (Eds.), Proceedings of the 6th Meeting of the German Neuroscience Society. Neuroforum Suppl., 1, 65App. Kempermann, G., Kuhn, H.G. and Gage, F.H. (1997) More hippocampal neurons in adult mice living in an enriched environment. Nature, 386: 493–495. Kim, C.B., Pier, L.P. and Spear, P.D. (1997) Effects of aging on numbers and sizes of neurons in histochemically defined subregions of monkey striate cortex. Anat. Rec., 247: 119–128. Kim, C.B., Tom, B.W. and Spear, P.D. (1996) Effects of aging on the densities, numbers, and sizes of retinal ganglion cells in rhesus monkey. Neurobiol. Aging, 17: 431–438.
Kornatz, K.W., Christou, E.A. and Enoka, R.M. (2005) Practice reduces motor unit discharge variability in a hand muscle and improves manual dexterity in old adults. J. Appl. Physiol., 98: 2072–2080. Kramer, A.F., Hahn, S., Cohen, N.J., Banich, M.T., McAuley, E., Harrison, C.R., Chason, J., Vakil, E., Bardell, L., Boileau, R.A. and Colcombe, A. (1999) Ageing, fitness and neurocognitive function. Nature, 400: 418–419. Lee, H.K., Min, S.S., Gallagher, M. and Kirkwood, A. (2005) NMDA receptor-independent long-term depression correlates with successful aging in rats. Nat. Neurosci., 8: 1657–1659. Lee, C.J. and Whitsel, B.L. (1992) Mechanisms underlying somatosensory cortical dynamics: I. In vivo studies. Cereb. Cortex, 2: 81–106. Liepert, J., Tegenthoff, M. and Malin, J.P. (1995) Changes of cortical motor area size during immobilization. Electroencephalogr. Clin. Neurophysiol., 97: 382–386. Mendelson, J.R. and Ricketts, C. (2001) Age-related temporal processing speed deterioration in auditory cortex. Hear. Res., 158: 84–94. Miller, R.J. (1987) Multiple calcium channels and neuronal function. Science, 235: 46–52. Mohammed, A.H., Henriksson, B.G., Soderstrom, S., Ebendal, T., Olsson, T. and Seckl, J.R. (1993) Environmental influences on the central nervous system and their implications for the aging rat. Behav. Brain Res., 57: 183–191. Morrison, J.H. and Hof, P.R. (1997) Life and death of neurons in the aging brain. Science, 278: 412–419. Newell, F.N., Ernst, M.O., Tjan, B.S. and H.H. and Bu¨lthoff, H.H. (2001) Viewpoint dependence in visual and haptic object recognition. Psychol. Sci., 12: 37–42. Orgel, L.E. (1963) The maintenance of the accuracy of protein synthesis and its relevance to aging. Proc. Nat. Acad. Sci. USA, 49: 517–521. Palmore, E. (1988) The acts on aging quiz: a handbook of uses and results. Springer, New York. Pantev, C., Oostenveld, R., Engelien, A., Ross, B., Roberts, L.E. and Hoke, M. (1998) Increased auditory cortical representation in musicians. Nature, 392: 811–814. Pascual-Leone, A. and Torres, F. (1993) Plasticity of the sensorimotor cortex representation of the reading finger in Braille readers. Brain, 116: 39–52. Pearl, R. (1924) Studies in Human Biology. Williams and Williams, Baltimore, MD. Perls, T.T. (1995). The oldest old. Sci. Am., 272: 70–75. Peters, A. (2002) The effects of normal aging on myelin and nerve fibers: a review. J. Neurocytol., 31: 581–593. Peters, A., Morrison, J.H., Rosene, D.L. and Hyman, B.T. (1998a) Feature article: are neurons lost from the primate cerebral cortex during normal aging? Cereb. Cortex, 8: 295–300. Peters, A., Nigro, N. and McNally, K.J. (1997) A further evaluation of the effects of age on the striate cortex of the rhesus monkey. Neurobiol. Aging,, 18: 29–36. Peters, A., Sethares, C. and Moss, M.B. (1998b) The effects of aging on layer 1 in area 46 of prefrontal cortex in the rhesus monkey. Cereb. Cortex, 8: 671–684.
79 Pleger, B., Dinse, H.R., Ragert, P., Schwenkreis, P., Malin, J.P. and Tegenthoff, M. (2001) Shifts in cortical representations predict human discrimination improvement. Proc. Natl. Acad. Sci. USA, 98: 12255–12260. Pleger, B., Foerster, A.F., Ragert, P., Dinse, H.R., Schwenkreis, P., Nicolas, V. and Tegenthoff, M. (2003) Functional imaging of perceptual learning in human primary and secondary somatosensory cortex. Neuron, 40: 643–653. Porciatti, V., Burr, D.C., Morrone, C. and Fiorentini, A. (1992) The effects of ageing on the pattern electroretinogram and visual evoked potential in humans. Vis. Res., 32: 1199–1209. Priplata, A.A., Niemi, J.B., Harry, J.D., Lipsitz, L.A. and Collins, J.J. (2003) Vibrating insoles and balance control in elderly people. Lancet, 362: 1123–1124. Ragert, P., Kalisch, T., Bliem, B., Franzkowiak, S. and Dinse, H.R. (submitted) Frequency-specific changes in human tactile discrimination behavior induced by tactile LTP- and LTD stimulation protocols. Ragert, P., Kalisch, T. and Dinse, H.R. (2005) Perceptual changes in human tactile discrimination behavior induced by coactivation using LTP and LTD protocols. Soc. Neurosci. Abstr., 173.6. Ragert, P., Pleger, B., Vo¨lker, B., Meier, C., Schwenkreis, P., Tegenthoff, M. and Dinse, H.R. (2003) Impaired tactile performance in patients with hand immobilisation. Soc. Neurosci. Abstr., 29: 379.13. Ragert, P., Schmid, A., Altenmueller, E. and Dinse, H.R. (2004) Superior tactile performance and learning in professional pianists: Evidence for meta-plasticity in musicians. Europ. J. Neurosci., 19: 473–478. Rapp, P.R. and Amaral, D.G. (1992) Individual differences in the cognitive and neurobiological consequences of normal aging. Trends Neurosci., 15: 340–345. Rapp, P.R. and Gallagher, M. (1996) Preserved neuron number in the hippocampus of aged rats with spatial learning deficits. Proc. Natl. Acad. Sci. USA, 93: 9926–9930. Raz, N., Lindenberger, U., Rodrigue, K.M., Kennedy, K.M., Head, D., Williamson, A., Dahle, C., Gerstorf, D. and Acker, J.D. (2005) Regional brain changes in aging healthy adults: general trends, individual differences and modifiers. Cereb. Cortex, 15: 1676–1689. Recanzone, G.H., Merzenich, M.M., Jenkins, W.M., Grajski, K.A. and Dinse, H.R. (1992) Topographic reorganization of the hand representation in cortical area 3b owl monkeys trained in a frequency-discrimination task. J. Neurophysiol., 67: 1031–1056. Recanzone, G.H., Schreiner, C.E. and Merzenich, M.M. (1993) Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkeys. J. Neurosci., 13: 87–103. Reinke, H. and Dinse, H.R. (1996) Functional characterization of cutaneous mechanoreceptor properties in aged rats. Neurosci. Lett., 216: 171–174. Rosenzweig, M.R. and Bennett, E.L. (1996) Psychobiology of plasticity: effects of training and experience on brain and behavior. Behav. Brain Res., 78: 57–65.
Rowe, J.W. and Kahn, R.L. (1987) Human aging: usual and successful. Science, 237: 143–149. Rubner, M. (1908) Das Problem der Lebensdauer und seine Beziehung zu Wachstum und Erna¨hrung. Oldenbourg, Mu¨nchen. Saito, S., Kobayashi, S., Ohashi, Y., Igarashi, M., Komiya, Y. and Ando, S. (1994) Decreased synaptic density in aged brains and its prevention by rearing under enriched environment as revealed by synaptophysin contents. J. Neurosci. Res., 39: 57–62. Sawaki, L., Yaseen, Z., Kopylev, L. and Cohen, L.G. (2003) Age-dependent changes in the ability to encode a novel elementary motor memory. Ann. Neurol., 53: 521–524. Schmolesky, M.T., Wang, Y., Pu, M. and Leventhal, A.G. (2000) Degradation of stimulus selectivity of visual cortical cells in senescent rhesus monkeys. Nat. Neurosci., 3: 384–390. Schuurman, T., Klein, H., Beneke, M. and Traber, J. (1987) Nimodipine and motor deficits in the aged rat. Neurosci. Res. Comm., 1: 9–15. Schuurman, T. and Traber, J. (1989a) Effects of nimodipine on behavior of old rats. In: Traber, J. and Gispen, W.H. (Eds.), Nimodipine and Central Nervous System Function: New Vistas. Schattauer, Stuttgart, New York, pp. 96–208. Schuurman, T. and Traber, J. (1989b) Old rats as an animal model for senile dementia: behavioral effects of nimodipine. In: Reisberg, W. and Bergener, B. (Eds.), Diagnosis and Treatment of Senile Dementia. Springer, Berlin, Heidelberg, pp. 295–307. Smith, T.D., Adams, M.M., Gallagher, M., Morrison, J.H. and Rapp, P.R. (2000) Circuit-specific alterations in hippocampal synaptophysin immunoreactivity predict spatial learning impairment in aged rats. J. Neurosci., 20: 6587–6593. Smith, C.D., Walton, A., Loveland, A.D., Umberger, G.H., Kryscio, R.J. and Gash, D.M. (2005) Memories that last in old age: motor skill learning and memory preservation. Neurobiol. Aging,, 26: 883–890. Sohal, R.S. and Weindruch, R. (1996) Oxidative stress, caloric restriction, and aging. Science, 273: 59–63. Sowell, E.R., Peterson, B.S., Thompson, P.M., Welcome, S.E., Henkenius, A.L. and Toga, A.W. (2003) Mapping cortical change across the human life span. Nat. Neurosci., 6: 309–315. Spear, P.D. (1993) Neural bases of visual deficits during aging. Vision Res., 33: 2589–2609. Spengler, F., Godde, B. and Dinse, H.R. (1995) Effects of aging on topographic organization of somatosensory cortex. NeuroReport, 6: 469–473. Sprott, R.L. (1997) Diet and calorie restriction. Exp. Gerontol., 32: 205–214. Stensaas, S.S., Eddington, D.K. and Dobelle, W.H. (1974) The topography and variability of the primary visual cortex in man. J. Neurosurg., 40: 747–755. Sterr, A., Mu¨ller, M.M., Elbert, T., Rockstroh, B., Pantev, C. and Taub, E. (1998) Perceptual correlates of changes in cortical representation of fingers in blind multifinger Braille readers. J. Neurosci., 18: 4417–4423.
80 Stevens, J.C. (1992) Aging and spatial acuity of touch. J. Gerontol., 47: 35–40. Stevens, J.C. and Choo, K.K. (1996) Spatial acuity of the body surface over the life span. Somatosens. Motor Res., 13: 153–166. Stevens, J.C., Foulke, E. and Patterson, M.Q. (1996) Tactile acuity, aging, and Braille reading in long-term blindness. J. Exp. Psychol., 2: 91–106. Stoll, S., Dorner, H., Blo¨sch, M. and Platt, R. (1990) Agedependent differences in the gait of rats. Arch. Gerontol. Geriatr., 10: 216–268. Tegenthoff, M., Ragert, P., Pleger, B., Schwenkreis, P., Forster, A.F., Nicolas, V. and Dinse, H.R. (2005) Persistent improvement of tactile discrimination performance and enlargement of cortical somatosensory maps after 5 Hz rTMS. PloS Biol., 3: e362. Tommerdahl, M., Delemos, K.A., Favorov, O.V., Metz, C.B., Vierck Jr., C.J. and Whitsel, B.L. (1998) Response of anterior parietal cortex to different modes of same-site skin stimulation. J. Neurophysiol., 80: 3272–3283. Tommerdahl, M., Delemos, K.A., Vierck Jr., C.J., Favorov, O.V. and Whitsel, B.L. (1996) Anterior parietal cortical response to tactile and skin-heating stimuli applied to the same skin site. J. Neurophysiol., 75: 2662–2670. Trick, G.L., Trick, L.R. and Haywood, K.M. (1986) Altered pattern evoked retinal and cortical potentials associated with human senescence. Curr. Eye Res., 5: 717–724.
Varela, J.A., Sen, K., Gibson, J., Fost, J., Abbott, L.F. and Nelson, S.B. (1997) A quantitative description of short-term plasticity at excitatory synapses in layer 2/3 of rat primary visual cortex. J. Neurosci., 17: 7926–7940. Verdu, E., Ceballos, D., Vilches, J.J. and Navarro, X. (2000) Influence of aging on peripheral nerve function and regeneration. J. Peripher. Nerv. Syst., 5: 191–208. Verzar, F. (1963) The aging of collagen. Sci. Am., 208: 104–114. Walford, R.L. (1967) The role of autoimmune phenomena in the ageing process. Symp. Soc. Exp. Biol., 21: 351–373. Walford, R.L. (1985) The extension of maximum life span. Clin. Geriatr. Med., 1: 29–35. Wanagat, J., Allison, D.B. and Weindruch, R. (1999) Caloric intake and aging: mechanisms in rodents and a study in nonhuman primates. Toxicol. Sci., 52: 35–40. Wohlert, A.B. (1996) Tactile perception of spatial stimuli on the lip surface by young and older adults. J. Speech Hear. Res., 39: 1191–1198. Woodward, K.L. (1993) The relationship between skin compliance, age, gender, and tactile discriminative thresholds in humans. Somatosens. Motor Res., 10: 63–67. Xerri, C., Stern, J.M. and Merzenich, M.M. (1994) Alterations of the cortical representation of the rat ventrum induced by nursing behavior. J. Neurosci., 14: 1710–1721. Zucker, R.S. (1989) Short-term synaptic plasticity. Annu. Rev. Neurosci., 12: 13–31.
a.
b.
c.
d.
Plate 5.3. Specific effects of age on receptive fields of the hindpaw recorded in somatosensory cortex of aged rats. In addition, representative examples of behavioral changes of walking pattern derived from footprint analysis are shown: (a) young, control animal; (b) old animal. On the left: hindpaw; on the right: forepaw. Note selectivity of walking impairment restricted to hind-leg. At bottom of (a) and (b) are examples of receptive fields recorded in the hindpaw representation ((a) and (b), left) and in the forepaw representation ((a) and (b), right) in a young (a) and an old animal (b). Age-related changes are limited to the behaviorally impaired extremity. To visualize the effects of aging on the topography of the underlying cortical maps, we reconstructed somatosensory maps using a computer-based interpolation-algorithm based on a linear least square approximation of sampling coordinates of penetration sites and corresponding receptive field centers. Reconstructions of a cortical hindpaw representation are shown for a control (c) and for an old rat (d). Left: examples of cortical topographies represented as a regular lattice within somatosensory cortex. Middle: the extrapolated cortical representation of a schematic and standardized drawing of the hindpaw. Dashed lines indicate horizontal, solid lines the vertical components of the lattice. One square of the lattice represents 1 mm2 skin area. Diamonds indicate penetration sites; squares give the interpolated RF centers. Dotted lines give the deviation between them. Right: back projection of the regular lattice of the cortical map onto the hindpaw. Squares give the interpolated, asterisks the measured RF centers. One square of the lattice represents the skin portion that is represented by 0.01 mm2 cortical areas. According to these reconstructions, maps of the hindpaw representation recorded in old animals, which are characterized by a selective impairment of the hind-limbs, show a dramatic distortion of their representational maps and a loss of topographic order (modified from Spengler et al., 1995; Ju¨rgens and Dinse, 1997b).
Plate 5.4. Local field potential (LFP) maps constructed on the basis of LFP recorded by multiple penetrations showing the spatial activity distributions evoked by tactile stimulation in the hindpaw representation of the primary somatosensory cortex of a young (left) and an old rat (right). Shown are activity distributions following stimulation of a digit. Normalized LFP-amplitudes are color-coded and projected onto the cortical area of the hindpaw representation reconstructed from [x, y] coordinates of penetration sites. Scale bar: 1 mm; each plot is oriented with rostral left and lateral up (reprinted from Spengler et al. (1995) with permission from Lippincott Williams & Wilkins).
Plate 5.7. Tactile 2-point discrimination thresholds of the tip of the right index finger as a function of age (total of 120 participants). According to univariate ANOVA (with age F(2,118) ¼ 347.785; po0.001) the differences across the three age groups were significant. After coactivation (pink symbols), thresholds of the coactivated participants (young control group and group of elderly) were significantly reduced. Coactivation-induced improvement in the group aged 66–86 was several fold stronger in magnitude as compared to the young subject. As a result, after coactivation thresholds of the elderly resembled those found in the participants 47–55-years old (reprinted from Dinse et al. (2006) with permission from Wiley-Liss, Inc.).
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 6
Brain plasticity and functional losses in the aged: scientific bases for a novel intervention Henry W. Mahncke2, Amy Bronstone2 and Michael M. Merzenich1, 1
Keck Center for Integrative Neurosciences, University of California, San Francisco, CA 94143-0732, USA 2 Posit Science Corporation, San Francisco, CA 94104, USA
Abstract: Aging is associated with progressive losses in function across multiple systems, including sensation, cognition, memory, motor control, and affect. The traditional view has been that functional decline in aging is unavoidable because it is a direct consequence of brain machinery wearing down over time. In recent years, an alternative perspective has emerged, which elaborates on this traditional view of age-related functional decline. This new viewpoint — based upon decades of research in neuroscience, experimental psychology, and other related fields — argues that as people age, brain plasticity processes with negative consequences begin to dominate brain functioning. Four core factors — reduced schedules of brain activity, noisy processing, weakened neuromodulatory control, and negative learning — interact to create a selfreinforcing downward spiral of degraded brain function in older adults. This downward spiral might begin from reduced brain activity due to behavioral change, from a loss in brain function driven by aging brain machinery, or more likely from both. In aggregate, these interrelated factors promote plastic changes in the brain that result in age-related functional decline. This new viewpoint on the root causes of functional decline immediately suggests a remedial approach. Studies of adult brain plasticity have shown that substantial improvement in function and/or recovery from losses in sensation, cognition, memory, motor control, and affect should be possible, using appropriately designed behavioral training paradigms. Driving brain plasticity with positive outcomes requires engaging older adults in demanding sensory, cognitive, and motor activities on an intensive basis, in a behavioral context designed to reengage and strengthen the neuromodulatory systems that control learning in adults, with the goal of increasing the fidelity, reliability, and power of cortical representations. Such a training program would serve a substantial unmet need in aging adults. Current treatments directed at age-related functional losses are limited in important ways. Pharmacological therapies can target only a limited number of the many changes believed to underlie functional decline. Behavioral approaches focus on teaching specific strategies to aid higher order cognitive functions, and do not usually aspire to fundamentally change brain function. A brain-plasticity-based training program would potentially be applicable to all aging adults with the promise of improving their operational capabilities. We have constructed such a brain-plasticity-based training program and conducted an initial randomized controlled pilot study to evaluate the feasibility of its use by older adults. A main objective of this initial study was to estimate the effect size on standardized neuropsychological measures of memory. We found that older adults could learn the training program quickly, and could use it entirely unsupervised for the majority of the time required. Pre- and posttesting documented a significant improvement in memory within the training group (effect size 0.41, po0.0005), with no significant Corresponding author. Tel.: +1 415-476-0490; Fax: +1 415-476-1941; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57006-2
81
82
within-group changes in a time-matched computer using active control group, or in a no-contact control group. Thus, a brain-plasticity-based intervention targeting normal age-related cognitive decline may potentially offer benefit to a broad population of older adults. Keywords: brain plasticity; cognitive rehabilitation; computer-based training
Introduction This chapter reviews the scientific bases of a novel approach intended to improve the functional performance of older adults by slowing, halting, or reversing large-scale and progressive losses in brain functioning commonly experienced in later life. This hypothesis-driven approach is envisioned to be much like an exercise program for the brain that, ideally, should be initiated early in the aging process to enhance brain health and cognitive fitness before significant losses develop, but also could be effective later in the aging process when significant losses have already emerged. The core of this chapter introduces a new perspective about the root causes of functional decline in aging that is based on decades of research on brain plasticity, experimental psychology, and other related fields. Brain plasticity refers to the lifelong capacity for physical and functional brain change enjoyed by humans and other animals and is inherently bidirectional: through the same mechanisms and plasticity processes, brain function can either be strengthened or degraded, depending on the circumstances. During normal aging, individuals typically undergo physical, behavioral, and environmental changes that, in the aggregate, promote negative plastic changes that degrade brain function. Four interrelated factors are proposed as the core causes of deterioration of functioning in older adults. These root causes of functional decline involve a complex interplay of physical brain deterioration, behavioral and environmental changes, and brain plasticity processes. Just as brain plasticity processes with negative consequences can contribute to age-related functional decline, plasticity processes that strengthen brain function can provide a foundation for a therapy to restore sensory, cognitive, memory, motor, and affect systems in aging. This chapter focuses particularly on age-related cognitive decline,
though the concepts and principles discussed here should apply to other areas of functioning (e.g., motor control) known to deteriorate with age. The principles governing such brain plasticity processes are now sufficiently well understood to develop a new approach to maximize the quality and extend the duration of healthy aging. A brainplasticity-based approach should be significantly more effective than current interventions for healthy aging, and could conceivably work in conjunction with a variety of other behavioral and pharmaceutical advances. When clinically validated, this science-based approach, which explicitly targets the underlying causes of long, slow functional decline, could signify a revolution in aging therapeutics.
Cognitive decline in aging is progressive and can become pathological Cognitive decline is a universal aspect of the aging process. Memory decline during aging is pervasive (Park and Gutchess, 2003; Reuter-Lorenz and Sylvester, 2003; West, 2004). It may begin as early as age of 30 and, on the average, worsens slowly but steadily thereafter (Fig. 1) (Park et al., 1996). In addition, virtually all older adults will eventually develop a reduction in speed of processing (Salthouse, 1996). Various other cognitive abilities (e.g., visuospatial skill, executive functions, speech comprehension) have been found to commonly deteriorate with age (Harvey and Mohs, 2000; Zacks and Hasher, 2000; Schneider et al., 2002; Buckner, 2004). A number of labels have been used to describe normal age-related cognitive decline, including age-associated memory impairment, age-consistent memory impairment, benign senescent forgetfulness, late-life forgetfulness, and aging-associated cognitive decline (Ritchie et al., 2001; Bischkopf et al., 2002; Fillit et al., 2002). In
83
Fig. 1. Memory changes with age. Items recalled from a 16-word list in the immediate period (A, California verbal learning test CVLTII trial 5) and in the delay period (B, CVLT-II delayed free recall), and number of digits recalled in a digit-span backwards task (C, Weschler memory scale III) (CVLT-II data courtesy of the University of California, San Francisco Memory and Aging Center).
normal aging, the extent of cognitive decline gradually increases with age, although there is considerable variability across individuals in the nature, degree, and timing of cognitive loss (Ylikoski et al., 1999; Park et al., 2003). Despite this variability, normal cognitive decline is an inevitable consequence of age; should individuals live long enough, virtually all will eventually lose a degree of cognitive efficacy. Normal cognitive decline is distinct from pathological cognitive loss, which affects a sizeable proportion of older individuals and culminates in dementia. Pathological cognitive decline may look much like healthy aging in the beginning stages, but at some point a precipitous decrease in functioning, particularly in memory, typically occurs. About one in four older adults will experience a decline now generally diagnosed as mild cognitive impairment (MCI), in which function in a specific cognitive domain (e.g., memory) is impaired while activities of daily living generally remain intact (Unverzagt et al., 2001). Individuals with MCI show extreme losses in neuromodulatory activity crucial for sustaining learning operations and vivifying memory (AsDEaRC, 2001–2002, 2002). MCI may be a transitional or high-risk state between normal cognition and dementia, as 80% of individuals diagnosed with MCI are diagnosed with dementia within 5–8 years (Craft et al., 2003). In contrast, only 1 2% of adults with normal agerelated cognitive decline develop dementia each year (Craft et al., 2003).
Some argue that if adults were to live long enough, the progressive physical deterioration of the brain would eventually cause dementia in all cases (Terry and Katzman, 2001). Others argue that cases of dementia (and perhaps MCI) are specific pathological conditions that are not the expected end states of healthy aging (Fillit et al., 2002). Data exist to support both theories regarding the inevitability of dementia. On one hand, older adults become more vulnerable to developing dementia with increasing age, such that almost half of the adults of age 85 and above have Alzheimer’s disease (AD) (National Institute on Aging, 1998). On the other hand, many individuals remain cognitively vital even into extreme old age, evidencing only minor changes in speed of processing and the most attention-demanding tasks (Fillit et al., 2002). In either case, it is clear that normal cognitive decline is a universal phenomenon, and that a clinically significant amount of cognitive decline is increasingly common with age.
Root causes of age-related cognitive decline Changes in the brain occur with aging Thousands of studies have documented the physical, anatomical, physiological, and chemical changes that occur in the brain with aging (Bussiere and Hof, 2000; Magistretti et al., 2000; Raz, 2000; Mattson, 2003; Backman and Farde, 2005).
84
In aggregate, this large body of research has established five fundamental principles: (1) neurons and the strengths and richness of their interconnections progressively atrophy as individuals age; (2) the deteriorating brain machinery includes cortical areas and subcortical nuclei that are specifically related to sensation, cognition, memory, motor control, and affect; (3) the metabolic decline and down-regulation of key neuronal populations commonly precede cell death; (4) many aspects of physical and chemical deterioration and emergent neuropathology are correlated with general and specific behavioral losses; and (5) although there is substantial variability in the time of onset, course, and magnitude of functional and physical deterioration, these changes are a virtually universal outcome of the later years of an extended human life. The above observations have led to a viewpoint that is often termed the ‘‘wear and tear’’ hypothesis of cognitive decline in aging (Aldwin and Gilmer, 2004). This hypothesis suggested that brain machinery simply wears down over time. The observed anatomical changes (e.g., cell death, metabolic status, connectivity) and consequent functional changes (e.g., memory deficits, reduced processing speed, impaired spatial abilities) are the consequences of any biological or mechanical machine that has been in operation for multiple decades. The natural conclusion of the wear and tear hypothesis is that cognitive decline is normal, inevitable, and irreversible (Baron and Cerella, 1993). The physical aging of the brain obviously plays an important role in age-related cognitive decline. However, it is increasingly clear that the inevitable physical deterioration of the aging brain cannot completely account for many of the changes in functioning observed in older adults. The extensive literatures on brain plasticity and the perceptual psychophysics of aging strongly suggest that brain plasticity with negative consequences is a crucial contributor to age-related cognitive decline. As individuals age, their schedules and strengths of brain engagement substantially change, and are paralleled by active degradation of brain function. It is believed that such changes in brain use and engagement are direct and critical contributors to age-related cognitive decline.
Learning changes the brain through brain plasticity Brain plasticity refers to the brain’s lifelong capacity for physical and functional change; it is this capacity that explains how experience induces learning throughout life. The concept of brain plasticity is more than a century old (WoodruffPak, 1993), and its study has been ongoing for several decades. Historically, brain plasticity has been more often discussed in the contexts of early child development, stroke recovery, and perceptual learning than in regard to aging. Before the concept of lifelong brain plasticity was introduced, many researchers believed that the human brain was hard-wired in early life (Woodruff-Pak, 1993). Evidence supported this view by demonstrating that the brain developed the physical structures and long-range interconnections that determine neurological functioning during an early ‘‘critical period’’ of child development. It was established that, during this critical period, the brain was capable of substantial remodeling in response to alterations in input; but after the critical period closed, it was generally observed that the brain was not capable of further significant remodeling, elaboration, or growth. This notion that the brain developed its immutable long-range interconnections in early life contributed to the belief that age-related cognitive decline was inevitable and irreversible (Baron and Cerella, 1993). Today, after decades of accumulated cross-disciplinary research, a new and very different view has emerged about the origin and maintenance of human abilities. This view holds that the brain is plastic; that is, the brain is capable of reorganization, including developing new short-range interconnections, at any age throughout adult life. Brain plasticity experiments have documented a number of important ways in which progressive learning changes brain machinery. In aggregate, this research demonstrates that the adult brain continuously adapts to disproportionately represent relevant sensory stimuli and behavioral outputs with well-coordinated populations of neurons. This is achieved by engaging competitive processes in brain networks that refine the selective representations of sensory inputs or motor actions, typically resulting in increased strengths of cortical resources devoted
85
to, and enhanced representational fidelity (or ‘‘precision’’) of, the learned stimulus or behavior. Brain plasticity with positive consequences Competitive processes underlie all brain plasticity. In perceptual, cognitive, and motor skill learning tasks, competitive processes result in the narrowing of time and space constants that define the selectivity of processing in cortical networks. In this way, the selective responses of cortical neurons specialize to meet the specific demands of the task. The representation of input timing is dramatically elaborated, and the cortex’s ability to respond accurately in fast, precisely measured time as a receiver and a controller of action are dramatically advanced. For example, in monkeys trained to detect a specific pattern of stimulation to the fingers, the somatosensory cortex reorganizes to represent that specific input pattern with large, well-organized, and spatiotemporally coherent responses (Recanzone et al., 1992a, b; Wang et al., 1995). Similarly, in monkeys trained to perform demanding motor tasks (e.g., retrieving food pellets, turning bolts), the primary motor cortex reorganizes to represent the specific required motions of the digits and hand with larger cortical areas (Nudo et al., 1996). These types of changes have now been well documented in humans as well; for example, violin players have been shown to have stronger and more distinct representations of the fingers in the right hemisphere, corresponding to the individuated finger movements required by their left hands (Elbert et al., 1995). This example illustrates that beautifully elaborated and highly differentiated cortical representations develop in the learning of any highly skilled behavior. In learning skilled behaviors, such as playing a musical instrument, as brain machinery is progressively refined in its specificity, selectivity, and fidelity through competitive processes, it increases the representational power of behaviorally important sensory stimuli and motor outputs, as manifested by increased response magnitude and distributed response coherence. This is achieved, in large part, by plasticity processes that increase cortico cortical connectional strengths between neurons in nearly simultaneously excited cortical networks.
A key effect of this learning-induced change is to strengthen the signal-to-noise ratio of relevant cortical activity. Cortical systems operate against a constant background of internal noise from high spontaneous network activity levels; detecting the signal in this noise is a key challenge to such systems. Enhancement of the signal-to-noise ratio is likely to be a key mechanism by which learning improves brain function. The outcome of these competitive processes is positive because, through the locally adaptive processes by which the brain specializes to represent salient input, the brain’s processing machinery becomes more locally and globally adapted to perform important behavioral tasks. As a general principle, brain plasticity with positive consequences is likely to underlie virtually all forms of perceptual and skill learning in the brain. Brain plasticity with negative consequences We have just described how brain plasticity underlies all learning (e.g., perceptual, cognitive, motor). Because plasticity processes are inherently competitive, there always will be a competitive ‘‘winner’’ and ‘‘loser’’ (i.e., excitatory and inhibitory synaptic changes); thus, plastic changes with negative consequences are just as common as those with positive outcomes. Plasticity can be manipulated by adjusting the learning context; it is possible to actively degrade and weaken the brain processing machinery just as easily as it is possible to refine, elaborate, and strengthen the processing machinery. One example of plasticity with negative consequences is seen in monkeys trained under conditions in which heavy synchronous input is delivered across fingers (Jenkins et al., 1990; Allard et al., 1991; Wang et al., 1995) or the entire hand (Byl et al., 1996). In response to this type of sensory stimulation, the somatosensory cortex reorganizes to adaptively represent the undifferentiated spatiotemporal characteristics of the trained input. This results in an undifferentiated map — with abnormally large, overlapping receptive fields and degraded spatial and temporal response characteristics. While this map is adaptive in that it represents the use conditions of the hand, it is maladaptive in that the map does not support the
86
use of the hand under conditions that require the accurate processing of sensory inputs and motor outputs with high degrees of spatiotemporal complexity. Although this example shows how negative plastic changes can be actively induced, these changes more often occur naturally (i.e., without conscious effort) in later life. For example, as people age they commonly begin to stereotype and simplify behaviors that previously were quite complex and elaborated. The brain is likely to automatically adjust to these less complex behaviors by simplifying its representations that support them. We refer to these changes as brain plasticity with negative consequences because, through the locally adaptive processes by which the brain specializes to represent salient input, the brain’s processing machinery becomes less locally and globally adapted to perform important behavioral tasks. Brain plasticity with negative consequences is very likely to underlie specific pathological conditions (e.g., focal dystonia of the hand) (Byl et al., 1996) as well as general sensory or cognitive dysfunctions (e.g., learning impairments in children) (Tallal et al., 1996a). Based on a growing literature in the fields of psychophysics, neurology, neuropsychology, and brain plasticity, it is almost certain that the problems of age-related cognitive decline are substantially caused by negative dimensions of brain plasticity as well.
Age-related cognitive decline is a problem of brain plasticity with negative consequences Our forebrain processing machinery is sustained in a refined, powerful, and efficient operational state
by its intensive use under challenging conditions. In adulthood, continuous active interaction with environments that are demanding to sensory, cognitive, and motor systems is necessary to maintain brain health and cognitive fitness. As people age, a self-reinforcing, downwards spiral of reduced interaction with challenging environments and reduced brain health significantly contributes to cognitive decline (see Table 1). This downward spiral might begin either from a reduction in the schedule and engagement of brain activity or from an initial small loss in brain function driven by degraded sensory inputs (or, more likely, from both). In either case, once such a spiral begins, it continues through a sequence of interrelated events that reinforces a cascade of negative interactions, resulting in worsened cognitive fitness and brain health. We identify four interrelated factors as central and mutually reinforcing: 1. 2. 3. 4.
reduced schedules of activity noisy processing weakened neuromodulatory control negative learning
Reduced schedules of activity As people age, they typically change their activity patterns, such that the level of engagement in cognitively demanding activities is lessened (Hultsch et al., 1999). Even people with historically high levels of cognitive activity typically reduce their level of stimulation, either by conscious choice (e.g., retirement) or by unconsciously ‘‘resting on their laurels’’ and pursuing only activities at which they already excel. This results in less overall
Table 1. Root causes of functional decline in aging Reduced Schedules of Activity – Reduction in the schedules of inputs and actions that engage the brain that are required to continuously refine existing skills and drive new learning. Often referred to as ‘‘brain disuse.’’ Noisy Processing – Brain processing that produces low-fidelity, unreliable, and weakly-salient cortical representations of sensory inputs and actions. This occurs because the deteriorated brain produces poor signal quality, and must adjust its time and space constants to process these degraded signals, thus creating a noisy processing machine. Weakened Neuromodulatory Control – Down-regulation of metabolism and connectivity of neuromodulatory control systems caused by age-related physical deterioration and reduced schedules of activity. Negative Learning – Changes in behavior that accelerate cognitive decline, typically chosen because ordinary behaviors have become more difficult.
87
stimulation for sensory, cognitive, and motor systems, and importantly reduces stimulation for attention, reward, and novelty-detecting neuromodulatory systems. Through animal models, scientists have shown that the physical and functional consequences of brain disuse (engendered by exposing animals to impoverished environments) parallel the signature changes in the aged human brain. These studies have documented that a lack of brain engagement causes negative changes in neuronal metabolism (e.g., the production and function of neurotransmitters, receptors, and other key functional biochemical constituents of neurons), and in neuronal architecture (e.g., the elaboration of dendrites, axonal arbors, spines and synapses, cortical and subcortical neuropil, and gray matter) (Diamond et al., 1975; Katz and Davies, 1984; Sirevaag and Greenough, 1985; Beaulieu and Colonnier, 1989; Park et al., 1992; Melendez et al., 2004). These negative physical changes are accompanied by impaired learning and memory capacities that are thought to be the result of long-term alterations in neuronal plasticity driven by exposure to impoverished, nonstimulating, and noncomplex environments (Lewis, 2004). The parallels between the physical and functional changes seen in these models of brain disuse and those seen in the aged human brain are unmistakable, and the fact that such changes are reversible through environmental enrichment (Winocur, 1998) suggests that age-related cognitive decline may be slowed, arrested, or even reversed. Noisy processing Another consequence of aging is that sensory input from all systems (e.g., auditory, visual, tactile, proprioceptive) is degraded as a result of basic deterioration of peripheral sensory organs (e.g., loss of hair cells in the cochlea, loss of photoreceptors in the retina, changes in skin properties). The brain must adjust to these degraded sensory inputs by lengthening space and time integration constants in an effort to detect relevant signals. These adaptive changes are made at a cost — brain systems with long space and time integration constants cannot accurately represent the details of spatiotemporally complex signals. This inaccuracy
manifests as temporally and spatially noisy responses to relevant stimuli. These adaptive changes necessarily slow the speed of information processing as well. Weakened neuromodulatory control A further consequence of aging is that the metabolism, connectivity, and eventually, structure of neuromodulatory control systems, which regulate learning and plasticity in adults, become degraded. The key neuromodulators controlling plasticity are ACh (Bartus et al., 1982), which modulates synaptic plasticity in the hippocampus, cerebral cortex, and striatum (Doya, 2002), and controls memory and the rate of learning (Gu, 2002); dopamine, which mediates many aspects of cognitive, emotive, and motor functions (Gu, 2002), and is implicated in the prediction of reward and in action learning (Doya, 2002); serotonin, which regulates the time scale of reward prediction (Doya, 2002); and norepinephrine, which controls mental alertness and attentional focus (Usher et al., 1999). In aggregate, degraded neuromodulatory control systems weaken the brain’s control over its own plasticity, lowering learning rates and trapping the brain in potentially inappropriate or unhelpful patterns of activation. Negative learning As reduced schedules of activity, noisy processing, and weakened neuromodulatory control interact to make novel or demanding activities more challenging to perform, individuals naturally adapt their behaviors in ways that can reinforce negative aspects of the sensory input and motor output. For example, as it becomes harder to follow the rapid speech of a child on the telephone, an older adult might turn up the volume on the phone (increasing signal distortion along with loudness), find it more frustrating to have such conversations (decreasing neuromodulatory responses required to maintain high brain function), or simply choose to have fewer of such conversations (further reducing the schedule of brain activity). Substantial reorganizations in the responses of older brains to sensory and cognitive tasks relative
88
to younger brains have been measured using functional magnetic resonance imaging (fMRI). In a variety of such tasks, the changes seen in older brains can be interpreted as a dedifferentiation of response properties, including the recruitment of contralateral or new brain regions or the substitution of different brain regions to support task performance (Park et al., 2001). This neurological dedifferentiation is most likely a manifestation of the physiological brain plasticity with negative consequences we describe here. In aggregate, these four factors create a brain that is substantially less capable of representing the spatiotemporal detail of incoming stimuli, less able to represent such stimuli with strong, coherent, and salient neural activity, less able to actively modulate its own activity and capacity for change, and less able to support rapid interactions across relevant brain systems. Such a brain will manifest longer time and space constants, a slower processing speed, and integrated sensory, cognitive, and motor dysfunction. Below, we review the data from a wealth of psychophysical, neuropsychological, and cognitive studies that demonstrate that aged brains manifest reduced accuracy and speed of information processing as well as integrated dysfunction.
Cognitive decline is driven by changes across brain systems Although sensory and cognitive systems are often discussed and studied as separate entities, a large body of anatomical, physiological, and behavioral evidence suggests that, in fact, these systems are very tightly interrelated (Schneider and PichoraFuller, 2000). Information continuously flows both forward and backward through the brain’s sensory, cognitive, and motor systems. Sensory systems detect and analyze fundamental stimulus properties and feed this information forward to cognitive systems that store, manipulate, and act on it. Cognitive systems feedback to influence sensory processing through attention, expectation, memory, and context, while directly driving motor systems to execute planned activities. Motor systems are tightly integrated with cognitive systems through premotor areas involved in movement
planning, and indirectly provide feedback to sensory systems through proprioceptive and vestibular systems. Because sensory, cognitive, and motor systems are parts of a highly integrated information processing system, disruption in any one system would be expected to cause disruption in the others, and degrade the overall accuracy and speed of information processing (Schneider and PichoraFuller, 2000). Indeed, deficits in sensory, cognitive, and motor functioning are common in older adults (Harvey and Mohs, 2000). Although this chapter focuses on sensory and cognitive deficits, the principles and science underlying these issues are also directly relevant to motor function in older adults. Researchers have begun to explicate the complex ways in which these systems interact. It is now clear that sensory systems with degraded function negatively affect cognitive function. The source of such degraded sensory input has typically been assumed to be in the periphery (e.g., loss of hair cells in the cochlea, loss of photoreceptors in the retina) given the well-documented changes that occur there (Scialfa, 2002; Madden et al., 2003). However, a growing literature has shown that central sensory processing deficits play a significant role in the reduced cognitive performance of older adults as well (Schneider and Pichora-Fuller, 2000; Faubert, 2002). In the auditory system of older adults, the negative effect of sensory losses on cognitive performance has been extensively documented. Similar findings are emerging in the study of the visual system. Other systems (e.g., somatosensory, vestibular) also decline with age, although their relationship to associated cognitive systems is not yet well understood. Below, we summarize the deficits in the auditory and visual systems, and the research that explores how these deficits contribute to cognitive decline in older adults. Changes in the central auditory system contribute to cognitive deficits in aging Many adults experience a decline in auditory sensitivity with age, called presbycusia, which is commonly experienced as a sensory loss in the highfrequency range of hearing, and is caused by the deterioration of inner hair cells in the cochlea. However, many other age-related auditory sensory
89
deficits have been shown to exist independent of, or in combination with, high-frequency hearing loss, suggesting that these deficits cannot be solely attributed to deterioration of the peripheral sensory system and must be rooted in the central auditory system (Schneider and Pichora-Fuller, 2000). The ability to temporally resolve an auditory signal is critical for accurate speech perception (Drullman, 1995a, b) and decreases with age (Abel et al., 1990; Moore and Peters, 1992; Fitzgibbons and Gordon-Salant, 1994; Schneider et al., 1994). By comparing the temporal resolution abilities of young adults with good hearing, hearing-impaired older listeners, and older listeners with good hearing, researchers have determined the extent to which hearing loss and age may mediate temporal resolution. These studies consistently have shown that older adults with good hearing have reduced temporal resolution compared to younger adults with good hearing (i.e., temporal resolution declines with age), and that there is no difference in temporal resolution abilities of older listeners with good hearing and those with hearing loss (i.e., reduced temporal resolution in older adults is unrelated to hearing loss) (Abel et al., 1990; Moore and Peters, 1992; Fitzgibbons and Gordon-Salant, 1994; Schneider et al., 1994). Older adults also experience challenges with speech perception. A common complaint among older individuals is that everyday conversations are hard to understand — speakers seem to mumble or speak too fast, and cannot be understood in noisy situations (Schneider et al., 2002). Even when adults with good hearing sensitivity are in quiet conditions, they may not fully understand all words or speech sounds (Schneider et al., 2002). Consistent with this subjective loss of cognitive efficacy, experimental studies have shown that older adults make more errors than younger adults in recognizing and remembering fast speech (Pichora-Fuller et al., 1995; Wingfield et al., 1999; Schneider et al., 2002), speech under noisy conditions (Humes and Roberts, 1990; Humes and Christopherson, 1991; Murphy et al., 2000), and speech lacking contextual cues (Gordon-Salant and Fitzgibbons, 2001). Moreover, these deficits are apparent even when controlling for hearing loss (Dubno et al., 1984; Cheesman et al., 1995;
Gordon-Salant and Fitzgibbons, 2001; GordonSalant and Fitzgibbons, 1993), suggesting that peripheral sensory loss is not the only factor in this aging deficit. Although the neurological origins of these deficits in speech perception are not yet well understood, significant insights have been gained from studies comparing younger and older adults. In studies of speech rate and background noise on speech recognition in older and younger adults, the performance of younger adults when listening to rapid (Wingfield and Lindfield, 1995; Wingfield, 1996) or noisy (Schneider et al., 2002) speech was similar to that of older adults listening to slower speech or speech under quiet conditions. These results suggest that the neurological dysfunctions in older adults act to lower the temporal fidelity of and add noise to auditory input. These and other studies have gone on to demonstrate that these perceptual deficits have deleterious consequences for memory and cognitive performance in older adults. Several studies of the role of noise in speech processing have shown that when young adults performed verbal memory tasks under signal-to-noise conditions that matched their sensory performance to the relatively poor performance of older adults, their memory abilities were equivalent to those of older adults (Murphy et al., 2000; Schneider et al., 2002). These results demonstrate that the poor sensory function of older adults can significantly impair their memory for speech. In aggregate, there is a large and increasingly deep body of knowledge from studies of auditory psychophysics, perception, and cognition in older individuals, which argues that degraded representational fidelity and noise in the central auditory system is responsible for crucial auditory processing deficits seen in older adults, and that these sensory processing deficits can in turn cause meaningful deficits in memory and cognitive functions (Schneider and Pichora-Fuller, 2000). The logical implication of this literature and the literature of brain plasticity is that a training program designed to improve the fidelity of the representation of auditory stimuli in older adults should lead to substantially improved cognitive and memory performance in tasks involving the auditory system.
90
This conclusion is not in conflict with studies that have shown that peripheral sensory loss also contributes to cognitive deficits in older adults. However, we contend that central processing deficits could be remediated with a plasticity-based training program while there is no known remedy for peripheral sensory loss. Changes in the central visual system contribute to cognitive deficits in aging Age-related changes in the eye, whether caused by specific pathologies (e.g., cataracts, glaucoma, macular degeneration) or generalized issues in aging (e.g., decline in the number of rods), can reduce visual acuity, contrast sensitivity, color vision, and light sensitivity. This decline in the visual peripheral sensory system without question contributes to a less accurate and more noisy representation of the visual world in older adults. However, as in the auditory system, there is a growing body of evidence that shows that the decline in various basic visual abilities occurring with age is independent of optical factors (Morrison and McGrath, 1985; Owsley et al., 1985; Nameda et al., 1989), suggesting that a deteriorated central visual sensory system produces a noisy representation of the visual world that substantially contributes to a decline in visual cognitive processing (Schneider and Pichora-Fuller, 2000). Impairments in central visual perception include a difficulty in detecting and discriminating between static peripheral targets; problems with motion perception; an impaired ability to track and visually process moving objects (Sharpe and Sylvester, 1978; Scialfa and Kline, 1988; Kline, 1994; Olincy et al., 1997), difficulty in identifying and discriminating letters (Akutsu et al., 1991), trouble inferring threedimensional structure from two-dimensional images (Plude et al., 1986; Robins-Wahlin et al., 1993), and difficulty in mental rotation (Dollinger, 1995); and poorer face discrimination abilities (Owsley et al., 1981; Eslinger and Benton, 1983; Koss et al., 1991; Cronin-Golomb et al., 2000). Older adults show deficits in backward masking tasks involving visual stimuli (Kline and Birren, 1975; Kline and Szafran, 1975; Walsh, 1976) and in flicker fusion tasks (Kim and Mayer, 1994), both of
which suggest abnormalities in temporal integration. They are also less sensitive to object movement (Elliot et al., 1989; Kline et al., 1994), and are less able to detect coherent motion (Trick and Silverman, 1991; Wojciechowski et al., 1995). In combination, these spatial and temporal psychophysical deficits suggest that problems in the central visual system substantially impair even the earliest stages of visual processing in older observers. As in the auditory system, a growing number of studies suggest that age-related problems in visual cognition, visual memory, and visuospatial skills can be traced to degraded central visual system processing. Lowering the contrast of stimuli in visual neuropsychological tasks to mimic the contrast deficit of older adults decreases the cognitive performance of younger adults to that of a typical 50–55-year-old (Spinks et al., 1996). Improving the contrast of stimuli significantly increases the performance of older adults in reading comprehension (Echt and Pollack, 1998). When noise is added to visual stimuli to mimic the poor discrimination abilities of older participants, the ability of younger participants to identify visually presented words declines, such that it can no longer be distinguished from the performance of older adults (Speranza et al., 2000). Although this literature is less well developed than research in the auditory system, these findings argue that degraded representational fidelity and noise in the central visual system are responsible for crucial visual processing deficits seen in older adults, and that these sensory processing deficits can in turn cause meaningful deficits in memory and cognitive functions. Together, these literatures suggest that degraded representational fidelity and consequent cognitive and memory deficits are general operating principles of the aging brain, and that training programs targeting each sensory system in turn should lead to substantially improved cognitive and memory performance. Physical and functional deterioration in the brain can be slowed, arrested, and reversed As already mentioned, there now exists ample scientific support for the idea that the brain has a lifelong capacity for plasticity. We have just
91
described how reduced schedules of activity, noisy processing, weakened neuromodulatory control, and negative learning work in concert to significantly impair cognition in older adults. If these conditions were reversed, would it be possible to restore cognitive functioning in older adults that had experienced significant decline? Evidence from human and animal research strongly indicates that substantial physical, sensory, cognitive, and motor recovery is possible. Human studies have documented that cognitive activity wards off future decline with aging, perhaps by building what others refer to as a ‘‘cognitive reserve,’’ which may be a euphemism for strengthened brain processing machinery (Whalley et al., 2004). In the past few years, various welldesigned, prospective studies have shown that participation in cognitively stimulating activities (Wilson et al., 2002, 2003), intellectually complex work (Schooler et al., 1999), and leisure activities (Scarmeas et al., 2001; Verghese et al., 2003) during adulthood reduces the risk of loss of cognitive abilities in later life. Involvement in cognitive activity would clearly counter each of the four conditions believed to contribute to functional decline in the aged: cognitive activity is the opposite of brain disuse; it would strengthen the brain processing machinery to ensure less noisy processing; it is likely to be done in behavioral contexts (e.g., attention, reward, novelty) that strengthen neuromodulatory control; and it disrupts the downwards spiral of negative learning. Human behavioral studies have shown that losses in sensory, cognitive, and motor processing can be reversed. Specific training can refine degraded representations in the sensory and motor cortices (Bao et al., 2003; Byl et al., 2003), improve signal-to-noise conditions for neuronal representations and distributed neuronal response coherence (Deutsch et al., 2000; Nagarajan et al., 2000; Nagarajan and Merzenich, unpublished manuscript), and restore the effectiveness of long-range feed-forward connections (Temple et al., 2000, 2003; Olesen et al., 2004). Neglected cognitive skills can be strengthened and refined by use (Wolf et al., 2001; Dick et al., 2003). A large and growing body of animal studies have shown that an enriched environment
designed to be cognitively stimulating promotes positive plastic changes in the brain and can reverse the negative physical, sensory, and cognitive aspects of aging. (For reviews, see Diamond, 2001; Mohammed et al., 2002; Lewis, 2004; Li and Tang, 2005). These studies have shown that new neuron production can be increased in areas where cell division and proliferation are possible (e.g., the hippocampus) (Kempermann et al., 1997, 2002; Lemaire et al., 1999) and apoptotic cell death can be reduced (Young et al., 1999). Gray matter can be thickened: dendrites, spines, and synapses in the cortical neuropil can be elaborated (Diamond et al., 1975; Greenough et al., 1978, 1985; Floeter and Greenough, 1979; Green et al., 1983; Diamond et al., 1985; Mohammed et al., 1993; Rosenzweig and Bennett, 1996; Mattson et al., 2001; Kleim et al., 2002; Mohammed et al., 2002; Frick and Fernandez, 2003; Frick et al., 2003). Even myelination, which had previously been thought to be irrecoverable in the adult brain, can probably be restored (Stevens et al., 2002; Saleh et al., 2003; Piraino et al., 2005). Neuromodulatory control systems, weakened during aging, can be strengthened by behavioral training. Such training can up-regulate the metabolic states and the production and release of key neurotransmitters of limbic system and basal ganglion neurons (Nakamura, 1991; Bezard et al., 2003; Cohen et al., 2003; Tillerson et al., 2003). Through their more effective and intense reactivation, cortical and subcortical terminals of modulatory control nuclei can be elaborated (Wolfman et al., 1994; Spengler et al., 1995). Under optimal environmental conditions, almost every physical aspect of the brain can recover from age-related losses. A degradation of vascular dynamics attributable to ACh control of nitric-oxide-related enzymes believed to presage AD pathology may be at least partially overcome (Hilbig et al., 2002). A recent study demonstrated that even certain pathological hallmarks of AD (e.g., amyloid bodies) could be ameliorated by exposure to an enriched environment (Lazarov et al., 2005). Numerous studies have shown that behavioral training or exposure to a novel environment can refine representations in the sensory, somatosensory, and motor cortices of animals (Wang et al.,
92
1995; Nudo et al., 1996; Xerri et al., 1996, 1998; Byl et al., 1997; Coq and Xerri, 2001; Nudo et al., 2003). In addition, exposure to an enriched environment can improve memory and learning in older animals (Doty, 1972; Cummins et al., 1973; Berman et al., 1988; Kobayashi et al., 2002; Frick and Fernandez, 2003; Frick et al., 2003; Fernandez et al., 2004), and those that are cognitively impaired (Rampon et al., 2000; Arendash et al., 2004) perhaps by inducing synaptic structural changes that enhance memory and learning (Kempermann et al., 1997; Rampon et al., 2000). The negative effects of earlier exposure to an impoverished environment can be at least partially reversed by exposure to an enriched environment later in life (Winocur, 1998). This very large group of studies shows that substantial recovery is possible in the aging brain: physical losses can be reversed, and many aspects of sensory, motor, neuromodulatory, and cognitive systems can be restored to optimal levels of functioning. Additional experimental and applied studies conducted over the next several decades will more clearly define the extent of positive plastic changes that may be achieved, and the conditions under which these changes can be maintained over time. For brevity’s sake, we have selected four examples of brain plasticity research for more in-depth discussion. In each example, we describe how negative plasticity processes degrades sensory, cognitive, or motor functioning and how positive plasticity has restored behavioral and neurological functioning.
Age-related decline in rat Negative plasticity in elderly rat As a rat nears the end of its life, it loses control of its forepaws. This loss is manifested, for example, by increasing functional difficulty in food object retrieval and manipulation. Across the same period of time, the rat’s mobility becomes progressively degraded: its gait becomes slow and clumsy. If the rat lives long enough, it will lose the ability to control its hind legs in locomotion; turn its feet
over so that their hairy dorsal surface is touching the cage surface (apparently because contact of the glabrous surface with the ground is painful); and drag itself around its environment using its forelegs. The aged rat’s difficulty in feeding itself contributes to the rat’s death shortly into its third year of life. The neuronal basis of the rat’s loss of control of its paws and limbs is revealed through reconstructing the cortical maps of the rat’s paw surfaces in the somatosensory cortex and movement representations in the motor cortex (Godde et al., 2002). The cortical maps of the paw surfaces and movement representations are profoundly dedifferentiated and noisy in the old rat, with cortical neurons responding weakly and unreliably (Coq and Xerri, 2000; Godde et al., 2002). In such a cortex, inputs have weak salience and poorly engage nondeclarative memory processes crucial for sustaining complex, normal forepaw grasp behaviors. In the 2-year-old rat, physical signs of deterioration are broadly expressed in the somatosensory and motor cortices, and in subcortical thalamic and limbic system nuclei (Godde et al., 2002). The rat’s gray matter is thin; the neuropil is shrunken; neurons have less complex dendrites and fewer spines; synapses are less elaborate; intracortical axons are less complexly branched; and myelination is reduced. In sum, the cerebral cortex, and the subcortical thalamic and modulatory control nuclei that support it, are slowly dying. This deterioration in the cortex and the modulatory subcortical nuclei is paralleled by slower learning rates and lower learning ceilings in aged animals. A degradation of dynamic control of vascular perfusion almost certainly is due to the inexorable degradation of modulatory control system (nucleus basalis, locus coeruleus) function. Positive plasticity-based training reverses age-related decline Many of these destructive functional and physical changes in the aged rat can be reversed through appropriate, targeted, intensive retraining of the rat’s forepaws, and postural and mobility control (Churs et al., 1996; Reinke and Dinse, 1999). The
93
training that was applied in this model engages the rat in behaviors that progressively reestablish refined representations of the paw surfaces and of paw and limb movements. Rats were trained to cross-rotating bumpy rods to retrieve a food reward. The difficulty of the task was progressively increased by narrowing the bar and increasing the rate of rotation. Successfully crossing the bar delivers significant spatiotemporally complex input to the somatosensory system while demanding substantial attentional resources and requiring complex motor output. Following training, the reversal of these physical changes is revealed by the restoration of relatively normal forepaw maps in the primary somatosensory and motor cortices. The restoration of the cortical maps in these rats translated into functional recovery: trained rats recovered their ability to manipulate food objects with their forepaws and control their limbs in locomotion. With these functions restored, the rats lived 4–5 months longer than would otherwise be the case. Although the documentation of physical changes following such training is still incomplete, the limited studies conducted to date and other related experiments show that the cortical gray matter can be thickened, largely through an increase of neuropil; dendrites can be elaborated; the resting metabolism of cortical and subcortical areas can be increased; the reversal of the loss of dynamic brain perfusion and the up-regulation of a nitric-oxide-related enzyme indicates that ACh production and nucleus basalis function becomes more normal. This model shows that appropriately structured and intensive behavioral engagement can substantially reverse both physical and functional deterioration of complex forebrain-mediated behaviors in older animals. While the changes induced in these experiments do not completely physically or functionally restore youthful behaviors, the positive, normalizing changes that do occur are expressed on a very large scale. Moreover, the scope of the training employed in these experiments has been very limited. More intensive and elaborate training should produce considerably more powerful and complete reversals of functional and physical losses.
Remediation of acquired hand movement disorders Negative plasticity in overtrained humans and monkeys Acquired hand movement disorders (e.g., focal hand dystonia) often arise from specific forms of occupational hand use in humans (e.g., playing the piano, keyboard data entry). We have induced acquired movement disorders in monkeys through a negative plasticity scenario, and shown that the acquired loss of motor control is a consequence of learning-driven dedifferentiation of sensory and motor cortex representations in the forebrain. Humans with acquired hand movement disorders show the same pattern of degraded hand cortical maps as were seen in the primates. This is a predictable consequence for any behavior in which there is a competitive ‘‘winner’’ achieved through very stereotypic excitation of skin surfaces, or in which nearly identical, larger sectors of skin are consistently simultaneously activated (Wang et al., 1995; Byl et al., 1996; Nudo et al., 1996; Xerri et al., 1996; Elbert et al., 1998; Merzenich, 2001). Inputs that are nearly simultaneously engaged by a very stereotypic or broadly engaging stimulus will be mutually costrengthened, and their integration will result in larger receptive fields. By natural plasticity processes, the separate, normally differentiated representations of sensory inputs from fingers and palmar surfaces can be largely subsumed by an undifferentiated cortex in which almost any stimulus excites almost any hand-zone neuron (Wang et al., 1995; Byl et al., 1996; Elbert et al., 1998). The behaviors that generate dedifferentiation of cortical hand representations in monkeys are exactly the kinds of hand-use behaviors that generate acquired hand movement disorders in humans (Byl et al., 1996; Byl, 2004). Positive plasticity-based training reverses acquired hand movement disorders Using strategies that have much in common with the training conducted on rats described earlier, we developed a behavioral training program to reverse chronic hand dystonias in humans. The training program uses multiple tasks designed to
94
engage the broad types of sensory inputs that provide feedback for fine hand motor control (Byl et al., 2000). Training begins with sensory discrimination tasks and progresses to graded movements, sensorimotor activities, motor control activities of daily living, fine motor practice, and finally targetspecific task practice. Training tasks are progressively difficult and modified by the context, content, force, and mobility of the stimulus while being targeted to a restricted skin surface. All training activities require substantial attention, and success is actively rewarded. With training applied for about one hour per day for 1–2 months, most patients (the majority of whom have been professional musicians) recover relatively normal hand use (Byl and McKenzie, 2000; Byl et al., 2003). Gains with training were independent of the age of the individual. Initial brainimaging studies indicate that training results in more normal representational topographies and sensory-evoked responses in participants. These studies strongly indicate that training programs can reverse functional and behavioral losses in fine motor control and sensory input from the hand in adult humans. As in trained rats, such training rerefines grossly dedifferentiated maps of sensory inputs and movement governing sensoryguided motor behaviors.
Up-regulating dopamine cell function in adult rats Limb disuse exacerbates Parkinsonian symptoms in rat Parkinson’s disease is characterized by progressive motor impairment caused by degeneration of dopaminergic (DA) neurons in the nigrostriatal system (Zigmond and Burke, 2002). Researchers can induce hemi-Parkinsonian symptoms (e.g., slowness or loss of movement on the affected side, preferential use of the nonaffected side) in the rat by unilaterally destroying a percentage of DA neurons in the basal ganglion. These Parkinsonian symptoms are exacerbated by restraining use of the rat’s impaired forelimb (Tillerson et al., 2002), suggesting that a decrease in physical activity not only is a symptom of the disease but also
contributes to the disease process, perhaps through negative plasticity processes. Reversing behavioral and neurochemical losses through exercise and exposure to a complex environment A series of studies has demonstrated that behavioral and environmental conditions can reverse the behavioral and neurochemical losses in Parkinson’s-induced rats. One study forced the rat to use one of its forelimbs for a period of time before inducing behavioral and neurochemical deficits in the forced-use limb (Cohen et al., 2003). This forced exercise attenuated the loss of striatal DA neurons and its metabolites in response to inducing degeneration of DA neurons in the rat. A second study forced the rat to exercise its impaired limb after it had been injured, and showed that this exercise reversed the induced behavioral deficits (Tillerson et al., 2003). This recovery from lesioninduced behavioral deficits is paralleled by an attenuation of the depletion of striatal DA neurons in rats (Tillerson et al., 2003). The gains made during the exercise program are lost once the exercise is discontinued, indicating that continuous ‘‘therapy’’ is needed to maintain improvements (Woodlee and Schallert, 2004). Other studies have shown that a motor learning environment (a ‘‘rich’’ environment in which objects in the rat’s cage are changed daily) can prevent the progression into Parkinson’s-like symptoms in rats whose DA neurons have been destroyed (Bezard et al., 2003; Faherty et al., 2005). This prophylaxis is almost certainly attributable to the heavy, learning-activity-based engagement of surviving DA neurons. Together, these studies of behavior and neurodegeneration in rat models of Parkinson’s disease strongly indicate that the health and vitality of the DA neuromodulatory system is regulated by its own functional activity. A training program of forced use blocks the progression of cell loss and the exacerbation of the down-regulation of DA neuron production and release attributable to motor dysfunction, while symptom progression is reversed. This result opens the door for the investigation of active training programs that go
95
beyond forced use to revivify DA as well as other neuromodulatory control systems in the brain.
Reversing language learning and reading impairments in children and young adults Negative plasticity in children with learning and reading impairments Over the past two decades, brain plasticity studies in monkeys and rats have led to the hypothesis that impaired language development commonly leading to reading problems is often a consequence of an early plasticity outcome by which the cortex organizes its aural speech processing machinery to specialize for the representation of a degraded (noisy) speech model (Merzenich et al., 1993, 1998a, b; Merzenich and Jenkins, 1995). Many inherited neurological faults, as well as consistently degraded inputs from the inner ear, would be expected to result in a muffled or noisy speech model. The ‘‘fuzzy’’ phonemic representations behaviorally and physiologically documented in language-impaired children are consistent with this scenario. Moreover, theoretical and animal models of this developmental scenario predict problems in speech representation and language function that are consistent with the behavioral deficit picture presented in most language-impaired and readingimpaired children. Positive plasticity-based training reverses and young adults with learning and reading impairments More than a decade ago, researchers posited that these processing deficiencies might be corrected in individuals of any age by appropriate, targeted, and intensive behavioral training. A seven-exercise training program was developed, based on the principles of brain plasticity cited above, to adaptively renormalize speech feature representation and generalize improved processing abilities to all of the syntactic relationships (contexts) that are needed for facile speech reception. About five thousand individuals spanning from about 4 to 18 years of age were given standard assessments testing all aspects of speech reception and language
usage before and after training. The results demonstrated large-scale improvements on virtually every language-related cognitive or memory task (Merzenich et al., 1996a, 1998a, b; Tallal et al., 1996a, b ). Benefits generalized to aural speech assessments of memory, cognition, and ‘‘processing efficiency,’’ and to language usage (Tallal et al., 1996b; Merzenich et al., 1998a). Benefits from training were independent of the age of the individuals (Tallal et al., 1996b; Merzenich et al., 1998a, b). More than 600,000 children and young adults have now been trained with this program. The conclusion that the training increases representational salience was confirmed by neurological studies that longitudinally reconstructed dynamic cortical responses. fMRI studies revealed that the originally very abnormal response patterns recorded while these children performed key reading and language behaviors could be consistently restored to a more normal form after training. Brain imaging and human recording studies have demonstrated that the neuronal representations of aural speech inputs are substantially more salient, powerful, and reliable after training (Deutsch et al., 2000; Nagarajan et al., 2000; Temple et al., 2001, 2003; Hayes et al., 2003 Nagarajan and Merzenich, unpublished manuscript). Trained children had higher amplitude and more coordinated magnetically recorded responses to the sound parts of words represented within the primary auditory cortex. Event-related potential studies have shown that the discriminable differences that distinguish confusable speech phonemes were renormalized in the average trained child. The level of coherent gamma activity evoked in memory-related tasks, which was grossly abnormal before training, became normal after training in children for both quiet and noisy background conditions. The four studies described above provide compelling proof-of-principle demonstrations that brain-plasticity-based training programs reverse noisy processing, renormalize temporal and spatial integration constants, enhance functioning of neuromodulatory systems relevant to learning and memory, and improve cognitive performance in a variety of animal and human models of neurological dysfunction. Each of these dysfunctions is
96
directly relevant to the cognitive challenges faced by older individuals, and each suggests specific aspects of training programs that would be relevant to slow, arrest, or reverse age-related cognitive decline.
Current approaches to treating cognitive decline have limited applicability and efficacy Two general strategies have been pursued to ameliorate the cognitive changes seen in aging: pharmacological and behavioral. Pharmacological approaches have focused on blocking and possibly reversing the pathological processes that contribute to the physical and functional deterioration of the brain in clinically defined conditions, typically AD (and now, MCI). By targeting one of these hypothesized pathological processes, such as vascular changes, amyloid deposits, or the levels of important neuromodulatory transmitters, it is hoped that adults diagnosed with such diseases can retain memory abilities, cognition, executive functions, and movement control. Behavioral approaches have generally focused on teaching specific strategies in memory and attention to healthy older individuals as well as to those diagnosed with AD or MCI.
Pharmacological approaches The most frequently applied drugs in patients with MCI and AD are acetylcholine esterase (AChE) inhibitors, which are designed to enhance levels of ACh in the brain by blocking the normal ACh breakdown. Positive benefits provided by AChE inhibitors have been modest. A recent systematic review of clinical trials of AChE inhibitors found that only 10–20% of patients with AD benefited from these drugs, and that high rates of noncompliance among treated patients were commonplace (Kaduszkiewicz et al., 2005). Additionally, the benefit from treatment is only temporary; on an average, patients improve over 3–6 months and return to pretreatment status 9–12 months after treatment initiation (Johannsen, 2004). In patients with MCI, donezepil slows the progression to AD for 12 months, after which progressive functional
decline continues at the same rate as control patients (Peterson et al., 2005). Even with the potentially dramatic costs savings that come from delaying AD onset, the cost effectiveness derived from AChE inhibitors is uncertain because of the high cost of these drugs and the fact that all patients still advance, with perhaps a brief delay, to the more advanced stages of AD (Foster and Plosker, 1999). Research scientists have been working intensely for more than a decade to improve the therapeutic landscape for MCI and AD treatment. There are more than 100 drugs now in the pipeline targeting many different pathological processes believed to contribute to cognitive decline. Other research approaches have investigated strategies for promoting neuron regeneration or replacement using genetic modification or stem cell-based approaches designed to reinvigorate, protect, or grow more new neurons, or to provide new sources of neurons for deteriorating brains. While these approaches are hopeful, no practical strategy is in hand, and Food and Drug Administration (FDA) approval for their use likely will not happen for a number of years. These investigational paths are promising and will almost certainly lead to improved treatments for AD and perhaps also for MCI. At the same time, none of these future therapies addresses the tremendous problem of normal age-related cognitive decline. In addition, even novel drugs may be only marginally helpful for patients with AD or MCI because they usually address only a single dimension of the complex multidimensional processes of brain deterioration in aging. Behavioral approaches A number of studies have suggested that cognitive activity or stimulation could be a protective factor against the functional losses of aging. Because many of these studies were cross sectional, the causal relationship between stimulation and cognitive performance was difficult to establish. In the past few years, however, a number of well-controlled longitudinal studies have shown that participation in cognitively stimulating activities (Wilson et al., 2002, 2003), intellectually complex
97
work (Schooler et al., 1999), and leisure activities (Scarmeas et al., 2001; Verghese et al., 2003) during adulthood reduces the risk of loss of cognitive abilities in later life. For example, one study measured older individuals’ self-reported frequency of participation in a variety of activities, with each assigned an objective level of cognitive stimulation, over a period of 5 years. Older individuals at the highest level of cognitive activity (90th percentile) experienced a 35% less decline in their cognitive abilities than individuals with low levels of cognitive activity (10th percentile) (Wilson et al., 2003). The results of these studies are entirely consonant with the negative brain plasticity viewpoint on age-related cognitive decline, as the cognitive stimulation quantified in these studies would directly affect issues of disuse, noisy processing, weakened neuromodulatory control, and negative learning. To date, various behavioral training strategies have been proposed to remediate age-related cognitive or memory impairment. The studies evaluating these approaches typically have applied strategy learning to enhance memory function in healthy older adults, with some evidence for success (Yesavage, 1983, 1989; Verhaeghen et al., 1992; Caprio-Prevette and Fry, 1996; Verhaeghen and Marcoen, 1996; Mohs et al., 1998; Glisky and Glisky, 1999; McDougall, 1999). However, these approaches are limited by their approach, and have generally shown small effect sizes, poor maintenance over time, and no generalization beyond the trained skill (Gatz, 2005). More recently, several small trials have been completed that assessed the impact of behavioral training in patients with AD (Davis et al., 2001; Clare et al., 2002; Loewenstein et al., 2004). These studies demonstrated that significant short-term improvements in certain cognitive functions were achievable even in this severely impaired population, although the extent to which these changes generalize more broadly to cognitive and everyday functioning has not yet been established. It is difficult to fully evaluate the promise of any single behavioral approach as few large, rigorous studies have been conducted to evaluate proposed training programs. A notable exception is the ACTIVE (advanced cognitive training in vital elderly)
study, a randomized, controlled trial that evaluated three behavioral training programs (in speed of processing, memory, and reasoning) in older adults (Ball et al., 2002; Edwards et al., 2002). Training in any one of these areas improved performance in that area, but this did not translate to improved everyday functioning possible due to a ceiling effect. These studies clearly demonstrate the promise of training-based approaches; however, we believe that a more intensive and comprehensive training program explicitly based on the principles of brain plasticity would most likely achieve more robust benefits than have been seen to date. In general, we do not expect that compensatory strategies, or training that targets only higher order cognitive functions, will achieve powerful, sustained effects because such strategies do not address the fact that age-related decline in sensation, memory, cognition, and guided motor control has more fundamental roots in degraded brain processing. Until optimal programs are developed, we believe it is unlikely that a standard of care for a behavioral approach to cognitive decline in aging will emerge. A novel training program to enhance memory and cognition in the aged The negative plasticity perspective on the origins of cognitive decline in older individuals immediately suggests a novel approach to treating such cognitive losses. We have built an initial version of a brain-plasticity-based training program explicitly designed to intervene in the downward cycle of negative plasticity by enhancing signal-to-noise ratios and improving neuromodulatory function, while also increasing overall brain stimulation and correcting negative learning. This brain-plasticitybased training program operates on four basic principles. Strongly engage the brain To reverse underlying disuse and drive brain plasticity, the program strongly engages the brain with demanding exercises and a daily training schedule. Thousands of trials are required to ensure that the representations of behaviorally important inputs
98
are coselected or integrated to create robust and complex stimulus-specific and action-event-specific neuronal responses in the cortex. In addition, program exercises employ an adaptive training approach that begins with simplest tasks where there is a high likelihood of success, and proceeds adaptively and incrementally with a series of exercises in which the task demands are made gradually more difficult. Performance within each component is overlearned through repetitive, successful practice with rewards.
requiring temporally focused periods of attended behavior with the goal of stimulating the nucleus basalis in every training exercise cycle (Kilgard and Merzenich, 1998a). Rewards are delivered several thousand times in each daily training session to exercise DA systems in the ventral tegmental area and the substantia nigra (Backman and Farde, 2005). Serotonergic and noradrenergic novelty detection systems are targeted with similar frequency to stimulate the locus coeruleus and the dorsal raphe nucleus.
Renormalize noisy processing
Strengthen critical life skills
The training program aims to improve the ability of the brain’s auditory and speech systems to engage memory and cognitive systems by enhancing their representational fidelity. Training tasks and stimuli are designed to sharply increase the fidelity and power of representations of complex, dynamic inputs; decrease spatial and temporal integration constants; and directly assure the effective generalization of highly spatially and temporally refined processing to all of the contexts for facile and efficient ‘‘complex’’ (i.e., real) signal reception and memory.
Besides targeting fundamental aspects of brain plasticity, the program aims to guide users out of learned behaviors with negative consequences for brain health and into new behaviors that positively reinforce their enhanced brain function. Structure of the training program The overall program is composed of six interrelated training exercises that in aggregate span the acoustic organization of speech. The exercises include the following:
Enhance neuromodulatory function
Program exercises are also designed to strengthen the basic function of each neuromodulatory system component essential for the regulation of learning and memory. Dimensions of behavioral context (arousal, attention, reward, novelty) affect the release of specific neurotransmitters (ACh, dopamine, serotonin, norepinephrine, endogenous opioids) that in turn enable, amplify, and shape plasticity in the adult brain (Merzenich and Jenkins, 1993; Merzenich et al., 1996b; Cahill and McGaugh, 1998; Kilgard and Merzenich, 1998a, b; Bao et al., 2001; Merzenich, 2001; Gibbs and Summers, 2002; Kilgard and Merzenich, 2002; Weinberger, 2003; Schweighofer et al., 2004). For training to be maximally efficient, attentional, reward, and novelty detection system engagement must be closely controlled to achieve near-optimum learning rates. Exercises are specifically designed to engage cholinergic attention systems by
‘‘High or Low’’: frequency-modulated sweeps (time-order judgment task) ‘‘Tell Us Apart’’: syllables (discrimination task) ‘‘Match It’’: short words with confusable stop-consonants (spatial-match task) ‘‘Sound Replay’’: short words with confusable stop-consonants (forward-span task) ‘‘Listen and Do’’: complete spoken sentences (instruction-following task) ‘‘Story Teller’’: complete spoken narratives (narrative-memory task)
Each exercise employs a combination of acoustically emphasized stimuli, adaptive training procedures, and intensive engagement of attentional, reward, and novelty-detection systems. In aggregate, these exercises are designed to improve the accuracy and the speed with which the brain processes speech information, and reengage the neuromodulatory systems that gate learning and
99
memory. By doing so, we hypothesize that the representational salience of speech input is improved in older brains, that the functional connectivity (feed forward and feedback) between sensory and memory systems would be improved, and that as a result, speech reception accuracy, speed of processing, and memory for speech would improve. Pilot results from randomized, controlled study As an initial test of this training program, we conducted a randomized, controlled, pilot study designed to assess the usability of this kind of demanding, intensive program in a classroom environment by older individuals, and to estimate the effect size of intervention on standardized neuropsychological measures of memory. We recruited 94 individuals from a local active living community (Rossmoor, CA; aged 63–94, mean age 79.9, mean 16.3 years of education) and, under the authority of an Institutional Review Board, enrolled them into a randomized three-arm study. Participants in the first study (intervention) arm used the program in a classroom setting at its recommended dosing (60 min/day, 5 days/ week, 8 weeks). Trainers supervised the classroom to provide technical assistance and general encouragement. Participants in the second study arm (active control) used the same computers and classrooms to watch and listen to educational material presented on DVD. This activity was comparable to program as an active control, in that it engaged participants in an engaging auditory and visual learning activity, was time- and intensity-matched to the intervention, and kept participants blind as to their active control status. The active control attempt explicitly to control neither for the adaptively and progressively challenging nature of program exercises nor for their intense reward and attentional engagement, as we consider those key ‘‘active ingredients’’ in the intervention. Participants in the third study arm (no-contact control) engaged in no-study activities during the training period. All participants completed identical neuropsychological assessment batteries before and after the training period. The primary
instrument in this battery was the RBANS (repeatable battery for the assessment of neuropsychological status), a standardized instrument composed of 12 individual subtests covering areas of immediate and delayed memory, attention, visuospatial function, and spoken language. The RBANS has alternate forms designed and tested to be equivalent; we used these alternate forms in the pretraining and posttraining visits to minimize test retest effects. The assessments in the neuropsychological battery were very different from the training exercises in the remedial program, ensuring that any changes seen in the assessments would represent true generalization of improvement rather than training to the assessment. Study participants were required to be 60 years of age or older, have a mini-mental state examination (MMSE) of 24 or higher, and have RBANS overall scores to be generally representative of normal aging (taken to be 2 standard deviations within the normative population range, 70 130), and not self-identify with dementia. Fifty-one participants meeting these criteria were randomized into the experimental group, and 41 completed all training and assessment activities. Four participants withdrew during training citing schedule conflicts; however, no withdrawals cited dissatisfaction with the training programs as a reason for withdrawal (six participants had protocol violations in the posttest condition and were excluded from analysis). All participants were able to learn the usage of the exercises with the built-in training and minor assistance from the classroom trainers; by two weeks into training all participants were using this experimental program without assistance. Twenty-five participants were randomized into the active control group, and 16 completed all training and assessment activities. There were nine withdrawals over the few days of training citing dissatisfaction with the active control; however, participants completing the active control program were generally satisfied with the material. Eighteen participants were randomized into the no-contact control, and 15 completed all assessment activities with withdrawals due to scheduling
100 conflicts or moves. The pretraining groups were equivalent in age (one-way ANOVA, p40.4) and educational level (p40.4). Given that the training program focused on renormalizing the auditory system, the primary outcome measure was a global auditory memory score based on the six auditory tests of the RBANS (list learning, story memory, digit span forwards, delayed list recall, delayed list recognition, and delayed story recall). The global auditory memory score was calculated by using the normative RBANS population data to construct age-normed (by decade) look-up tables allowing the conversion of raw score data on each test (which generally showed a strong skew) to scaled score data (optimally normally distributed with a population mean of 10 and a standard deviation of 3). Delayed list recall and delayed list recognition were summed before scaling to allow the inclusion of the significantly skewed and otherwise unscalable delayed list recognition data. The five scaled scores can then be summed to yield a global auditory memory score. These look-up tables were then used to calculate scaled scores and global auditory memory scores for each participant in this study for pre- and posttraining assessments.
Evaluation of the neuropsychological data showed a significant improvement in the global auditory memory score within the trained group (Fig. 2, po0.0005, two-tailed paired t-test) and nonsignificant trend toward improvement in the active control group (p40.1) and no significant effect in the no-contact control group (p40.4). The magnitude of the effect size in this assessment was 0.41, or slightly higher than 1/3 of a standard deviation of enhancement relative to the distribution in the normal population (the standard deviation of the global auditory memory score in the normative RBANS population is 9.0). The improvement in the global auditory memory score was driven by changes in each of the five scaled score assessments of auditory memory function (Fig. 3). This suggests that the effects of training are broadly distributed across cognitive systems that relay on speech input, as we would predict from the design of the training function. Another approach for quantifying the effect size is to examine the percentile change in the group relative to the normal population. Using the distribution of global auditory memory scores from the normative RBANS population, prior to training, the trained group scored at the 35th percentile. Following training, the group scored at the 59th percentile.
Fig. 2. Brain-plasticity-based training enhances global auditory memory scores in older adults. Pre- and post-training global auditory memory scores for the intervention group (3.7 point change, significant), the active control group (2.4 point change, not significant), and the no-contact control group (1.8 point change, not significant).
101
Fig. 3. Memory enhancement is distributed broadly across neuropsychological measures. Pre- and posttraining auditory memoryscaled scores for the intervention group.
Fig. 4. Age-associated course of decline of memory as assessed with the RBANS. Decline of memory over time was estimated from RBANS normative data by developing scaled score look-up tables for list learning, story memory, digit span, the sum of delayed list recall and delayed list recognition, and delayed story recall based on the entire normative data set including individuals from age 20 to 89. Global auditory memory score was calculated as the sum of the scaled scores. Global auditory memory scores were averaged across a moving window with a width of 10 years to plot the rate of decline of RBANS memory function with age.
To roughly translate this effect size into a measure more relevant for populations undergoing the natural course of normal aging, we might estimate the rate of cognitive decline per year in the normative population and compare the effect size in this study to this rate of decline. We estimated the rate of decline of memory over time from the RBANS normative data by developing scaled scores and a global auditory memory score as described above but based on the entire normative
data set including individuals from age 20 to 89 (i.e., not age-stratified by decade as described above). We then averaged this global auditory memory score data set with a moving window with a width of 10 years to plot the rate of decline of RBANS memory function with age (Fig. 4). This function shows an initial decline from 25 to 40 years of age, followed by a broad plateau, which is followed by a subsequent decline from the age of 62 onwards. The shape of this function suggests
102
that the RBANS, which was designed for mildly impaired populations, is not likely to be sensitive to the known cognitive changes that occur in middle age and early old age. The slope of the decline in the 62+ period is 0.35 points per year (through age 84, the last year for which the data for full 10 year window is available in the normative data set). The training-induced change in this global auditory memory score is 3.5 points, suggesting that average improvement in the trained group was roughly equivalent to 10 years of memory performance as assessed with the RBANS. We note that this rough approach is only appropriate at the group level and cannot be used to assess changes at the individual level due to the meaningful test retest variance for any given individual in the RBANS (and virtually all neuropsychological tests). In summary, these pilot data demonstrate the promise of this training-based intervention and provide a proof-of-principle to guide larger studies with a wider array of assessments that are fully statistically powered.
Conclusions The losses in sensory, cognitive, memory, and motor abilities during aging can profoundly affect everyday functioning and quality of life. Because the brain experiences physical deterioration coinciding with the onset of cognitive deficits, it has long been assumed that this atrophy is the sole cause of the loss of cognitive and memory abilities in the aged. The science of brain plasticity suggests a different model of origin of age-related cognitive decline in which the role of physical atrophy is complemented by the interactions of brain disuse, noisy processing, weakened neuromodulatory control, and negative learning. We have developed an initial version of a brainplasticity-based training program designed to address the four potential causes of age-related cognitive decline, and in doing so, to enhance auditory perception, memory, and cognition in normally aging individuals. An initial pilot randomized controlled trial demonstrated the feasibility of the approach, in that the intervention was usable, learnable, and well accepted by the target
population, and showed substantial promise in the effect size of the memory enhancement. Going forward, further studies with this training program are required to establish more completely the functional areas and magnitude of enhancement, and in particular to quantitatively assess what broader impact the training has on self-report of everyday functioning measures. Structural and functional brain imaging studies to document the types and magnitudes of brain plasticity underlying the behavioral changes following training will be important as well. Finally, we believe that this training program represents only the first step in the development of a complete suite of training programs that, in aggregate, should target the broad array of sensory, cognitive, memory, and motor problems that emerge with aging.
Acknowledgments We thank William Boschin, Anne Bruce, Bradley Brummett, Jill Damon, Danielle Doan, Lisa Faille, Amy Gentile, Jason Minow, and Amy Walthall, for their work in collecting the data in the study; Jed Appelman, Patrick Brannelly, Jane Chang, David Cheng, Laurel Cox, Miriam Hashimi, Tisha Hilario, Mark Johnson, Nicholas Joyce, Jaclyn Kohlriter, Kim Schilling, Cynthia Warren, Darrell Wayne, and Rick Wood for their contributions to the execution of the study; and Natasha Belfor, Bonnie Connor, Joseph Hardy and Kimberly Tanner for their helpful comments on the manuscript. We are grateful to Christopher Randolph for the normative RBANS data used to derive global auditory memory scores and for discussions regarding data analysis. We also thank Marghi Merzenich and Kathy Gowell for help preparing this manuscript.
References Abel, S.M., Krever, E.M. and Alberti, P.W. (1990) Auditory detection, discrimination and speech processing in ageing, noise-sensitive and hearing-impaired listeners. Scand. Audiol., 19(Suppl 1): 43–54. (ADEAR) AsDEaRC. (2002) 2001–2002 Alzheimer’s disease progress report. National Institute on Aging, Washington, DC.
103 Akutsu, H., Legge, G.E., Ross, J.A. and Schuebel, K.J. (1991) Psychophysics of reading–X. Effects of age-related changes in vision. J. Gerontol., 46(Suppl 6): P325–P331. Aldwin, C.M. and Gilmer, D.F. (2004) Health, Illness, and Optimal Aging. Sage Publications, Thousand Oaks, CA.. Allard, T., Clark, S.A., Jenkins, W.M. and Merzenich, M.M. (1991) Reorganization of somatosensory area 3b representations in adult owl monkeys after digital syndactyly. J. Neurophysiol., 66(Suppl 3): 1048–1058. Arendash, G.W., Garcia, M.F., Costa, D.A., Cracchiolo, J.R., Wefes, I.M. and Potter, H. (2004) Environmental enrichment improves cognition in aged Alzheimer’s transgenic mice despite stable beta-amyloid deposition. Neuroreport, 15(Suppl 11): 1751–1754. Backman, L. and Farde, L. (2005) The role of dopamine systems in cognitive aging. In: Cabeza, R., Nyberg, L. and Park, D. (Eds.), Cognitive Neuroscience of Aging. Oxford University Press, New York, pp. 58–84. Ball, K., Berch, D.B., Helmers, K.F., Jobe, J.B., Leveck, M.D., Marsiske, M., Morris, J.N., Rebok, 0G0.W., Smith, D.M., Tennstedt, S.L., Unverzagt, F.W. and Willis, S.L. (2002) Effects of cognitive training interventions with older adults: a randomized controlled trial. JAMA, 288(Suppl 18): 2271–2281. Bao, S., Chan, V.T. and Merzenich, M.M. (2001) Cortical remodelling induced by activity of ventral tegmental dopamine neurons. Nature, 412: 79–83. Bao, S., Chang, E.F., Davis, J.D., Gobeske, K.T. and Merzenich, M.M. (2003) Progressive degradation and subsequent refinement of acoustic representations in the adult auditory cortex. J. Neurosci., 23: 10765–10775. Baron, A. and Cerella, J. (1993) Laboratory tests of the disuse account of cognitive decline. In: Cerella, J.M., Rybash, J., Hoyer, W.J. and Commons, M.L. (Eds.), Adult Information Processing: Limits on Loss. Academic Press, San Diego, CA, pp. 175–203. Bartus, R.T., Dean 3rd, R.L., Beer, B. and Lippa, A.S. (1982) The cholinergic hypothesis of geriatric memory dysfunction. Science, 217(Suppl 4558): 408–414. Beaulieu, C. and Colonnier, M. (1989) Number and size of neurons and synapses in the motor cortex of cats raised in different environmental complexities. J. Comp. Neurol., 289(Suppl 1): 178–181. Berman, R.F., Goldman, H. and Altman, H.J. (1988) Age-related changes in regional cerebral blood flow and behavior in Sprague Dawley rats. Neurobiol. Aging,, 9(Suppl 5–6): 691–696. Bezard, E., Dovero, S., Belin, D., Duconger, S., Jackson-Lewis, V., Przedborski, S., Piazza, P.V., Gross, C.E. and Jaber, M. (2003) Enriched environment confers resistance to 1-methyl4-phenyl-1,2,3,6-tetrahydropyridine and cocaine: involvement of dopamine transporter and trophic factors. J. Neurosci., 23(Suppl 35): 10999–11007. Bischkopf, J., Busse, A. and Angermeyer, M.C. (2002) Mild cognitive impairment — a review of prevalence, incidence and outcome according to current approaches. Acta Psychiatr. Scand., 106(Suppl 6): 403–414.
Buckner, R.L. (2004) Memory and executive function in aging and AD: multiple factors that cause decline and reserve factors that compensate. Neuron, 44(Suppl 1): 195–208. Bussiere, T. and Hof, P.R. (2000) Morphological changes in human cerebral cortex during normal aging. In: Hof, P.R. and Mobbs, C.V. (Eds.), Functional Neurobiology of Aging. Academic Press, San Diego, CA, pp. 77–84. Byl, N.N. (2004) Focal hand dystonia may result from aberrant neuroplasticity. Adv. Neurol., 94: 19–28. Byl, N.N. and McKenzie, A. (2000) Treatment effectiveness for patients with a history of repetitive hand use and focal hand dystonia. J. Hand Ther., 14: 289–301. Byl, N., Merzenich, M., Cheung, S., Bedenbaugh, P., Nagarajan, S. and Jenkins, W. (1997) A primate model for studying focal dystonia and repetitive strain injury: effects on the primary somatosensory cortex. Phys. Ther. Case Rep., 77(Suppl 3): 269–283. Byl, N., Merzenich, M. and Jenkins, W. (1996) A primate genesis model of focal hand dystonia and repetitive strain injury. Neurology, 47: 508–520. Byl, N.N., Nagarajan, S. and McKenzie, A.M. (2000). Effectiveness of sensorimotor training on structure and function in musicians with focal hand dystonia: three case studies. In: Society for Neuroscience Annual Meeting, New Orleans, LA. Byl, N., Nagarajan, S. and McKenzie, A. (2003) Effect of sensory discrimination training on structure and function in patients with focal hand dystonia: three case studies. Arch. Phys. Med. Rehab., 84: 1505–1514. Cahill, L. and McGaugh, J.L. (1998) Mechanisms of emotional arousal and lasting declarative memory. Trends Neurosci., 21: 294–299. Caprio-Prevette, M.D. and Fry, P.S. (1996) Memory enhancement program for community-based older adults: development and evaluation. Exp. Aging Res., 22(Suppl 3): 281–303. Cheesman, M.F., Hepburn, D., Armitage, J.C. and Marshall, K. (1995) Comparison of growth of masking functions and speech discrimination abilities in younger and older adults. Audiology, 34(Suppl 6): 321–333. Churs, L., Spengler, F., Jurgens, M. and Dinse, H.R. (1996) Environmental enrichment counteracts decline of sensorimotor performance and deterioration of cortical organization in aged rats. Soc. Neurosci. Abstr., 22: 48. Clare, L., Wilson, B.A., Carter, G., Roth, I. and Hodges, J.R. (2002) Relearning face-name associations in early Alzheimer’s disease. Neuropsychology, 16(Suppl 4): 538–547. Cohen, A.D., Tillerson, J.L., Smith, A.D., Schallert, T. and Zigmond, M.J. (2003) Neuroprotective effects of prior limb use in 6-hydroxydopamine-treated rats: possible role of GDNF. J. Neurochem., 85: 299–305. Coq, J.O. and Xerri, C. (2000) Age-related alteration of the forepaw representation in the rat primary somatosensory cortex. Neuroscience, 99: 403–411. Coq, J.O. and Xerri, C. (2001) Sensorimotor experience modulates age-dependent alterations of the forepaw representation in the rat primary somatosensory cortex. Neuroscience, 104: 705–715.
104 Craft, S., Cholerton, B. and Reger, M. (2003). Aging and cognition: what is normal? In: W.R. Hazzard, J.P. Blass, J.B. Halter, J.G.Ouslander, M. Tinetti (Eds.), Principles of Geriatric Medicine and Gerontology (5th ed.), pp. 1355–1372. Cronin-Golomb, A., Cronin-Golomb, M., Dunne, T.E., Brown, A.C., Jain, K., Cipolloni, P.B. and Auerbach, S.H. (2000) Facial frequency manipulation normalizes face discrimination in AD. Neurology, 54(Suppl 12): 2316–2318. Cummins, R.A., Walsh, R.N., Budtz-Olsen, O.E., Konstantinos, T. and Horsfall, C.R. (1973) Environmentally-induced changes in the brains of elderly rats. Nature, 243(Suppl 5409): 516–518. Davis, R.N., Massman, P.J. and Doody, R.S. (2001) Cognitive intervention in Alzheimer disease: a randomized placebo-controlled study. Alzheimer Dis. Assoc. Disord., 15(Suppl 1): 1–9. Deutsch, G.K., Linn, N., Taylor, H.L., Miller, S.L., Temple, E., Gabrieli, J.D.E., Nagarajan, S.S. and Merzenich, M.M. (2000). Treatment for developmental dyslexia: improvements in language associated with improvements in reading decoding. In: Society for Neuroscience’s 30th Annual Meeting, Nov 4–9, New Orleans. Diamond, M.C. (2001) Response of the brain to enrichment. An. Acad. Bras. Cienc., 73(Suppl 2): 211–220. Diamond, M.C., Johnson, R.E., Protti, A.M., Ott, C. and Kajisa, L. (1985) Plasticity in the 904-day-old male rat cerebral cortex. Exp. Neurol., 87(Suppl 2): 309–317. Diamond, M.C., Lindner, B., Johnson, R., Bennett, E.L. and Rosenzweig, M.R. (1975) Differences in occipital cortical synapses from environmentally enriched, impoverished, and standard colony rats. J. Neurosci. Res., 1(Suppl 2): 109–119. Dick, M.B., Hsieh, S., Bricker, J. and Dick-Muehlke, C. (2003) Facilitating acquisition and transfer of a continuous motor task in health adults and patients with Alzheimer’s disease. Neuropsychology, 17: 202–212. Dollinger, S.M.C. (1995) Mental rotation performance: age, sex and visual field differences. Dev. Neuropsychol., 11: 215–222. Doty, B.A. (1972) The effects of cage environment upon avoidance responding of aged rats. J. Gerontol., 27(Suppl 3): 358–360. Doya, K. (2002) Metalearning and neuromodulation. Neural Netw., 15(Suppl 4–6): 495–506. Drullman, R. (1995a) Speech intelligibility in noise: relative contribution of speech elements above and below the noise level. J. Acoust. Soc. Am., 98(Suppl 3): 1796–1798. Drullman, R. (1995b) Temporal envelope and fine structure cues for speech intelligibility. J. Acoust. Soc. Am., 97(Suppl 1): 585–592. Dubno, J.R., Dirks, D.D. and Morgan, D.E. (1984) Effects of age and mild hearing loss on speech recognition in noise. J. Acoust. Soc. Am., 76(Suppl 1): 87–96. Echt, K.V. and Pollack, R.H. (1998). The effect of illumination, contrast, and age on text comprehension performance. In: Cognitive Aging, Atlanta, GA. Edwards, J.D., Wadley, V.G., Myers, R.S., Roenker, D.L., Cissell, G.M. and Ball, K.K. (2002) Transfer of a speed of processing intervention to near and far cognitive functions. Gerontology, 48(Suppl 5): 329–340.
Elbert, T., Candia, V., Altenmuller, E., Rau, H., Sterr, A., Rockstroh, B., Pantev, C. and Taub, E. (1998) Alteration of digital representations in somatosensory cortex in focal hand dystonia. Neuroreport, 9: 3571–3575. Elbert, T., Pantev, C., Wienbruch, C., Rockstroh, B. and Taub, E. (1995) Increased cortical representation of the fingers of the left hand in string players. Science, 270(Suppl 5234): 305–307. Elliot, D.B., Whitaker, D. and Thompson, P. (1989) Use of displacement threshold hyperacuity to isolate the neural component of senile vision loss. Appl. Opt., 28: 1914–1918. Eslinger, P.J. and Benton, A.L. (1983) Visuoperceptual performances in aging and dementia: clinical and theoretical implications. J. Clin. Neuropsychol., 5(Suppl 3): 213–220. Faherty, C.J., Raviie Shepherd, K., Herasimtschuk, A. and Smeyne, R.J. (2005) Environmental enrichment in adulthood eliminates neuronal death in experimental Parkinsonism. Brain Res. Mol. Brain Res., 134(Suppl 1): 170–179. Faubert, J. (2002) Visual perception and aging. Can. J. Exp. Psychol., 56(Suppl 3): 164–176. Fernandez, C.I., Collazo, J., Bauza, Y., Castellanos, M.R. and Lopez, O. (2004) Environmental enrichment-behavior-oxidative stress interactions in the aged rat: issues for a therapeutic approach in human aging. Ann. N.Y. Acad. Sci., 1019: 53–57. Fillit, H.M., Butler, R.N., O’Connell, A.W., Albert, M.S., Birren, J.E., Cotman, C.W., Greenough, W.T., Gold, P.E., Kramer, A.F., Kuller, L.H. and Perls, T.T. (2002) Achieving and maintaining cognitive vitality with aging. Mayo Clin. Proc., 77(Suppl 7): 681–696. Fitzgibbons, P.J. and Gordon-Salant, S. (1994) Age effects on measures of auditory duration discrimination. J. Speech Hear. Res., 37(Suppl 3): 662–670. Floeter, M.K. and Greenough, W.T. (1979) Cerebellar plasticity: modification of Purkinje cell structure by differential rearing in monkeys. Science, 206(Suppl 4415): 227–229. Foster, R.H. and Plosker, G.L. (1999) Donepezil: pharmacoeconomic implications of therapy. Pharmacoeconomics, 16(Suppl 1): 99–114. Frick, K.M. and Fernandez, S.M. (2003) Enrichment enhances spatial memory and increases synaptophysin levels in aged female mice. Neurobiol. Aging,, 24: 615–626. Frick, K.M., Stearns, N.A., Pan, J.Y. and Berger-Sweeney, J. (2003) Effects of environmental enrichment on spatial memory and neurochemistry in middle-aged mice. Learn. Mem., 10: 187–198. Gatz, M. (2005) Educating the brain to avoid dementia: can mental exercise prevent Alzheimer disease? PLoS Med., 2(Suppl 1): e7. Gibbs, M.E. and Summers, A. (2002) Role of adrenoceptor subtypes in memory consolidation. Prog. Neurobiol., 67: 345–391. Glisky, E.L. and Glisky, M.L. (1999) Memory rehabilitation in the elderly. In: Stuss, D.J., Winocur, G. and Robertson, I. (Eds.), Cognitive Rehabilitation. Cambridge University Press, Cambridge, pp. 347–361.
105 Godde, B., Berkefeld, T., David-Jurgens, M. and Dinse, H.R. (2002) Age-related changes in primary somatosensory cortex of rats: evidence for parallel degenerative and plastic-adaptive processes. Neurosci. Biobehav. Rev., 26: 743–745. Gordon-Salant, S. and Fitzgibbons, P.J. (1993) Temporal factors and speech recognition performance in young and elderly listeners. J. Speech Hear. Res., 36(Suppl 6): 1276–1285. Gordon-Salant, S. and Fitzgibbons, P.J. (2001) Sources of agerelated recognition difficulty for time-compressed speech. J. Speech Lang. Hear. Res., 44(Suppl 4): 709–719. Green, E.J., Greenough, W.T. and Schlumpf, B.E. (1983) Effects of complex or isolated environments on cortical dendrites of middle-aged rats. Brain Res., 264(Suppl 2): 233–240. Greenough, W.T., Hwang, H.M. and Gorman, C. (1985) Evidence for active synapse formation or altered postsynaptic metabolism in visual cortex of rats reared in complex environments. PNAS, 82(Suppl 13): 4549–4552. Greenough, W.T., West, R.W. and DeVoogd, T.J. (1978) Subsynaptic plate perforations: changes with age and experience in the rat. Science, 202(Suppl 4372): 1096–1098. Gu, Q. (2002) Neuromodulatory transmitter systems in the cortex and their role in cortical plasticity. Neuroscience, 111(Suppl 4): 815–835. Harvey, P.D. and Mohs, R.C. (2000) Memory changes with aging and dementia. In: Hof, P.R. and Mobbs, C.V. (Eds.), Functional Neurobiology of Aging. Academic Press, San Diego, CA., pp. 53–63. Hayes, E.A., Warrier, C.M., Nicol, T.G., Zecker, S.G. and Kraus, N. (2003) Neural plasticity following auditory training in children with learning problems. Clin. Neurophysiol., 114: 673–684. Hilbig, H., Holler, J., Dinse, H.R. and Bidmon, H.J. (2002) In contrast to neuronal NOS-I, the inducible NOS-II expression in aging brains is modified by enriched environmental conditions. Exp. Toxicol. Pathol., 53: 427–431. Hultsch, D.F., Hertzog, C., Small, B.J. and Dixon, R.A. (1999) Use it or lose it: engaged lifestyle as a buffer of cognitive decline in aging? Psychol. Aging,, 14(Suppl 2): 245–263. Humes, L.E. and Christopherson, L. (1991) Speech identification difficulties of hearing-impaired elderly persons: the contributions of auditory processing deficits. J. Speech Hear. Res., 34(Suppl 3): 686–693. Humes, L.E. and Roberts, L. (1990) Speech-recognition difficulties of the hearing-impaired elderly: the contributions of audibility. J. Speech Hear. Res., 33(Suppl 4): 726–735. Jenkins, W., Merzenich, M., Ochs, M., Allard, T. and GuicRobles, E. (1990) Functional reorganization of primary somatosensory cortex in monkeys after behaviorally controlled tactile stimulation. J. Neurophysiol., 63(Suppl 1): 82–104. Johannsen, P. (2004) Long-term cholinesterase inhibitor treatment of Alzheimer’s disease. CNS Drugs, 18(Suppl 12): 757–768. Kaduszkiewicz, H., Zimmermann, T., Beck-Bornholdt, H.P. and van den Bussche, H. (2005) Cholinesterase inhibitors for patients with Alzheimer’s disease: systematic review of randomised clinical trials. Br. Med. J., 331(Suppl 7512): 321–327.
Katz, H.B. and Davies, C.A. (1984) Effects of differential environments on the cerebral anatomy of rats as a function of previous and subsequent housing conditions. Exp. Neurol., 83(Suppl 2): 274–287. Kempermann, G., Gast, D. and Gage, F.H. (2002) Neuroplasticity in old age: sustained fivefold induction of hippocampal neurogenesis by long-term environmental enrichment. Ann. Neurol., 52: 135–143. Kempermann, G., Kuhn, H.G. and Gage, F.H. (1997) More hippocampal neurons in adult mice living in an enriched environment. Nature, 386(Suppl 6624): 493–495. Kilgard, M.P. and Merzenich, M.M. (1998a) Cortical map reorganization enabled by nucleus basalis activity. Science, 279: 1714–1718. Kilgard, M.P. and Merzenich, M.M. (1998b) Plasticity of temporal information processing in the primary auditory cortex. Nat. Neurosci., 1: 727–731. Kilgard, M.P. and Merzenich, M.M. (2002) Order-sensitive plasticity in adult primary auditory cortex. PNAS, 99: 3205–3209. Kim, C.B.Y. and Mayer, M.J. (1994) Foveal flicker sensitivity in healthy aging eyes: II. Cross-sectional aging trends from 18 through 77 years of age. J. Opt. Soc. Am., 11: 1958–1969. Kleim, J.A., Barbay, S., Cooper, N.R., Hogg, T.M., Reidel, C.N., Remple, M.S. and Nudo, R.J. (2002) Motor learningdependent synaptogenesis is localized to functionally reorganized motor cortex. Neurobiol. Learn. Mem., 77: 63–77. Kline, D., Culham, J., Bartel, P. and Lynk, L. (1994). Aging and hyperacuity thresholds as a function of contrast and oscillation rate. In: Candian Psychological Association. Penticton, British Columbia. Kline, D.W. (1994) Optimizing the visibility of displays for older observers. Exp. Aging Res., 20(Suppl 1): 11–23. Kline, D.W. and Birren, J.E. (1975) Age differences in backward dichoptic masking. Exp. Aging Res., 1(Suppl 1): 17–25. Kline, D.W. and Szafran, J. (1975) Age differences in backward monoptic visual noise masking. J. Gerontol., 30(Suppl 3): 307–311. Kobayashi, S., Ohashi, Y. and Ando, S. (2002) Effects of enriched environments with different durations and starting times on learning capacity during aging in rats assessed by a refined procedure of the Hebb-Williams maze task. J. Neurosci. Res., 70(Suppl 3): 340–346. Koss, E., Haxby, J.V., DeCarli, C., Schapiro, M.B., Friedland, R.P. and Rapoport, S.I. (1991) Patterns of performance preservation and loss in healthy aging. Dev. Neuropsychol., 7: 99–113. Lazarov, O., Robinson, J., Tang, Y.P., Hairston, I.S., KoradeMirnics, Z., Lee, V.M., Hersh, L.B., Sapolsky, R.M., Mirnics, K. and Sisodia, S.S. (2005) Environmental enrichment reduces Abeta levels and amyloid deposition in transgenic mice. Cell, 120(Suppl 5): 701–713. Lemaire, V., Aurousseau, C., Le Moal, M. and Abrous, D.N. (1999) Behavioural trait of reactivity to novelty is related to hippocampal neurogenesis. Eur. J. Neurosci., 11: 4006–4014. Lewis, M.H. (2004) Environmental complexity and central nervous system development and function. Ment. Retard. Dev. Disabil. Res. Rev., 10(Suppl 2): 91–95.
106 Li, L. and Tang, B.L. (2005) Environmental enrichment and neurodegenerative diseases. Biochem. Biophys. Res. Commun., 334(Suppl 2): 293–297. Loewenstein, D.A., Acevedo, A., Czaja, S.J. and Duara, R. (2004) Cognitive rehabilitation of mildly impaired Alzheimer disease patients on cholinesterase inhibitors. Am. J. Geriatr. Psychiatry,, 12(Suppl 4): 395–402. Madden, D.J., Whiting, W.L. and Huettel, S.A. (2003) Age-related changes in neural activity during visual perception and attention. In: Cabeza, R., Nyberg, L. and Park, D. (Eds.), Cognitive Neuroscience of Aging: Linking Cognitive and Cerebral Aging. Oxford University Press, New York, pp. 157–185. Magistretti, S., Joray, S. and Pellvin, L. (2000) Brain energy metabolism: cellular aspects and relevance to functional brain imaging. In: Hof, P.R. and Mobbs, C.V. (Eds.), Functional Neurobiology of Aging. Academic Press, San Diego, CA, pp. 203–209. Mattson, M.P. (2003) Cellular and neurochemical aspects of the aging brain. In: Hazzard, W.R., Blass, J.P., Halter, J.B., Ouslander, J.G. and Tinetti, M. (Eds.), Principles of Geriatric Medicine and Gerontology (5th Edn). McGraw-Hill, New York, pp. 1341–1354. Mattson, M.P., Duan, W., Lee, J. and Guo, Z. (2001) Suppression of brain aging and neurodegenerative disorders by dietary restriction and environmental enrichment: molecular mechanisms. Mech. Ageing Dev., 122: 757–778. McDougall Jr., G.J. (1999) Cognitive interventions among older adults. Annu Rev Nurs Res., 17: 219–240. Melendez, R.I., Gregory, M.L., Bardo, M.T. and Kalivas, P.W. (2004) Impoverished rearing environment alters metabotropic glutamate receptor expression and function in the prefrontal cortex. Neuropsychopharmacology, 29(Suppl 11): 1980–1987. Merzenich, M.M. (2001) Cortical plasticity contributing to child development. In: McClelland, J. and Siegler, R. (Eds.), Mechanisms in Cognitive Development. L. Ehrlbaum Associates, Mahwah, NJ, pp. 67–96. Merzenich, M.M. and Jenkins, W.M. (1993) Cortical representation of learned behaviors. In: Andersen, P. (Ed.), Memory Concepts. Elsevier, Amsterdam, pp. 437–453. Merzenich, M.M. and Jenkins, W.M. (1995) Cortical plasticity, learning and learning dysfunction. In: Julesz, B. and Kovacs, I. (Eds.), Maturational Windows and Adult Cortical Plasticity. Addison-Wesley, New York, pp. 247–272. Merzenich, M.M., Schreiner, C., Jenkins, W. and Wang, X. (1993) Neural mechanisms underlying temporal integration, segmentation, and input sequence representation: some implications for the origin of learning disabilities. Ann. N. Y. Acad. Sci., 682: 1–22. Merzenich, M.M., Jenkins, W.M., Johnston, P., Schreiner, C., Miller, S.L. and Tallal, P. (1996a) Temporal processing deficits of language-learning impaired children ameliorated by training. Science, 271: 77–81. Merzenich, M.M., Spengler, F., Byl, N., Wang, X. and Jenkins, W. (1996b) Representational plasticity underlying learning; contributions to the origins and expressions of neurobehavioral disabilities. In: Ono, T., McNaughton, B.L., Molotchnikoff, S., Rolls, E.T. and Nishijo, H. (Eds.), Perception, Memory and Emotion: Frontiers in Neuroscience. Pergamon, Cambridge.
Merzenich, M.M., Miller, S., Jenkins, W., Saunders, G., Protopapas, A., Peterson, B. and Tallal, P. (1998a) Amelioration of the acoustic reception and speech reception deficits underlying language-based learning impairments. In: von Euler, C. (Ed.), Basic Neural Mechanisms in Cognition and Language. Elsevier, Amsterdam, pp. 143–172. Merzenich, M.M., Tallal, P., Peterson, B., Miller, S.L. and Jenkins, W.M. (1998b) Some neurological principles relevant to the origins of — and the cortical plasticity based remediation of — language learning impairments. In: Grafman, J. and Cristen, Y. (Eds.), Neuroplasticity: Building a Bridge from the Laboratory to the Clinic. Springer-Verlag, New York, pp. 169–187. Mohammed, A.H., Henriksson, B.G., Soderstrom, S., Ebendal, T., Olsson, T. and Seckl, J.R. (1993) Environmental influences on the central nervous system and their implications for the aging rat. Behav. Brain Res., 57: 183–191. Mohammed, A.H., Zhu, S.W., Darmopil, S., Hjerling-Leffler, J., Ernfors, P., Winblad, B., Diamond, M.C., Eriksson, P.S. and Bogdanovic, N. (2002) Environmental enrichment and the brain. Prog. Brain Res., 138: 109–133. Mohs, R.C., Ashman, T.A., Jantzen, K., Albert, M., Brandt, J., Gordon, B., Rasmusson, X., Grossman, M., Jacobs, D. and Stern, Y. (1998) A study of the efficacy of a comprehensive memory enhancement program in healthy elderly persons. Psychiatry Res., 77: 183–195. Moore, B.C. and Peters, R.W. (1992) Pitch discrimination and phase sensitivity in young and elderly subjects and its relationship to frequency selectivity. J. Acoust. Soc. Am., 91(Suppl 5): 2881–2893. Morrison, J.D. and McGrath, C. (1985) Assessment of the optical contributions to the age-related deterioration in vision. Q. J. Exp. Physiol., 70(Suppl 2): 249–269. Murphy, D.R., Craik, F.I., Li, K.Z. and Schneider, B.A. (2000) Comparing the effects of aging and background noise on short-term memory performance. Psychol. Aging,, 15(Suppl 2): 323–334. Nagarajan, S.S., Deutsch, G., Houde, J.F., Szymanski, M. and Merzenich, M.M. (2000). Plasticity of auditory cortical magnetic field responses in reading-impaired children. In: Society for Neuroscience’s 30th Annual Meeting. Nov 4–9, New Orleans. Nagarajan, S.S. and Merzenich, M.M. Recovery of function in the auditory/aural speech cortex resulting from intensive listening training. Unpublished manuscript. Nakamura, S. (1991) Axonal sprouting of noradrenergic locus coeruleus neurons following repeated stress and antidepressant treatment. Prog. Brain Res., 88: 587–598. Nameda, N., Kawara, T. and Ohzu, H. (1989) Human visual spatio-temporal frequency performance as a function of age. Optom. Vis. Sci., 66(Suppl 11): 760–765. National Institute on Aging (1998). Progress Report on Alzheimer’s Disease. National Institute on Aging, Bethesda, MD. Report No.: NIH Publication No. 99–3616. Nudo, R.J., Larson, D., Plautz, E.J., Friel, K.M., Barbay, S. and Frost, S.B. (2003) A squirrel monkey model of poststroke motor recovery. ILAR J., 44: 161–174.
107 Nudo, R.J., Milliken, G.W., Jenkins, W.M. and Merzenich, M.M. (1996) Use-dependent alterations of movement representations in primary motor cortex of adult squirrel monkeys. J. Neurosci., 16(Suppl 2): 785–807. Olesen, P.J., Westerberg, H. and Klingberg, T. (2004) Increased prefrontal and parietal activity after training of working memory. Nat. Neurosci., 7: 75–79. Olincy, A., Ross, R.G., Young, D.A. and Freedman, R. (1997) Age diminishes performance on an antisaccade eye movement task. Neurobiol. Aging,, 18(Suppl 5): 483–489. Owsley, C., Gardner, T., Sekuler, R. and Lieberman, H. (1985) Role of the crystalline lens in the spatial vision loss of the elderly. Invest. Ophthalmol. Vis. Sci., 26(Suppl 8): 1165–1170. Owsley, C., Sekuler, R. and Boldt, C. (1981) Aging and lowcontrast vision: face perception. Invest. Ophthalmol. Vis. Sci., 21(Suppl 2): 362–365. Park, D.C. and Gutchess, A.H. (2003) Long-term memory and aging: a cognitive neuroscience perspective. In: Cabeza, R., Nyberg, L. and Park, D. (Eds.), Cognitive Neuroscience of Aging: Linking Cognitive and Cerebral Aging. Oxford University Press, New York, pp. 218–245. Park, D.C., Polk, T., Mikles, J.A., Taylor, S. and Marshetz, C. (2001) Cerebral aging: the integration of behavioral and neurobiological models of cognitive function. Dialogues Clin. Neurosci., 3: 151–165. Park, D.C., Smith, A.D., Lautenschlager, G., Earles, J.D., Frieske, D., Zwahr, M. and Gaines, C.L. (1996) Mediators of long-term memory performance across the life span. Psychol. Aging,, 11(Suppl 4): 621–637. Park, G.A., Pappas, B.A., Murtha, S.M. and Ally, A. (1992) Enriched environment primes forebrain choline acetyltransferase activity to respond to learning experience. Neurosci. Lett., 143(Suppl 1–2): 259–262. Park, H.L., O’Connell, J.E. and Thomson, R.G. (2003) A systematic review of cognitive decline in the general elderly population. Int. J. Geriatr. Psychiatry,, 18(Suppl 12): 1121–1134. Peterson, R.C., Thomas, R.G., Grundman, M., Bennett, D., Doody, R., Ferris, S., Galasko, D., Jin, S., Kaye, J., Levey, A., Pfeiffer, E., Sano, M., van Dyck, C.H. and Thal, L.J. (2005) Vitamin E and donepezil for the treatment of mild cognitive impairment. N. Engl. J. Med., 352(Suppl 23): 2379–2388. Pichora-Fuller, M.K., Schneider, B.A. and Daneman, M. (1995) How young and old adults listen to and remember speech in noise. J. Acoust. Soc. Am., 97(Suppl 1): 593–608. Piraino, P.S., Yednock, T.A., Messersmith, E.K., Pleiss, M.A., Freedman, S.B., Hammond, R.R. and Karlik, S.J. (2005) Spontaneous remyelination following prolonged inhibition of alpha(4) integrin in chronic EAE. J. Neuroimmunol., 167(Suppl 1–2): 53–63. Plude, D.J., Milberg, W.P. and Cerella, J. (1986) Age differences in depicting and perceiving tridimensionality in simple line drawings. Exp. Aging Res., 12(Suppl 4): 221–225. Rampon, C., Tang, Y.P., Goodhouse, J., Shimizu, E., Kyin, M. and Tsien, J.Z. (2000) Enrichment induces structural changes and recovery from nonspatial memory deficits in CA1 NMDAR1-knockout mice. Nat. Neurosci., 3(Suppl 3): 238–244.
Raz, N. (2000) Aging of the brain and its impact on cognitive performance: integration of structural and functional findings. In: Craik, F.I.M. and Salthouse, T.A. (Eds.), The Handbook of Aging and Cognition. Lawrence Erlbaum Associates, Mahwah, NJ, pp. 1–90. Recanzone, G.H., Merzenich, M.M., Jenkins, W.M., Grajski, K.A. and Dinse, H.R. (1992a) Topographic reorganization of the hand representation in cortical area 3b owl monkeys trained in a frequency-discrimination task. J. Neurophysiol., 67(Suppl 5): 1031–1056. Recanzone, G.H., Merzenich, M.M. and Schreiner, C.E. (1992b) Changes in the distributed temporal response properties of SI cortical neurons reflect improvements in performance on a temporally based tactile discrimination task. J. Neurophysiol., 67(Suppl 5): 1071–1091. Reinke, H. and Dinse, H.R. (1999) Plasticity in the somatosensory and motor cortex of rats: impact of age and housing conditions. In: Elsner, N. and Eysel, U. (Eds.), From Molecular Neurobiology to Clinical Neuroscience. Thieme, Stuttgart, p. 409. Reuter-Lorenz, P.A. and Sylvester, C.C. (2003) The cognitive neuroscience of working memory and aging. In: Cabeza, R., Nyberg, L. and Park, D. (Eds.), Cognitive Neuroscience of Aging: Linking Cognitive and Cerebral Aging. Oxford University Press, New York, pp. 186–217. Ritchie, K., Artero, S. and Touchon, J. (2001) Classification criteria for mild cognitive impairment: a population-based validation study. Neurology, 56(Suppl 1): 37–42. Robins-Wahlin, T., Ba¨ckman, L., Wahlin, A. and Winblad, B. (1993) Visuospatial functioning and spatial orientation in a community-based sample of healthy very old persons. Arch. Gerontol. Geriatr., 17: 165–177. Rosenzweig, M.R. and Bennett, E.L. (1996) Psychobiology of plasticity: effects of training and experience on brain and behavior. Behav. Brain Res., 78: 57–65. Saleh, M.C., Espinosa de los Monteros, A., de Arriba Zerpa, G.A., Fontaine, I., Piaud, O., Djordjijevic, D., Baroukh, N., Garcia Otin, A.L., Ortiz, E., Lewis, S., Fiette, L., Santambrogio, P., Belzung, C., Connor, J.R., de Vellis, J., Pasquini, J.M., Zakin, M.M., Baron, B. and Guillou, F. (2003) Myelination and motor coordination are increased in transferrin transgenic mice. J. Neurosci. Res., 72(Suppl 5): 587–594. Salthouse, T.A. (1996) The processing speed theory of adult age differences in cognition. Psychol. Rev., 103: 403–428. Scarmeas, N., Levy, G., Tang, M.X., Manly, J. and Stern, Y. (2001) Influence of leisure activity on the incidence of Alzheimer’s disease. Neurology, 57: 2236–2242. Schneider, B.A., Daneman, M. and Pichora-Fuller, M.K. (2002) Listening in aging adults: from discourse comprehension to psychoacoustics. Can. J. Exp. Psychol., 56(Suppl 3): 139–152. Schneider, B.A. and Pichora-Fuller, M.K. (2000) Implications of perceptual deterioration for cognitive aging research. In: Craik, F.I.M. and Salthouse, T.A. (Eds.), The Handbook of Aging and Cognition. Lawrence Erlbaum Associates, Mahwah, NJ, pp. 155–219.
108 Schneider, B.A., Pichora-Fuller, M.K., Kowalchuk, D. and Lamb, M. (1994) Gap detection and the precedence effect in young and old adults. J. Acoust. Soc. Am., 95(Suppl 2): 980–991. Schooler, C., Mulatu, M.S. and Oates, G. (1999) The continuing effects of substantively complex work on the intellectual functioning of older workers. Psychol. Aging,, 14(Suppl 3): 483–506. Schweighofer, N., Doya, K. and Kuroda, S. (2004) Cerebellar aminergic neuromodulation: towards a functional understanding. Brain Res. Rev., 44: 103–116. Scialfa, C.T. (2002) The role of sensory factors in cognitive aging research. Can. J. Exp. Psychol., 56(Suppl 3): 153–163. Scialfa, C.T. and Kline, D.W. (1988) Effects of noise type and retinal eccentricity on age differences in identification and localization. J. Gerontol., 43(Suppl 4): P91–P99. Sharpe, J.A. and Sylvester, T.O. (1978) Effect of aging on horizontal smooth pursuit. Invest Ophthalmol. Vis. Sci., 17(Suppl 5): 465–468. Sirevaag, A.M. and Greenough, W.T. (1985) Differential rearing effects on rat visual cortex synapses. II. Synaptic morphometry. Brain Res., 351(Suppl 2): 215–226. Spengler, F., Godde, B. and Dinse, H.R. (1995) Effects of ageing on topographic organization of somatosensory cortex. Neuroreport, 6: 469–473. Speranza, F., Daneman, M. and Schneider, B.A. (2000) How aging affects the reading of words in noisy backgrounds. Psychol. Aging,, 15(Suppl 2): 253–258. Spinks, R., Gilmore, G.C. and Thomas, C. (1996). Age simulation of a sensory deficit does impair cognitive test performance. In: Cognitive Aging. Atlanta, GA. Stevens, B., Porta, S., Haak, L.L., Gallo, V. and Fields, R.D. (2002) Adenosine: a neuron-glial transmitter promoting myelination in the CNS in response to action potentials. Neuron, 36: 855–868. Tallal, P., Miller, S.L., Bedi, G., Byma, G., Wang, X., Nagarajan, S.S., Schreiner, C., Jenkins, W.M. and Merzenich, M.M. (1996a) Language comprehension in language-learning impaired children improved with acoustically modified speech. Science, 271(Suppl 5245): 81–84. Tallal, P., Miller, S., Jenkins, W. and Merzenich, M. (1996b) Role of temporal processing in developmental languagebased learning disorders: research and clinical implications. In: Blachman, B. (Ed.), Foundations of Reading. MIT Press, Boston, pp. 71–88. Temple, E., Deutsch, G.K., Poldrack, R.A., Miller, S.L., Tallal, P., Merzenich, M.M. and Gabrieli, J.D. (2003) Neural deficits in children with dyslexia ameliorated by behavioral remediation: evidence from functional MRI. PNAS, 100: 2860–2865. Temple, E., Poldrack, R.A., Deutsch, G.K., Salidis, J., Tallal, P., Merzenich, M.M. and Gabrieli, J.D.E. (2001). Dyslexic children show neural and behavioral effects of remediation: evidence from fMRI. In: Society for Neuroscience. Temple, E., Poldrack, R.A., Protopapas, A., Nagarajan, S., Saltz, T., Tallal, P., Merzenich, M.M. and Gabrieli, J.D. (2000) Disruption of the neural response to rapid acoustic stimuli in dyslexia: evidence from functional MRI. PNAS, 97: 13907–13912.
Terry, R.D. and Katzman, R. (2001) Life span and synapses: will there be a primary senile dementia? Neurobiol. Aging, 22(Suppl 3): 347–354. Tillerson, J.L., Caudle, W.M., Reveron, M.E. and Miller, G.W. (2003) Exercise induces behavioral recovery and attenuates neurochemical deficits in rodent models of Parkinson’s disease. Neuroscience, 119: 899–911. Tillerson, J.L., Cohen, A.D., Caudle, W.M., Zigmond, M.J., Schallert, T. and Miller, G.W. (2002) Forced nonuse in unilateral parkinsonian rats exacerbates injury. J. Neurosci., 22(Suppl 15): 6790–6799. Trick, G.L. and Silverman, S.E. (1991) Visual sensitivity to motion: age-related changes and deficits in senile dementia of the Alzheimer type. Neurology, 41(Suppl 9): 1437–1440. Unverzagt, F.W., Gao, S., Baiyewu, O., Ogunniyi, A.O., Gureje, O., Perkins, A., Emsley, C.L., Dickens, J., Evans, R., Musick, B., Hall, K.S., Hui, S.L. and Hendrie, H.C. (2001) Prevalence of cognitive impairment: data from the Indianapolis Study of Health and Aging. Neurology, 57(Suppl 9): 1655–1662. Usher, M., Cohen, J.D., Servan-Schreiber, D., Rajkowski, J. and Aston-Jones, G. (1999) The role of locus coeruleus in the regulation of cognitive performance. Science, 283(Suppl 5401): 549–554. Verghese, J., Lipton, R.B., Katz, M.J., Hall, C.B., Derby, C.A., Kuslansky, G., Ambrose, A.F., Sliwinski, M. and Buschke, H. (2003) Leisure activities and the risk of dementia in the elderly. N. Engl. J. Med., 348(Suppl 25): 2508–2516. Verhaeghen, P. and Marcoen, A. (1996) On the mechanisms of plasticity in young and older adults after instruction in the method of loci: evidence for an amplification model. Psychol. Aging,, 11(Suppl 1): 164–178. Verhaeghen, P., Marcoen, A. and Goossens, L. (1992) Improving memory performance in the aged through mnemonic training: a meta-analytic study. Psychol. Aging,, 7(Suppl 2): 242–251. Walsh, D.A. (1976) Age differences in central perceptual processing: a dichoptic backward masking investigation. J. Gerontol., 31(Suppl 2): 178–185. Wang, X., Merzenich, M., Sameshima, K. and Jenkins, W. (1995) Remodelling of hand representation in adult cortex determined by timing of tactile stimulation. Nature, 378(Suppl 2): 71–75. Weinberger, N.M. (2003) The nucleus basalis and memory codes: auditory cortical plasticity and the induction of specific, associative behavioral memory. Neurobiol. Learn. Mem., 80: 268–284. West, R. (2004) The neural basis of age-related declines in prospective memory. In: Cabeza, R., Nyberg, L. and Park, D. (Eds.), Cognitive Neuroscience of Aging: Linking Cognitive and Cerebral Aging. Oxford University Press, New York, pp. 246–264. Whalley, L.J., Deary, I.J., Appleton, C.L. and Starr, J.M. (2004) Cognitive reserve and the neurobiology of cognitive aging. Ageing Res. Rev., 3(Suppl 4): 369–382. Wilson, R.S., Bennett, D.A., Bienias, J.L., Aggarwal, N.T., Mendes de Leon, C.F., Morris, M.C., Schneider, J.A. and Evans, D.A. (2002) Cognitive activity and incident AD in a population-based sample of older persons. Neurology, 59(Suppl 12): 1910–1914.
109 Wilson, R.S., Bennett, D.A., Bienias, J.L., Mendes de Leon, C.F., Morris, M.C. and Evans, D.A. (2003) Cognitive activity and cognitive decline in a biracial community population. Neurology, 61(Suppl 6): 812–816. Wingfield, A. (1996) Cognitive factors in auditory performance: context, speed of processing, and constraints of memory. J. Am. Acad. Audiol., 7(Suppl 3): 175–182. Wingfield, A. and Lindfield, K.C. (1995) Multiple memory systems in the processing of speech: evidence from aging. Exp. Aging Res., 21(Suppl 2): 101–121. Wingfield, A., Tun, P.A., Koh, C.K. and Rosen, M.J. (1999) Regaining lost time: adult aging and the effect of time restoration on recall of time-compressed speech. Psychol. Aging,, 14(Suppl 3): 380–389. Winocur, G. (1998) Environmental influences on cognitive decline in aged rats. Neurobiol. Aging,, 19(Suppl 6): 589–597. Wojciechowski, R., Trick, G.L. and Steinman, S.B. (1995) Topography of the age-related decline in motion sensitivity. Optom. Vis. Sci., 72(Suppl 2): 67–74. Wolf, B., Fehs, H., De Weerdt, J., van der Meer, J., Noom, M. and Aufdemkampe, G. (2001) Effect of a physical therapeutic intervention for balance problems in the elderly: a singleblind, randomized, controlled multicentre trial. Clin. Rehabil., 15: 626–636. Wolfman, C., Abo, V., Calco, D., Medina, J., Dajas, F. and Silveira, R. (1994) Recovery of central noradrenergic neurons one year after the administration of the neurotoxin DSP4. Neurochem. Int., 25: 395–400. Woodlee, M.T. and Schallert, T. (2004) The interplay between behavior and neurodegeneration in rat models of Parkinson’s disease and stroke. Restor. Neurol. Neurosci., 22(Suppl 3–5): 153–161. Woodruff-Pak, D.S. (1993) Neural plasticity as a substrate for cognitive adaptation in adulthood and aging. In: Cerella, J.,
Rybash, J., Hover, W. and Commons, M.L. (Eds.), Adult Information Processing: Limits on Loss. Academic Press, San Diego, CA, pp. 13–35. Xerri, C., Coq, J., Merzenich, M. and Jenkins, W. (1996) Experience-induced plasticity of cutaneous maps in the primary somatosensory cortex of adult monkeys and rats. J. Physiol. Paris,, 90(Suppl 3–4): 277–287. Xerri, C., Merzenich, M.M., Peterson, B.E. and Jenkins, W. (1998) Plasticity of primary somatosensory cortex paralleling sensorimotor skill recovery from stroke in adult monkeys. J. Neurophysiol., 79: 2119–2148. Yesavage, J.A. (1983) Imagery pretraining and memory training in the elderly. Gerontology, 29(Suppl 4): 271–275. Yesavage, JA. (1989) Techniques for cognitive training of memory in age-associated memory impairment. Arch. Gerontol. Geriatr. Suppl., 1: 185–190. Ylikoski, R., Ylikoski, A., Keskivaara, P., Tilvis, R., Sulkava, R. and Erkinjuntti, T. (1999) Heterogeneity of cognitive profiles in aging: successful aging, normal aging, and individuals at risk for cognitive decline. Eur. J. Neurol., 6(Suppl 6): 645–652. Young, D., Lawlor, P.A., Leone, P., Dragunow, M. and During, M.J. (1999) Environmental enrichment inhibits spontaneous apoptosis, prevents seizures and is neuroprotective. Nat. Med., 5(Suppl 4): 448–453. Zacks, R.T. and Hasher, L. (2000) Human memory. In: Craik, F.I.M. and Salthouse, T.A. (Eds.), The Handbook of Aging and Cognition. Lawrence Erlbaum Associates, Mahwah, NJ, pp. 293–357. Zigmond, M.J. and Burke, R.E. (2002) Pathophysiology of Parkinson’s disease. In: Davis, K.L., Coyle, J., Charney, D. and Nemeroff, C. (Eds.), Fifth Generation of Progress. Lippincott, Williams and Wilkins, Philadelphia, PA pp. 1781–1194.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 7
Cortical plasticity and rehabilitation Raluca Moucha and Michael P. Kilgard Neuroscience Program, School of Brain and Behavioral Sciences, University of Texas at Dallas, Dallas, TX, USA
Abstract: The brain is constantly adapting to environmental and endogenous changes (including injury) that occur at every stage of life. The mechanisms that regulate neural plasticity have been refined over millions of years. Motivation and sensory experience directly shape the rewiring that makes learning and neurological recovery possible. Guiding neural reorganization in a manner that facilitates recovery of function is a primary goal of neurological rehabilitation. As the rules that govern neural plasticity become better understood, it will be possible to manipulate the sensory and motor experience of patients to induce specific forms of plasticity. This review summarizes our current knowledge regarding factors that regulate cortical plasticity, illustrates specific forms of reorganization induced by control of each factor, and suggests how to exploit these factors for clinical benefit. Keywords: Cortical plasticity; Experience-dependent plasticity; Cortical reorganization; Neuromodulators; Cholinergic; Rehabilitation significance, conveyed by release of modulatory neurotransmitters (Fig. 1).
Factors that regulate plasticity Plasticity is the remarkable ability of developing, adult, and aging brains to adapt to a changing world. This potential is revealed whenever an organism must meet a new environmental demand or recover from nervous system damage. Plasticity occurs in sensory and motor systems following deprivation of input or overstimulation, increased or decreased usage, learning of new skills, and injury. These experience-dependent changes can be as subtle as a change in neuronal excitability (Engineer et al., 2004) or as dramatic as the rewiring of auditory cortex to process visual information (Sur et al., 1988). Topographic maps, receptive field (RF) size, neuronal firing rate, temporal precision, and combination sensitivity can all be modified by our experiences. The types of plasticity activated by specific situations depend on the nature of the experiences and their behavioral
Attentional modulation Neural plasticity is essential for adapting to changes in the environment but plasticity can be destabilizing if not well regulated. Limiting plasticity prevents meaningless events from driving changes that could degrade previously acquired memories and skills. Attention plays a key role in the regulation of plasticity associated with sensory experience. Repeated sensory stimulation alters topography in primary sensory cortex only when monkeys use the stimuli to make behavioral judgments (Recanzone et al., 1992, 1993). Many studies have shown that cortical neurons respond differently to attended versus unattended stimuli. Neurons in secondary somatosensory cortex, for example, exhibit greater response synchronization when monkeys are engaged in a tactile task (Steinmetz et al., 2000). Attention can also directly
Corresponding author. Tel.: +1 (972) 883-2345; Fax: +1 (972) 883-2491; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57007-4
111
112
Fig. 1. Several neurotransmitter systems which project widely into the cortex are implicated in learning and experience-dependent plasticity: Acetylcholine from the cholinergic nucleus basalis (NB), dopamine from the ventral tegmentum (VTA), noradrenaline from the locus coeruleus (LC), and serotonin from the raphe nuclei (RN). In addition to these major neurotransmitters, GABA-ergic projections, histamine, and neuro-hormones also play a role in modulating plasticity. Release of these transmitters is normally regulated by behavioral state but can also be triggered by drugs or direct electrical stimulation. Cortical plasticity results when release of these transmitters is repeatedly associated with the occurrence of a sensory stimulus. (Source: McEwen, 2003). See Plate 7.1 in Colour Plate Section.
affect firing rates of cortical neurons (Treue and Maunsell, 1999; Recanzone and Wurtz, 2000). Results from several psychophysical studies support the hypothesis that attention regulates cortical plasticity and learning. Distinct forms of perceptual learning result when subjects attend to different features of an otherwise identical sensory input (Ahissar and Hochstein, 1993). Exposure to moving dot patterns can improve motion direction discrimination ability even if the motion is undetectable (due to low coherence), as long as the subjects are actively engaged in a visual task (Watanabe et al., 2001; Seitz and Watanabe, 2003). These results suggest that directed attention facilitates the learning of associated sensory features. Neuromodulatory influences Several neuromodulators, such as dopamine, norepinephrine, and acetylcholine, are known to
regulate learning and memory in humans (Hasselmo, 1995). The observation that synaptic plasticity is also enhanced by the presence of these neurotransmitters supports the relationship between learning and plasticity (Singer, 1986; Brocher et al., 1992). Injection of acetylcholine or norepinephrine directly into visual, somatosensory, or auditory cortex during sensory stimulation can promote expression of neural plasticity in the intact brain (Greuel et al., 1988; McKenna et al., 1989; Delacour et al., 1990). Pairing sensory inputs with electrical activation of the nucleus basalis (NB), locus coeruleus (LC), or ventral tegmental area (VTA) also results in plasticity that is specific to features of the associated input (Kilgard and Merzenich, 1998; Bouret and Sara, 2002; Bao et al., 2003). Stimulation of neuromodulatory neurotransmitter release by amphetamine enhances cortical plasticity in human subjects (Dinse et al., 2003; Tegenthoff et al., 2004). Since release
113
of these neurotransmitters is normally triggered by behaviorally arousing events, it is likely they contribute to the regulation of cortical plasticity. Patterns of sensory activation Many studies have shown that sensory input determines the form of cortical reorganization. When animals or humans repeatedly practice a skill that engages a limited region of the sensory epithelium, the regions of the cortical map that respond to task-specific inputs are enlarged (Jenkins et al., 1990; Recanzone et al., 1992, 1993; Elbert et al., 1995; Sterr et al., 1998). Cortical RFs can narrow or broaden and response latency can increase or decrease depending on the spatial and temporal pattern of sensory activation encountered during training. Owl monkeys trained on a tone frequency discrimination task have A1 neurons with smaller RFs and longer response latencies than untrained controls (Recanzone et al., 1993). Monkeys trained to detect changes in the rate of a tactile vibration exhibit larger RFs and faster response latencies (Recanzone et al., 1992). In contrast, training on a task with stimuli that move across the skin cause RFs to shrink (Jenkins et al., 1990). Training on a visual orientation task increased the steepness of orientation tuning in the trained region of the visual field (Schoups et al., 2001). These studies support the hypothesis that perceptual learning and cortical plasticity are specific to attended sensory features. The rodent whisker system has proven particularly useful for directly comparing how cortical plasticity is shaped by different spatial patterns of activity. If all but one whisker is cut, for example, the responsiveness of the spared whisker is increased (Glazewski et al., 1998). Cutting a single whisker reduces input to the corresponding region of barrel cortex, decreases the responsiveness of the deprived neurons, and increases the responsiveness to neighboring whiskers. If all the whiskers are cut, the reduction in the response to the principle whisker is more modest. A checkerboard deprivation pattern causes responses to the deprived whiskers to decrease, but does not increase the response to the spared whiskers (Wallace and Fox, 1999). Finally, cutting all but two neighboring
whiskers causes the RF of neurons in each region to shift toward the other spared whisker (Diamond et al., 1993). These results suggest that competition between sensory inputs induces the different forms of changes in responsiveness. Timing of sensory inputs The temporal coincidence of sensory stimulation can be just as important as its spatial pattern in determining the direction and magnitude of cortical plasticity. Inputs that are correlated in time are more likely to cause a change in neural responses than uncorrelated inputs. Simultaneous activation of an area of skin with a vibrating disc increased RF size in primary somatosensory cortex, while stimulation of a single point on the skin does not cause any change (Godde et al., 1996). Simultaneous activation of the developing auditory system by repeated exposure to broadband noise causes increased cortical RFs and degraded tonotopic maps (Chang and Merzenich, 2003). Such changes are not seen after equivalent exposure to tones. Increased simultaneous activation of the fingers due to surgical fusion or operant training leads to large, multidigit RFs in somatosensory cortex (Allard et al., 1991; Wang et al., 1995). This finding suggests that the usual segregation of each digit’s cortical representation reflects the normally asynchronous activation of each digit. In vitro and more recently in vivo studies have further demonstrated that the time window for correlated inputs to induce plasticity is on the order of tens of milliseconds (Tsodyks, 2002; Dan and Poo, 2004). These results indicate that the precise spatial and temporal pattern of inputs shape cortical networks due to operation of Hebbian synaptic plasticity. Duration of experience Many factors regulate the time course of learning and plasticity (Ebbinghaus, 1885; Dubnau et al., 2003). Fear conditioning can induce rapid and long lasting shifts of neuronal tuning toward the frequency of the conditioned tone (Bjordahl et al., 1998; Weinberger, 2003). In contrast, plasticity following skill learning or use-dependent plasticity develops gradually over time. The magnitude
114
of effects often depends on duration of training and correlates with performance accuracy (Pleger et al., 2003). Motor map reorganization, which is accompanied by synaptogenesis and believed to underlie consolidation of motor skills, occurs during the late phase (after 10 days) of motor skill learning (Kleim et al., 2004). The schedule of inputs can also determine the induction of stable versus reversible synaptic modifications (Mauelshagen et al., 1998). Spaced repetition of LTP inducing stimuli prevents the reversal of LTP due to subsequent spontaneous activity that occurs after massed repetition (Zhou et al., 2003). Stable synaptic modifications are also induced by visual experience when the exposure to unidirectional moving bars occurs in a spaced pattern (three sets of 60 flashes separated by 5 min) versus massed pattern (180 flashes continuously). If persistent synaptic changes are important for learning and memory, the effective use of training strategies that prevent their reversal is important. In behaviorally trained mice temporally spaced training more effectively recruits protein synthesis and enhanced long-term memory of contextual conditioning, while massed training triggers greater protein phosphatase 1 activity which suppresses memory formation (Genoux et al., 2002; Scharf et al., 2002). These results suggest that the schedule of training determines the duration of neural plasticity and learning.
(Kapadia et al., 1995). This improvement in human performance is paralleled an enhancement of neuronal responses in monkey V1 when equivalent visual stimuli are presented (Kapadia et al., 1995). When foot shock is paired with a tone, the presence of unpaired background tones determines whether auditory cortex neurons shift their frequency tuning toward or away from the paired tone (Bakin and Weinberger, 1990; Ohl and Scheich, 1996; Dimyan and Weinberger, 1999). While all of these studies suggest that many factors regulate plasticity and learning, direct comparison of the interactions between these factors has proven difficult. Differences in the behavioral response, task difficulty, task goal, motivation, modality and species often confound the influence of the discussed factors on plasticity. Because these factors are so tightly interdependent it has been difficult to tease apart their relative importance in directing different forms of plasticity. Varying sensory patterns or adding a complex background, for example, would also affect task difficulty in most cases. Currently, reduced preparations provide the best opportunities to study the interactions between each of the factors that regulate neural plasticity. Experimental paradigms that directly stimulate modulatory systems have proven particularly valuable in documenting the influence of stimulus pattern, timing, and background conditions on cortical plasticity.
Influence of background stimuli on plasticity
Sensory input paired with controlled release of neuromodulators
Psychologists and psychophysicists have known for decades that unattended background stimuli (context) influence perceptual learning. Studies of sensory plasticity have typically been conducted in environments stripped of context, by using soundproof booths or gray backgrounds. Recent experiments in more naturalistic and complex settings have shown that context also influences plasticity. In many cases, adding complex backgrounds actually improves learning. Contrast discrimination learning, for example, can be facilitated by fixed contrast stimuli flanking the target stimulus (Adini et al., 2002). Dim line objects are easier to detect when flanked with a second collinear bar
Pairing electrical activation of the cholinergic NB with different sounds generates changes in cortical map and RF properties in rats that closely parallel the different forms of plasticity resulting from operant training in monkeys. For example, temporally modulated stimuli tend to increase RF size, while stimuli that activate different regions of the receptor surface tend to decrease RF size (Kilgard et al., 2002). While the differential plasticity observed in operant studies could be attributed to any number of technical differences, in the NB stimulation experiments the only explanation for the differential plasticity was the temporal and
115
spectral properties of the sounds associated with NB stimulation. The observation that similar sensory inputs lead to comparable plasticity even in the absence of operant training supports the conclusion that sensory features determine the form of cortical plasticity. During natural learning, changing task contingencies are known to alter the type, amount, and timing of neuromodulator release. For example, novel sounds activate cholinergic NB neurons for a few trials, but habituate rapidly (Richardson and DeLong, 1990, 1991). The response can later be reinstated if the sound is associated with a reward or punishment. NB releases acetylcholine onto the cortex only during the learning phase of a lever press task, but not after the task is well learned (Orsetti et al., 1996). Electrical stimulation bypasses the natural triggers of NB activity and eliminates the natural brake on cortical plasticity. The consistency of electrical activation makes it possible to systematically compare how the type, amount, and timing of neuromodulator release influence cortical plasticity when associated with sensory stimuli of differing spatial and temporal properties. Patterns of activation determine type of reorganization Distinct types of cortical reorganization are generated when NB stimulation is associated with different sensory inputs. Cortical topography, RF size, and response timing are altered as a function of the temporal modulation and spatial distribution of inputs associated with NB stimulation. The focal activation caused by presentations of a single
tone frequency results in expansion of the area responsive to the tone, and modest RFs broadening. Distributing the activation over more frequency sectors (i.e., seven tone frequencies) prevents the map reorganization but results in a narrowing of RFs (Kilgard et al., 2001). Rapidly modulated tone trains cause map expansion and dramatic RF broadening when activation is focal (i.e., one carrier frequency) and less extreme RF broadening and no map plasticity when the tone trains activate several regions (i.e., seven different carrier frequencies). These results document how different activation patterns direct cortical plasticity: (1) sensory map expansion only results when sensory activation is focal. (2) Distributing inputs across the cochlea tends to reduce RF size. (3) Modulated stimuli tend to increase RF size compared to unmodulated stimuli (Table 1, Kilgard et al., 2002). The observation that RF size is increased by stimuli with high degree of temporal modulation and little spatial variability (tone trains) and decreased by stimuli with high spatial variability and no temporal modulation (unmodulated tones of varying frequency) is consistent with earlier observations of plasticity in operant trained monkeys (Recanzone et al., 1992, 1993). These results indicate that NB stimulation directs changes that are similar to operant induced plasticity even though the rats did not use the stimuli in any way. Natural sounds usually vary both in spatial and temporal structure and create more complex activity patterns than tones. Pairing frequency modulated sweeps and complex acoustic sequences leads to forms of plasticity that are unpredictable from earlier studies with simple tones (Kilgard and Merzenich, 2002; Moucha et al., 2005). FM sweeps
Table 1. Plasticity induced by one month of nucleus basalis stimulation paired with different sounds NB stimulation paired with
Plasticity observed
References
Single tone Tone train Distributed tones Distributed tone trains Noise burst trains Frequency modulated tones Complex acoustic sequence Background sounds
Map expansion+decrease latency Map expansion+decrease latency+RF broadening RF narrowing+increase latency RF broadening+temporal plasticity RF broadening+map degradation RF broadening+decrease latency+decreased thresholds Combination sensitivity+decrease latency+decreased thresholds Alters plasticity generated in silence
Kilgard and Merzenich (1998) Kilgard et al. (2001) Kilgard and Merzenich (1998) Kilgard et al. (2001) Bao et al. (2003) Moucha et al. (2005) Kilgard and Merzenich (2002) Moucha et al. (2005)
116
result in a moving pattern of activation across the cochlea. Although tones and FM sweeps may share many features in common (including starting frequency, bandwidth, intensity, duration), pairing FM sweeps with NB stimulation causes different plasticity compared with unmodulated tones. Pairing FM sweeps with NB decreases response latency, broadens RFs, and increased sensitivity to quiet tones. These changes are restricted to the region of A1 activated by the sweep, but no map expansion results. When the starting frequency of the FM sweeps is varied no plasticity is observed in any region of A1 (Moucha et al., 2005). While repeated exposure to FM’s does not cause any preference for FM direction (increasing or decreasing), pairing a sequence of sounds with identical NB stimulation can result in the development of responses sensitive to tone order. Although plasticity mechanisms have presumably evolved to increase cortical processing capacity for behaviorally relevant inputs, it is not immediately obvious why the plasticity associated with each spatial and temporal input pattern is beneficial. Correlation of sensory inputs Studies in auditory, visual, and somatosensory cortex have suggested that input correlations strongly influence neural plasticity (Buonomano and Merzenich, 1998). In the developing visual system, for example, alternating asynchronous electrical stimulation of the optic nerve prevents normal development of binocular visual responses (Stryker and Strickland, 1984). In auditory cortex, sounds designed to decrease or increase correlation across the frequency map lead to very different forms of plasticity (Pandya et al., 2005). Alternating activation of two nonoverlapping auditory neuron populations by two tones of distant frequencies (2 and 14 kHz) results in map segregation, decreased excitability, and longer response latencies of the activated neurons. These changes do not occur when NB-stimulation is paired with a modulated noise burst that synchronously activate large populations of A1 neurons. Pairing pulsed noises with NB stimulation disrupts tonotopic maps and reduces spontaneous discharge correlation in the primary auditory cortex (Bao et al.,
2003). These finding are in agreement with the Hebbian postulate that inputs with decreased correlation weaken cortical responses and supports other observations that primary sensory cortices segregate inputs that are asynchronous and integrate correlated inputs (Allard et al., 1991; Wang et al., 1995). Duration of associative sensory pairing The duration of NB-induced plasticity depends on the schedule of the pairing protocol. Repetitively pairing NB stimulation with a tone for several minutes causes a shift in frequency tuning that reverses within 5 h (Zhang et al., 2005). Cortical map expansion builds with repeated pairings. One month of 300 NB-tone pairings per day increases the A1 representation of the paired frequency by twice as much as a week of pairing (Kilgard and Merzenich, 1998). After a month of pairing, NBinduced map plasticity endures for at least 20 days (Carrasco et al., 2004). NB stimulation also increases the duration of cortical and subcortical plasticity induced by cortical microstimulation (Ma and Suga, 2003). These results support earlier observations that cholinergic modulation contributes to both short-term and long-term plasticity. Background stimuli influence plasticity outcomes Although background stimuli are known to influence task performance and plasticity (Kapadia et al., 1995; Adini et al., 2002), it has not been clear whether the differences are due to altered task difficulty or to some specific influence of the distracters. By directly pairing sensory stimuli with NB stimulation in different contexts, we have shown that background stimuli can influence plasticity independent of any influence on task performance. Background sounds can alter bandwidth, threshold, and map plasticity. The 20% increase in RF size that occurs after pairing a single tone with NB stimulation does not occur if the same tone-NB pairing is interleaved with flanking tones that are not associated with NB stimulation (Kilgard et al., 2001). Repeated presentation of
117
the word /SASH/ paired with NB stimulation causes expansion of the high frequency region of A1 (Pandya et al., 2003), presumably because the first element of the word is the high frequency /S/ sound. This map plasticity is eliminated when each phoneme of the word /S/, /A/, and /SH/ are also presented, but not paired with NB stimulation. Finally, the addition of unpaired FM sweeps that contrast the duration and direction of the paired FM sweeps results in threshold and latency plasticity not observed if the identical FM’s sweeps are paired with NB stimulation in a silent background (Moucha et al., 2005). These results indicate that background conditions, previously thought to be irrelevant, are likely to shape many forms of cortical plasticity.
Clinical conclusions It was proposed two decades ago that cortical reorganization after injury may be the neural substrate for recovery of function after brain damage (Jenkins and Merzenich, 1987). More recent studies in primates have shown that rehabilitative training can direct reorganization to benefit recovery (Nudo et al., 1996). There is no longer a doubt that reorganization after brain lesions is shaped by the sensorimotor experiences in the weeks to months following injury. Hence it is important to effectively manage plasticity after brain damage. Many of the factors that influence plasticity can be manipulated in clinical settings to enhance therapeutic outcomes. Attention is often impaired after brain injury and likely plays a critical role in directing traininginduced plasticity. Patients with the highest vigilance scores typically receive greatest benefit from the rehabilitation therapy (Sohlberg et al., 2000). Some strategies, such as constraint therapy, that increase arousal (and even frustration) can be more effective than traditional occupational therapies (Taub and Uswatte, 2003). The diffuse modulatory systems including the cholinergic NB are particularly vulnerable to dysregulation after brain damage. Experimental damage to the NB prevents map reorganization and retards skill learning in rats (Fig. 2). The recent
observation that NB damage also prevents recovery from brain damage suggests many of the same mechanisms that regulate normal learning also regulate recovery from injury (Conner et al., 2005). In some patients, medication may be beneficial for normalizing attentional and neuromodulatory mechanisms. Agents that stimulate neuromodulators known to place the brain in a permissive state for experience-dependent changes are most likely to be effective. Drugs that act on noradrenergic, dopaminergic, serotonergic, and cholinergic systems have been shown in laboratory and clinical research to be pharmacological adjuvants in neurorehabilitation (Phillips et al., 2003). Amphetamines lead to a diffuse increase of several modulators and can have a positive influence even when administered only as a single dose at the beginning of therapy (Feeney et al., 1982). It is important to note that drug administration only aids recovery when paired with practice. Amphetamine administration during speech language therapy increases the rate of improvement of aphasic patients during the early recovery period after stroke (Walker-Batson et al., 2004). Amphetamine also facilitates speech training in adult cochlear implant users (Tobey et al., 2005) and second language acquisition in normal subjects (Breitenstein et al., 2004). More research is needed to evaluate how best to facilitate neurological recovery using nervous system stimulants and other psychoactive compounds. Since sensory and motor experiences (associated with release of modulatory neurotransmitters) determine the form of plasticity generated, it is critical to develop targeted rehabilitation techniques designed to stimulate adaptive plasticity following brain damage. Motor maps are altered by skill acquisition not by repetitive use alone (Nudo, 1997). In somatosensory cortex postlesion changes are related to individual strategies and sensorimotor experience resulting from idiosyncratic behavior. The type of reorganization often depends on the strategy used by individual monkeys to reacquire an object retrieval skill after an experimentally-induced stroke (Xerri et al., 1998). These findings imply that cortical map plasticity can be influenced by the pattern of sensorimotor stimulation during behavioral treatment. In dysphagic
118
Fig. 2. Rats can improve their grasping skill with repeated practice. After practice more neurons in caudal motor cortex control the trained forepaw (a, b). Without the normal input from the cholinergic nucleus basalis (i.e., after lesion of NB cholinergic neurons) rats cannot improve their accuracy with training and the motor map of the trained forepaw remains unchanged (c). This result indicates that practice alone without appropriate levels of neuromodulators does not result in learning or map plasticity. Cortical lesions of the caudal forepaw representation after training (d) results in loss of accuracy that can be recovered after retraining and expansion of the rostral motor map of the trained forepaw (e). If nucleus basalis is lesioned during retraining, recovery of reaching accuracy is impaired and the rostral forepaw representation does not change (f). This result indicates that appropriate levels of neuromodulators are also required to promote compensatory plasticity and recovery of function after brain damage. (Source: Results illustrated are from experiments by Conner et al., 2005). See Plate 7.2 in Colour Plate Section.
stroke patients electrical stimulation of the pharynx results in motor cortex plasticity that is dependent on the pattern of stimulation (frequency, intensity, and duration of stimulation) and correlates with improvement in swallowing function (Fraser et al., 2002). Several treatment strategies now effectively combine modulation of somatosensory input, administration of pharmacological adjuvants, and cortical stimulation to improve outcomes of rehabilitation (Hummel and Cohen, 2005). The influence of background has not been wellstudied in the context of neurorehabilitation.
However, studies have documented beneficial effects of general environmental enrichment in recovery after experimental brain infarcts (review Johansson, 2004). Enriched environments further enhance recovery when combined with training or drug therapy (Biernaskie and Corbett, 2001; Puurunen et al., 2001). Our results from plasticity experiments indicate that adding complex backgrounds during rehabilitative training may aid in emphasizing and facilitating performance on specific tasks. In conclusion, therapies that optimize neural plasticity by integrating all the concepts described
119 Table 2. Factors regulating plasticity Factors regulating plasticity
Effect
References
Attention
Enhances stimulus driven plasticity via internal trigger of neuromodulator release Achieve optimal levels of neuromodulators required for plasticity Determines form of plasticity (reorganization of sensory representations, temporal precision, spatial selectivity, etc.) Potentiation and stabilization of changes by stimulating protein synthesis mechanisms, and reducing phosphatases that prevent longterm changes Consolidation of changes via synaptogenesis
Hasselmo (1995)
Drugs Pattern of stimuli Temporal delivery (spaced vs. massed training) Duration of training
above are likely to improve patient outcomes (Table 2). Optimal modulator release can be accomplished by modulating attention and arousal either through task requirements or stimulating drugs. Stimuli used in training can be selected to address specific changes (rewiring) needed to direct recovery of function in individual patients. The proper timing of training sessions (i.e., spaced rather than massed training) and duration should also be optimized for training to be effective and long lasting. The addition of background stimuli may prove beneficial in many situations. This context can be used to emphasize aspects of a task or to incrementally increase task difficulty to maintain the patient’s motivation and arousal. Ideally, the progress and efficacy of therapy should be monitored (and adjusted) in each patient using brain imaging or evoked potentials. We are now beginning to understand how many factors interplay in directing different forms of plasticity. Manipulation of the many parameters known to shape brain plasticity, including the pattern, timing, and duration of events associated with attention and release of modulatory neurotransmitters, is essential to improving neurorehabilitation.
Acknowledgments We thank Cherie Percaccio, Amanda Puckett, Vikram Jakkamsetti, Dr. Owen Floddy and Dr. Pritesh Pandya for insightful comments and review of the manuscript. This work was supported by grants from the National Institute for Deafness
Phillips et al. (2003) Buonomano and Merzenich (1998) Genoux et al. (2002); Scharf et al. (2002); Zhou et al. (2003) Kleim et al. (2004)
and Other Communication Disorders, the James S. McDonnell Foundation, and Cure Autism Now.
References Adini, Y., Sagi, D. and Tsodyks, M. (2002) Context-enabled learning in the human visual system. Nature, 415: 790–793. Ahissar, M. and Hochstein, S. (1993) Attentional control of early perceptual learning. Proc. Natl. Acad. Sci. USA, 90: 5718–5722. Allard, T., Clark, S.A., Jenkins, W.M. and Merzenich, M.M. (1991) Reorganization of somatosensory area 3b representations in adult owl monkeys after digital syndactyly. J. Neurophysiol., 66: 1048–1058. Bakin, J.S. and Weinberger, N.M. (1990) Classical conditioning induces CS-specific receptive field plasticity in the auditory cortex of the guinea pig. Brain Res., 536: 271–286. Bao, S., Chang, E.F., Davis, J.D., Gobeske, K.T. and Merzenich, M.M. (2003) Progressive degradation and subsequent refinement of acoustic representations in the adult auditory cortex. J. Neurosci., 23: 10765–10775. Biernaskie, J. and Corbett, D. (2001) Enriched rehabilitative training promotes improved forelimb motor function and enhanced dendritic growth after focal ischemic injury. J. Neurosci., 21: 5272–5280. Bjordahl, T.S., Dimyan, M.A. and Weinberger, N.M. (1998) Induction of long-term receptive field plasticity in the auditory cortex of the waking guinea pig by stimulation of the nucleus basalis. Behav. Neurosci., 112: 467–479. Bouret, S. and Sara, S.J. (2002) Locus coeruleus activation modulates firing rate and temporal organization of odourinduced single-cell responses in rat piriform cortex. Eur. J. Neurosci., 16: 2371–2382. Breitenstein, C., Wailke, S., Bushuven, S., Kamping, S., Zwitserlood, P., Ringelstein, E.B. and Knecht, S. (2004) Damphetamine boosts language learning independent of its cardiovascular and motor arousing effects. Neuropsychopharmacology, 29: 1704–1714.
120 Brocher, S., Artola, A. and Singer, W. (1992) Agonists of cholinergic and noradrenergic receptors facilitate synergistically the induction of long-term potentiation in slices of rat visual cortex. Brain Res., 573: 27–36. Buonomano, D.V. and Merzenich, M.M. (1998) Cortical plasticity: from synapses to maps. Annu. Rev. Neurosci., 21: 149–186. Carrasco, R.A., Puckett, A., Pandya, P.K., McMenamy, A., Gibbons, J., Heydrick, C., Moucha, R., Jain, R. and Kilgard, M.P. (2004) Frequency map reorganization in rat primary auditory cortex enabled by basal forebrain activation endures for at least 20 days. Program No. 987.14. Abstract Viewer and Itinerary Planner. San Diego: Soc. Neurosci., Abst. 2004. Online. Chang, E.F. and Merzenich, M.M. (2003) Environmental noise retards auditory cortical development. Science, 300: 498–502. Conner, J.M., Chiba, A.A. and Tuszynski, M.H. (2005) The basal forebrain cholinergic system is essential for cortical plasticity and functional recovery following brain injury. Neuron, 46: 173–179. Dan, Y. and Poo, M.M. (2004) Spike timing-dependent plasticity of neural circuits. Neuron, 44: 23–30. Delacour, J., Houcine, O. and Costa, J.C. (1990) Evidence for a cholinergic mechanism of ‘‘learned’’ changes in the responses of barrel field neurons of the awake and undrugged rat. Neuroscience, 34: 1–8. Diamond, M.E., Armstrong-James, M. and Ebner, F.F. (1993) Experience-dependent plasticity in adult rat barrel cortex. Proc. Natl. Acad. Sci. USA, 90: 2082–2086. Dimyan, M.A. and Weinberger, N.M. (1999) Basal forebrain stimulation induces discriminative receptive field plasticity in the auditory cortex. Behav. Neurosci., 113: 691–702. Dinse, H.R., Ragert, P., Pleger, B., Schwenkreis, P. and Tegenthoff, M. (2003) Pharmacological modulation of perceptual learning and associated cortical reorganization. Science, 301: 91–94. Dubnau, J., Chiang, A.S. and Tully, T. (2003) Neural substrates of memory: from synapse to system. J. Neurobiol., 54: 238–253. Ebbinghaus, H. (1885) Uber das Gedachtnis. Untersuchungen zur Experimentellen Psychologie. Duckner & Humbolt, Leipzig. Elbert, T., Pantev, C., Wienbruch, C., Rockstroh, B. and Taub, E. (1995) Increased cortical representation of the fingers of the left hand in string players. Science, 270: 305–307. Engineer, N.D., Percaccio, C.R., Pandya, P.K., Moucha, R., Rathbun, D.L. and Kilgard, M.P. (2004) Environmental enrichment improves response strength, threshold, selectivity, and latency of auditory cortex neurons. J. Neurophysiol., 92: 73–82. Feeney, D.M., Gonzalez, A. and Law, W.A. (1982) Amphetamine, haloperidol, and experience interact to affect rate of recovery after motor cortex injury. Science, 217: 855–857. Fraser, C., Power, M., Hamdy, S., Rothwell, J., Hobday, D., Hollander, I., Tyrell, P., Hobson, A., Williams, S. and Thompson, D. (2002) Driving plasticity in human adult motor cortex is associated with improved motor function after brain injury. Neuron, 34: 831–840.
Genoux, D., Haditsch, U., Knobloch, M., Michalon, A., Storm, D. and Mansuy, I.M. (2002) Protein phosphatase 1 is a molecular constraint on learning and memory. Nature, 418: 970–975. Glazewski, S., McKenna, M., Jacquin, M. and Fox, K. (1998) Experience-dependent depression of vibrissae responses in adolescent rat barrel cortex. Eur. J. Neurosci., 10: 2107–2116. Godde, B., Spengler, F. and Dinse, H.R. (1996) Associative pairing of tactile stimulation induces somatosensory cortical reorganization in rats and humans. Neuroreport., 8: 281–285. Greuel, J.M., Luhmann, H.J. and Singer, W. (1988) Pharmacological induction of use-dependent receptive field modifications in the visual cortex. Science, 242: 74–77. Hasselmo, M.E. (1995) Neuromodulation and cortical function: modeling the physiological basis of behavior. Behav. Brain Res., 67: 1–27. Hummel, F.C. and Cohen, L.G. (2005) Drivers of brain plasticity. Curr. Opin. Neurol., 18: 667–674. Jenkins, W.M. and Merzenich, M.M. (1987) Reorganization of neocortical representations after brain injury: a neurophysiological model of the bases of recovery from stroke. Prog. Brain Res., 71: 249–266. Jenkins, W.M., Merzenich, M.M., Ochs, M.T., Allard, T. and Guic-Robles, E. (1990) Functional reorganization of primary somatosensory cortex in adult owl monkeys after behaviorally controlled tactile stimulation. J. Neurophysiol., 63: 82–104. Johansson, B.B. (2004) Brain plasticity in health and disease. Keio J. Med., 53: 231–246. Kapadia, M.K., Ito, M., Gilbert, C.D. and Westheimer, G. (1995) Improvement in visual sensitivity by changes in local context: parallel studies in human observers and in V1 of alert monkeys. Neuron, 15: 843–856. Kilgard, M.P. and Merzenich, M.M. (1998) Cortical map reorganization enabled by nucleus basalis activity. Science, 279: 1714–1718. Kilgard, M.P. and Merzenich, M.M. (2002) Order sensitive plasticity in adult primary auditory cortex. Proc. Natl. Acad. Sci. USA, 99: 3205–3209. Kilgard, M.P., Pandya, P.K., Engineer, N.D. and Moucha, R. (2002) Cortical network reorganization guided by sensory input features. Biol. Cybern., 87: 333–343. Kilgard, M.P., Pandya, P.K., Vazquez, J., Gehi, A., Schreiner, C.E. and Merzenich, M.M. (2001) Sensory input directs spatial and temporal plasticity in primary auditory cortex. J. Neurophysiol., 86: 326–338. Kleim, J.A., Hogg, T.M., VandenBerg, P.M., Cooper, N.R., Bruneau, R. and Remple, M. (2004) Cortical synaptogenesis and motor map reorganization occur during late, but not early, phase of motor skill learning. J. Neurosci., 24: 628–633. Ma, X. and Suga, N. (2003) Augmentation of plasticity of the central auditory system by the basal forebrain and/or somatosensory cortex. J. Neurophysiol., 89: 90–103. Mauelshagen, J., Sherff, C.M. and Carew, T.J. (1998) Differential induction of long-term synaptic facilitation by spaced and massed applications of serotonin at sensory neuron synapses of Aplysia californica. Learn. Mem., 5: 246–256.
121 McEwen, B.S. (2003) Estrogen effects on the brain: Much more than sex. Karger Gazette, 66: 10–12. McKenna, T.M., Ashe, J.H. and Weinberger, N.M. (1989) Cholinergic modulation of frequency receptive fields in auditory cortex: I. Frequency-specific effects of muscarinic agonists. Synapse, 4: 30–43. Moucha, R., Pandya, P.K., Engineer, N.D., Rathbun, D.L. and Kilgard, M.P. (2005) Background sounds contribute to spectrotemporal plasticity in primary auditory cortex. Exp. Brain Res., 162: 417–427. Nudo, R.J. (1997) Remodeling of cortical motor representations after stroke: implications for recovery from brain damage. Mol. Psychiatry, 2: 188–191. Nudo, R.J., Wise, B.M., SiFuentes, F. and Milliken, G.W. (1996) Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science, 272: 1791–1794. Ohl, F.W. and Scheich, H. (1996) Differential frequency conditioning enhances spectral contrast sensitivity of units in auditory cortex (field Al) of the alert Mongolian gerbil. Eur. J. Neurosci., 8: 1001–1017. Orsetti, M., Casamenti, F. and Pepeu, G. (1996) Enhanced acetylcholine release in the hippocampus and cortex during acquisition of an operant behavior. Brain Res., 724: 89–96. Pandya, P.K., Engineer, N., Moucha, R., Rathbun, D.L., Vazquez, J. and Kilgard, M.P. (2003) Background stimuli control cortical map expansion in primary auditory cortex. In: International Conference on Auditory Cortex, Magdeburg, Germany. Pandya, P.K., Moucha, R., Engineer, N.D., Rathbun, D.L., Vazquez, J. and Kilgard, M.P. (2005) Asynchronous inputs alter excitability, spike timing, and topography in primary auditory cortex. Hear. Res., 203: 10–20. Phillips, J.P., Devier, D.J. and Feeney, D.M. (2003) Rehabilitation pharmacology: bridging laboratory work to clinical application. J. Head Trauma Rehabil., 18: 342–356. Pleger, B., Foerster, A.F., Ragert, P., Dinse, H.R., Schwenkreis, P., Malin, J.P., Nicolas, V. and Tegenthoff, M. (2003) Functional imaging of perceptual learning in human primary and secondary somatosensory cortex. Neuron, 40: 643–653. Puurunen, K., Jolkkonen, J., Sirvio, J., Haapalinna, A. and Sivenius, J. (2001) Selegiline combined with enriched-environment housing attenuates spatial learning deficits following focal cerebral ischemia in rats. Exp. Neurol., 167: 348–355. Recanzone, G.H., Merzenich, M.M., Jenkins, W.M., Grajski, K.A. and Dinse, H.R. (1992) Topographic reorganization of the hand representation in cortical area 3b owl monkeys trained in a frequency-discrimination task. J. Neurophysiol., 67: 1031–1056. Recanzone, G.H., Schreiner, C.E. and Merzenich, M.M. (1993) Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkeys. J. Neurosci., 13: 87–103. Recanzone, G.H. and Wurtz, R.H. (2000) Effects of attention on MT and MST neuronal activity during pursuit initiation. J. Neurophysiol., 83: 777–790.
Richardson, R.T. and DeLong, M.R. (1990) Context-dependent responses of primate nucleus basalis neurons in a go/nogo task. J. Neurosci., 10: 2528–2540. Richardson, R.T. and DeLong, M.R. (1991) Electrophysiological studies of the functions of the nucleus basalis in primates. Adv. Exp. Med. Biol., 295: 233–252. Scharf, M.T., Woo, N.H., Lattal, K.M., Young, J.Z., Nguyen, P.V. and Abel, T. (2002) Protein synthesis is required for the enhancement of long-term potentiation and long-term memory by spaced training. J. Neurophysiol., 87: 2770–2777. Schoups, A., Vogels, R., Qian, N. and Orban, G. (2001) Practising orientation identification improves orientation coding in V1 neurons. Nature, 412: 549–553. Seitz, A.R. and Watanabe, T. (2003) Psychophysics: is subliminal learning really passive? Nature, 422: 36. Singer, W. (1986) The brain as a self-organizing system. Eur. Arch. Psychiatry Neurol. Sci., 236: 4–9. Sohlberg, M.M., McLaughlin, K.A., Pavese, A., Heidrich, A. and Posner, M.I. (2000) Evaluation of attention process training and brain injury education in persons with acquired brain injury. J. Clin. Exp. Neuropsychol., 22: 656–676. Steinmetz, P.N., Roy, A., Fitzgerald, P.J., Hsiao, S.S., Johnson, K.O. and Niebur, E. (2000) Attention modulates synchronized neuronal firing in primate somatosensory cortex. Nature, 404: 187–190. Sterr, A., Muller, M.M., Elbert, T., Rockstroh, B., Pantev, C. and Taub, E. (1998) Changed perceptions in Braille readers. Nature, 391: 134–135. Stryker, M. and Strickland, S. (1984) Physiological segregation of ocular dominance columns depends on the pattern of afferent electrical activity. Invest. Ophtamol. Vis. Sci. Suppl., 25: 278. Sur, M., Garraghty, P.E. and Roe, A.W. (1988) Experimentally induced visual projections into auditory thalamus and cortex. Science, 242: 1437–1441. Taub, E. and Uswatte, G. (2003) Constraint-induced movement therapy: bridging from the primate laboratory to the stroke rehabilitation laboratory. J. Rehabil. Med., 41: 34–40. Tegenthoff, M., Cornelius, B., Pleger, B., Malin, J.P. and Schwenkreis, P. (2004) Amphetamine enhances training-induced motor cortex plasticity. Acta. Neurol. Scand., 109: 330–336. Tobey, E.A., Devous Sr, M.D., Buckley, K., Overson, G., Harris, T., Ringe, W. and Martinez-Verhoff, J. (2005) Pharmacological enhancement of aural habilitation in adult cochlear implant users. Ear Hear., 26: 45S–56S. Treue, S. and Maunsell, J.H. (1999) Effects of attention on the processing of motion in macaque middle temporal and medial superior temporal visual cortical areas. J. Neurosci., 19: 7591–7602. Tsodyks, M. (2002) Spike-timing-dependent synaptic plasticity — the long road towards understanding neuronal mechanisms of learning and memory. Trends Neurosci., 25: 599–600. Walker-Batson, D., Smith, P., Curtis, S. and Unwin, D.H. (2004) Neuromodulation paired with learning dependent practice to enhance post stroke recovery? Restor. Neurol. Neurosci., 22: 387–392.
122 Wallace, H. and Fox, K. (1999) The effect of vibrissa deprivation pattern on the form of plasticity induced in rat barrel cortex. Somatosens. Mot. Res., 16: 122–138. Wang, X., Merzenich, M.M., Sameshima, K. and Jenkins, W.M. (1995) Remodelling of hand representation in adult cortex determined by timing of tactile stimulation. Nature, 378: 71–75. Watanabe, T., Nanez, J.E. and Sasaki, Y. (2001) Perceptual learning without perception. Nature, 413: 844–848. Weinberger, N.M. (2003) The nucleus basalis and memory codes: auditory cortical plasticity and the induction of specific, associative behavioral memory. Neurobiol. Learn Mem., 80: 268–284.
Xerri, C., Merzenich, M.M., Peterson, B.E. and Jenkins, W. (1998) Plasticity of primary somatosensory cortex paralleling sensorimotor skill recovery from stroke in adult monkeys. J. Neurophysiol., 79: 2119–2148. Zhang, Y., Hamilton, S.E., Nathanson, N.M. and Yan, J. (2005) Decreased input-specific plasticity of the auditory cortex in mice lacking M1 muscarinic acetylcholine receptors. Cereb. Cortex, In press. Zhou, Q., Tao, H.W. and Poo, M.M. (2003) Reversal and stabilization of synaptic modifications in a developing visual system. Science, 300: 1953–1957.
Plate 7.1. Several neurotransmitter systems which project widely into the cortex are implicated in learning and experience-dependent plasticity: Acetylcholine from the cholinergic nucleus basalis (NB), dopamine from the ventral tegmentum (VTA), noradrenaline from the locus coeruleus (LC), and serotonin from the raphe nuclei (RN). In addition to these major neurotransmitters, GABA-ergic projections, histamine, and neuro-hormones also play a role in modulating plasticity. Release of these transmitters is normally regulated by behavioral state but can also be triggered by drugs or direct electrical stimulation. Cortical plasticity results when release of these transmitters is repeatedly associated with the occurrence of a sensory stimulus. (Source: McEwen, 2003.)
Plate 7.2. Rats can improve their grasping skill with repeated practice. After practice more neurons in caudal motor cortex control the trained forepaw (a, b). Without the normal input from the cholinergic nucleus basalis (i.e., after lesion of NB cholinergic neurons) rats cannot improve their accuracy with training and the motor map of the trained forepaw remains unchanged (c). This result indicates that practice alone without appropriate levels of neuromodulators does not result in learning or map plasticity. Cortical lesions of the caudal forepaw representation after training (d) results in loss of accuracy that can be recovered after retraining and expansion of the rostral motor map of the trained forepaw (e). If nucleus basalis is lesioned during retraining, recovery of reaching accuracy is impaired and the rostral forepaw representation does not change (f). This result indicates that appropriate levels of neuromodulators are also required to promote compensatory plasticity and recovery of function after brain damage. (Source: Results illustrated are from experiments by Conner et al., 2005).
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 8
Neural mechanisms of prefrontal cortical function: implications for cognitive rehabilitation Mark D’Esposito1, and Anthony J.-W. Chen1,2 1
Helen Wills Neuroscience Institute and Department of Psychology, University of California, Berkeley, CA, USA 2 Neurology and Geriatrics, University of California, San Francisco, CA, USA
Abstract: Understanding the role of the frontal lobes in cognition remains a challenge for neurologists and neuroscientists. It is proposed that goal-directed behavior, at the core of what we consider human, depends critically on the function of the frontal lobes, and, specifically, the prefrontal cortex (PFC). In this chapter, we put forth the hypothesis that further insight into the neural mechanisms underlying normal PFC function may ultimately help us understand the frontal-lobe syndrome, and importantly, potentially lead to effective therapeutic interventions for frontal-lobe dysfunction. Thus, the aim of this chapter is to review current hypotheses and knowledge about the neural mechanisms underlying the normal function of the PFC in cognition that could guide the development of therapeutic interventions. Keywords: prefrontal cortex; working memory; executive function; fMRI physicians since they observed that his basic language, motor, and sensory functions seemed relatively intact. Today, any clinician who has had the experience of performing a mental status examination on a patient with a frontal lobe injury similar to the one Gage suffered is also likely to struggle with describing the essence of his or her neurologic dysfunction. Frontal lobe dysfunction may not lead to failures in basic abilities, such as speech or ambulation. Patients with frontal-lobe injuries, such as Phineas Gage, are therefore often able to walk out of the hospital but their functional deficits emerge when they are challenged by the complexities of real life. Altered interactions at home and failures at work occur for reasons that most patients, as well as clinicians, have difficulty defining. In recent years, significant advances have been made in understanding the function of the frontal lobes with a virtual explosion in the field of cognitive neuroscience and the development of new neuroscientific methods to study brain function of
Introduction The frontal lobes occupy nearly one-third of the human brain, and yet understanding the role of the frontal lobes in cognition remains a challenge for neurologists and neuroscientists. The symptoms of the famous patient, Phineas Gage, who was described over 100 years ago, is an example of the consequences of damage to this brain region. Prior to his accident he was a ‘‘religious, family-loving, honest, and hard-working man’’ who was described after his frontal injury as ‘‘fitful, irreverent, indulging at times in the grossest profanity y impatient of restraint or advice when it conflicts with his desires y obstinate y devising many plans of operation, which are no sooner arranged than they are abandoned in turn for others appearing more feasible’’ (Harlow, 1868). The consequences of Gage’s brain injury were perplexing to his Corresponding author. Tel.: 510-643-3340; Fax: 510-642-3192; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57008-6
123
124
both human and nonhuman primates. It is proposed that goal-directed behavior, at the core of what we consider human, depends critically on the function of the frontal lobes, and, specifically, the prefrontal cortex (PFC). The extensive reciprocal connections between the PFC to virtually all cortical and subcortical structures place the PFC in a unique neuroanatomical position to monitor and manipulate diverse cognitive processes. There are at least two major neural networks that interact with the PFC (Goldman-Rakic and Friedman, 1991). The first network involves reciprocal cortical–cortical connections between the PFC and the posterior parietal cortex as well as connections with the anterior and posterior cingulate, and medial temporal-lobe regions including the entorhinal and parahippocampal cortex (Selemon and Goldman-Rakic, 1988). The second network involves cortical–subcortical connections between the PFC and the striatum, globus pallidus, substantia nigra, and mediodorsal nucleus of the thalamus (Ilinisky et al., 1985). Each of these networks likely subserves different component processes necessary for goal-directed behavior such as maintenance of goals (cortical network) and response selection and motor control (subcortical network). The prevalence of the frontal-lobe syndrome, in one form or another is greater than it appears. For example, a wide range of pathology such as traumatic brain injury, stroke, and neoplasms commonly affect the PFC, and many other conditions such as attention-deficit disorder, substance addiction, schizophrenia, and , schizophrenia, and – 4normal aging, are proposed to involve selective dysfunction of frontal brain systems. Cognitive deficits resulting from PFC damage have been particularly challenging to rehabilitate (D’Esposito and Gazzaley, 2006). Despite considerable effort by clinicians and researchers to develop rehabilitation strategies for such individuals, attempts to treat such patients have often yielded disappointing results. It is challenging to develop such therapeutic interventions for several reasons. First, a solid theoretical basis on which to develop interventions has been lacking. Such a foundation is crucial not only for designing therapies, but also for deciding how to measure and test the efficacy of therapies. Inappropriate outcome measures are
as likely to result in ‘negative’ treatment trials as inappropriate treatments, and it is difficult to distinguish between these two causes of negative results. Secondly, there is a significant range of cognitive processes that may be altered by injuries to the PFC (e.g. deficits in planning, response inhibition, initiation, self-awareness). This seemingly diverse array of processes, typically described at a behavioral or cognitive level, makes for a confusing array of therapeutic targets. We suggest that some of this confusion may be reduced by developing a unifying theoretical framework that emphasizes the underlying neural processes. Third, many patients with PFC dysfunction exhibit behavioral deficits such as a lack in self-awareness, poor motivation, or mood-regulation disorders that cause a fundamental impediment to any rehabilitation process. It becomes clear to any practitioner that these factors must be taken into account for rehabilitation of dysfunction in any domain, including noncognitive domains such as motor function. The diverse cognitive and behavioral deficits of patients with PFC dysfunction makes it difficult to develop generalized rehabilitation interventions for such patients. Many different techniques have been developed, and studies testing the validity of such techniques are usually presented as case studies describing interventions on individual patients or small series of cases. This leads to uncertainty that these techniques are generalizable. In this chapter, we put forth the hypothesis that further insight into the neural mechanisms underlying normal PFC function may ultimately help us understand the frontal lobe syndrome, and importantly, potentially lead to more consistently effective therapeutic interventions for this condition. We will review published hypotheses regarding the neural mechanisms underlying PFC function that could guide the development of therapeutic interventions in patients with frontal lobe syndromes. Current cognitive-rehabilitation treatments, and clinical trials, rarely consider the normal neural mechanisms of the PFC. Thus, this approach may but complement result in treatment prescriptions that differ from more traditional behavioral approaches. Furthermore, by delineating neural targets of therapy, this approach will help fuse
125
training-based rehabilitation approaches with newer biologic tools to ultimately maximize the effects of rehabilitation therapy.
Neural mechanisms underlying prefrontal cortical function Active maintenance of task-relevant representations The results of experiments in behaving monkeys using recordings from single units in the lateral PFC Figure 1(a), have consistently found persistent, sustained levels of neuronal firing during the retention interval in tasks which require a monkey to retain information for a brief time (e.g. seconds) (Fuster and Alexander, 1971; Kubota and Niki, 1971; Funahashi et al., 1989). This sustained
activity is thought to provide a bridge between the stimulus cue, for instance, the location of a flash of light, and its contingent response, for instance, a saccade to the remembered location. These results have been supported by functionalimaging studies in humans and there is now a critical mass of studies of neural activity in the lateral PFC in humans during delay tasks (for review, see Curtis and D’Esposito, 2003). For example, in a fMRI study using a oculomotor delay task identical to that used in monkey studies, not only did we observe PFC activity during the retention interval (see Fig. 1(b)) but the magnitude of the activity correlated positively with the accuracy of the memory-guided saccade that followed later. This relationship suggests that the fidelity of the stored location is reflected in the delay period activity (Curtis et al., 2004). Thus, the existence of
Fig. 1. Neural activity in the monkey and human lateral PFC during the retention interval of a spatial oculomotor delayed response (ODR) task. (A) Average of single-unit recordings from 46 neurons with delay period activity from the monkey DLPFC (area 46) (adapted from Funahashi et al., 1989). C ¼ cue; D ¼ delay; R ¼ response. (B) Significant delay period activity (left) and average (7SE) fMRI signal (right) from right lateral PFC (area 46; circled) in a human performing an ODR task (unpublished data). The gray bar represents the length of the delay interval. Notice how in both cases the level of PFC activity persists throughout the delay, seconds after the stimulus cue has disappeared. See Plate 8.1 in Colour Plate Section.
126
persistent neural activity during blank memory intervals of delay tasks is a powerful empirical finding that lends strong support for the hypothesis that the lateral PFC is a critical node that supports active maintenance of task-relevant representations. The necessity of this region for active maintenance of task-relevant representations has been demonstrated by studies that have found impaired performance on delay tasks in monkeys with selective lesions of the lateral PFC (Bauer and Fuster, 1976; Funahashi et al., 1993). These physiological and lesion studies provided evidence that suggests that the primary function of the lateral PFC is to create and maintain internal representations of relevant sensory information necessary for guiding behavior. Miller and Cohen (2001) extended this hypothesis by suggesting that in addition to recent sensory information, integrated representations of task contingencies and even abstract rules (e.g., if this object then this later response) are also maintained in the PFC. This is similar to what Fuster (1997) has long emphasized, namely, that the PFC is critically responsible for temporal integration and the mediation of events that are separated in time but contingent on one another. Thus, sustained delay-period activity likely reflects not only the maintenance of many goal-directed representations such as past sensory events (i.e., a retrospective code), but also representations of anticipated action and preparatory set (i.e., prospective codes) (Quintana and Fuster, 1993; D’Esposito et al., 2000). For example, we found evidence in the fMRI study previously mentioned (Curtis et al., 2004), that different brain regions are involved in the storage of retrospective versus prospective codes. In the oculomotor delayed response (ODR) task, participants in such studies were biased toward or against the use of a prospective motor code. In one condition (match trials), participants were able to plan a saccade to the target as soon as the cue appeared and then they could simply postpone the initiation of the saccade until after the delay. Delay period activity should reflect this strategy, that is, the maintenance of a prospective motor code or motor intention. In a comparison condition (nonmatch trials), a saccade was made after the retention interval to an unpredictable location that did not match the location of the
sample. The participants still had to remember the location of the sample so that they could discern between the matching and nonmatching targets. Since a saccade was never made to the sample location and the nonmatching location was unpredictable, we concluded that this manipulation biased the participant away from maintaining a motor code during the delay. Instead, it encouraged the maintenance of a retrospective sensory code, or sustained spatial attention. We found that delay period activity was greater for the match compared to nonmatch trials within oculomotor regions whereas delay period activity for nonmatch trials was greater in frontal parietal regions (see Fig. 2).
Enhancement and suppression of task-relevant representations Goal-directed behavior requires an interaction of top-down and bottom-up processes. By bottomup, we mean those processes that guide automatic behavior and are determined by the nature of sensory input. By top-down, we mean those processes that guide behavior that is determined by internal states such as knowledge from previous experience, expectations, and goals (Miller and Cohen, 2001). It is proposed that the PFC plays a critical role in goal-directed behavior by providing direct feedback signals to posterior association cortex, that is processing incoming sensory input from a particular modality (e.g., visual or auditory). For example, when a person is looking at a crowd of people, the visual scene presented to the retina may include a myriad of angles, shapes, people, and objects. However, if that person is a police officer looking for an armed robber escaping through the crowd, some mechanism of suppressing irrelevant visual information while enhancing task-relevant information is necessary for an efficient and effective search. Thus, neural activity throughout the brain that is generated by input from the outside world, may be differentially enhanced or suppressed, presumably from top-down signal emanating from the PFC, based on the context of the situation.
127
Fig. 2. Statistical parametric t-maps contrasting oculomotor delayed matching-to-sample versus nonmatching-to-sample delay period-specific activity. Activity during the (a) early and late (b) delay period are shown. Warm colors depict regions with greater delay period activity on matching than nonmatching trials. Cool colors depict regions with greater delay period activity on nonmatching than matching trials. FEF ¼ frontal eyefields; SEF ¼ supplementary eye-fields; MFG ¼ middle frontal gyrus; pIFS ¼ posterior inferior frontal sulcus; iPCS ¼ inferior precentral sulcus; IPS ¼ intraparietal sulcus. See Plate 8.2 in Colour Plate Section.
We used a delay task during fMRI scanning to directly study top-down modulation by investigating the processes involved when participants were required to enhance relevant and suppress irrelevant information (Gazzaley et al., 2005). During each trial, participants observed sequences of two faces and two natural scenes presented in a randomized order. The tasks differed in the instructions informing the participants how to process the stimuli: (1) Remember Faces and Ignore Scenes, (2)
Remember Scenes and Ignore Faces, or (3) Passively View faces and scenes without attempting to remember them. In each task, the period in which the cue stimuli were presented was balanced for bottom-up visual information, thus allowing us to probe the influence of goal-directed behavior on neural activity (top-down modulation). In the two memory tasks, the encoding of the task-relevant stimuli requires selective attention and thus permits the dissociation of physiological measures of enhancement and suppression relative to the passive baseline. Also in the memory tasks, after a short delay period, the participants were tested on their ability to recognize a probe stimulus as being one of the task-relevant cues, yielding a behavioral measure of memory performance. These experiments were performed using both event-related fMRI and electroencephalography (event-related potentials, ERP) to record correlates of neural activity while the participants performed the task. This allowed us to capitalize on the high spatial resolution of the fMRI and the high temporal resolution of the ERP. We investigated activity measures of enhancement and suppression obtained from the visual association cortex of young healthy individuals. For fMRI, we used an independent functional localizer to identify both stimulus-selective face regions and scene regions in the fusiform gyrus (called the fusiform face area or FFA) and the parahippocampal/lingual gyrus (called the parahippocampal place area or PPA), respectively. For ERP, we utilized a face-selective ERP, the N170, a component localized to posterior occipital electrodes and reflecting visual association cortex activity with face specificity (Bentin et al., 1996). Our fMRI and ERP data revealed topdown modulation of both activity magnitude and processing speed that occurred above and below the perceptual baseline depending on task instruction (see Fig. 3). That is, during the encoding period of the delay task, FFA activity was enhanced, and the N170 occurred earlier, when faces had to be remembered as compared to a condition where they were passively viewed. Likewise, FFA activity was suppressed, and the N170 occurred later, when faces had to be ignored compared to a condition where they were passively viewed.
128
Fig. 3. Functional MRI (left) and ERP data (right) during the performance of a face/scene delay task in healthy human individuals. The left-hand graph shows fMRI signal from the right fusiform face area (FFA) during the three behavioral conditions is plotted. fMRI signal is greatest during the Remember Faces condition and least during the Ignore Faces condition. The right- hand graphs show the average N170 peak latency values during the three behavioral conditions are plotted. N170 latency is earliest during the Remember Faces condition and latest during the Ignore Faces condition.
Thus, there appeared to be two types of topdown signals, one that serves to enhance task-relevant information, and the other that serves to suppress task-relevant information. It is well documented that the nervous system utilizes interleaved inhibitory and excitatory mechanisms throughout the neuroaxis (e.g., spinal reflexes, cerebellar outputs, and basal-ganglia movement control networks). It is thus not surprising that top-down modulation would utilize enhancement and suppression to control cognition, providing a powerful contrast for sculpting these neural processes (Knight et al., 1999; Shimamura, 2000). Thus, by generating contrast via enhancing and suppressing activity magnitude and processing speed, top-down signals bias the likelihood of successful representation of relevant information in a competitive system. Modulation of the processing speed as reflected by a shift in the latency of the N170 was also a novel finding that revealed another aspect of topdown modulation. It suggested that in addition to modifying activity magnitude, top-down influences can modulate the time-course of neural activity, as reflected by a shorter time to reach maximal synchronized neural activity (Silva, 1991). It has been proposed that amplification of activity magnitude improves signal/noise ratio, allowing more information to be extracted from relevant stimuli (Hillyard et al., 1998). Likewise, faster processing speed
reflects an augmentation in the efficiency of neural processing, further facilitating extraction of important information. Though theories of PFC function propose that it is the source of the types of top-down signals that we have described, this hypothesis largely originates from suggestive findings rather than direct empirical evidence. However, a few studies lend direct support to this hypothesis. For example, Fuster et al. (1985) investigated the effect of inactivation of specific parts of the PFC by cooling on spiking activity in inferotemporal cortex (ITC) neurons during a delayed-match-to-sample color task. During the delay interval in this task — when persistent stimulus-specific activity in ITC neurons is observed — inactivation caused attenuated spiking profiles and a loss of stimulus-specificity of ITC neurons. These two alterations of ITC signaling strongly implicate the PFC as a source of topdown signals necessary for maintaining robust sensory representations in the absence of bottomup sensory activity. Tomita et al. (1999) isolated top-down signals during the retrieval of pairedassociates in a visual-memory task. Spiking activity was recorded from stimulus-specific ITC neurons as cue stimuli we re presented to the ipsilateral hemifield. This experiment’s unique feature was the ability to separate bottom-up sensory signals from a top-down mnemonic reactivation by using a posterior split-brain procedure that
129
limited hemispheric crosstalk to the anterior corpus callosum connecting each PFC. When a probe stimulus was presented ipsilaterally to the recording site, thus restricting bottom-up visual input to the contralateral hemisphere, stimulus-specific neurons became activated at the recording site approximately 170 ms later. Since these neurons received no bottom-up visual signals of the probe stimulus and the only route between the two hemispheres was via the PFC, this experiment showed that PFC neurons were sufficient to trigger the reactivation of object-selective representations in ITC regions in a top-down manner. The combined lesion/electrophysiological approach in humans has rarely been implemented. However, Chao and Knight (1998) studied patients with lateral PFC lesions during delayed match-to-sample tasks. It was found that when distracting stimuli are presented during the delay period the amplitude of the recorded ERP recorded from posterior electrodes was markedly increased in patients compared to controls. These results were interpreted to show disinhibition of sensory processing and the results supports a role of the PFC in suppressing the representation of stimuli that are irrelevant for current behavior.
Accessing task-relevant representations In the previous two sections, we presented data that provides evidence that the PFC likely provides top-down signals that can either enhance task-relevant representations that must be subsequently maintained to guide behavior, or suppress task-irrelevant representations. The PFC is also likely to be involved in the decision processes that are necessary to utilize task-relevant representations that are being actively maintained. To address this, we performed an event-related fMRI study that was designed to determine the relative contributions of cue, delay, and probe task periods to PFC activation in a verbal delayed recognition task (Rypma and D’Esposito, 1999). In this study, we asked participants to maintain either 2 or 6 letters across an unfilled delay period, followed by a single probe letter during which they indicated with a button press whether the probe letter was a
part of the memory set. Thus, to assess the role of the PFC in accessing task-relevant representations, we were specifically interested in PFC activity during the probe period of the task. For the analysis of the data in this experiment, we measured behavioral performance of each individual participant in terms of their memory retrieval rate, which is the interpolated slope obtained when plotting reaction time (RT) against memory load (2- vs. 6-letter trials). The RT slope is assumed to be a valid index for memory retrieval rate when participants must make a yes/no decision about the membership of a probe stimulus in the memory set (Sternberg, 1966). Further, memory retrieval rate may vary with the efficiency of memory-scanning processes. Linear regression analyses were applied to data from each individual participant from each task stage (e.g., cue, delay, and probe periods) and from lateral PFC regions of interest to test for relationships between behavioral performance and PFC activity. The results of these individual differences analyses indicated that, in the dorsolateral PFC, retrieval rate and activation were positively correlated, but only during the probe period (see Fig. 4). That is, those individuals with faster retrieval rates had less PFC activation whereas those individuals with slower retrieval rates had more PFC activation. These findings lead to several conclusions. First, the finding of a significant brain–behavior link in dorsolateral PFC, only during the probe period suggests that this PFC region is involved in memory scanning, a retrieval process that is initiated with the onset of the probe stimulus. Thus, in addition to being involved in encoding and maintaining task-relevant representations, the results indicated that the PFC is also involved in accessing these representations. Second, the findings suggest that the reduction in information retrieval rate, possibly reflecting the rate of memory scanning (e.g., Sternberg, 1969), is related to increases in dorsolateral PFC activation. This result is intriguing as it suggests one possible model for the neural correlates of processing efficiency. It may be that differences in processing speed are related to decreases in the efficiency of cortical processing, as reflected in the present data by the extent of activation during memory retrieval. Since decreased
130
Fig. 4. Scatter plots of the normalized regional fMRI signal during the Response Period in dorsolateral PFC plotted against normalized reaction times (RT) in younger participants (black squares) and older participants (circles).
retrieval rate corresponds to less efficient workingmemory scanning, poorer performers on this task may have recruited broader networks within lateral PFC to compensate for inefficient working memory scanning processes.
mechanisms underlying failure of frontal systems. To examine the effects of such failure on the types of PFC cognitive processes discussed so far in this chapter, we have studied older individuals, as well as patients with frontal injury using the same paradigms as those previously administered to young healthy control participants.
Dysfunction of frontal systems in aging and disease It is well-established that many aspects of cognition decline with normal aging (Cabeza et al., 2000). It has been postulated that cognitive aging is due to frontal systems impairment (West, 1996). Support for this hypothesis derives from evidence of agerelated changes in structure (e.g., de Brabander et al., 1998) and function (e.g., Nielsen-Bohlman and Knight, 1995; Cabeza et al., 2002) of the PFC. For example, in vivo and post-mortem studies of humans and primates have shown that the strongest age-related cerebral cortical changes are reductions in the PFC gray matter and white matter (e.g., Raz et al., 1998; Tisserand et al., 2002). In addition, functional neuroimaging studies have reported agerelated changes in PFC activation and studies of a variety of cognitive tasks (e.g., Grady et al., 1999; McIntosh et al., 1999; Rypma and D’Esposito, 2000) have indicated that changes occur in the functional connectivity with posterior cortical regions with aging. Thus, normal aging serves as a reasonable model of the neural
Effects of aging and frontal lesions on top-down modulation Behavioral evidence suggests that age-related memory impairments are associated with increased sensitivity to interference from task-irrelevant information (Hasher and Zacks, 1988; West, 1996; May et al., 1999). Thus, we hypothesized that on memory tasks that require top-down modulation, older individuals may have a selective deficit in their ability to suppress task-irrelevant information. We used the face/scene delay task described earlier in the chapter and compared the ability of a group of younger (19–30 years of age) and older individuals (60–77 years of age) to remember one type of stimulus (e.g., faces) and ignore the other type (e.g., scenes) when both types of stimuli were presented at the same time. Direct comparisons of fMRI’s across age groups revealed a significantly greater signal magnitude within the scene-selective region in the older group than in the younger
131
group in the Ignore Scenes condition, while no age-related differences existed between the Remember Scenes or Passive View conditions. These comparisons reveal a selective age-related deficit in the suppression of task-irrelevant information analysis using three modulation indices: overall modulation index (Remember Scenes –Ignore Scenes), enhancement index (Remember Scenes–Passive View), and suppression index (Passive view–Ignore scenes) confirmed an age-related decrease in the degree of overall modulation. Critically, this age-related decrease in modulation can be attributed to a selective decrease in the subcomponent process of suppression, as there was no significant difference in the enhancement subcomponent. Preliminary data from a study of patients with focal frontal lesions due to strokes using the same experimental paradigm show that these patients have a lower overall modulation index, that is the magnitude of enhancement and suppression of category-specific visual association areas is reduced, relative to healthy control participants (see Fig. 5(a)). Although these findings are preliminary, these data may provide strong evidence that the PFC is the source of top-down signals that enhance and suppress task-relevant information. Intriguingly, some of these patients exhibited activation of PFC in the hemisphere opposite to their lesion, while healthy participants typically had unilateral activation. Similar to the finding in studies of normal aging (Rajah and D’Esposito, 2005), this finding may reflect compensatory activity for an injured system (see Fig. 5(b)).
Effects of aging on neural efficiency The accuracy of the performance of young (mean age ¼ 25) and older participants (mean age ¼ 69) in the delay task described earlier in this chapter was not significantly different between younger and older participants but younger participants were faster than older participants. During each trial, the participants encoded either 2 or 6 letters and retained them across an unfilled delay interval (Rypma and D’Esposito, 2000). Regression analyses of the participants’ RT and PFC activity
(Rypma and D’Esposito, 1999) indicated that the activity in the dorsolateral PFC in younger participants showed a significant positive correlation between mean RT and cortical activity. In contrast, in dorsolateral PFC in older participants, the response period regression coefficients showed a significant negative correlation between mean RT and cortical activity (Fig. 4). We propose that changes in the relationship between neural activity and performance that we observed in these data reflect an age-related decrease in neural efficiency, which provides a plausible explanation for the observed age-related changes in performance on a wide range of tasks. Specifically, we found that decreased speed of information retrieval during delay tasks (possibly reflecting less efficient memory-scanning processes (Sternberg, 1969)) is related to increases in dorsolateral PFC activation for younger participants, but to decreases in dorsolateral PFC activation for older participants. There is convincing evidence that suggest that reductions in informationprocessing speed is related to age-related decreases in the overall efficiency of cognitive processing (e.g., Myerson et al., 1990; Salthouse, 1996). The current results suggest that there may be age-related differences in the neural correlates of processing efficiency, thus reductions in neural efficiency may lead to slowing of cognitive processes; specifically, the speed with which task-relevant representations can be accessed.
Consequences of frontal systems dysfunction In the healthy young brain, information from the environment drives early processing activity within the posterior sensory regions from the ‘bottom-up’ (see Fig. 6(a)). In order to guide subsequent processing for task- or goal-relevance, the PFC modulates posterior activity in a top-down fashion (e.g., guided by internal states such as knowledge from previous experience, expectations and goals). We propose that relevant networks are up-modulated relative to nonrelevant networks, resulting in an integration of the relevant networks to achieve a coherent output. As in our example of a person looking into a crowd of people, there are likely
132
Fig. 5. The top shows top-down modulation index of three patients with frontal lesions (red diamonds) and 18 healthy control individuals (black diamonds). Note that patients with frontal lesions have the lowest modulation index. Below are fMRI scans of healthy control participants and a patient with frontal lesion during performance of a face/scene delay task. Note that the healthy participants exhibit PFC activation in the site of the patient’s PFC lesion. However, during performance of the task, the patient exhibited PFC activation in the hemisphere opposite to the site of the lesion. See Plate 8.5 in Colour Plate Section.
neural mechanisms, mediated by the PFC, for downmodulating processing of irrelevant visual information (e.g., hot-dog vendor) while up-modulating taskrelevant information (e.g., police officer chasing a robber) necessary for an efficient and effective search. When PFC networks are not functioning optimally (either by direct injury of the nodes of the network or damage to the pathways that connect them), the changes may be understood as a loss of functional integration (see Fig. 6(b)). In contrast
to the example above, someone with PFC dysfunction may be left with visual processing (or any other modality) that is driven by bottom-up propagation of neural activity, with poor ability to selectively up-modulate networks that process the relevant information in preference to the many distractors in the environment. Similarly, these individuals may have difficulty selectively attending to other important information, keeping task-relevant information in mind, or prioritizing
133
Fig. 6. Illustration showing how the flow of information in sensory processing is guided by PFC function in the healthy and injured brain. (A) In the healthy brain, information from the external world enters the brain from the ‘bottom-up’ through sensory activation. Relevant information (e.g., face) is surrounded by irrelevant information (e.g., scene) with little intrinsic differentiation. Later sensory processing is modulated by the PFC based on task- or goal-relevance and information is differentially processed (symbolized by large face which is relevant and small scene which is not relevant).(B) When the PFC is injured, we propose that there is a loss of functional integration within the PFC networks. Weakened anterior-to-posterior functional connectivity (indicated by smaller size arrow) may result from structural disconnection such as axonal injury or subcortical lesions, or other failures of communication. The end result is a loss of top-down modulation of posterior activity, making posterior processing poorly modulated for task-relevance.
134
her activities to achieve important goals. They are left with disorganized behavior, mirrored by neural activity that is not organized by top-down guidance. The type of behavior described above corresponds accurately with what is observed in patients with frontal-lobe dysfunction. For example, patients with frontal-lobe lesions may display a remarkable tendency to imitate the examiner’s gestures and behaviors, even without instruction to do so, and even when this imitation entails considerable personal embarrassment (Lhermitte, 1986; Lhermitte et al., 1986). The mere sight of an object may elicit the compulsion to use it, although there was no request to use and the context is inappropriate, as in a patient who sees a tongue depressor and proceeds to give the physician a checkup or who puts on a pair of glasses despite having a pair on already. Without frontal function, autonomy from the environment and abstract thinking are impaired or impossible. A given stimulus automatically calls up a predetermined response regardless of the context. Likewise, Mesulam (2002) describes the PFC as necessary for overcoming the default mode — a state of reflexively triggered instinctive behaviors that are impervious to the context of the situation. Cognitive rehabilitation of PFC function The PFC plays a central role in the integration of neuronal activity distributed over multiple brain regions in order to facilitate processes that are relevant to the context. Thus, goal-directed behavior emerges from an integrated network of brain regions. Any analysis of plasticity in PFC function after brain injury should take into account this integrative function of the PFC, including the relationships between the PFC and functionally interconnected regions. Based on these hypotheses regarding PFC function, we discuss how certain key aspects of rehabilitation interventions may work to improve function of the PFC. Cognitive training tasks should challenge patients to engage ‘top-down’ modulatory processes mediated by the PFC. Most activity-based training regimens start with the very simple principle that training must effectively engage specific systems in order to strengthen them. Not all cognitive training tasks will engage the modulatory control
processes mediated by PFC networks. Functional MRI studies aimed at investigating normal brain–behavioral relationships can provide guidance for the key ingredients of tasks that can engage PFC networks. Specific cognitive processes have been shown to preferentially involve the PFC. For example, tasks that involve working memory (e.g., the maintenance of information over a short period of time), selective processing of competing information based on task-relevance, and performance of dual tasks all preferentially activate lateral PFC networks ( D’Esposito et al., 1995; Banich et al., 2000; Curtis et al., 2004). Each of these tasks shares the requirement to recruit top-down control processes for successful performance. During the performance of these tasks, it is the processing demands, and not the specific contents of stimuli per se, that activate PFC networks. For example, the PFC is engaged during working memory tasks regardless of the type of information (e.g., words or objects) that must be remembered (D’Esposito et al., 1998). As a general principle, training of specific top-down control processes (e.g., maintenance and updating processes), and not content, is important in training tasks that aim to target PFC function. This principle is particularly important for effectively transfer to new contexts, in which the specific content or material of the tasks has changed. Training should enhance the transfer and generalization of training effects to new materials or contexts. Transfer of training effects to nontrained tasks, and generalization of training effects to ecologically relevant contexts are key goals of rehabilitation training. We propose targeting the common underlying substrates of PFC integrative functions for that purpose. We hypothesize that if PFC networks are a substrate for transfer, then training that effectively improves indices of functional integration in PFC networks will be more likely to improve behavioral performance on nontrained tasks, including, in particular, tasks that require similar top-down control processes, applied to new situations that were not specifically trained. This contrasts in a subtle way with training to improve performance of specific tasks, such as preparing a specific meal (e.g., a peanut-butter sandwich in the rehabilitation center kitchen),
135
rather than the processes that help a person plan and cook meals. A patient may be able to learn this specific task well enough to return home and prepare this same meal, achieving an important functional goal, but she may not be able to apply the same processes of idea-generation, goal maintenance, planning and prioritizing necessary to prepare a different meal when the peanut butter runs out. Even greater challenges requiring flexibility in control processes are presented in many workplaces. Features of cognitive training that target and engage PFC networks are more likely to enhance the effectiveness and efficiency of topdown control processing in these challenging environments. Cognitive training material should cross sensory modalities. PFC is multimodal association cortex, serving to integrate information from all modalities. Thus, the processes subserved by PFC networks may involve any sensory modality. This is illustrated in functional imaging studies of working memory processes that reveal similar activation of PFC networks whether visual, auditory, or even olfactory content is used (Schumacher et al., 1996; Zelano et al., 2005). Thus, training across multiple modalities may maximize engagement of these underlying PFC networks, and may enhance a patient’s ability to recruit PFC networks in different contexts. Cognitive training should include a goal-oriented approach. Training of top-down control processes should focus on improving PFC function in the context of achieving specific personally relevant goals. The content of the goals, and thus the content of the tasks, may be individualized. Neural subprocesses involved in goal management may be trained regardless of the specific content. For example, in one goal-management training program (GMT) formalized for research purposes, patients are asked to go through five main steps (Levine et al., 2000). First, patients are asked to stop, and explicitly outline the goals of their actions. Patients are guided in generating personally-relevant goals, which may include achieving everyday tasks such as planning a meal or making a doctor’s appointment. Subsequent management of goal-generated tasks would require steps that would further engage PFC networks. These steps include
generation of subgoals and listing of associated tasks; learning and recalling goals and subgoals; and executing the goal-oriented tasks. These steps may require processes including sustaining attention, holding information in working memory, and self-evaluation of performance through comparing the intended outcomes with actual outcomes. All of these processes should engage the PFC with the objective of improving top-down control and goal management processing even in novel situations. Automatize the performance of basic, repetitive tasks that do not require flexibility. Training prescriptions that target top-down modulatory processes may be contrasted with training that encourages a reduction in demands for top-down modulatory control. The hallmark of automatization is an improvement in the speed and consistency of performance of a specific task that is repeated numerous times (Logan, 1988). This is a familiar principle in most current rehabilitation therapies, including, for example, repetitive lowerlimb motions to entrain automatic movements. A disengagement of PFC controlled processes has been found with learning in cognitive domains as well, reflected in reduced PFC activation in functional imaging studies (Raichle et al., 1994; Petersen et al., 1998; Buchel et al., 1999; Ramsey et al., 2004). Training with the goal of automatizing task processing may serve a therapeutic goal complementary to that of strengthening PFC networks — to reduce the demands for PFC modulatory control in order to ‘unload’ limited control resources. For example, a patient who is able to lift a fork to feed herself as an automatic process is more likely to be able to pay attention to an ongoing conversation. Cognitive training tasks should adaptively challenge the patient. Even tasks that engage PFC control processes may become less challenging with practice, and thus less effective at encouraging learning. As a patient’s proficiency with a PFC function improves, tasks may be adjusted such that demands for PFC processes are increased. This is more specific than simply increasing the ‘difficulty’ of the task, as parameters that are adjusted should quantitatively vary the level of engagement of specific PFC processes and not simply vary the level of general attentional, arousal, or
136
motivational processes. For example, to progressively increase working memory demands, stimulus load (the number of items to remember) or the length of the delay period (during which patients need to remember the items) can be increased in training tasks. Olesen and colleagues recently showed that an adaptive training program can increase working-memory capacity as measured behaviorally with an accompanied increase in PFC network activity as measured by fMRI (Olesen et al., 2004).
Development of biomarkers to assess rehabilitation of PFC function Rehabilitation of PFC dysfunction through cognitive training may be considered a process of guiding mechanisms of plasticity for the ‘re-integration’ of functional PFC networks. That is, mechanisms of plasticity following brain injuries include the possibilities of re-organization of available network components, or the generation of new network components. Mechanisms of plasticity at the cellular level that may support re-organization or regeneration include alterations in metabolism, synaptogenesis, and synaptic pruning, including growth of new long-distance projections, and perhaps neurogenesis (Keyvani and Schallert, 2002; Bach-y-Rita, 2003; Carmichael, 2003; Rossini and Dal Forno, 2004). These cellular changes do not fully describe the process of functional recovery, as the same changes could, if unguided, lead to tumor growth, for example. Ultimately, for these neuronal changes to affect neurological function, they must be translated into changes in the functioning of networks of neurons. It is suggested that effective training guides these neuronal changes to achieve functionally integrated networks and coherent behavioral output. Though there are as yet few studies of these mechanisms of plasticity in the PFC, there is much evidence to support this principle in motor and sensory cortices (Kilgard and Merzenich, 1998; Bao et al., 2001; Blake et al., 2002; Beitel et al., 2003). When brain injury disconnects cortical regions in the anterior–posterior networks illustrated in Fig. 6, cognitive rehabilitation treatments may
guide the cellular mechanisms discussed above to enhance the functional re-integration of networks. At least three different levels of change may support reintegration of network function. First, integration of residual intact PFC regions with relevant posterior regions may be supported by synaptic reorganization and synaptogenesis. Second, with respect to functional recovery, it is possible that there is some redundancy in PFC circuitry, such that residual intact areas may be able to reorganize to take over function previously supported through other regions. Cognitive training would, in essence, help in making damaged, poorly integrated collections of neurons into more efficient, better integrated functional networks for the performance of relevant tasks. Third, reorganization across the different PFC functional subdivisions is possible. In this chapter, we have only discussed the function lateral PFC, yet medial and orbital PFC regions also contribute to top-down control (Fuster, 1997). Given the extensive connectivity between this triad of PFC subdivisions (i.e., lateral, medial, orbital), when one region is damaged, it is possible that other regions in this highly interconnected network may also reorganize, interacting with the residual neurons of the damaged region to provide the deficient function via a newly integrated functional network. Given these potential mechanisms for recovery of function, it should be possible to develop biomarkers that would not only give insight into these mechanisms but could guide cognitive therapy and assess the effectiveness of such treatments. For example, we propose that after effective rehabilitation training aimed at enhancing PFC function, activity within the PFC should become better integrated, and there should be evidence of increased anterior–posterior functional connectivity. As evidence of improved functional integration, there should be increased task-relevant modulation of posterior brain activity as measured by fMRI. Task-related activity in the PFC may be actually increase with training, relating to increased strength or capacity to exert modulatory control (Olesen et al., 2004; Erickson et al., 2006). Examining plasticity in PFC function at this level of theoretical detail is a new frontier that bridges cognitive neuroscience and clinical rehabilitation.
137
Conclusions The PFC is a critical node in a network of brain regions that supports goal-directed behavior. Using a theoretical framework that incorporates an understanding of neural mechanisms underlying PFC function to develop cognitive rehabilitation, and combining training with neurophysiologic outcome measures may provide a powerful approach for improving such training for patients with brain injuries. Cognitive training alone may not be enough to overcome many of the deficits experienced by patients with brain injuries. A better understanding of plasticity in PFC function may provide new targets for rehabilitation therapies. Learning-related changes in PFC networks may be modified by pharmacotherapies (Kimberg et al., 1997; McDowell et al., 1998), more direct extrinsic targeting of PFC networks with methods such as transcranial magnetic stimulation (Butefisch et al., 2004), and adjunctive biologic modifiers of plasticity such as physical exercise, stem cells, and growth factors (Chopp and Li, 2002; Cotman and Berchtold, 2002; Bang et al., 2005). Since modulatory control processes mediated by the PFC are crucial for numerous learning processes, remediation of PFC function is expected to have a broad impact on learning and recovery from brain injuries. Addressing dysfunction in PFC control will enhance rehabilitation of motor functions, as well as improve an individual’s ability to adjust to residual motor deficits. Thus, improved rehabilitation of PFC functions should be the first-line goal of rehabilitation neuroscientists and practitioners.
References Bach-y-Rita, P. (2003) Theoretical basis for brain plasticity after a TBI. Brain Inj., 17: 643–651. Bang, O.Y., Lee, J.S., Lee, P.H. and Lee, G. (2005) Autologous mesenchymal stem cell transplantation in stroke patients. Ann. Neurol., 57: 874–882. Banich, M.T., Milham, M.P., Atchley, R., Cohen, N.J., Webb, A., Wszalek, T., Kramer, A.F., Liang, Z.P., Wright, A., Shenker, J. and Magin, R. (2000) fMRI studies of Stroop tasks reveal unique roles of anterior and posterior brain systems in attentional selection. J. Cogn. Neurosci., 12: 988–1000.
Bao, S., Chan, V.T. and Merzenich, M.M. (2001) Cortical remodelling induced by activity of ventral tegmental dopamine neurons. Nature, 412: 79–83. Bauer, R.H. and Fuster, J.M. (1976) Delayed-matching and delayed-response deficit from cooling dorsolateral prefrontal cortex in monkeys. J. Comp. Physiological Psychol., 90: 293–302. Beitel, R.E., Schreiner, C.E., Cheung, S.W., Wang, X. and Merzenich, M.M. (2003) Reward-dependent plasticity in the primary auditory cortex of adult monkeys trained to discriminate temporally modulated signals. Proc. Natl. Acad. Sci. USA, 100: 11070–11075. Bentin, S., Allison, T., Puce, A., Perez, E. and McCarthy, G. (1996) Electrophysiological studies of face perception in humans. J. Cogn. Neurosci., 8: 551–565. Blake, D.T., Byl, N.N. and Merzenich, M.M. (2002) Representation of the hand in the cerebral cortex. Behav. Brain Res., 135: 179–184. Buchel, C., Coull, J.T. and Friston, K.J. (1999) The predictive value of changes in effective connectivity for human learning. Science, 283: 1538–1541. Butefisch, C.M., Khurana, V., Kopylev, L. and Cohen, L.G. (2004) Enhancing encoding of a motor memory in the primary motor cortex by cortical stimulation. J. Neurophysiol., 91: 2110–2116. Cabeza, R., Anderson, N.D., Houle, S., Mangels, J.A. and Nyberg, L. (2000) Age-related differences in neural activity during item and temporal-order memory retrieval: a positron emission tomography study. J. Cogn. Neurosci., 12: 197–206. Cabeza, R., Anderson, N.D., Locantore, J.K. and McIntosh, A.R. (2002) Aging gracefully: compensatory brain activity in high-performing older adults. Neuroimage, 17: 1394–1402. Carmichael, S.T. (2003) Plasticity of cortical projections after stroke. Neuroscientist, 9: 64–75. Chao, L. and Knight, R. (1998) Contribution of human prefrontal cortex to delay performance. J. Cogn. Neurosci., 10: 167–177. Chopp, M. and Li, Y. (2002) Treatment of neural injury with marrow stromal cells. Lancet Neurol., 1: 92–100. Cotman, C.W. and Berchtold, N.C. (2002) Exercise: a behavioral intervention to enhance brain health and plasticity. Trends Neurosci., 25: 295–301. Curtis, C.E. and D’Esposito, M. (2003) Persistent activity in the prefrontal cortex during working memory. Trends Cogn. Sci., 7: 415–423. Curtis, C.E., Rao, V.Y. and D’Esposito, M. (2004) Maintenance of spatial and motor codes during oculomotor delayed response tasks. J. Neurosci., 24: 3944–3952. D’Esposito, M., Aguirre, G.K., Zarahn, E. and Ballard, D. (1998) Functional MRI studies of spatial and non-spatial working memory. Cogn. Brain Res., 7: 1–13. D’Esposito, M., Ballard, D., Zarahn, E. and Aguirre, G.K. (2000) The role of prefrontal cortex in sensory memory and motor preparation: an event-related fMRI study. Neuroimage, 11: 400–408. D’Esposito, M., Detre, J.A., Alsop, D.C., Shin, R.K., Atlas, S. and Grossman, M. (1995) The neural basis of the central executive system of working memory. Nature, 378: 279–281.
138 D’Esposito, M. and Gazzaley, A. (2006) Neurorehabilitation and executive function. In: Selzer, M.E., Cohen, L., Gage, F.H., Clarke, S. and Duncan, P.W. (Eds.), Neural Rehabilitation and Repair. Cambridge University Press, Cambridge, UK, pp. 475–487. de Brabander, J.M., Kramers, R.J. and Uylings, H.B. (1998) Layerspecific dendritic regression of pyramidal cells with ageing in the human prefrontal cortex. Eur. J. Neurosci., 10: 1261–1269. Erickson, K.I., Colcombe, S.J., Wadhwa, R., Bherer, L., Peterson, M.S., Scalf, P.E., Kim, J.S., Alvarado, M., and Kramer, A.F., 2006. Training-induced functional activation changes in dual-task processing: an fMRI study. Cereb Cortex. Funahashi, S., Bruce, C.J. and Goldman-Rakic, P.S. (1989) Mnemonic coding of visual space in the monkey’s dorsolateral prefrontal cortex. J. Neurophysiol., 61: 331–349. Funahashi, S., Bruce, C.J. and Goldman-Rakic, P.S. (1993) Dorsolateral prefrontal lesions and oculomotor delayed-response performance: evidence for mnemonic ‘‘scotomas’’. J. Neurosci., 13: 1479–1497. Fuster, J. (1997) The Prefrontal Cortex: Anatomy, Physiology, and Neuropsychology of the Frontal Lobes. Raven Press, New York. Fuster, J.M. and Alexander, G.E. (1971) Neuron activity related to short-term memory. Science, 173: 652–654. Fuster, J.M., Bauer, R.H. and Jervey, J.P. (1985) Functional interactions between inferotemporal and prefrontal cortex in a cognitive task. Brain Res., 330: 299–307. Gazzaley, A., Cooney, J.W., McEvoy, K., Knight, R.T. and D’Esposito, M. (2005) Top-down enhancement and suppression of the magnitude and speed of neural activity. J. Cogn. Neurosci., 17: 1–11. Goldman-Rakic, P.S. and Friedman, H.R. (1991) The circuitry of working memory revealed by anatomy and metabolic imaging. In: Levin, H., Eisenberg, H. and Benton, A. (Eds.), Frontal Lobe Function and Dysfunction. Oxford University Press, New York, pp. 72–91. Grady, C.L., McIntosh, A.R., Rajah, M.N., Beig, S. and Craik, F.I. (1999) The effects of age on the neural correlates of episodic encoding. Cereb Cortex, 9: 805–814. Harlow, J. (1868) Recovery from the passage of an iron bar through the head. Proc. Mass. Med. Soc., 2: 725–728. Hasher, L. and Zacks, R.T. (1988) Working memory, comprehension, and aging: a review and a new view. Psychol. Learn. Motivation Adv. Res. Theory, 22: 193–225. Hillyard, S.A., Vogel, E.K. and Luck, S.J. (1998) Sensory gain control (amplification) as a mechanism of selective attention: electrophysiological and neuroimaging evidence. Philos Trans. R Soc. Lond. B Biol. Sci., 353: 1257–1270. Ilinisky, I.A., Jouandet, M.L. and Goldman-Rakic, P.S. (1985) Organization of the nigrothalamocortical system in the rhesus monkey. J. Comp. Neurol., 236: 315–330. Keyvani, K. and Schallert, T. (2002) Plasticity-associated molecular and structural events in the injured brain. J. Neuropathol. Exp. Neurol., 61: 831–840. Kilgard, M.P. and Merzenich, M.M. (1998) Cortical map reorganization enabled by nucleus basalis activity. Science, 279: 1714–1718.
Kimberg, D.Y., D’Esposito, M. and Farah, M. (1997) Effects of bromocriptine on human subjects depend on working memory capacity. NeuroReport, 8: 3581–3585. Knight, R.T., Staines, W.R., Swick, D. and Chao, L.L. (1999) Prefrontal cortex regulates inhibition and excitation in distributed neural networks. Acta Psychol. (Amst)., 101: 159–178. Kubota, K. and Niki, H. (1971) Prefrontal cortical unit activity and delayed alternation performance in monkeys. J. Neurophysiol., 34: 337–347. Levine, B., Robertson, I.H., Clare, L., Carter, G., Hong, J., Wilson, B.A., Duncan, J. and Stuss, D.T. (2000) Rehabilitation of executive functioning: an experimental-clinical validation of goal management training. J. Int. Neuropsychol. Soc., 6: 299–312. Lhermitte, F. (1986) Human autonomy and the frontal lobes Part II: patient behavior in complex and social situations: the ‘‘environmental dependency syndrome’’. Ann. Neurol., 19: 335–343. Lhermitte, F., Pillon, B. and Serdaru, M. (1986) Human autonomy and the frontal lobes. Part I: imitation and utilization behavior: a neuropsychological study of 75 patients. Ann. Neurol., 19: 326–334. Logan, G.D. (1988) Toward an instance theory of automatization. Psychol. Rev., 95: 492–527. May, C.P., Hasher, L. and Kane, M.J. (1999) The role of interference in memory span. Memory Cogn., 27: 759–767. McDowell, S., Whyte, J. and D’Esposito, M. (1998) Differential effect of a dopamine agonist on prefrontal function in traumatic brain injury patients. Brain, 121: 1155–1164. McIntosh, A.R., Sekuler, A.B., Penpeci, C., Rajah, M.N., Grady, C.L., Sekuler, R. and Bennett, P.J. (1999) Recruitment of unique neural systems to support visual memory in normal aging. Curr. Biol., 9: 1275–1278. Mesulam, M.M. (2002) The human fronal lobes: transcending the default mode through contingent encoding. In: Stuss, D. and Knight, R.T. (Eds.), Principles of Frontal Lobe Function. Oxford University Press, New York, pp. 8–30. Miller, E.K. and Cohen, J.D. (2001) An integrative theory of prefrontal cortex function. Annu. Rev. Neurosci., 24: 167–202. Myerson, J., Hale, S., Wagstaff, D., Poon, L.W. and Smith, G.A. (1990) The information-loss model: a mathematical theory of age-related cognitive slowing. Psychol. Rev., 97: 475–487. Nielsen-Bohlman, L. and Knight, R.T. (1995) Prefrontal alterations during memory processing in aging. Cereb Cortex, 5: 541–549. Olesen, P.J., Westerberg, H. and Klingberg, T. (2004) Increased prefrontal and parietal activity after training of working memory. Nat. Neurosci., 7: 75–79. Petersen, S.E., van Mier, H., Fiez, J.A. and Raichle, M.E. (1998) The effects of practice on the functional anatomy of task performance. Proc. Natl. Acad. Sci. USA, 95: 853–860. Quintana, J. and Fuster, J.M. (1993) Spatial and temporal factors in the role of prefrontal and parietal cortex in visuomotor integration. Cerebral Cortex, 3: 122–132.
139 Raichle, M.E., Fiez, J.A., Videen, T.O., MacLeod, A.M., Pardo, J.V., Fox, P.T. and Petersen, S.E. (1994) Practice-related changes in human brain functional anatomy during nonmotor learning. Cereb. Cortex, 4: 8–26. Rajah, M.N. and D’Esposito, M. (2005) Region-specific changes in prefrontal function with age: a review of PET and fMRI studies on working and episodic memory. Brain, 128: 1964–1983. Ramsey, N.F., Jansma, J.M., Jager, G., Van Raalten, T. and Kahn, R.S. (2004) Neurophysiological factors in human information processing capacity. Brain, 127: 517–525. Raz, N., Gunning-Dixon, F.M., Head, D., Dupuis, J.H. and Acker, J.D. (1998) Neuroanatomical correlates of cognitive aging: evidence from structural magnetic resonance imaging. Neuropsychology, 12(1): 95–114. Rossini, P.M. and Dal Forno, G. (2004) Integrated technology for evaluation of brain function and neural plasticity. Phys. Med. Rehabil. Clin. N. Am., 15: 263–306. Rypma, B. and D’Esposito, M. (1999) The roles of prefrontal brain regions in components of working memory: effects of memory load and individual differences. Proc Natl Acad Sci USA, 96: 6558–6563. Rypma, B. and D’Esposito, M. (2000) Isolating the neural mechanisms of age-related changes in human working memory. Nat. Neurosci., 3: 509–515. Salthouse, T.A. (1996) The processing-speed theory of adult age differences in cognition. Psychol. Rev., 103: 403–428. Schumacher, E.H., Lauber, E., Awh, E., Jonides, J., Smith, E.E. and Koeppe, R.A. (1996) PET evidence for an amodal verbal working memory system. NeuroImage, 3: 79–88.
Selemon, R.D. and Goldman-Rakic, P.S. (1988) Common cortical and subcortical targets of the dorsolateral prefrontal and posterior parietal cortices in the rhesus monkey: evidence for a distributed neural network serving spatially guided behavior. J. Neurosci., 8: 4049–4068. Shimamura, A.P. (2000) The role of the prefrontal cortex in dynamic filtering. Psychobiology, 28: 207–218. Silva, L.D. (1991) Neural mechanisms underlying brain waves: from neural membranes to networks. Electroenceph. Clin. Neurophysiol., 79: 81–93. Sternberg, S. (1966) High-speed scanning in human memory. Science, 153: 652–654. Sternberg, S. (1969) Memory-scanning: mental processes revealed by reaction-time experiments. Am. Sci., 57: 421–457. Tisserand, D.J., Pruessner, J.C., Sanz Arigita, E.J., van Boxtel, M.P., Evans, A.C., Jolles, J. and Uylings, H.B. (2002) Regional frontal cortical volumes decrease differentially in aging: an MRI study to compare volumetric approaches and voxel-based morphometry. Neuroimage, 17: 657–669. Tomita, H., Ohbayashi, M., Nakahara, K., Hasegawa, I. and Miyashita, Y. (1999) Top-down signal from prefrontal cortex in executive control of memory retrieval. Nature, 401: 699–703. West, R.L. (1996) An application of prefrontal cortex function theory to cognitive aging. Psychol. Bull., 2: 272–292. Zelano, C., Bensafi, M., Porter, J., Mainland, J., Johnson, B., Bremner, E., Telles, C., Khan, R. and Sobel, N. (2005) Attentional modulation in human primary olfactory cortex. Nat. Neurosci., 8: 114–120.
Plate 8.1. Neural activity in the monkey and human lateral PFC during the retention interval of a spatial oculomotor delayed response (ODR) task. (A) Average of single-unit recordings from 46 neurons with delay period activity from the monkey DLPFC (area 46) (adapted from Funahashi et al., 1989). C ¼ cue; D ¼ delay; R ¼ response. (B) Significant delay period activity (left) and average (7SE) fMRI signal (right) from right lateral PFC (area 46; circled) in a human performing an ODR task (unpublished data). The gray bar represents the length of the delay interval. Notice how in both cases the level of PFC activity persists throughout the delay, seconds after the stimulus cue has disappeared.
Plate 8.2. Statistical parametric t-maps contrasting oculomotor delayed matching-to-sample versus nonmatching-to-sample delay period-specific activity. Activity during the (a) early and late (b) delay period are shown. Warm colors depict regions with greater delay period activity on matching than nonmatching trials. Cool colors depict regions with greater delay period activity on nonmatching than matching trials. FEF ¼ frontal eye-fields; SEF ¼ supplementary eye-fields; MFG ¼ middle frontal gyrus; pIFS ¼ posterior inferior frontal sulcus; iPCS ¼ inferior precentral sulcus; IPS ¼ intraparietal sulcus.
Plate 8.5. The top shows top-down modulation index of three patients with frontal lesions (red diamonds) and 18 healthy control individuals (black diamonds). Note that patients with frontal lesions have the lowest modulation index. Below are fMRI scans of healthy control participants and a patient with frontal lesion during performance of a face/scene delay task. Note that the healthy participants exhibit PFC activation in the site of the patient’s PFC lesion. However, during performance of the task, the patient exhibited PFC activation in the hemisphere opposite to the site of the lesion.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 9
Recovery from aphasia following brain injury: the role of reorganization Elisabeth B. Marsh and Argye E. Hillis Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD21287, USA
Abstract: Language is predominantly a left hemisphere function, yet patients with extensive damage to known language areas often recover quite well in the days to weeks to even years following focal brain injury. This recovery period can be divided into three overlapping stages: acute, subacute, and chronic, each with different underlying neural mechanisms. Reorganization of structure and function through the expression of neural plasticity plays a crucial role in recovery of language at least during the subacute phase of weeks to months after the occurrence of an injury. In this chapter we review the evidence for reorganization of language function after injury, the role it plays in the recovery of language following brain damage, and how knowledge of the mechanisms of recovery will allow design of more effective methods of rehabilitation. Keywords: language; aphasia; reorganization; recovery; stroke; MRI different language disorders. While localization of language function is complex and controversial, a few widely agreed upon generalizations can be stated. Damage or dysfunction of the left posterior inferior frontal cortex (Broca’s area) generally leads to impaired programming or planning of speech articulation, agrammatic speech, and difficulty in processing morphosyntactic processes for comprehension of syntactically complex sentences. Dysfunction of left temporal or parietal areas can disrupt representations of the meanings of words, and dysfunction of left posterior, superior temporal gyrus (Wernicke’s area) disrupts the linkage between words and their meanings.
Introduction It is commonly accepted that language is predominantly a left hemisphere function, yet patients with extensive damage to the left hemisphere are often able to recover, at least somewhat, over the days to weeks to even years following their injury. Reorganization of structure/function of the nervous system in which an undamaged part of the brain takes over the function of the damaged portion has long been assumed to be a crucial component of the recovery process. In this chapter we review the evidence for such reorganization, the role it plays in the recovery of language following brain damage, and how knowledge of mechanisms of recovery can lead to more effective methods of rehabilitation. To orient readers outside of the specialty of aphasiology, we should note that different areas of brain damage (or dysfunction) generally lead to
How does reorganization fit into recovery? We have proposed that recovery of language after stroke occurs in three overlapping phases, each with a unique set of underlying neural mechanisms (Hillis and Breese, 2003; Hillis, 2005a). Acute recovery, which is primarily due to restoration of
Corresponding author. Tel.: +1 410 614 2381; Fax: +1 410 614 9807.; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57009-8
143
144
tissue function, occurs in the hours to days following brain damage. Subacute recovery, which is primarily due to neural reorganization, is a longer and more complex process. It follows the acute phase and lasts for several weeks to months during which new connections are formed and synaptic efficacy changes. Finally, chronic recovery consists of compensation and reorganization of cognitive function. This phase begins months to years after injuries such as stroke or other forms of injury to the brain and it may continue for the remainder of the person’s life. The acute phase Some aphasic patients see resolution of their deficits very quickly, during the acute stage of stroke. There are two potential explanations for this rapid recovery in the first hours to days after onset of the lesion. The first is assumed to be the result of expression of neural plasticity through which reorganization occurs as compensatory means of damaged or absent functions, and reestablishment of damaged connections. The second, and probably more common, mechanism involves restoration of the function of damaged neural tissue largely brought about by reperfusion of the dysfunctional area. In 1983, Skyhøj Olsen and colleagues first reported the existence of the ischemic penumbra, an area of the brain surrounding the ischemic infarct that remains viable for some time, but is dysfunctional because it is receiving low blood flow (Skyhøj Olsen et al., 1983). The ischemic penumbra can be estimated using two different forms of magnetic resonance imaging (MRI). Diffusionweighted imaging (DWI) shows the core of infarcted or densely ischemic tissue that is generally unsalvageable. Perfusion-weighted imaging (PWI) shows the entire area of hypoperfusion. Subtracting the area of the core infarct from the area of hypoperfusion represents the area of salvageable tissue that is dysfunctional due to low blood flow, but not yet dead. If blood flow is restored to the penumbra, it frequently recovers function. Several recent studies support the proposal that immediate recovery of language after stroke is more likely due to reperfusion of the ischemic
penumbra than due to reorganization. In 2002, we studied recovery of auditory comprehension in 18 patients following ischemic stroke. Participants underwent language testing and imaging within the first 24 h of their stroke, and again 3–5 days later. The study showed that those who had reperfused Brodmann’s Area 22 (posterior, superior temporal gyrus, commonly referred to as Wernicke’s area) had recovered, while those who had not reperfused Brodmann’s area 22 continued to have deficits (Hillis and Heidler, 2002). Several other studies have demonstrated that reperfusion of certain areas of the brain is associated with recovery of specific language functions (Hillis et al., 2002, 2003), and in each case improvement of the specific task immediately followed an intervention to restore blood flow (e.g., carotid stenting or blood pressure augmentation). These studies illustrate that reperfusion of the hypoperfused area is a critical component underlying acute recovery because the language functions recovered only if particular areas were reperfused. If reorganization were responsible for the recovery, one would expect to see at least some improvement in patients who showed no signs of reperfusion. It must also be taken into account that reperfusion in itself can cause damage to neural tissue. There is some evidence to support that early reorganization of motor or sensory functions may also occur through expression of neural plasticity and contribute to recovery of language tasks that require these functions. Single cell recording studies have demonstrated that immediate reorganization of spatial representations occurs in the somatosensory cortex of rats and primates following injury (Jenkins and Merzenich, 1987; Coq and Xerri, 1999). While these results indicate a role of reorganization in the acute phase of recovery, reorganization of sensory and motor functions likely plays a limited role in the recovery of language (although they may contribute to recovery of speech articulation in patients with Broca’s aphasia). Language, the representations and processes underlying meaning and syntax, relies on a more complex network of connections that probably takes more time to reorganize than do basic sensory or motor functions. Additionally, other studies have found that, even in the recovery of motor
145
function following stroke in hemiplegic patients, reperfusion appears to be the primary mechanism of recovery (Binkofski et al., 1996). The subacute phase In most cases, reperfusion can only salvage the ischemic penumbra for the first few days following ischemia. Eventually, the hypoperfused area often progresses to infarction in the absence of intervention. However, language recovery continues in the subacute phase, often at a rapid rate, throughout the following weeks to months after stroke. Different neural mechanisms must be responsible for this phase of recovery. Again, there are two possible components: recovery from diaschisis and reorganization. The term diaschisis was first proposed by von Monakow in 1914 to describe the loss of function in a portion of the brain that receives input, but is distant, from the site of injury (von Monakow, 1914). The theory is that the dysfunction is due to hypometabolism that occurs downstream in a neural network as a result of the loss of neural input from the damaged area. Recovery occurs when there is additional input from other, undamaged, areas of the brain, restoring the function. While it is difficult to obtain direct evidence, it is likely that recovery from diaschisis has a role in subacute recovery of language function. Alternatively, or additionally, there is a wealth of evidence suggesting that expression of neural plasticity causing reorganization of both structure and function plays a crucial role in subacute recovery following brain damage. In the following sections we will discuss in depth its role in the recovery process. The chronic phase Following the end of reorganization in the subacute phase, many patients are still able to improve their language skills. During the chronic phase (years after stroke), recovery of language is achieved by learning new ways to retrieve language representations and establishing compensatory strategies. This distinction can be illustrated by the chronic recovery of two patients. ‘‘Jeanette’’
suffered a stroke, and three years later was still unable to retrieve the written form of verbs, although her ability to retrieve the spoken form was intact. During rehabilitation, she learned to say the initial sound of the verb aloud and convert that sound into a letter (i.e., she relearned ‘‘phonics’’). Retrieval of the first letter was usually sufficient to access the written form (Hillis and Caramazza, 1995), indicating that she found a new ‘‘pathway’’ to previously stored language representations. ‘‘Holly,’’ in contrast, had intact written naming, but was severely impaired at retrieval of the spoken names of objects, eight years after a brain injury. During treatment, she learned to write the names of objects and then sound out what she had written. This compensatory strategy led to numerous regularization errors, for example mispronouncing the word steak as ‘‘steek,’’ indicating that she relied on the phonics (the learned compensatory strategy) rather than accessing a previously learned language representation (Hillis, 1991). Both these patients are examples of how recovery of language is still possible years after stroke, but must be achieved by finding new ways to access intact representations (in the case of Jeanette) or effective compensation strategies (in the case of Holly), rather than relying on reorganization. In summary, recovery of language occurs in three overlapping stages, each with different underlying mechanisms. Reorganization plays a crucial role in the subacute recovery phase, days to weeks after the onset of deficits. Too much time has passed for improvement to be due to reperfusion of the ischemic penumbra, yet the patient displays remarkable recovery of language function that likely represents new structure/function relationships. Below we review further evidence of reorganization, studies that provide insight into the brain areas involved, factors that influence the location and extent of reorganization, and the implications for treatment and rehabilitation.
Evidence for reorganization in aphasia We have known for over a century that, in normal, healthy individuals, language is predominantly
146
represented in the left hemisphere (Broca, 1865). Bilateral activation during verb generation in normal individuals undergoing functional MRI (fMRI) or positron emission tomography (PET) indicates that it is not uncommon for the right hemisphere to also become engaged during language tasks (Demonet et al., 1992; Weiller et al., 1995; Noppeney et al., 2005). However, the left hemisphere is dominant for language in 92% (Knecht et al., 2003) of right-handed individuals (and at least 50% of left-handed individuals). Furthermore, studies of the lesions associated with specific language deficits (Mohr et al., 1978; Alexander et al., 1987; Hart and Gordon, 1990; Raymer et al., 1997) have uncovered left hemisphere areas critical for performing various language functions. Despite the integral role of these areas in language, individuals with extensive damage to their left hemisphere are often able to recover quite well. As mentioned previously, reorganization is one of the underlying mechanisms responsible for this recovery. In 1874, Wernicke first postulated that regions in the contralateral cortex are responsible for the return of language following left hemisphere injury (Wernicke, 1874). There are also several sources of more recent evidence for hemispheric transfer. Patients with damage to most of their left hemisphere are often able to recover a substantial amount of language function. In these patients, the right hemisphere must assume the lost function (Basso et al., 1989; Willmes and Poeck, 1993; Hillis, 2005a). Moreover, children who have their entire left hemisphere removed for the treatment of intractable epilepsy are able to acquire/relearn language skills following surgery (Smith, 1966; Vargha-Khadem et al., 1997). How do we know that the improvement we see following damage or removal of the left hemisphere truly represents reorganization, and patients are not simply uncovering latent right hemisphere lexical processes that have been present all along? The first clue is the timing of recovery following injury. Whereas improvement that is due to reperfusion is immediate, recovery in the absence of reperfusion takes weeks to years, indicating a separate process. Presumably, it takes time for new connections to be made and
pathways formed. The patients previously described with recovery after large areas of damage exhibited initial deficits. This would not be expected if language function was intact in the right hemisphere from the very start. Instead, we would expect the right hemisphere to immediately assume function when the left hemisphere is damaged. There are multiple other, and more direct, sources of evidence for inter-hemispheric reorganization after stroke or other forms of damage to the brain, including surgery. Patients who recover language after large left hemisphere strokes sometimes show aphasia after a new right hemisphere stroke, indicating that language has crossed over to the right hemisphere (Nielson, 1946; Levine and Mohr, 1979; Basso et al., 1989). Most individuals do not display deficits in basic language tasks like naming and repetition following right hemisphere strokes, indicating that it does not normally play an essential role in these basic language tasks (although it has a role in understanding verbal humor, abstractions, and multiple meanings of words; Tompkins et al., 2002). Similarly, temporarily anesthetizing the right hemisphere (during Wada testing) of recovered aphasics can cause the recurrence of language deficits, indicating that language function had shifted to the right hemisphere. Wada testing involves injection of sodium amytal into one carotid artery. When injected in the left carotid artery in normal right-handed individuals, it disrupts language but not when injected into the right carotid artery (Kinsbourne, 1971; Czopf, 1972). Repetitive transcranial magnetic stimulation (rTMS) that suppresses the homologous right hemisphere language areas in aphasic stroke or tumor patients yields similar results (Martin et al., 2004; Thiel et al., 2005). In the latter case, we see that the left hemisphere is still important for language in these patients (individuals make an increased number of errors when rTMS is applied over left hemisphere language areas also). However, in these patients (but not in normal individuals) language function will cease, or become impaired, when rTMS is applied over the right hemisphere as well. This indicates a decreased lateralization of language, presumably due to reorganization. Since anesthetizing the right hemisphere of normal individuals does not result
147
in any language impairment, the results of Wada testing and rTMS results provide further evidence that improvement in stroke or tumor patients is due to new, rather than normal residual, function in the right hemisphere, and thus is an expression of neural plasticity. In light of all of this evidence, there is little doubt that reorganization occurs after brain injury. In the next section, we discuss evidence for reorganization from functional imaging studies, and how functional imaging can help determine where and when reorganization occurs after focal brain lesions.
The location of reorganization — evidence from functional imaging While it is clear that reorganization occurs, allowing at least partial recovery of language following injury, the areas that are responsible for assuming the lost function remain controversial. When fMRI and PET imaging became available, there was hope within the community that these techniques would reveal with certainty the areas that assume the language function for each damaged region of the brain. Unfortunately, results of numerous subsequent studies appear somewhat conflicting. Many functional imaging studies have shown activation in the homologous areas of the right hemisphere during language tasks in patients with lesions in the left hemisphere who have recovered from aphasia (Kertesz, 1989; Weiller et al., 1995; Ohyama et al., 1996; Cappa et al., 1997; Thulborn et al., 1999; Thompson et al., 2000; Weiller, 2000; Leff et al., 2002). For example, a study done by Weiller and colleagues in 1995 of six patients who suffered fluent aphasia after stroke and subsequently recovered showed increased activation in the right hemisphere homologs of Broca’s and Wernicke’s areas. Activation in these six patients was similar to that seen in normal left-handed individuals (Krams et al., 1994), providing additional evidence that these homologous cortical areas are capable of language function. However, many other studies report that a variety of frontal, temporal, and parietal regions surrounding lesions
in the left hemisphere assume function after damage to language areas (Karbe et al., 1995, 1998; Heiss et al., 1997, 1999; Cao et al., 1999; Warburton et al., 1999; Thiel et al., 2001; Perani et al., 2003). There are at least three plausible (and nonexclusive) explanations of this apparent contradiction. First, it has been proposed that both the right hemisphere and spared left hemisphere areas may contribute to recovery, but the location and extent of recovery depend on the extent of the left hemisphere damage, the duration of the injury, and which language functions are affected (Hillis, 2002, 2005b). These topics will be discussed in detail later in this chapter. A second possibility is that both the right and left hemispheres are always very much involved in the recovery process. Functional imaging in epilepsy patients after temporal lobectomy shows that language recovery hinges on both integrating the remaining working parts of the network in the damaged hemisphere and new activation in homologous right hemisphere areas (Noppeney et al., 2005). In fact, most studies acknowledge that there is increased activation in both the right hemisphere areas homologous to the lesion and left hemisphere areas that surround the lesion, although they attribute varying degrees of significance to these two forms of activation. A final theory is that language initially switches to the contralateral hemisphere following damage until the left hemisphere areas can be reintegrated into the language network. Heiss and colleagues (1997, 1999) reported activation in the nondominant hemisphere in patients 2 weeks after stroke, but reactivation of the left hemisphere speech areas surrounding the areas of infarction in those patients with a favorable long-term outcome. Several studies propose that those patients who are able to shift language function back over to the left hemisphere have the best chance of good recovery (Small et al., 1998; Heiss et al., 1999; Rosen et al., 2000; Marshall, 2003). While the explanations above can account for some of the controversy about functional imaging studies, other complexities make the location of reorganization ambiguous. One recent study showed that suppression of areas of right hemisphere activation using rTMS did not always
148
interfere with language (Naeser et al., 2005). It has also been reported that, in some cases, such rTMS actually facilitates language, raising the possibility that right hemisphere activation may actually be maladaptive and interfere with language recovery (Martin et al., 2004). These results raise an important question: what does right hemisphere activation actually represent? There are many plausible answers to this difficult question. Consider the following illustrative results. A recent study by Vandenbulcke and colleagues (2005) of 19 patients with primary progressive aphasia who underwent fMRI while performing semantic association tasks revealed increased activation in the right temporal lobe of patients when compared to controls, with a greater shift in laterality for patients with word comprehension deficits. The preferred explanation by the authors was that this activation represents reorganization of lexical processes to the right hemisphere. However, if their results reflected interhemispheric reorganization, the right hemisphere would have to be less capable of performing the semantic association task than the left, as those who appeared less lateralized to the left made more errors. Another possible explanation of their results is that the right hemisphere activation reflects a maladaptive strategy that involves engagement of the right hemisphere. This would explain the not as good as performance of individuals with increased right hemisphere activation, yet it is unlikely that the main mechanism underlying language errors in primary progressive aphasia would be increased activation of a maladaptive right hemisphere. The third possibility is that the right temporal lobe is actually normally involved in semantic tasks, to a lesser extent than the left, and that damage to the left hemisphere forces the patients to rely more heavily on this right hemisphere processing. If right hemisphere semantics is cruder, increased reliance on the right hemisphere would result in more errors, as well as increased right hemisphere activation on fMRI. This same critique can be made of all fMRI and PET studies. It is difficult to be certain that increased signs of activation from imaging studies reflect reorganization of structure and functional relationships, rather than increased reliance on normal right
hemisphere functions, or the release from inhibition in regions of the brain that are normally engaged but not essential for language. Alterations in hemodynamics after stroke further confound results of fMRI studies in patients with cerebrovascular disease. The results of functional imaging studies in participants with poststroke aphasia may be affected by vascular disease in the left hemisphere (Hillis et al., in press). The BOLD (Blood Oxygen Level Dependent) effect seen on fMRI depends on the brain’s hemodynamic response to neural activation that is disproportionate to the oxygen extraction fraction in the area. However, in patients with vascular disease, the reactivity of the vessels may be impaired, altering the vascular response to neural activation (Marshall, 2004; Rossini et al., 2004). This may cause there to be no BOLD effect, or even negative BOLD effect, even though the area is activated. Patients with stroke commonly have cerebrovascular disease. Hemodynamics is also altered in acute ischemia, head trauma, and large vessel cerebrovascular disease without infarct. Even longitudinal studies within individuals (e.g., done acutely after infarct and then repeated to determine what changes are correlated with recovery of function) may be confounded by changes in vascular reactivity over time. As previously stated, most studies acknowledge that in left hemisphere stroke there is increased activation in both right hemisphere areas and areas around the lesion in recovering aphasia. There is, however, debate over what the contralateral activation represents. Some authors (e.g., Rosen et al., 2000) have suggested that right hemisphere activation in left-sided lesions probably reflects the adoption of a new, less efficient strategy by the right hemisphere. This proposal fits with the theory that the right hemisphere may be capable of performing some, but not all language tasks. For example, we know that the right hemisphere has some ability to process semantic information; however, orthographic to phonologic conversion is most likely a left-hemisphere-only process. An alternative explanation is that there is a loss of mechanisms that normally regulate the right hemisphere’s level of activation, when the left hemisphere is damaged (Kinsbourne, 1977). This would
149
explain why right hemisphere activation in patients with left-sided lesions has no correlation to performance. Unfortunately, there have been conflicting results regarding correlation between language performance and level of activation as determined by imaging techniques (Belin et al., 1996; Karbe et al., 1998; Heiss et al., 1999; Rosen et al., 2000). Some authors have reported that performance does correlate with the degree of activation in the right hemisphere (Noppeney et al., 2005). Questions have also been raised as to what activation in areas around a lesion represents. Some propose that in small strokes, areas around a lesion correspond to areas that are normally active in language and activation is simply the restoration of normal function, not actual remapping (Nudo et al., 1996; Heiss et al., 1999). When areas around a lesion show increased activation over time, these left hemisphere areas may be taking on new functions or being reperfused and thereby restored to their normal function. Although it is widely accepted that reorganization occurs, there is clearly controversy regarding the nature and localization of the reorganized function. In an attempt to overcome some of the complexities described above, some groups have designed creative experiments or analyses. One study showed decremental activation in the right hemisphere with practice during a word retrieval task to demonstrate that contralateral regions are responsible for recovered language function (Blasi et al., 2002; Weiller, 2000 is another example). Similar decremental activation is seen in the left hemisphere language areas of normal individuals following practice. Other studies have paired PET or fMRI with either rTMS or the Wada test. The majority of these studies have found good, but not perfect correlation between results regarding hemisphere lateralization of language (Carpentier et al., 2001; Adcock et al., 2003). Unfortunately, determining the location of the underlying network that is critical for other language processes is far more complex than determining lateralization. However, multimodality imaging holds promise for this endeavor as well. In summary, when used in concert with techniques like rTMS, MEG, electrophysiology,
DWI/PWI lesion studies, and lesion-deficit correlations, fMRI and PET imaging can be valuable tools that have the potential to provide great insight into which neural substrates underlie language function. Both the contralateral right hemisphere regions and regions around the lesions of the left hemisphere have been implicated both acutely and chronically in recovery after brain injury. Some studies show not only that the best recovery ultimately involves the regained function of the left hemisphere, but also that the right can take over until the left can resume function.
Factors affecting the location and extent of reorganization Many factors may influence the anatomical location where language processing occurs after damage to the left hemisphere. It is therefore difficult to predict the exact locations in the brain that will be essential for recovery in an individual patient. Several important factors will be discussed below. Size and location of the lesion The size and the anatomical location of the initial lesion are two major factors believed to play a role in whether reorganization ultimately occurs in the right or left hemisphere. Many studies have concluded that language tends to shift to the right when there is extensive damage to the left hemisphere (Karbe et al., 1998; Cao et al., 1999). Others have gone one step further and proposed that the brain attempts to reintegrate left hemisphere language areas into the network, and that patients who are able to do so have fuller recoveries (Heiss et al., 1999). It is likely that the specialization of the areas surrounding the damaged tissue also plays a key role in whether language shifts to the side that is contralateral to the location of the lesion. If the tissue that surrounds the lesion plays an active role in language (for example, Broca’s or Wernicke’s area), then that tissue is more likely to assume lost language functions. When the tissue surrounding a lesion cannot become part of the new language network, language processing is likely to shift to the right hemisphere.
150
Time after onset of symptoms Time may also play a role in which areas become involved in language processing after injuries to areas normally involved in language. Single cell recording studies show that areas of somatosensory cortex begin to reorganize very quickly following damage, with adjacent regions taking over the function of the area that has been damaged. The same may occur in humans who have small lesions in sensory and motor cortices, and who regain full motor or sensory function after a couple of days, without detectable increase in cortical perfusion. However, such rapid reorganization appears not to occur for language, since only patients who showed reperfusion showed recovery of word comprehension a few days after the lesion occurred (Hillis and Heidler, 2002). As noted previously, there are several lines of evidence to suggest that language initially shifts to the right hemisphere, but often shifts back to areas around a lesion in the left hemisphere in patients who recover well. Therefore, the anatomical location of regained functions of damaged areas varies for different times after injury. Rate of damage In recent years, investigators have suggested that the rate with which damage occurs influences the anatomical location of the recovered language function. We have already discussed which areas in the brain are responsible for reorganized language after stroke (in which damage occurs rapidly). Patients with intractable epilepsy or with slow growing brain tumors provide an opportunity to study reorganization when damage occurs progressively over an extended period of time. In 2001, Thiel (Thiel et al., 2001, 2005) and colleagues studied a series of patients with brain tumors and found that patients with slow-growing tumors on the left side developed compensatory right hemisphere language mechanisms and recovered more completely than patients with rapidly growing tumors who were not able to shift language to the opposite hemisphere (Thiel et al., 2001, 2005). Patients with mesial temporal lobe epilepsy who underwent temporal lobectomy for treatment of
the epilepsy often had preoperative interhemispheric shift of language (Billingsley et al., 2001; Carpentier et al., 2001; Gaillard et al., 2002; Adcock et al., 2003), likely caused by chronic dysfunction of the left temporal lobe, even in the absence of detectable structural lesions. In some cases, left temporal lobectomy causes no language deficits, presumably because language has already shifted to the right hemisphere. This finding can also explain the excellent recovery observed in children who undergo left hemispherectomy for intractable epilepsy with a left hemisphere focus. In these children, language has presumably already shifted partially to the right hemisphere due to the damage caused by repeated epileptic seizures. Nature of the language function lost and recovered There are two final important considerations regarding the extent to which language shifts to the right in patients with injuries of the left hemisphere: (1) the specific language task that is affected in each individual patient (corresponding to the site of damage within the left hemisphere); and (2) how much of that function the patient is able to regain. In one study (Kertesz et al., 1979), the size of the lesion correlated positively with recovery of comprehension, yet it was inversely correlated with rates of improvement of word generation or sentence production. These findings lead to the hypothesis that the right hemisphere was able to compensate for word comprehension deficits but not for articulation or grammatical impairment in these patients (Kertesz et al., 1979; see Gainotti, 1993 as another example). Furthermore, another study showed that rightward shift of activation on fMRI correlated inversely with picture naming performance (Cao et al., 1999) and was associated with poorer recovery than ipsilateral activation (Karbe et al., 1998; Heiss et al., 1999), indicating that the left hemisphere is critical for naming. It is possible that the right hemisphere can accomplish a few language tasks, but not others, such as orthographic to phonologic conversion (Saffran et al., 1980). Studies of split brain patients indicate that phonologic codes are localized solely in the left hemisphere while orthographic and semantic are localized more bilaterally
151
(Zaidel, 1976). It is also commonly accepted that the right hemisphere is better at comprehension than at production. Thus, whether or not language shifts to the right hemisphere following brain damage is likely to depend on the normal engagement and capacity of the portion of the right hemisphere homologous to the lesion for a particular task. If the right hemisphere cannot assume a disrupted language function, the patient must rely instead on shifting responsibility more heavily to other portions of the left hemisphere network (Hillis, 2002).
Factors that facilitate reorganization Rehabilitation The role of cognitive rehabilitation in stimulating reorganization of brain/language relationships is controversial. By better understanding what occurs in the brain during reorganization, we will undoubtedly be better able to design rehabilitation programs that facilitate the process. Many of the techniques currently employed by speech language pathologists and other clinicians focus on treatment of the damaged component of the system, while others focus on increasing the utilization of spared components. One method aims at increasing the use of spared components of language through ‘‘stimulation therapy’’ (Duffy and Coelho, 2001). The patient becomes able to produce the desired response via a number of different facilitation approaches. Duffy and Coelho’s methods presumably help promote reorganization of language to areas of the brain that can assume the lost function. Better understanding of how and where reorganization occurs would also contribute to our understanding of the mechanisms underlying these rehabilitation techniques. The neural networks underlying various language tasks and their use often change dramatically following damage. We can determine which tasks are most likely to improve with treatment by knowing which cognitive processes are most likely to change following injury, what these changes are, and how they occur. Treatment may also be guided by better understanding the neural mechanisms underlying
reorganization and recovery. Understanding recovery on a molecular level might have important implications for deciding when to begin treatment, how often treatment should be given, and which medications might augment response to therapy. Recent studies have indicated that learning occurs in the brain by changing the strength of connections between neurons. This process is known as long-term potentiation (LTP) (see Chapter 3). Assuming that LTP underlies learning (and relearning, in the case of patients with brain damage), we can design treatment programs that induce this mechanism. Intense practice would be the most effective form of treatment (Hillis, 2005c). Proof of this concept is visible in everyday life. Children learn to play soccer faster and better after a week of intense training than a few months of weekly practices. Likewise, we would not expect someone taking weekly French classes for the past year to speak as fluently as someone who had just returned from a month in Paris, staying with native speakers of French. The same is true for patients recovering from aphasia. A patient improved dramatically while participating in daily 2 h rehabilitation sessions, but not when offered the same therapy only twice a week (Hillis, 1998). Medications Enhanced knowledge of the underlying mechanisms of expression of neural plasticity and a better understanding of how reorganization can be promoted would help us to design more effective rehabilitation programs following brain injury. The brain’s chemical milieu undoubtedly has a tremendous impact on synaptic plasticity, and that raises the possibility of manipulating reorganization with medications and thereby augmenting therapy. As discussed above, LTP plays an important role in reorganization of neural circuits, and we now know that LTP depends on the presence of multiple neurotransmitters. Kirkwood and colleagues have shown that norepinephrine and acetylcholine are both necessary for such synaptic plasticity to occur (Kirkwood et al., 1999). It has, therefore, been proposed that it would be possible to enhance plasticity and facilitate reorganization
152
by increasing the amount of these neural transmitters that are available in the synaptic cleft. The role of acetylcholine in cognitive functioning has already been extensively studied in patients with Alzheimer’s Disease. Acetylcholinesterase inhibitors are frequently given to slow deterioration in cognition. Several studies have found that acetylcholinesterase inhibitors may have a positive effect on recovery during language rehabilitation in aphasic patients as well (Goldstein, 1999, 2000a; Berthier et al., 2003). There is even stronger evidence to support the use of norepinephrine, a catecholamine released during periods of excitement, challenge, or in response to rewards. Its importance in learning (McGaugh and Roozendaal, 2002) helps to explain why feedback, or praise, has been shown to be effective during rehabilitation (Hillis, 2005c). Numerous studies have indicated that administering stimulant medications, in particular dextro-amphetamine, a noradrenergic agonist, to aphasic patients undergoing speech therapy results in an increased improvement in language function compared to therapy alone (Walker-Batson et al., 1991, 2000, 2001). Dextro-amphetamine has been widely studied as a pharmacotheraputic agent in the treatment of aphasia. It has nonspecific effects on the transmission of catecholamines, affecting not only norepinephrine, but dopamine as well (see Chapter 20). This has led researchers to examine the role of dopamine in recovery of language. Dopamine is also involved in the reward pathway, and it has been suggested that bromocriptine, a dopamine agonist, may stimulate LTP when combined with behavioral therapy. Unfortunately, studies have yielded mixed results. One study involving seven patients with nonfluent aphasia showed increased improvement in language function as doses were increased. The improvement deteriorated when the medication was decreased (Sabe et al., 1992), supporting dopamine’s importance. Other studies have shown limited or no benefit (MacLennan et al., 1991; Gupta and Milcoch, 1992). The fact that antidepressants, which affect the availability of catecholemines and/or serotonin within the synaptic cleft, have been shown to be effective in language recovery when combined with behavioral therapy appears to support the beneficial effect of
these medications (Gillen et al., 2001). However, it is unclear whether these drugs exert their effect by decreasing depression or increasing motivation or enhancing LTP (Gillen et al., 2001; Hillis, 2005c). Wilson (1997) has shown that a patient’s emotional state has a tremendous influence on the effectiveness of their rehabilitation. Whether it is the impact of dopamine on mood, or the impact of dopamine on synaptic plasticity remains unclear. Considering that mood is intricately related to a person’s cognitive ability, it is likely that dopamine may prove useful regardless of the exact mechanism. A medication with a rather different action, piracetam, has been reported to aid in recovery of language. Piracetam is believed to work by improving cerebral blood flow or by neuroprotection, and it has been proclaimed to enhance cognitive functions. Piracetam has been used both acutely and chronically after stroke to improve function and/or response to rehabilitation. Several studies have shown some benefit, but others have reported little or no effect on aphasia (Enderby et al., 1994; De Deyn et al., 1997; Coq and Xerri, 1999; Orgogozo, 1999; Kessler et al., 2000). Since catecholamines and acetylcholine are involved in LTP, it would make sense that medications that block norepinephrine, acetylcholine, and/or dopamine could impede reorganization and have a negative effect on recovery; this hypothesis has been supported by several studies (Goldstein, 1995, 1998, 2000b; Small, 2002). This finding has important implications in the treatment of patients following stroke because many drugs that are likely to have ill effects are frequently prescribed to stroke patients, especially during hospitalization. These medications include clonidine, prazosin, and antipsychotics (e.g., haloperidol). The evidence against the use of these drugs is not yet strong enough to prohibit their use when patients are undergoing rehabilitation, but the increasing evidence against these drugs may ultimately influence our selection of medications for recovering stroke patients (Goldstein and the Syngen in Acute Stroke Study Investigators, (1995)). It was pointed out earlier in this chapter that the acute phase of recovery following stroke depends
153
on the restoration of blood flow to dysfunctional, hypoperfused areas. During this phase, medications that promote reperfusion may be most effective in facilitating recovery. However, during the subacute phase (reorganization), there is mounting evidence that, when combined with cognitive rehabilitation, the use of small molecules to alter brain chemistry may have a significant effect on the brain’s synaptic plasticity, as described above. It is important to note, however, that no medication has been shown to be sufficiently effective to become standard of care following stroke. Nevertheless, there are several promising candidate medications that may augment rehabilitation (WalkerBatson, 2000; Walker-Batson et al., 1991, 2001).
lead to improved recovery following damage to one hemisphere (Frackowiak et al., 1997). Other studies, however, have not found that gender affects the laterality of the representation of language (Adcock et al., 2003). Finally, the size of the lesion and length of time over which the damage is inflicted (both discussed previously in this chapter) may also play a role in the degree of reorganization. Clinicians and researchers are still looking for the answers to these questions. Fortunately, there remains great interest in the field. With new technologies and methodologies for investigating brain/language relationships becoming available, answers may come soon.
References Unanswered questions and areas of future research It is currently an exciting era in the field of neurology. By better understanding the ways in which the brain recovers and is able to reorganize following injury, it is possible to begin designing better methods of treatment for our patients. There is accumulating evidence that rehabilitation can be effectively augmented with medications that increase synaptic plasticity. Nevertheless, many questions remain. There is a need for identification of all the factors that influence when and where in the brain reorganization occurs. One of the most important and perplexing issues is why some people recover faster and more completely than others. It is well known that different individuals have different amounts of reserves, and many studies have shown that the general ability of the brain to reorganize as a result of expression of neural plasticity decreases with age. While language recovery/acquisition has been documented in children up to 9 years of age following left-sided hemispherectomy (Vargha-Khadem et al., 1997), there is evidence for a substantial decrease in neural plasticity following puberty (Adcock et al., 2003). Some studies find that gender also plays a role. Some studies have found more bilateral activation when performing language tasks in women compared to men (Shaywitz et al., 1995; Pugh et al., 1996). This normal engagement of both hemispheres during language tasks might
Adcock, J.E., Wise, R.G., Oxbury, J.M. and Matthews, P.M. (2003) Quantitative fMRI assessment of the differences in lateralization of language-related brain activation in patients with temporal lobe epilepsy. Neuroimage, 18: 423–438. Alexander, M.P., Naeser, M.A. and Palumbo, C.L. (1987) Correlations of subcortical CT lesion sites and aphasia profiles. Brain, 110: 961–991. Basso, A., Gardelli, M., Grassi, M.P. and Mariotti, M. (1989) The role of the right hemisphere in recovery from aphasia. Two case studies. Cortex, 25: 555–566. Belin, P., van Eeckhout, P., Zilbovicius, M., Remy, P., Francois, C., Guillaume, S., Chain, F., Rancurel, G. and Samson, Y. (1996) Recovery from nonfluent aphasia after melodic intonation therapy: a PET study. Neurology, 47: 1504–1511. Berthier, M.L., Hinajosa, J., del Carmen Martin, M. and Fernandez, I. (2003) Open-label study of donepezil in chronic poststroke aphasia. Neurology, 60: 1218–1219. Billingsley, R.L., McAndrews, M.P., Crawley, A.P. and Mikulis, D.J. (2001) Functional MRI of phonological and semantic processing in temporal lobe epilepsy. Brain, 124: 1218–1227. Binkofski, F., Seitz, R., Arnold, S., Claben, J., Benecke, R. and Freund, H.J. (1996) Thalamic hypometabolism and corticospinal tract integrity determine motor recovery in stroke. Ann. Neurol., 39: 460–470. Blasi, V., Young, A.C., Tansy, A.P., Petersen, S.E., Snyder, A.Z. and Corbetta, M. (2002) Word retrieval learning modulates right frontal cortex in patients with left frontal damage. Neuron, 36: 159–170. Broca, P. (1865) Sur la faculte du langage articule. Paris Bull. Soc. Anthr,, 6: 337–393. Cao, Y., Vikingstad, E.M., George, K.P., Johnson, A.F. and Welch, K.M. (1999) Cortical language activation in stroke patients recovering from aphasia with functional MRI. Stroke, 30: 2331–2340.
154 Cappa, S.F., Perani, D., Grassi, F., Bressi, S., Alberoni, M., Franceschi, M., Bettinardi, V., Todde, S. and Fazio, F. (1997) A PET follow-up study of recovery after stroke in acute aphasics. Brain and Language, 56: 55–67. Carpentier, A., Pugh, K.R., Westerveld, M., Studholme, C., Skrinjar, O., Thompson, J.L., Spencer, D.D. and Constable, R.T. (2001) Functional MRI of language processing: dependence on input modality and temporal lobe epilepsy. Epilepsia, 42: 1241–1254. Coq, J. and Xerri, C. (1999) Acute reorganization of forepaw representation in the rat SI cortex after focal injury: neuroprotective effects of piracetam treatment. Euro. J. Neurosci., 11: 2597–2608. Czopf, J. (1972) Role of the non-dominant hemisphere in the restitution of speech in aphasia. Arciv fur Psychiatrie und Nervenkrankheiten, 216: 162–171. De Deyn, P.P., Reuck, J.D., Deberdt, W., Vlietinck, R., Orgogozo, J.M. and Members of the Piracetam in Acute Stroke Study (PASS) Group., (1997). Treatment of acute ischemic stroke with piracetam. Stroke, 28, 2347–2352. Demonet, J.F., Chollet, F., Ramsey, S., Cardebat, D., Nespoulous, J.L., Wise, R., Rascol, A. and Frackowiack, R. (1992) The anatomy of phonological and semantic processing in normal subjects. Brain, 115: 1753–1768. Duffy, R.J. and Coelho, C.A. (2001) Schuell’s stimulation approach to rehabilitation. In: Chapey, R. (Ed.), Language Intervention Strategies in Aphasia and Related Neurogenic Communication Disorders (4thEd.). Williams and Wilkens, Baltimore, MD. Enderby, P., Broeckx, J., Hospers, W., Schildermans, F. and Deberdt, W. (1994) Effect of piracetam on recovery and rehabilitation after stroke: a double-blind, placebo-controlled study. Clin. Neuropharmacol, 17: 320–331. Frackowiak, R.S., Friston, K.J., Frith, C.D., Dolan, R.J. and Mazziotta, J.C. (1997) The cerebral basis of functional recovery. In: Frackowiack, R.S.J. (Ed.), Human Brain Function. Academic Press, San Diego, CA, pp. 275–299. Gaillard, W.D., Balsamo, L., Xu, B., Grandin, C.B., Braniecki, S.H., Papero, P.H., Weinstein, S., Conry, J., Pearl, P.L., Sachs, B., Sato, S., Jabbari, B., Vezina, L.G., Frattali, C. and Theodore, W.H. (2002) Language dominance in partial epilepsy patients identified with an fMRI reading task. Neurology, 59: 256–265. Gainotti, G. (1993) The riddle of the right hemisphere’s contribution to the recovery of language. Euro. J. Discord Commun., 28: 227–246. Gillen, R., Tennen, H., Mckee, T.E., Gernert-Dott, P. and Affleck, G. (2001) Depressive symptoms and history of depression predict rehabilitation efficiency in stroke patients. Arch. Phys. Med. Rehab, 82: 1645–1649. Goldstein, L.B. (1998) Potential effects of common drugs on stroke recovery. Arch.Neurol, 55: 454–456. Goldstein, L.B. (1999) Pharmacological approach to functional reorganization: the role of norepinephrine. Rev. Neurol. (Paris), 155: 731–736. Goldstein, L.B. (2000a) Effects of amphetamines and small related molecules on recovery after stroke in animals and man. Neuropharmacology, 39: 852–859.
Goldstein, L.B. (2000b) Should antihypertensive therapies be given to patients with acute ischemic stroke? Drug Safety, 22: 13–18. Goldstein, L.B. and the Syngen in Acute Stroke Study Investigators (1995) Common drugs may influence motor recovery after stroke. Neurology, 45: 865–871. Gupta, S.R. and Milcoch, A.G. (1992) Bromocriptine treatment of nonfluent aphasia. Arch. Phys Med Rehab, 73: 373–376. Hart, J. and Gordon, B. (1990) Delineation of single-word semantic comprehension deficits in aphasia, with anatomical correlation. Ann. Neurol., 27: 226–231. Heiss, W.D., Karbe, H., Weber-Luxenburger, G., Herholz, K., Kessler, J., Pietrzyk, U. and Pawlik, G. (1997) Speech-induced cerebral metabolic activation reflects recovery from aphasia. J. Neurol. Sci., 145: 213–217. Heiss, W.D., Kessler, J., Thiel, A., Ghaemi, M. and Karbe, H. (1999) Differential capacity of left and right hemisphere areas for compensation of poststroke aphasia. Ann. Neurol., 45: 430–438. Hillis, A.E. (1991) Effects of separate treatments for distinct impairments within the naming process. In: Prescott, T. (Ed.) Clinical Aphasiology, Vol. 19. Pro-Ed, Austin, TX, pp. 255–265. Hillis, A.E. (1998) Treatment of naming disorders: new issues regarding old therapies. J. Int. Neuropsychol. Soc., 4: 648–660. Hillis, A.E. (2002) Does the right make it right? Questions about recovery of language after stroke. Ann. Neurol., 51: 537–538. Hillis, A.E. (2005a) Stages and mechanisms of recovery from aphasia. Jpn J. Neuropsychol., 21: 35–43 (published in English and Japanese). Hillis, A.E. (2005b) Can shift to the right be a good thing? Ann. Neurol, 58: 346. Hillis, A.E. (2005c) For a theory of rehabilitation: progress in the decade of the brain. In: Halligan, P. and Wade, D. (Eds.), Effectiveness of Rehabilitation of Cognitive Deficits. Oxford University Press, Oxford, pp. 271–280. Hillis, A.E. and Breese, E.L. (2003) Aphasia recovery: stages and mechanisms. In: Freddi, A. (Ed.), Stroke Today. San Raffaele Termini, Rome, pp. 288–297. Hillis, A.E. and Caramazza, A. (1995) I know it but I can’t write it: selective deficits in long and short-term memory. In: Campbell, R. (Ed.), Broken Memories: Neuropsychological Case Studies. Blackwell, London, pp. 344–365. Hillis, A.E. and Heidler, J. (2002) Mechanisms of early aphasia recovery: evidence from MR perfusion imaging. Aphasiology, 16: 885–896. Hillis, A.E., Kane, A., Tuffiash, E., Ulatowski, J.A., Barker, P., Beauchamp, N., Oh, S. and Wityk, R. (2002) Reperfusion of specific brain regions by raising blood pressure restores selective language functions in subacute stroke. Brain Lang., 79: 495–510. Hillis, A.E., Wityk, R., Barker, P.B. and Caramazza, A. (2003) Neural regions essential for writing verbs. Nat. Neurosci., 6: 19–20.
155 Jenkins, W.M. and Merzenich, M.M. (1987) Reorganization of neocortical representations after brain injury: a neurophysiological model of the bases of recovery from stroke. Prog. Brain Res., 71: 249–266. Karbe, H., Kessler, J., Herholz, K., Fink, G.R. and Heiss, W.D. (1995) Long-term prognosis of poststroke aphasia studied with positron emission tomography. Arch. Neurol., 52: 186–190. Karbe, H., Thiel, A., Weber-Luxenburger, G., Herholz, K., Kessler, J. and Heiss, W.D. (1998) Brain plasticity in poststroke aphasia: what is the contribution of the right hemisphere? Brain and Lang, 64: 215–230. Kertesz, A. (1989) Recovery of language disorders: homologous contralateral or connected ipsilateral compensation? Brain Inj. Recovery, pp. 27–43. Kertesz, A., Harlock, W. and Coates, R. (1979) Computer tomographic localization, lesion size and prognosis in aphasia and nonverbal impairment. Brain Lang, 8: 34–50. Kessler, J., Thiel, A., Karbe, H. and Heiss, W.D. (2000) Piracetam improves activated blood flow and facilitates rehabilitation of poststroke aphasic patients. Stroke, 31: 2112–2116. Kinsbourne, M. (1971) The minor cerebral hemisphere as a source of aphasic speech. Arch. Neurol., 25: 302–306. Kinsbourne, M. (1977) Hemi-neglect and hemisphere rivalry. In: Weinstein, E.A. and Friedland, R.L. (Eds.), Hemi-inattention and hemispheric specialization. Raven Press, New York, pp. 41–52. Kirkwood, A., Rozas, C., Kirkwood, J., Perez, F. and Bear, M. (1999) Modulation of long-term synaptic depression in visual cortex by acetylcholine and norepinephrine. J. Neurosci., 19: 1599–1609. Knecht, S., Jansen, A., Frank, A., van Randenborgh, J., Sommer, J., Kanowski, M. and Heinze, H.J. (2003) How atypical is atypical language dominance? Neuroimage, 18: 917–927. Krams, M., Weiller, C., Rijntjes, M., Mueller, S.P. and Diener, H.C. (1994) Speech representation and handedness in normal subjects: a positron emission tomography study. Neurology, 44(Suppl. 2): A232 (Abstract). Leff, A., Crinion, J., Scott, S., Turkheimer, F., Howard, D. and Wise, R. (2002) A physiological change in homotopic cortex following left posterior temporal lobe infarction. Ann. Neurol., 51: 553–558. Levine, D.M. and Mohr, J.P. (1979) Language after bilateral cerebral infarctions: role of the minor hemisphere. Neurology, 29: 927–938. MacLennan, D.L., Nicholas, L.E., Morley, G.K. and Brookshire, R.H. (1991) The effects of bromocriptine on speech and language function in a man with transcortical motor aphasia. In: Prescott, T. (Ed.) Clinical Aphasiology, Vol. 21. Pro-ed, Austin, TX. Marshall, R.S. (2003) Neuroimaging correlates of stroke recovery. Paper presented at the 28th International Stroke Meeting, Phoenix, AZ. Marshall, R.S. (2004) The functional relevance of cerebral hemodynamics: why blood flow matters to the injured recovering brain. Curr. Opin. Neurol., 17: 705–709.
Martin, P.I., Naeser, M.A., Theoret, H., Tormos, J.M., Nicholas, M., Kurland, J., Fregni, F., Seekins, H., Doron, K. and Pascual-Leone, A. (2004) Transcranial magnetic stimulation as complementary treatment for aphasia. Semin. Speech Lang, 25: 181–191. McGaugh, J.L. and Roozendaal, B. (2002) Role of adrenal stress hormones in forming lasting memories in the brain. Curr. Opin. Neurobiol., 12(2): 205–210. Mohr, J., Pessin, M., Finkelstein, S., Funkenstein, H., Duncan, G. and Davis, K. (1978) Broca’s Aphasia: pathological and clinical. Neurology, 28: 311–324. Naeser, M.A., Martin, P.I., Nicholas, M., Baker, E.H., Seekins, H., Kobayashi, M., Theoret, H., Fregni, F., Maria-Tormos, J., Kurland, J., Doron, K.W. and Pascual-Leone, A. (2005) Improved picture naming in chronic aphasia after TMS to part of right Broca’s area: an open protocol study. Brain Lang, 93: 95–105. Nielson, J.M. (1946) Agnosia, Apraxia, and Aphasia. Hoeber, New York. Noppeney, U., Price, C.J., Duncan, J.S. and Koepp, M.J. (2005) Reading after left anterior temporal lobe resection: an fMRI study. Brain, 128: 1377–1385. Nudo, R.J., Wise, B.M., SiFuentes, F. and Milliken, G.W. (1996) Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science, 272: 1791–1794. Ohyama, M., Senda, M., Kitamura, S., Ishii, K., Mishina, M. and Terashi, A. (1996) Role of the nondominant hemisphere and undamaged area during word repetition in poststroke aphasics. A PET activation study. Stroke, 27: 897–903. Orgogozo, J.M. (1999) Piracetam in the treatment of acute stroke. Pharmacopsychiatry, 32(Suppl 1): 25–32. Perani, D., Cappa, S.F., Tettamanti, M., Rosa, M., Scifo, P., Miozzo, A., Basso, A. and Fazio, F. (2003) An fMRI study of word retrieval in aphasia. Brain and Lang., 85: 357–368. Pugh, K.R., Shaywitz, B.A., Shaywitz, S.E., Constable, R.T., Skudlarski, P., Fulbright, R.K., Bronen, R.A., Shankweiler, D.P., Katz, L., Fletcher, J.M. and Gore, J.C. (1996) Cerebral organization of component processes in reading. Brain, 119: 1221–1238. Raymer, A., Foundas, A.L., Maher, L.M., Greenwald, M.L., Morris, M., Rothi, L.G. and Heilman, K.M. (1997) Cognitive neuropsychological analysis and neuroanatomical correlates in a case of acute anomia. Brain Lang, 58: 137–156. Rosen, H.J., Petersen, S.E., Linenweber, M.R., Snyder, A.Z., White, D.A., Chapman, L., Dromerick, A.W., Fiez, J.A. and Corbetta, M. (2000) Neural correlates of recovery from aphasia after damage to left inferior frontal cortex. Neurology, 55: 1883–1894. Rossini, P.M., Altamura, C., Ferretti, A., Vernieri, F., Zappasodi, F., Caulo, M., Pizzella, V., Del Gratta, C., Romani, G.L. and Tecchio, F. (2004) Does cerebrovascular disease affect the coupling between neuronal activity and local hemodynamics? Brain, 127: 99–110. Sabe, L., Leiguarda, R. and Starkstein, S.E. (1992) An openlabel trial of bromocriptine in nonfluent aphasia. Neurology, 42: 1637–1638.
156 Saffran, E.M., Bogyo, L.C., Schwartz, M. and Marin, O.S.M. (1980) Does deep dyslexia reflect right hemisphere reading? In: Coltheart, M., Patterson, K. and Marshall, J. (Eds.), Deep Dyslexia. Routeledge and Kegan Paul, London, pp. 381–406. Shaywitz, B.A., Shaywitz, S.E., Pugh, K.R., Constable, R.T., Skudlarski, P., Fulbright, R.K., Bronen, R.A., Fletcher, J.M., Shankweiler, D.P., Katz, L. and Gore, J.C. (1995) Sex differences in the functional organization of the brain for language. Nature, 6515: 607–609. Skyhøj Olsen, T., Larsen, B., Herning, M., Skriver, E.B. and Lassen, N. (1983) Blood flow and vascular reactivity in collaterally perfused brain tissue. Stroke, 14: 332–401. Small, S.L. (2002) Biological approaches to the treatment of aphasia. In: Hillis, A.E. (Ed.), Handbook of Adult Language Disorders: Integration of cognitive neuropsychology, neurology, and rehabilitation. Psychology Press, New York. Small, S.L., Flores, D.K. and Noll, D.C. (1998) Different neural circuits subserve reading before and after therapy for acquired dyslexia. Brain Lang, 62: 298–308. Smith, A. (1966) Speech and other functions after left (dominant) hemispherectomy. J Neurol. Neurosurg. Psychiatry, 29: 467–471. Thiel, A., Habedank, B., Winhuisen, L., Herholz, K., Kessler, J., Haupt, W.F. and Heiss, W.D. (2005) Essential language function of the right hemisphere in brain tumor patients. Ann. Neurol., 57: 128–131. Thiel, A., Herholz, K., Koyuncu, A., Ghaemi, M., Kracht, L.W., Habedank, B. and Heiss, W.D. (2001) Plasticity of language networks in patients with brain tumors: a positron emission tomography activation study. Ann. Neurol., 50: 620–629. Thompson, C.K., Fix, S.C., Gitelman, D.R., Parrsih, T.B. and Mesulam, M.M. (2000) fMRI studies of agrammatic sentence comprehension before and after treatment. Brain Lang., 74: 387–391. Thulborn, K.R., Carpenter, P.A. and Just, M.A. (1999) Plasticity of language-related brain function during recovery from stroke. Stroke, 30: 749–754. Tompkins, C.A., Fassbinder, W., Lehman-Blake, M.J. and Baumgaertner, A. (2002) The nature and implications of right hemisphere language disorders: issues in search of answers. In: Hillis, A.E. (Ed.), Handbook of Adult Language Disorders: Integration of cognitive neuropsychology, neurology, and rehabilitation. Psychology Press, New York, pp. 429–448.
Vandenbulcke, M., Peeters, R., Van Hecke, P. and Vandenberghe, R. (2005) Anterior temporal laterality in primary progressive aphasia shifts to the right. Ann. Neurol., 58: 362–370. Vargha-Khadem, F., Carr, L.J., Isaacs, E., Brett, E., Adams, C. and Mishkin, M. (1997) Onset of speech after left hemispherectomy in a nine-year-old boy. Brain, 120: 159–182. von Monakow, C. (1914) Die Lokalisation im Grosshirn und der Abbau der Funktionen durch corticale Herde. Bergmann, Wiesbaden. Walker-Batson, D. (2000) Use of pharmacotherapy in the treatment of aphasia. Brain Lang., 71: 252–254. Walker-Batson, D., Curtis, S., Natarajan, R., Ford, J., Dronkers, N., Salmeron, E., Lai, J. and Unwin, D.H. (2001) A double-blind placebo-controlled use of amphetamine in the treatment of aphasia. Stroke, 32: 2093–2098. Walker-Batson, D., Devous, M.D., Curtis, S., Unwin, D.H. and Greenlee, R.G. (1991) Amphetamine paired with physical therapy accelerates motor recovery after stroke. Further evidence. Stroke, 26: 2254–2259. Warburton, E., Swinburn, K., Price, C.J. and Wise, R.J. (1999) Mechanisms of recovery from aphasia: evidence from positron emission tomographic studies. J. Neurol. Neurosurg. Psychiatry, 66: 155–161. Weiller, C. (2000) Brain imaging in recovery from aphasia. Brain Lang., 74: 385–387. Weiller, C., Isensee, C., Rijntjes, M., Huber, W., Muller, S., Bier, D., Dutschka, K., Woods, R.P., Noth, J. and Diener, H.C. (1995) Recovery from Wernicke’s aphasia: a positron emission tomographic study. Ann. Neurol., 37: 723–732. Wernicke, K. (1874) Der aphasische symptomkomplex. Kohn and Neigat, Breslau. Willmes, K. and Poeck, K. (1993) To what extent can aphasic syndrome be localized? Brain, 116: 1527–1540. Wilson, B.A. (1997) Cognitive rehabilitation: how it is and might be. J. Internat. Neuropsychol. Soc., 3: 487–496. Zaidel, E. (1976) Auditory vocabulary of the right hemisphere following brain bisection or hemidecortication. Cortex, 12: 191–211.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 10
Relocation of specific visual functions following damage of mature posterior parietal cortex Stephen G. Lomber1,2,, S. Keun Yi1 and Erin M. Woller1 1
School of Behavioral and Brain Sciences, The University of Texas at Dallas, Richardson, TX 75080, USA 2 Centre for Brain and Mind, University of Western Ontario, London, ON N6A 5K8, Canada
Abstract: Many visual deficits have been reported following damage to specific cerebral sites within posterior parietal cortex. These deficits generally involve aspects of vision including spatial or motion perception and visuomotor control. One characteristic of many of these deficits is that they tend to attenuate over time. Presumably, other cortical regions possess adaptive neuroplastic mechanisms that allow them to accommodate functions that were previously dominated by the damaged region. This report summarizes a series of experiments that examined adaptive cortical plasticity following cerebral cortex damage sustained in maturity. Following bilateral lesions of posterior middle suprasylvian sulcal (pMSs) cortex in the cat, deficits were identified in both visual orienting and landmark discrimination tasks. However, the deficits on the visual orienting task were only profound for the first few days following the lesion and orienting abilities returned to normal levels within the first 2 weeks postlesion. In contrast, no such attenuation of the effect of the lesion was evident on the landmark discrimination task. Following recovery of function on the visual orienting task, individual cortical areas flanking the lesion were bilaterally deactivated with cooling. Reversible deactivation of anterior middle suprasylvian sulcal (aMSs) cortex, but none of the other adjacent cortices, yielded visual orienting deficits that are not found in intact animals during deactivation of aMSs cortex. Therefore, we concluded that the visual orienting functions normally mediated by pMSs cortex were able to relocate to aMSs cortex following lesion of pMSs cortex. Finally, bilateral lesion of both pMSs and aMSs cortices yielded visual orienting deficits that did not attenuate. Overall, this series of experiments demonstrates that certain visual functions may relocate to specific cortical loci following damage to discrete areas within posterior parietal cortex. Keywords: adaptive neuroplasticity; cortical plasticity; deficit attenuation; recovery of function; cat 1989, 1994; Yamasaki and Wurtz, 1991; Zeki, 1991; Pasternak and Merigan, 1994; Watson et al., 1994; Beckers and Zeki, 1995; Heilman et al., 1995, 2000; Rudolph and Pasternak, 1999; Lomber, 2001). These deficits generally involve aspects of vision including spatial or motion perception and visuomotor control. Cortex lining the posterior middle suprasylvian sulcus (pMSs; Fig. 1), a portion of the posterior parietal cortex of the cat, is perhaps the most extensively studied region of cat extrastriate visual cortex. The pMSs cortex has
Introduction Lesions of specific cerebral sites within posterior parietal cortex of the cat, monkey, or human often result in deficits of visual perception and cognition (e.g., Mesulam, 1981; Zihl et al., 1983; Newsome et al., 1985; Luh et al., 1986; Newsome and Pare´, 1988; Marshall and Halligan, 1988, 1995; Vaina, Corresponding author. Tel.: 519-663-5777; Fax: 519-663-3193; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57010-4
157
158
Fig. 1. Dorsal view of the cat cerebral cortex showing the positions of the regions of extrastriate visual cortex that were either permanently or reversibly deactivated. For all studies described, all lesions or cooling deactivations were performed bilaterally. Regions studied include: the posterior middle suprasylvian sulcus (pMSs), the anterior middle suprasylvian sulcus (aMSs), the middle suprasylvian gyrus (MSg), the middle ectosylvian gyrus (MEg), and the dorsal posterior suprasylvian gyrus (dPSg). Top is anterior. Illustration adapted from the drawings of Reinoso-Sua´rez (1961). In subsequent figures, small versions of this schematic are used to indicate when a site was lesioned (permanently deactivated; shown in black with saw-tooth edges) or cooled (temporarily deactivated; shown in gray with smooth edges).
been identified to play a critical role in the processing of motion signals (Pasternak et al., 1989; Rudolph and Pasternak, 1996; Huxlin and Pasternak, 2004), in accurate visual orienting (Hardy and Stein, 1988; Lomber et al., 1996; Payne et al., 1996; Lomber and Payne, 2000b), spatial discriminations (Lomber and Payne, 2001), proficient optokinetic nystagmus (Ventre, 1985; Tusa et al., 1989), depth perception (Kru¨ger et al., 1993), and in visuomotor integration (Sherk and Fowler, 2002). However, one additional characteristic of many of these deficits is that, regardless of the species examined, they tend to attenuate over time
(Hyva¨rinen, 1982; Luh et al., 1986; Newsome and Pare´, 1988; Lynch and McLaren, 1989; Yamasaki and Wurtz, 1991; Pasternak and Merigan, 1994; Huxlin and Pasternak, 2004). It is often proposed that other cortical regions possess adaptive neuroplastic mechanisms that allow them to accommodate functions that were previously dominated by the damaged region. In order to test this hypothesis, we examined two functions that had been previously found to be mediated by pMSs cortex: visual orienting (Lomber and Payne, 2001) and landmark spatial discriminations (Lomber and Payne, 2000a). First, we used bilateral reversible cooling deactivation to examine the contributions of the pMSs, the anterior middle suprasylvian sulcus (aMSs), the middle suprasylvian gyrus (MSg), and the middle ectosylvian gyrus (MEg) to visual orienting and landmark discriminations (Fig. 1). The results from this study demonstrated that only pMSs cortex makes significant contributions to these two tasks. In the second experiment, we bilaterally lesioned pMSs cortex with ibotenic acid and tested the animals daily on both tasks. Overall, while the deficits on the landmark discrimination task remained stable over time, the deficits on the visual orienting task were only profound for the first few days following the lesion and recovered over the first 2 weeks postlesion. Third, following recovery of function on the visual orienting task, individual cooling loops in contact with aMSs cortex, MSg cortex, dorsal posterior suprasylvian gyrus (dPSg) cortex, and MEg cortex (Fig. 1) were bilaterally deactivated to assess if visual orienting functions had relocated to any of these areas. Reversible deactivation of aMSs cortex, but none of the other adjacent cortices, yielded visual orienting deficits that are not found in intact animals during deactivation of aMSs cortex. In the final experiment, in one group of cats aMSs cortex was lesioned and in another group of cats both aMSs and pMSs were lesioned. In the aMSs lesion group, no deficits were detectable on either of the two tasks examined. In the aMSs and pMSs lesion group, profound and stable deficits were identified on both the landmark discrimination and visual orienting tasks. Therefore, we concluded that the presence of aMSs cortex is both necessary and sufficient for
159
the recovery of normal visual orienting functions following permanent inactivation of pMSs cortex.
commercial cat food and their diet was supplemented with dry cat food (Purinas cat chow).
Protocols and procedures
Visual orienting A white semicircular arena (90 cm in diameter; Fig. 2A) was used to test the cat’s ability to redirect its attention toward a visual stimulus. The arena had a 60 cm high perimeter wall with thirteen 2 V (DC) light-emitting diode (LED) spaced 151 apart with respect to the arena center and 27 cm above the arena floor (Fig. 2A). Immediately below each LED was a 3.5 cm diameter opening for the delivery of food rewards. The overall illumination was set at a low photopic level. Each cat was first trained to stand at the center of the apparatus and attend to the 01 LED, and then to detect and orient toward an illuminated peripheral LED. Cats were rewarded with moist or dry cat food depending on whether they approached the peripheral stimulus or the central position, respectively. The rapid and accurate turning of the head, or head and body, and accurate approach toward the locus of the LED stimulus constituted a correct orienting response. Any response other than a prompt direct approach to the appropriate stimulus was scored as incorrect. The cat was conditioned to approach the 01 position when a stimulus could not be localized and receive the low incentive food. Premature responses were not scored and went unrewarded. Scoring the task in this manner yields consistent and reliable indices of behavior (Sprague and Meikle, 1965; Sherman, 1977; Stein et al., 1989; Shupert et al., 1993; Lomber and Payne, 1996).
Overview Data from 12 mature (41 year) female domestic cats are presented in this report. All cats were obtained from a USDA licensed commercial laboratory animal breeding facility (Liberty Laboratories, Waverly, NY). They were housed in an ‘‘enriched’’ laboratory environment and had free access to water. Caloric intake was restricted to the testing sessions and to 1 h at the conclusion of each day, when the animals had free access to cat chow. All procedures were in accord with the National Research Council’s Guide for the Care and Use of Mammals in Neuroscience and Behavioral Research (2003) and with the approval of the Animal Care and Use Committee of The University of Texas at Dallas. All the cats were trained to perform two tasks: visual orienting and a landmark spatial discrimination. Following training on the tasks, the cats were placed in one of four cohorts of three cats each. Cohort A then received bilateral pairs of cooling loops (Lomber et al., 1996, 1999) in contact with cortex of the pMSs, the aMSs, the MSg, and the MEg. Cohort B had pMSs cortex bilaterally lesioned by injection of ibotenic acid and they received bilateral pairs of cooling loops in contact with aMSs cortex, MSg cortex, dorsal posterior suprasylvian (dPS) cortex, and MEg cortex. Cohort C had aMSs cortex bilaterally lesioned by injection of ibotenic acid. Finally, Cohort D had both aMSs and pMSs cortex bilaterally lesioned by injection of ibotenic acid. Twenty-four hours after completion of the surgical procedures, daily testing on each of the two tasks resumed for not less than 90 consecutive days. Behavioral tasks All training of the experimental animals was done binocularly. During both behavioral training and testing, cats were rewarded with soft or moist
Landmark discrimination A modified two-alternative forced-choice apparatus was used for the landmark discrimination training and testing (Fig. 2B). In the modified testing apparatus, the center divider between goal compartments was removed to create one large open arena with white walls. The landmark task is a variation of the landmark discrimination first described by Pohl (1973) and we adapted it for use with cats (Lomber and Payne, 2000a; Fig. 2C). Cats were presented with two wells, one on the left side of the apparatus and one on the right side.
160
One well contained a food reward that could not be seen by the cat. The location of the food was cued by the positioning of a black and white checkered box (6 cm wide 8 cm high 4 cm deep, with 2 cm checks) at one of six landmark locations between the food wells. Three left and three right, mirror-symmetric, positions were centered at 5, 14, or 23 cm from each reward compartment. The cue was always presented on the same side of the center of the apparatus as the food reward. The location of the landmark was randomized across trials. A trial was scored as incorrect if the cat went to the unrewarded food compartment and a trial went unscored if the cat failed to make a choice.
Surgical procedures
Fig. 2. (A) Visual orienting arena. A light-emitting diode (LED, small circle) is located above a food tray at each of 13 regularly spaced (151) intervals (for sake of clarity, only 301 intervals are shown). The animal was required to orient to, and move directly toward, a visual stimulus (e.g., right 751) to receive a food reward. Figure adapted from Malhotra et al. (2004). (B) Two-alternative forced-choice apparatus. The start box (right) was separated from the gray decision area (where the cat is standing in the figure) by opaque and transparent sliding doors. The pumps for cooling the loops, and thermometers for measuring probe temperature are shown in the upper, right corner of the illustration. The lid of the start box was hinged, enabling the cat to be easily returned to the box for the next trial without interfering with its tether. Note that for the landmark discrimination task, the median divider between the goal compartments was removed. (C) Arrangement of the goal compartment (viewed from above) for the landmark discrimination task. The cats discriminated whether a black and white checkered (2 cm checks) box, viewed in a white arena, was closer to a food reward well on the left (LW) or the right (RW) side of the testing apparatus. Six landmark positions were used (dashed outlines): three to the left of the midline (centered at 5, 14, or 23 cm from the reward compartment) and three to the right of the midline in a mirror-symmetric configuration. Figure adapted from Lomber and Payne (2000a).
Detailed surgical procedures and descriptions of cryoloop implantations (Lomber et al., 1999; Lomber and Payne, 2004) and ibotenic acid lesions (Payne et al., 2003) are described in detail in earlier publications. Briefly, general anesthesia was induced with sodium pentobarbital (25 mg/kg, or to effect, i.v.). Craniotomies were made over the desired regions and the dura was incised and reflected to expose the cerebrum. For the three cats of Cohort B, approximately 200 mg of ibotenic acid were bilaterally injected into the medial and lateral banks of pMSs cortex. For Cohort C, similar bilateral lesions were made into the medial and lateral banks of aMSs cortex. For Cohort D, bilateral lesions of both pMSs and aMSs cortex were performed. After adjustment of final shape to conform closely to the actual gyri and sulci, cryoloops were bilaterally implanted in contact with pMSs, aMSs, MSg, and MEg cortices for the three cats of Cohort A and aMSs, MSg, dPSg, and MEg cortices for the three cats of Cohort B. Following attachment of the cryoloops to the skull, the dura mater was replaced and bone defects around the implanted cooling loops were repaired with bone, Gelfoams, and acrylic. Dermal incisions were repaired with silk sutures that were removed approximately 10 days later. During the initial period after awaking, the cats were also given buprenorphine analgesic (0.01 mg/kg, s.c.). In all cases, postsurgical recovery was uneventful.
161
Cryoloop placements The pMSs loops and aMSs loops were designed to deactivate the posterior and anterior halves of the middle suprasylvian sulcus, respectively. The pMSs loops were approximately 8 mm long and extended in the middle suprasylvian sulcus from P3 to A5. These loops were in contact with both the medial and lateral banks and the goal was to deactivate major portions or all of areas PMLS and PLLS of Palmer et al. (1978), respectively, or areas LS and PEV of Grant and Shipp (1991). The aMSs loops were about 8 mm long and placed in the middle suprasylvian sulcus from A7 to A15. These loops were also in contact with both the medial and lateral banks. These loops were designed to fully deactivate major portions of areas AMLS and ALLS of Palmer et al. (1978). The crown of the MSg, between areas 21a and 5, corresponds to classically described area 7 (Hassler and Muhs-Clement, 1964; Olson and Lawler, 1987). In this report, we only examined the posterior portion of area 7. Therefore, the MSg loops were 3 7 mm and placed on the crown of the MSg between P1 and A6. Just posterior to MSg cortex, dPSg cortex includes the dorsal half of the posterior suprasylvian gyrus and the entire posterior bend of the suprasylvian gyrus. Deactivation of this region includes all of retinotopically defined area 21a (Tusa and Palmer, 1980). The dPS cryoloops had an unusual shape; an inverted ‘L.’ Finally, the MEg cryoloop was 3 6 mm and placed over the posterior middle ectosylvian (ME) gyrus from P1 to A5. This region is occupied by the dorsal zone of A1 (Middlebrooks and Zook, 1983), the region previously described as the suprasylvian fringe (Woolsey, 1960; Paula-Barbosa et al., 1975; Niimi and Matsuoka, 1979; Beneyto et al., 1998), and dorsal portions of primary auditory cortex (Reale and Imig, 1980). Behavioral testing Twenty-four hours after completion of the surgical procedures, daily testing on each of the two tasks resumed for not less than 90 consecutive days. All testing of the experimental animals was binocular and all cortical deactivations were bilateral.
Previous studies have revealed that a cryoloop temperature of 31C reliably blocks synaptic transmission through the full thickness of the contacted cortex, thus rendering it inactive and unresponsive to afferent signals that continue to arrive (Be´nita and Conde´, 1972; Lomber et al., 1994, 1996, 1999; Chafee and Goldman-Rakic, 2000). Cooling was effected by passing cold methanol through the lumen of the stainless steel hypodermic tubing that transfers the cold to the underlying cortex. The cats’ ability to perform the two tasks was tested using the standard 3-step paradigm: (1) with all cortices at normal body temperature and active; (2) with one pair of loops cooled to 3 1C; and (3) again with all cortices warm and active. For the visual orienting task, testing blocks consisted of 28 trials (2 presentations at each of the 12 peripheral locations and 4 presentations at the central position). A testing session consisted of three testing blocks in each of the three phases of testing (252 trials total). For the landmark task, a testing session consisted of 36 trials (6 each of 6 landmark loci) in each of the three phases of testing (108 trials total).
Terminal procedures As we have done in the past, following behavioral testing, the cats were anesthetized (sodium pentobarbital, 25–30 mg/kg, i.v.), small craniotomies were made, and we measured temperatures beneath the cooling loops to determine the deactivated region during cooling of the cryoloops to 371 1C. The purpose of these measurements was to identify the position of the 20 1C thermocline. Twenty degrees Celsius is the critical temperature below which neuronal activity is silenced (Lomber et al., 1999). Positions between the 20 1C thermocline and the cooling loop were at temperatures below 20 1C and were silenced, whereas positions distal to the 20 1C thermocline, relative to the cooling loop, were warmer than 20 1C and partially or fully active. Cortical temperatures during cooling were measured simultaneously at four different coronal levels in the brain using multiple microthermocouples (150 mm in diameter) manufactured for us by Omega Engineering (Stamford,
162
CT). The microthermocouples were first positioned, and then the loop was cooled to several temperatures. One hundred to one hundred twenty sites were sampled at each of the coronal levels. This procedure ensured that temperature measurements for a given cryoloop temperature setting were taken at exactly the same sites in cortex. For each measurement, cortex was cooled for approximately 5 min prior to a recording being made, as occurred in the behavioral component of the study. This protocol was then repeated at multiple, sequentially sampled sets of sites. After completion of the mappings, the craniotomies were closed and each animal was deeply anesthetized with sodium pentobarbital (50 mg/kg, i.v.) and perfused with fixatives in accordance with the recommendations of the American Veterinary Medical Association Panel on Euthanasia (Beaver et al., 2001). The fixed brain was exposed, photographed, frozen and cut into sections in the coronal plane on a microtome. Sections were stained for Nissl substance, myelin, or reacted for the presence of cytochrome oxidase to verify the location of the lesion, the location of the cryoloop, and the absence of surgical or cooling-induced damage.
Posterior parietal loci contributing to visual orienting and landmark discrimination Visual Orienting Following behavioral training of Cohort A, proficiency of visual orienting to stimuli within the binocular field was consistently above 80% correct at each of the tested positions. For the most peripheral positions (751 and 901), visual orienting accuracy was slightly lower with performance generally between 65 and 75% correct. Following cryoloop implantation, visual detection and orienting performance levels were virtually identical to preimplantation response levels (Fig. 3, warm). Bilateral deactivation of pMSs cortex virtually abolished all accurate visual orienting responses to LED stimuli presented anywhere in the both the left and right hemifields (Fig. 3, cool, A). Performance at the 01 position appeared to be unaffected, but
this was likely an artifact of the testing protocol as the animals were training to go to the 01 position if they could not detect a stimulus in the periphery. Bilateral deactivation of aMSs, MSg, or MEg cortex had no impact on visual orienting performance (Fig. 3, cool, B–D). Therefore, the only cortical locus that yielded a deficit on the visual orienting task was pMSs cortex. Landmark Discrimination Prior to cryoloop implantation in Cohort A, the cats correctly identified the right–left position of the landmark with a high-level of proficiency for all six landmark positions, although accuracy for the L23 and R23 positions, those farthest from the reward compartments, was somewhat lower (80% correct) compared to more proximal positions (100% correct). Following cryoloop implantation, virtually identical performance levels were recorded prior to and after cortical cooling (Fig. 4, warm, open bars). During bilateral deactivation of pMSs cortex, cats went to both the left and right food compartments, but their choices appeared to be random with regard to the positions of the landmark. Therefore, overall performance for all landmark positions was reduced to chance levels (50% correct; Fig. 4A, gray bars) as if the cats could recognize the presence of the landmark but could not grasp whether it was closer to the right or left reward compartments. In other words, the cats were no longer skilled at using the spatial relations between the landmark and the locus of the reward. Finally, bilateral deactivation of aMSs, MSg, or MEg cortex had no impact on the cats’ ability to perform the landmark spatial discrimination task (Fig. 4B–D). Therefore, the only cortical locus that yielded a deficit on the landmark spatial discrimination task was pMSs cortex. Task-specific adaptive neuroplasticity following lesions of pMSs cortex In addition to revealing the deficits on the visual orienting and landmark discrimination tasks during bilateral cooling of pMSs cortex, the cats of
163
Fig. 3. Performance on the visual orienting task before (warm) and during (cool) bilateral reversible deactivation of: (A) pMSs; (B) aMSs; (C) MSg; or (D) MEg. Data shown are from the three cats of Cohort A. Results are shown as polar plots with the two concentric semicircles representing 50 and 100% correct response levels and the length of each bold line corresponds to the percentage of correct responses at each location tested. Icons to the right show locations deactivated. Overall, bilateral deactivation of pMSs cortex profoundly reduced accurate orienting responses.
164
165
Cohort A also demonstrated that the deficits induced by cooling and were stable over time. Fig. 5A shows daily performance on the visual orienting and landmark discrimination tasks. Following cryoloop implantation, daily cooling deactivation of pMSs cortex yielded consistently profound deficits, on both tasks, for over 90 days following implantation (Fig. 5A). In earlier studies, visual orienting deficits induced by cooling deactivation have been found to be stable for periods of over 2.5 years postimplant (Lomber et al., 1999). In contrast, the cats of Cohort B demonstrated a very different result following the permanent bilateral deactivation of pMSs cortex by ibotenic acid. For the first 4 days following the pMSs cortex lesion (Fig. 5B, VO), the deficit appeared to be as profound as that identified during cooling deactivation (Fig. 5A, VO). However, on the fifth day postsurgery, performance began to gradually recover (Fig. 5B, VO). A full recovery of function was complete by the end of the second week. For the landmark discrimination task, no such recovery of function was identified (Fig. 5B, LD). Ninety days postsurgery, the impairment on the landmark discrimination task was still profound and stable. Therefore, comparing the visual orienting data from the reversible (Fig. 5A) and permanent (Fig. 5B) deactivation of pMSs cortex reveals neuroplastic changes that support the complete recovery of function identified on the visual orienting task following permanent pMSs cortex lesions (Fig. 5B). These observations beg the question: Why do compensations, such as deficit attenuation or full recovery of function, emerge following lesions but not during cooling deactivation. According to Lomber et al. (1996) a dominant factor is likely to be the duration of the deactivation. Following the pMSs cortex lesions, the cats lived with the neural
defect and, because of interactions with the environment, there is a constant pressure for the cat’s nervous system to compensate for the functional deficits and reduce the handicap. In contrast, the pressure to compensate for cooling-induced deficits is low because pMSs cortex was deactivated for o10% of each day, and the brain works normally for the remainder of the day. Mechanistic factors that may contribute to the lesion-induced compensations, which are not factors in cooling experiments, include the great pressure to recruit secondary, and less normally employed pathways to guide the behavior.
Identifying the locus for cortical reprogramming We have previously demonstrated (Lomber, 2001; Lomber and Payne, 2004) that none of the cortices flanking pMSs cortex are involved, in any substantial way, in visual orienting behavior (Fig. 3). Therefore, following the lesion of pMSs cortex and resulting recovery of function (Fig. 5B), any visual orientation deficits identified during reversible deactivation of the adjacent cortices would be compelling evidence for reprogramming of an adjacent cortical area. With this in mind, the next step for the cats of Cohort B was to bilaterally deactivate the individual cooling loops in contact with aMSs cortex, MSg cortex, dPSg cortex, and MEg cortex to assess if visual orienting functions had relocated to any of these areas. Bilateral cooling deactivation of aMSs cortex profoundly reduced visual orienting performance to levels similar to that of bilaterally cooling pMSs cortex in an intact cat (Fig. 6A). In contrast, bilateral cooling deactivation of none of the other adjacent cortices (MSg, dPSg, MEg) yielded any visual orienting deficits (Fig. 6B–D). This finding
Fig. 4. Performance on the landmark discrimination task before (warm) and during (cool) bilateral reversible deactivation of: (A) pMSs; (B) aMSs; (C) MSg; or (D) MEg. Data shown are from the three cats of Cohort A. For the landmark discrimination task, six landmark positions were used: three to the left of the midline (centered at 5, 14, or 23 cm from the reward locus) and three to the right of the midline in a mirror-symmetric configuration. Performance is expressed as a percent correct for each of the six different positions tested. The open bars show normal, control data collected prior to and after cooling deactivation each day (W ¼ warm). The gray bars represent data collected during cooling deactivation (C ¼ cool). Horizontal line represents chance performance (50% correct). Error bars are one standard error of the mean. Icons to the right show locations deactivated. Overall, bilateral deactivation of pMSs cortex profoundly reduced performance to chance levels across all six tested positions.
166
Fig. 5. Performance on the visual orienting (VO) and landmark discrimination (LD) tasks during daily bilateral cooling deactivation of pMSs cortex (A) or following lesion of the pMSs cortex (B). Days 2 and 1 show performance on the 2 days prior to the surgical procedure. Day 0 was the surgical procedure. Days 1–20 and 90 show performance on those days postprocedure. Percent correct for the visual orienting task indicates mean performance across all 12 peripheral testing positions for the three cats examined. Percent correct for the landmark discrimination task indicates mean performance across all six tested landmark positions for the three cats examined. For the VO task use the left vertical scale and for the LD task use the right vertical scale. (A) Data shown is from the three cats of Cohort A. (B) Data shown is from the three cats of Cohort B. Error bars indicate one standard error of the mean.
167
Fig. 6. Performance on the visual orienting task following bilateral lesions of pMSs cortex and bilateral cooling deactivation of: (A) aMSs; (B) MSg; (C) dPSg; and (D) MEg. Data was collected from the three cats of Cohort B between postlesion days 20 and 90. Percent correct for the visual orienting task indicates mean performance across all 12 peripheral testing positions for the three cats examined. Error bars indicate one standard error of the mean.
demonstrates that the visual orienting functions that are normally mediated by pMSs cortex were able to relocate to aMSs cortex following lesion of pMSs cortex.
Removal of aMSs cortex prevents expression of adaptive neuroplasticity Two final questions need to be considered in this series of experiments:
What effect do permanent lesions of aMSs cortex have on the two tasks examined? Since the functional consequences of permanent and reversible deactivations of pMSs cortex are very different, it might be expected that the functional consequences of permanent and reversible deactivations of aMSs cortex might be different. Therefore, the three cats of Cohort C received bilateral lesions of aMSs cortex using ibotenic acid. Overall, the cats displayed no change in performance on either the visual orienting or landmark discrimination tasks (Fig. 7A). Therefore, regardless of the deactivation technique used, aMSs cortex does not make any substantial contributions to either of the tasks examined. If visual orienting functions are able to relocate to aMSs cortex following lesion of pMSs cortex, are visual orienting functions profoundly impaired when both aMSs and pMSs are lesioned together, or may visual orienting functions be displaced to another region of cortex? To answer this question, the cats of Cohort D received lesions of both aMSs and pMSs cortex. Similar to the cats of Cohort B, that received lesions of only pMSs cortex (Fig. 5B), for several days immediately following the lesion, performance was profoundly impaired on both tasks. Then, following the pMSs lesion, performance on the visual orienting task recovered to normal levels within 2 weeks (Fig. 5B). In contrast, following the combined aMSs and pMSs lesions, there was no evidence of any deficit attenuation, even 90 days following the lesion (Fig. 7B). Therefore, following combined aMSs and pMSs lesions we were unable to find any evidence for recovery of visual orienting functions that would indicate that the function had relocated to another cerebral area. Our evidence indicates that following a larger lesion, which includes both aMSs and pMSs, that there is no available cortex that can accommodate the displaced visual orienting functions.
Is there a ‘‘price to pay’’ for cerebral reprogramming? The data presented above show that at least one highly localizable function of normal posterior parietal cortex becomes relocated across the
168
Fig. 7. Performance on the VO and LD tasks during daily testing following bilateral lesion of aMSs cortex (A) or following bilateral lesion of the aMSs and pMSs cortex (B). (A) Data shown is from the three cats of Cohort C. (B) Data shown is from the three cats of Cohort D. For other conventions, see Fig. 5.
cerebral surface after a lesion of the mature extrastriate visual cortex. However, it is not known if there is a ‘‘price’’ to pay for this redistribution, and it is unknown if functions normally associated with the region demonstrating adaptive neuroplasticity regions are reduced, displaced, or ‘‘crowded out’’ as proposed by Teuber (1975) when the
cerebral regions make contributions to new aspects of visually-guided behavior not normally performed by them. Such a change in function is a possibility because visuospatial abilities are reduced when sparing of language functions can be demonstrated in humans (Milner, 1974; Teuber, 1975).
169
Unfortunately, little is known about the functions of aMSs cortex other than it is important in performing direction of motion discriminations (Lomber, 2001). In future studies, it would be worth examining other possible functions of aMSs cortex to see if there is any impairment in the functions normally performed by aMSs cortex following lesions of pMSs cortex. The present studies have demonstrated that expression of neuroplasticity in aMSs cortex can make it accommodate functions displaced from pMSs cortex, but it is not known if there is a price to pay for such cerebral reprogramming.
Relocation of cerebral functions in other systems In the present study, we have shown that a visual function can relocate from one extrastriate area to another following damage of mature extrastriate visual cortex. This finding is unique because, within sensory systems, there are few, if any, examples of cerebral functions relocating following damage to the mature cerebrum. Most studies of cortical plasticity in sensory systems have found expansion or modifications in primary sensory cortex maps following peripheral manipulations. These studies have concerned somatosensory cortex following amputation (Merzenich et al., 1984; Code et al., 1992; Manger et al., 1996), severing of peripheral nerves (Merzenich et al., 1983a, b; Garraghty and Kaas, 1991; Wall et al., 1992), and changes in visual cortex following retinal lesions (Kaas et al., 1990; Chino et al., 1992; Gilbert and Wiesel, 1992; Schmid et al., 1996). Other studies have shown expansion of cross-modal sensory maps in extrastriate visual areas following peripheral manipulations (Rauschecker and Korte, 1993; Rauschecker, 1999). Other changes in sensory cortices have been shown to occur following damage of the mature cerebrum. Jenkins and Merzenich (1987) demonstrated that a specific sensory representation could relocate following removal of the region of the cerebrum that normally represents it. Pons et al. (1988) reported significant alterations in somatosensory cortical representations of the hand following removal of
the regions of the cortex that normally represent the hand. Extensive cortical reorganization has been found in extrastriate visual cortex following cortical damage in the developing brain (Payne and Lomber, 2001, 2002). Studies of the effect of damage of visual area MT/MST in the mature cerebrum have shown many examples of attenuation of deficits and recovery of function (Newsome and Pare´, 1988; Yamasaki and Wurtz, 1991; Pasternak and Merigan, 1994). However, the cortical substrate responsible for these examples of adaptive neuroplasticity has not been identified. As there are many similarities between pMSs cortex of the cat and areas MT/MST in the monkey (Payne, 1993), we predict that the cortex adjacent to MT/ MST is likely to be responsible for the recovery of function commonly found following lesions of MT/MST.
Acknowledgments We thank the National Science Foundation and the National Institute of Neurological Disorders and Stroke and for their generous support of this project. We also thank Shveta Malhotra and Amee Hall for assistance at various phases of the study.
References Beaver, B.V., Reed, W., Leary, S., McKiernan, B., Bain, F., Schultz, R., Bennett, B.T., Pascoe, P., Shull, E., Cork, L.C., Francis-Floyd, R., Amass, K.D., Johnson, R.J., Schmidt, R.H., Underwood, W., Thornton, G.W. and Kohn, B. (2001) 2000 Report of the American Veterinary Medical Association Panel on Euthanasia. J. Am. Vet. Med. Assoc., 218: 669–696. Beckers, G. and Zeki, S. (1995) The consequences of inactivating areas V1 and V5 on visual motion perception. Brain, 118: 49–60. Beneyto, M., Winer, J.A., Larue, D.T. and Prieto, J.J. (1998) Auditory connections and neurochemistry of the sagulum. J. Comp. Neurol., 401: 329–351. Be´nita, M. and Conde´, H. (1972) Effects of local cooling upon conduction and synaptic transmission. Brain Res., 36: 133–151. Chafee, M.V. and Goldman-Rakic, P.S. (2000) Inactivation of parietal and prefrontal cortex reveals interdependence of neural activity during memory-guided saccades. J. Neurophysiol., 83: 1550–1566.
170 Chino, Y.M., Kaas, J.H., Smith, E.L., Langston, A.L. and Cheng, H. (1992) Rapid reorganization of cortical maps in adult cats following restricted deafferentation in retina. Vision Res., 32: 789–796. Code, R.A., Eslin, D.E. and Juliana, S.L. (1992) Expansion of stimulus-evoked metabolic activity in monkey somatosensory cortex after peripheral denervation. Exp. Brain Res., 88: 341–344. Garraghty, P.E. and Kaas, J.H. (1991) Large-scale functional reorganization in adult monkey cortex after peripheral nerve injury. Proc. Natl. Acad. Sci. USA, 88: 6976–6980. Gilbert, C.D. and Wiesel, T.N. (1992) Receptive field dynamics in adult primary visual cortex. Nature, 356: 150–152. Grant, S. and Shipp, S. (1991) Visuotopic organization of the lateral suprasylvian area and of an adjacent area of the ectosylvian gyrus of cat cortex: a physiological and connectional study. Vis. Neurosci., 6: 315–338. Hardy, S.C. and Stein, B.E. (1988) Small lateral suprasylvian cortex lesions produce visual neglect and decreased visual activity in the superior colliculus. J. Comp. Neurol., 273: 527–542. Hassler, R. and Muhs-Clement, K. (1964) Architektonischer Aufbau des sensomotorischen and parietalen Cortex der Katze. J. Hirnforsch., 6: 377–420. Heilman, K.M., Valenstein, E., Day, A. and Watson, R.T. (1995) Frontal lobe neglect in monkeys. Neurology, 45: 1205–1210. Heilman, K.M., Valenstein, E., Day, A. and Watson, R.T. (2000) Neglect and related disorders. Semin. Neurol., 20: 463–470. Huxlin, K.R. and Pasternak, T. (2004) Training-induced recovery of visual motion perception after extrastriate cortical damage in the adult cat. Cereb. Cortex, 14: 81–90. Hyva¨rinen, J. (1982) The Parietal Cortex of Monkey and Man. Springer-Verlag, Berlin. Jenkins, W.M. and Merzenich, M.M. (1987) Reorganization of neocortical representations after brain injury: a neurophysiological model of the bases of recovery from stroke. Prog. Brain Res., 71: 249–266. Kru¨ger, K., Kiefer, W. and Groh, A. (1993) Lesion of the suprasylvian cortex impairs depth perception of cats. Neuroreport, 4: 883–886. Kaas, J.H., Krubitzer, L.A., Chino, Y.M., Langston, A.L., Polley, E.H. and Blair, N. (1990) Reorganization of retinotopic cortical maps in adult mammals after lesions of the retina. Science, 248: 229–231. Lomber, S.G. (2001) Behavioral cartography of visual functions in cat parietal cortex: areal and laminar dissociations. Prog. Brain Res., 134: 265–284. Lomber, S.G. and Payne, B.R. (1996) Removal of two halves restored the whole: reversal of visual hemineglect during bilateral cortical or collicular inactivation in the cat. Vis. Neurosci., 13: 1143–1156. Lomber, S.G. and Payne, B.R. (2000a) Contributions of cat posterior parietal cortex to visuospatial discrimination. Vis. Neurosci., 17: 701–709. Lomber, S.G. and Payne, B.R. (2000b) Translaminar differentiation of visually-guided behaviors revealed by restricted cerebral cooling deactivation. Cereb. Cortex, 10: 1066–1077.
Lomber, S.G. and Payne, B.R. (2001) Task-specific reversal of visual hemineglect during reversible deactivation of posterior parietal cortex: a comparison with deactivation of the superior colliculus. Vis. Neurosci., 18: 487–499. Lomber, S.G. and Payne, B.R. (2004) Cerebral areas mediating visual redirection of gaze: cooling deactivation of 15 loci in the cat. J. Comp. Neurol., 474: 190–208. Lomber, S.G., Cornwell, P., Sun, J.S., MacNeil, M.A. and Payne, B.R. (1994) Reversible inactivation of visual processing operations in middle suprasylvian cortex of the behaving cat. Proc. Natl. Acad. Sci. USA, 91: 2999–3003. Lomber, S.G., Payne, B.R., Cornwell, P. and Long, K.D. (1996) Perceptual and cognitive visual functions of parietal and temporal cortices in the cat. Cereb. Cortex, 6: 673–695. Lomber, S.G., Payne, B.R. and Horel, J.A. (1999) The cryoloop: an adaptable reversible cooling deactivation method for behavioral and electrophysiological assessment of neural function. J. Neurosci. Meth., 86: 179–194. Luh, K.E., Butter, C.M. and Buchtel, H.A. (1986) Impairments in orienting to visual stimuli in monkeys following unilateral lesions of the superior sulcal polysensory cortex. Neuropsychologia, 24: 461–470. Lynch, J.C. and McLaren, J.W. (1989) Deficits of visual attention and saccadic eye movements after lesions of parietooccipital cortex in monkeys. J. Neurophysiol., 61: 74–90. Malhotra, S., Hall, A.J. and Lomber, S.G. (2004) Cortical control of sound localization in the cat: unilateral cooling deactivation of 19 cerebral areas. J. Neurophysiol., 92: 1625–1643. Manger, P.R., Woods, T.M. and Jones, E.G. (1996) Plasticity of the somatosensory cortical map in macaque monkeys after chronic partial amputation of a digit. Proc. R. Soc. Lond. Biol. Sci., 263: 933–939. Marshall, J.C. and Halligan, P.W. (1988) Blindsight and insight in visuo-spatial neglect. Nature, 336: 766–767. Marshall, J.C. and Halligan, P.W. (1995) Seeing the forest but only half the trees? Nature, 373: 521–523. Merzenich, M.M., Kaas, J.H., Wall, J.T., Sur, M., Nelson, R.J. and Felleman, D.J. (1983a) Progression of change following median nerve section in the cortical representation of the hand in areas 3b and 1 in adult owl and squirrel monkeys. Neuroscience, 10: 639–665. Merzenich, M.M., Kaas, J.H., Wall, J.T., Nelson, R.J., Sur, M. and Felleman, D.J. (1983b) Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation. Neuroscience, 8: 33–55. Merzenich, M.M., Nelson, R.J., Stryker, M.P., Cynader, M.S., Schoppmann, A. and Zook, J.M. (1984) Somatosensory cortical map changes following digit amputation in adult monkeys. J. Comp. Neurol., 258: 281–296. Mesulam, M.-M. (1981) A cortical network for directed attention and unilateral neglect. Ann. Neurol., 10: 309–325. Middlebrooks, J.C. and Zook, J.M. (1983) Intrinsic organization of the cat’s medial geniculate body identified by projections to binaural response-specific bands in the primary auditory cortex. J. Neurosci., 1: 203–224.
171 Milner, B., 1974. Sparing of language function after unilateral brain damage. In: Eidelberg, E., Stein, D.G. (Eds.), Functional recovery after lesions of the nervous system. Neurosci. Res. Prog. Bull., 2, 213–217. Newsome, W.T. and Pare´, E.B. (1988) A selective impairment of motion perception following lesions of the middle temporal visual area (MT). J. Neurosci., 8: 2201–2211. Newsome, W.T., Wurtz, R.H., Du¨rsteler, M.R. and Mikami, A. (1985) Deficits in visual motion processing following ibotenic acid lesions of the middle temporal visual area of the macaque monkey. J. Neurosci., 5: 825–840. Niimi, K. and Matsuoka, H. (1979) Thalamocortical organization of the auditory system in the cat studied by retrograde and axonal transport of horseradish peroxidase. Adv. Anat. Embryol. Cell. Biol., 57: 1–56. Olson, C.R. and Lawler, K. (1987) Cortical and subcortical afferent connections of a posterior division of feline area 7 (area 7p). J. Comp. Neurol., 259: 13–30. Palmer, L.A., Rosenquist, A.C. and Tusa, R.J. (1978) The retinotopic organization of lateral suprasylvian visual areas in the cat. J. Comp. Neurol., 177: 237–256. Pasternak, T., Horn, K.M. and Maunsell, J.H.R. (1989) Deficits in speed discrimination following lesions of the lateral suprasylvian cortex in the cat. Vis. Neurosci., 3: 365–375. Pasternak, T. and Merigan, W.H. (1994) Motion perception following lesions of the superior temporal sulcus in the monkey. Cereb. Cortex, 4: 247–259. Paula-Barbosa, M.M., Feyo, P.B. and Sousa-Pinto, A. (1975) The association connexions of the suprasylvian fringe (SF) and other areas of the cat auditory cortex. Exp. Brain Res., 23: 535–554. Payne, B.R. (1993) Evidence for visual cortical area homologs in cat and macaque monkey. Cereb. Cortex, 3: 1–25. Payne, B.R. and Lomber, S.G. (2001) Reconstructing functional systems after lesions of cerebral cortex. Nat. Rev. Neurosci., 2: 911–919. Payne, B.R. and Lomber, S.G. (2002) Plasticity of visual cortex after injury: What’s different about the young brain? Neuroscientist, 8: 174–185. Payne, B.R., Lomber, S.G., Geeraerts, S., van der Gucht, E. and Vandenbussche, E. (1996) Reversible visual hemineglect. Proc. Natl. Acad. Sci. USA, 93: 290–294. Payne, B.R., Lomber, S.G., Rushmore, R.J. and PascualLeone, A. (2003) Cancellation of visuoparietal lesion-induced spatial neglect. Exp. Brain Res., 150: 395–398. Pohl, W. (1973) Dissociation of spatial discriminations deficits following frontal and parietal lesions in monkeys. J. Comp. Physiol. Psych., 82: 227–239. Pons, T.P., Garraghty, P.E. and Mishkin, M. (1988) Lesioninduced plasticity in the second somatosensory cortex of adult macaques. Proc. Natl. Acad. Sci. USA, 14: 5279–5281. Rauschecker, J.P. (1999) Auditory cortical plasticity: a comparison with other sensory systems. Trends Neurosci., 22: 74–80. Rauschecker, J.P. and Korte, M. (1993) Auditory compensation for early blindness in cat cerebral cortex. J. Neurosci., 13: 4538–4548.
Reale, R.A. and Imig, T.J. (1980) Tonotopic maps of auditory cortex in the cat. J. Comp. Neurol., 192: 265–292. Reinoso-Sua´rez, F. (1961) Topograpical Atlas of the Cat Brain for Experimental-Physiological Research [in German]. Merck, Darmstad, Germany. Rudolph, K.K. and Pasternak, T. (1996) Lesions of cat lateral suprasylvian cortex affect the perception of complex motion. Cereb. Cortex, 8: 814–822. Rudolph, K.K. and Pasternak, T. (1999) Transient and permanent deficits in motion perception after lesions of cortical areas MT and MST in the macaque monkey. Cereb. Cortex, 9: 90–100. Schmid, L.M., Rosa, M.G., Calford, M.B. and Ambler, J.S. (1996) Visuotopic reorganization in the primary visual cortex of adult cats following monocular and binocular retinal lesions. Cereb. Cortex, 6: 388–405. Sherk, H. and Fowler, G.A. (2002) Lesions of extrastriate cortex and consequences for visual guidance during locomotion. Exp. Brain Res., 144: 159–171. Sherman, S.M. (1977) The effect of superior colliculus lesions upon the visual fields of cats with cortical ablations. J. Comp. Neurol., 172: 211–230. Shupert, C., Cornwell, P. and Payne, B.R. (1993) Differential sparing of depth perception, orienting, and optokinetic nystagmus after neonatal versus adult lesions of cortical areas 17, 18 and 19 in the cat. Behav. Brain Res., 107: 633–650. Sprague, J.M. and Meikle, T.H.J. (1965) The role of the superior colliculus in visually guided behavior. Exp. Neurol., 11: 115–146. Stein, B.E., Meredith, M.A., Huneycutt, W.S. and McDale, L. (1989) Behavioral indices of multisensory integration: orientation to visual cues is affected by auditory stimuli. J. Cog. Neurosci., 1: 12–24. Teuber, H.-L., (1975) Recovery of function after brain injury in man. In: Outcome of severe damage to the nervous system, Ciba Foundation Symposium 34, Elsevier North-Holland, Amsterdam. Tusa, R.J., Demer, J.L. and Herdman, S.J. (1989) Cortical areas involved in OKN and VOR in cats: cortical lesions. J. Neurosci., 9: 1163–1178. Tusa, R.J. and Palmer, L.A. (1980) Retinotopic organization of areas 20 and 21 in the cat. J. Comp. Neurol., 193: 147–164. Vaina, L.M. (1989) Selective impairment of visual motion interpretation following lesions of the right occipito-parietal area in humans. Biol. Cybern., 61: 347–359. Vaina, L.M. (1994) Functional segregation of color and motion processing in the human visual cortex: clinical evidence. Cereb. Cortex, 4: 555–572. Ventre, J. (1985) Cortical control of oculomotor functions. I. Optokinetic nystagmus. Behav. Brain Res., 15: 211–226. Wall, J.T., Huerta, M.F. and Kaas, J.H. (1992) Changes in the cortical map of the hand following postnatal ulnar and radial nerve injury in monkeys: modification of somatotopic aggregates. J. Neurosci., 12: 3445–3455. Watson, R.T., Valenstein, E., Day, A. and Heilman, K.M. (1994) Posterior neocortical systems subserving awareness
172 and neglect. Neglect associated with superior temporal sulcus but not area 7 lesions. Arch. Neurol., 51: 1014–1021. Woolsey, C.N. (1960) Organization of cortical auditory system: a review and synthesis. In: Rasmussen, G.L. and Windle, W.F. (Eds.), Neural Mechanisms of the Auditory and Vestibular Systems. Thomas, Springfield, pp. 165–180. Yamasaki, D.S. and Wurtz, R.H. (1991) Recovery of function after lesions in the superior temporal sulcus in the monkey. J. Neurophysiol., 66: 651–673.
Zeki, S. (1991) Cerebral akinetopsia (visual motion blindness). Brain, 114: 811–824. Zihl, J., von Cramon, D. and Mai, N. (1983) Selective disturbances of movement vision after bilateral brain damage. Brain, 106: 313–340.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 11
A SPECT study of language and brain reorganization three years after pediatric brain injury Stephanie B. Chiu Wong1, Sandra B. Chapman1,2,, Lori G. Cook1, Raksha Anand1, Jacquelyn F. Gamino1 and Michael D. Devous Sr.3 1 Center for Brain Health, The University of Texas at Dallas, Dallas, TX 75235,USA Institute of Biomedical Sciences and Technology, The University of Texas at Dallas, Dallas, TX, USA 3 Nuclear Medicine Center, The University of Texas Southwestern Medical Center at Dallas, Dallas, TX, USA 2
Abstract: Using single photon emission computed tomography (SPECT), we investigated brain plasticity in children 3 years after sustaining a severe traumatic brain injury (TBI). First, we assessed brain perfusion patterns (i.e., the extent of brain blood flow to regions of the brain) at rest in eight children who suffered severe TBI as compared to perfusion patterns in eight normally developing children. Second, we examined differences in perfusion between children with severe TBI who showed good versus poor recovery in complex discourse skills. Specifically, the children were asked to produce and abstract core meaning for two stories in the form of a lesson. Inconsistent with our predictions, children with severe TBI showed areas of increased perfusion as compared to normally developing controls. Adult studies have shown the reverse pattern with TBI associated with reduced perfusion. With regard to the second aim and consistent with previously identified brain-discourse relations, we found a strong positive association between perfusion in right frontal regions and discourse abstraction abilities, with higher perfusion linked to better discourse outcomes and lower perfusion linked to poorer discourse outcomes. Furthermore, brain-discourse patterns of increased perfusion in left frontal regions were associated with lower discourse abstraction ability. The results are discussed in terms of how brain changes may represent adaptive and maladaptive plasticity. The findings offer direction for future studies of brain plasticity in response to neurocognitive treatments. Keywords: children; plasticity; recovery; language; SPECT; functional brain imaging; cognition; brain injury networks throughout adolescence (Giedd et al., 1999) and expansive growth in complex cognitive functions, such as discourse processing (Chapman et al., 1999). Neural plasticity in children involves adaptive structural and functional changes that occur in the brain, not only through genetically determined processes but also as a child engages in the learning process. A disruption to normal brain development as a consequence of childhood disease or injury presents a complex problem in that we are not only dealing with recovery of lost functions but the adaptability of
Introduction What is most appealing about young folks, after all, is the changes, not the still photograph of finished character but the movie, the soul in flux. Thomas Pynchon Childhood is a period of rich development, manifested by extensive restructuring of neural Corresponding author. Tel.: +1 214/905-3007; Fax: +1 1214/ 905-3026; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57011-6
173
174
the brain to support later developing cognitive functions. Brain injury disrupts childhood at an alarming rate. Each year, approximately 1 in 500 children sustain an injury to the brain serious enough to require hospitalization (Langlois, 2001). By the age of 16, 4 out of every 100 boys and 2.5 out of every 100 girls have sustained a traumatic brain injury (TBI) (Christensen, 2001). By the age of 21, 1 in 30 children will have sustained a brain injury (Christensen, 2001). Brain plasticity in childhood Overestimation of brain plasticity in the developing brain has led to optimism that the young and developing brain possesses greater capacity to recover after injury than the adult brain (Kennard, 1938; Chugani et al., 1996). Despite this optimism, earlier age at injury can contribute to later emergent difficulties during cognitive-linguistic development (Chapman et al., 2001b, 2003, 2004). Early biological injury may perturb development of the neural networks that support acquisition of later cognitive skills (Ewing-Cobbs et al., 1987; Jaffe et al., 1995; Yorkston et al., 1997; Chapman et al., 1998; Brookshire et al., 2000; Chapman et al., 2001b). Because children with early injuries may ‘‘grow into’’ their cognitive, linguistic, and behavioral deficits as the injured brain matures (e.g., Ewing-Cobbs et al., 1987; Fletcher et al., 1987), it is crucial to examine long-term recovery of cognitive functions. Evidence indicates that language-related abilities that undergo rapid development are particularly vulnerable after brain injury (Ewing-Cobbs et al., 1987; Chapman et al., 1999, 2006a). Specifically, discourse (i.e., connected language) processing is sensitive to the higher-level deficits that continue to unfold years after pediatric brain injury (Biddle et al., 1996; Reilly et al., 1998; Brookshire et al., 2000). Research has shown that difficulties in discourse production are common in children with severe TBI as compared to typically developing children (Chapman et al., 1997, 2006a; Reilly et al., 1998). Due to a close relation between discourse abilities and learning capacity in children (Johnson, 1983; Stein and Kirby, 1992),
investigating long-term recovery of discourse after pediatric brain injury bears both clinical and theoretical importance. Functional brain imaging: window to brain plasticity Until recently, brain plasticity subsequent to pediatric TBI has been inferred based on recovery of basic cognitive functions. With the advent of advanced functional brain imaging techniques such as single photon emission computed tomography (SPECT), positron emission tomography (PET), and functional magnetic resonance imaging (fMRI), a more direct and objective measurement of brain function may further our understanding of brain plasticity after injury in childhood. In children with TBI, adaptability of the brain to successfully recover previously acquired skills stands in stark contrast to the reduced plasticity that occurs in acquiring new skills at later stages (Chapman et al., 2001a). In general, the injured brain undergoes significant reorganization, including morphologic, hemodynamic, and biochemical changes (Bigler, 1999). Static brain imaging, however, may not fully capture the pathological consequences that affect neurobehavioral outcomes. Although structural imaging studies using magnetic resonance imaging (MRI) have revealed predominantly frontal and temporal involvement after pediatric TBI (Mendelsohn et al., 1992; Levin et al., 1993; Graham, 1996; Bigler, 1996), the relationships between site, size, and number of acute focal lesions inadequately accounts for neuropsychological outcomes (Kurth et al., 1994; Anderson and Bigler, 1995; Bigler, 1996). Functional brain imaging tools offer promising adjuncts to more fully elucidate the complex relationships between brain plasticity and behavioral outcomes. Jacobs et al. (1994) found that initial perfusion deficit on SPECT predicted favorable outcome in adults with mild and moderate brain injury. Moreover, recovery of specific cognitive functions has been correlated to adaptive changes in brain function using SPECT (Ichise et al., 1994; Jacobs et al., 1994; Bonne et al., 2003). Application of SPECT is likely to further elaborate changes in brain and behavioral function at
175
chronic stages of brain recovery beyond what is revealed by structural brain imaging, as this technique reveals anatomy and can probe the functional consequences of the damaged brain (Ichise et al., 1994). The majority of existing brain injury research using SPECT imaging has examined cognitive recovery in adult populations. Researchers have reported a high prevalence of SPECT abnormalities in adults with TBI who exhibit cognitive deficits (Goldenberg, 1992; Ichise et al., 1994; Kant et al., 1997; Umile et al., 1998; Bonne et al., 2003). These functional brain abnormalities are more extensive than the structural lesions. Gross et al. (1996) and Bonne et al. (2003) found decreased metabolism in anterior temporal, mid-temporal, anterior cingulate, precuneus, and frontal regions. These findings cannot be directly translated to the pediatric TBI population, since perfusion and metabolism patterns in children differ from those of adults (Chugani et al., 1987; Casey et al., 2005). Studies investigating functional brain imaging in pediatric TBI have increasingly used fMRI to examine brain activation as related to specific cognitive tasks. However, few studies have utilized SPECT or PET in children, notwithstanding that these techniques offer an objective and direct measure of changes in perfusion subsequent to pediatric TBI (Bergsneider et al., 2001). The relative accessibility of SPECT technology adds to its appeal and value (Ichise et al., 1994). As stated previously, severe brain injury in childhood has been associated with later emerging cognitive deficits, especially in higher level skills such as strategic processing of discourse (Chapman et al., 2004, 2006b; Cook et al., 2006). Few functional brain imaging studies have examined the relationship between the brain and specific cognitive functions in pediatric TBI (Bigler, 1999). No known study to date has examined patterns of functional brain lesions in pediatric TBI as compared to normally developing children, nor has any known study examined associations between functional brain lesions with long-term recovery of discourse abilities after severe brain injury in children. Relating long-term discourse outcome to perfusion patterns may help elucidate the nature of brain plasticity after TBI in childhood.
Purpose of study The focus of the present study was twofold. First, the current study examined whether brain perfusion patterns in children 3 years after a severe TBI differ from brain perfusion patterns of healthy pediatric controls in any consistent way. Based on existing literature of decreased metabolic function after brain damage (Bergsneider et al., 2001), we hypothesized that participants with TBI would demonstrate decreased relative perfusion in several brain regions compared to typically developing children. Second, this experiment investigated the relation between functional brain lesions and discourse outcome only in the group of children with TBI. Specifically, the aim was to examine the feasibility of relating later stage discourse recovery after TBI to patterns of perfusion. Based on emerging literature (e.g., Nichelli et al., 1995; Chapman et al., 2005), we hypothesized that poor discourse outcomes in pediatric TBI would be associated with reduced perfusion patterns in the right hemisphere.
Methods Participants The two populations for the first goal of this study included 16 children between 8 and 12 years of age. Specifically, eight children suffered a severe TBI 3 years prior to testing and eight were normally developing controls. The same eight children with severe TBI comprised the group used to address the second goal of examining brain perfusion patterns in poor versus good recovery. See Tables 1 and 2 for demographic data. Severity of TBI was established using a Glasgow Coma Scale score of 8 or below at the time of injury, 3 years (74 months) prior to this study. Injury resulted from a variety of causes, as shown in Table 1. Exclusionary criteria for injured children included previous hospitalization for head injury; pre-existing neurological disorder associated with cerebral dysfunction and/or cognitive deficit; preexisting severe psychiatric disorder; history of
176 Table 1. Demographics for brain injury group TBI participant
Age at scana
Sex
Hand
GCS
Cause of injury
Primary site of injuryb
1 2 3 4 5 6 7 8
8; 11; 20 8; 8; 14 11; 5 8; 5; 2 12; 0; 19 10; 11; 27 11; 1; 19 10; 8; 8
F F M F F M M F
R R R R R R L R
7 6 5 7 4 7 5 6
MPA MPA MVA Sports injury MPA MPA MVA MPA
L parietal lobe R parietal lobe L and R frontal lobe R frontal lobe Diffuse injury R parietal lobe L frontal lobe R frontal lobe
Note: MPA, motor-pedestrian accident; MVA, motor-vehicle accident a
Age is reported as years; months; days b According to MRI taken at time of injury
Table 2. Demographics for controls Control participant
Age at scana
Sex
Hand
1 2 3 4 5 6 7 8
9; 1; 19 10; 2; 28 8; 9; 14 11; 4; 10 9; 7; 3 12; 6; 21 11; 8; 8 11; 5; 26
M F F M M F M M
R R R R R R R R
a
Age is reported as years; months; days.
child abuse; penetrating gunshot wound to the brain; illegal alien status; history of hypoxia/anoxia; history of hypotension; history of meningitis or encephalitis; history of chronic, serious physical disorder, such as cancer, uncontrolled diabetes, etc.; enrollment in ESL classes (English is not the primary language). All participants had a minimum gestation of 37 weeks and minimum birth weight of 5 lbs 8 oz. The eight typically developing control children had normal psychomotor development, and no known history of learning disability. Case histories revealed that the control participants were medication-free and had no DSM IV Axis 1 psychiatric disorders. See Table 2 for demographic data. Informed consent was obtained for all participants from a legal guardian as approved by and according to the guidelines for pediatric studies of the Institutional Review Board of the University of Texas Southwestern Medical Center.
SPECT scan evaluation All participants were injected with 99m-Tc HMPAO (Ceretec, Nycomed Amersham Corp., Chicago, IL) while they were at rest, seated comfortably in a dimly lit, quiet room with eyes open and ears unoccluded. Approximately, 90 min after the injection, participants were placed in a supine position within the SPECT scanner. SPECT data for 120 projections were acquired using Picker Prism 3000 triple-head gamma camera (Picker Int. Corp., Cleveland, OH) with ultra-high resolution fanbeam collimators (6.5 mm full width half maximum (FWHM), 128 128 matrix) in continuous rotation through 1201 in five steps, alternating clockwise and counterclockwise. Image rendering Using Picker Prism software, the projection data was upscaled four times, reconstructed using a
177
ramp filter, and then filtered with a 3-D lowpass filter of sixth order. Chang’s method was used for attenuation correction (Chang, 1978). The transaxial slices were then converted into Talairach space using the Pelizzari fitting algorithm (Turkington et al., 1993) to adjust for translation, rotation, and scaling. Systematic review of the final data indicated that registration to an adult MRI did not impede data collection or analysis. Image analysis Spatial normalization Analysis of the brain imaging data was performed using the MEDX 3.3 software including statistical parametric mapping (SPM’96) by Friston et al. (1995) on a Sun Ultra 60 computer (Sun Microsystems, Mountain View, CA). The images were spatially normalized using the SPM software such that the anatomic location of each voxel could be reported in the standardized anatomic space of Talairach (Talairach and Tournoux, 1988). The registered images were then smoothed using a 10 mm FWHM Gaussian filter. Images were visually inspected pre and postmorphing to check that spatial normalization of impacted brains retains localization information. Intensity normalization Individual differences in global cerebral blood flow were controlled by scaling each individual by his or her own grand mean in a normalizing process based on the ratio of regional to global CBF described by Gullion et al. (1996). A mean SPECT image was created using images from all the normal and TBI participants. From the mean image, a binary-mask was constructed using a gray matter threshold of 70% the maximal pixel value. Only those voxels of the mean SPECT image that exceeded that gray matter threshold were used in the computation of the grand mean for each individual volume. Following the calculation of the grand mean, each voxel value in each individual’s volume was adjusted by a scale factor (the inverse of its grand mean multiplied by 100). Consequently, after scaling, each voxel value was interpreted as a percentage of the global
blood flow (which is constrained to equal 100 units) and was thus a measure of regional cerebral blood flow (rCBF) relative to the global blood flow. Discourse outcome Discourse performance, specifically, the ability to synthesize an abstracted core meaning (or moral) from a narrative, was evaluated within 3 months of the 3-year postinjury SPECT evaluation. Participants were tested individually in a quiet room. Two didactic stories written at a second grade vocabulary level were read aloud to each child. Subsequently, the child was asked to provide a life lesson conveyed by the story. The story lesson was used to evaluate the ability to synthesize an abstract core meaning from the explicit content conveyed by the entire story. The responses were rated on a 7-point scale developed by Delis et al. (1990) along dimensions of abstraction, accuracy, and completeness of response. A high-level lesson, reflected in scores 5, 6, or the highest score of 7, conveyed an abstracted core meaning which is generalized to real life. In contrast, scores ranging from 0 to 4 reflected problems in synthesizing an abstract core meaning (in the form of a lesson) as the response related instead concrete isolated details from the story. The discourse outcome measure (i.e., abstracted core meaning) was derived by adding the ratings for lessons from two different stories with a total possible score of 14 points for each child. Statistical analysis Issue 1: perfusion patterns in severe TBI Versus normally developing children Statistical variation between injured participants and the control group at each voxel was estimated according to the general linear model. A t-statistic image denoting the contrast was constructed. The resulting set of t values for group comparison constituted the statistical parametric map. The SPM t values were transformed to a normal distribution (SPM(Z)). The threshold for Z scores was set at p ¼ 0.05.
178
Issue 2: brain markers in SPECT of discourse recovery after TBI Covariate analysis was performed on the same eight children with TBI to determine what brain regions, if any, vary with external discourse scores. No control group was used in this study because we were interested in brain plasticity and discourse outcomes in a pediatric brain injury population. The multistudy covariate analysis paradigm within MEDx software was applied to detect any significant relationship between discourse lesson scores and perfusion data within each voxel of the whole brain across all eight individuals. This comprehensive analysis included all voxels, regardless of perfusion level, and detected any existing links between discourse and brain. Regions of brain were considered significant at a significance level of po0.05, which was subjected to Resel correction for multiple comparisons based on the number of resolvable elements analyzed.
Results Issue 1: perfusion patterns in severe TBI Versus normally developing children A comparison of the TBI group with the normally developing group revealed only one major area of
reduced relative cerebral blood perfusion (rCBF) in TBI participants in the right cerebellum (Z ¼ 3.38; p ¼ 0.00036). In contrast, at least seven areas of significant hyperperfusion were identified in the TBI group as compared to the normally developing children. Specifically, comparison of the TBI group with the normally developing children revealed relative increase in rCBF in the left frontal lobe (left frontal medial gyrus, BA10, Z ¼ 2.53; p ¼ 0.00563; and superior frontal gyrus, BA11, Z ¼ 3.70; p ¼ 0.00011), right globus pallidus–putamen–thalamus (Z ¼ 3.89; p ¼ 0.00005), bilateral occipital lobe (cuneus and lingual gyrus, BA18 and 31; Z ¼ 3.61; p ¼ 0.00015), and some bilaterally in the parietal lobes (precuneus, BA7 and 40; Z ¼ 3.68; p ¼ 0.00012) (see Figs. 1 and 2). Issue 2: brain markers in SPECT of discourse recovery after TBI Performance on the discourse outcome measure (abstracted core meaning) covaried positively with resting rCBF in the right frontal lobe (medial frontal gyrus, BA9, Z ¼ 2.90, p ¼ 0.00189; middle frontal gyrus, BA 9 and 10, Z ¼ 2.53, p ¼ 0.00566), right temporal lobe (middle temporal gyrus; BA39, Z ¼ 3.57, 0.00018,), and right cerebellum (Z ¼ 3.62, p ¼ 0.00015) (see Fig. 3).
Fig. 1. T-test results showing regions of increased rCBF in injured group relative to normally developing controls, superimposed onto MRI transverse slices. See Plate 11.1 in Colour Plate Section.
179 110 HEALTHY
rCBF
100
INJURED
90 80 70 60
Cerebellar
L med. FL
Occipital
Putamen
brain regions Fig. 2. Brain regions of distinct perfusion patterns between normally developing and injured children.
Fig. 3. Regions showing a positive covariance between rCBF and discourse abstraction performance. See Plate 11.3 in Colour Plate Section.
Correlation of discourse outcome scores and perfusion values in the right frontal lobe region identified by covariate analysis was R2 ¼ 0.709, p ¼ 0.00877, and correlation of discourse with the region identified in the right temporal lobe was R2 ¼ 0.880, p ¼ 0.000568 (see Fig. 4). In other words, more successful discourse performers demonstrated greater perfusion in the right hemisphere (i.e., right frontal and temporal areas), whereas poorer performance was associated with less perfusion in these right hemispheric brain regions. Additionally, performance on the discourse outcome measure covaried negatively with resting rCBF in the left frontal lobe (left inferior frontal
gyrus, BA 47, Z ¼ 4.15, p ¼ 0.00002) and the left middle frontal gyrus (BA10, Z ¼ 2.86, p ¼ 0.00214); bilateral occipital lobes (fusiform gyrus, BA 19; Z ¼ 4.13, p ¼ 0.00002; lingual gyrus; BA18, Z ¼ 3.19, p ¼ 0.00071), and the right parietal lobe (precuneus, BA 7 and 31, Z ¼ 3.14, p ¼ 0.00084). The negative correlation of discourse outcome measure (abstracted core meaning) and perfusion values in the left frontal lobe region identified by covariate analysis was R2 ¼ 0.962, p ¼ 0.0000180, and correlation of discourse with the region identified in the occipital lobes was R2 ¼ 0.816, p ¼ 0.00208 (see Figs. 5 and 6 for elaboration). In other words, poorer
180 110
110 100
100
90
90
rCBF
rCBF
R2 = 0.7086
80 70
R2 = 0.9616
80 70
60 0
5 10 discourse score
15
Fig. 4. Correlation of discourse performance and rCBF in the right frontal lobe region of interest identified by covariate analysis.
60 0
5 10 discourse score
15
Fig. 6. Correlation of discourse performance and rCBF in the left frontal lobe region of interest identified by covariate analysis.
R2 ¼ 0.0673, and the relation between discourse score and injury severity was R2 ¼ 0.3537. Owing to the small sample size, we wanted to ensure that the significant relationships between perfusion and discourse outcomes could not be attributed to extreme values of an outlier. Therefore, we reapplied regression analysis after removing one outlier who exhibited extremely low frontal perfusion, extremely high occipital perfusion, and the lowest discourse lesson score of the group. The correlation between discourse and resting rCBF remained significant in the right frontal lobe (R2 ¼ 0.664, p ¼ 0.0255).
Discussion
Fig. 5. Regions showing a negative covariance between rCBF and discourse abstraction performance. See Plate 11.5 in Colour Plate Section.
discourse recovery at 3 years postinjury was associated with greater perfusion at rest in these left frontal, bilateral occipital, and right parietal areas, whereas higher discourse outcome was associated with less perfusion in the same brain regions. Since the variables of age and severity of injury may contribute to the brain-discourse results delineated above, we assessed the correlation between these variables and discourse scores. The relation between discourse score and age was
This study provides evidence for brain plasticity in childhood through measurements conducted 3 years after sustaining a severe TBI. Two important findings emerged from this study. The first significant finding was the long-term outcome of widespread hyperperfusion in children after TBI relative to typically developing children detected by at-rest SPECT studies. The second was disparate brain perfusion patterns in SPECT in good versus poor discourse recovery. With regard to the first finding, two aspects of the distinct perfusion patterns in the TBI group were unexpected. First, the immense heterogeneity among brain injuries (e.g., size and site of lesion, etiology) could significantly diminish the chances of finding any general pattern distinguishing the
181
TBI group from the group of normally developing children. Despite the low likelihood of quantifiable differences in SPECT, a distinct pattern emerged that distinguished the two groups as a whole. The consistent differences between the control and TBI groups may be due to a general effect of diffuse axonal shearing and damage from the calivarium (Graham, 1996). Shearing of white matter is a common mechanism of injury in trauma which may not be easily detectible at time of injury (Bigler, 1999). Moreover, shearing may influence longterm brain reorganizing of the frontal–striatal–temporal-cerebellum neural networks. Emerging research will shed light on the reorganization of specific networks by employing new brain imaging technology such as diffusion tensor imaging. These new studies will require long-term follow-up after injury to more fully understand the limits and potential of plasticity over time in the developing brain. Second, counter to expectations, the children with TBI showed brain patterns of increased perfusion relative to normally developing children. The focal regions of hyperperfusion at rest included the left medial frontal cortex, right globus pallidus–putamen–thalamus complex, and right visual cortex. In addition, the children with TBI demonstrated only one area of hypoperfusion relative to the healthy controls, that is, the cerebellum. We anticipated that if a consistent pattern emerged, the TBI group would demonstrate reduced perfusion, since in adults, the most commonly reported pattern has been bilateral areas of diminished brain blood flow after TBI (Bonne et al., 2003). Owing to the exploratory nature of this study, we can only speculate as to why the brain reorganized in such a way to have increased perfusion. We offer three possible explanations. First, hyperperfusion may reflect an adaptive strategy of increased recruitment of brain regions after brain injury. In other words, the heightened perfusion may indicate compensation as behavioral functions are displaced to brain regions other than the injured brain regions. Second, hyperperfusion may represent maladaptive changes of greater or inefficient effort. The pattern of greater resting perfusion than healthy controls in the left medial frontal
region and less in the cerebellum is consistent with fMRI findings of increased left medial frontal activation (Schulz et al., 2004) and MRI findings of cerebral atrophy (Montes, 2005) in Attention Deficit-Hyperactivity Disorder, which may reflect a disruption in inhibitory mechanisms (Kaufmann et al., 1993; Max et al., 2004). In both of the first two explanations, the greater areas of hyperperfusion may occur due to a brain that is working harder or working inefficiently. Third, hyperperfusion may indicate that the children with TBI are delayed in their developmental stage. The hyperperfusion of the children with TBI may be similar to healthy yet younger children, as the brain injury may have caused a lag in the development of their brain function (Chugani et al., 1991). Young children with immature performance on complex cognitive tasks have demonstrated hemispheric and localization distinctions from older and more successful children in FMRI activation studies (Bunge et al., 2002; Moses et al., 2002). Future studies using fMRI and cognitive activation paradigms to mark different stages of recovery will advance our limited understanding of increased and decreased perfusion in the recovering and developing brain. Evidence derived from the second aim of this study and discussed below provides some clarification that brain reorganization years after pediatric TBI may reflect both adaptive and maladaptive changes. As pointed out earlier, the second key finding of the current study was that greater resting perfusion of right frontal and temporal regions was positively associated with better discourse outcomes on complex tasks requiring abstraction of the core meaning in children with severe TBI. This brain–behavior relation is consistent with work by Nichelli et al. (1995) and Robertson et al. (2000) who found right frontal activation using PET and fMRI when healthy adult participants abstracted meaning from discourse. Similarly, using PET studies and discourse outcome measures of coherence, Postman and colleagues (2006) identified a significant relationship between right frontal regions and higher discourse coherence scores. Additionally, impaired discourse abstraction was significantly related to reduced perfusion on SPECT in an adult neurogenic population in the
182
early stages of a frontal lobe dementia (Chapman et al., 2005). In one study of elementary-aged children, discourse coherence seems to utilize the right hemisphere whereas processing detailed information may use the left hemisphere (Dapretto et al., 2005). An intriguing new brain–behavior relation identified in this current study was increased perfusion of left frontal regions at rest associated with poorer ability to abstract the core meaning of a narrative in the form of a lesson. Several issues make us interpret the SPECT perfusion and discourse patterns cautiously. First, this SPECT study scanned participants in a resting state so discourse cannot be directly associated with the resulting perfusion patterns. That is, any association between discourse and perfusion is speculative. Nonetheless, we believe the brain-discourse patterns warrant further study not only because of supporting literature but also because we were careful to consider the skewing of results due to a few patients. As cited above, the association of complex discourse processing with right-sided perfusion in the present study is consistent with previous findings of right hemisphere activation with abstract processing of language (Nichelli et al., 1995; Chapman et al., 2005; Postman et al., 2006). Second, a few participants within the small sample size may skew the results. For example, the poor discourse performance in a few participants with intense localized initial trauma may confound the group trend. To control for this possibility, the SPECT images of each individual were visually inspected and it was found that the brains displayed no region of extreme perfusion similar to discrete results. In addition, removal of an outlier with extreme perfusion values and extremely poor performance still resulted in significant covariate results. Regions identified by whole brain analysis appeared to be related to the discourse outcome and not idiosyncrasies within the TBI group. We speculate that children who show better recovery of discourse abstraction skills after brain injury are able to recruit the right frontal lobe during the course of recovery to support subsequent development of these more complex discourse skills that emerge in early adolescence. Whether or not the right frontal areas were less disrupted functionally with the initial injury
requires further study by employing functional brain imaging scans early and later after injury. Conversely, we have previously speculated that children who show poor recovery of discourse abstracting abilities may fail to recruit right frontal lobe functions. Rather, they may be stuck at lower levels of development, recalling more of the details from the discourse sample but being unable to abstract the core meaning (Cook et al., 2006). The push to recall information at a detail level after TBI may engage the left frontal region or even more temporal regions at the cost of failing to engage and recruit right frontal regions. In other words, perhaps the greater blood flow in the left frontal lobe of children with poor recovery occurs as a result of over-reliance, or working hard, to process the sequential, lower-level elements of language (i.e., focusing on isolated details) purported to be a supported by left neural networks in general and left frontal networks specifically (Robertson et al., 2000; Dapretto et al., 2005). Children who focus on details often fail to integrate the detailed information into a condensed, abstract meaning (e.g., produce a story lesson). The increased left frontal activity 3 years after injury may impede children’s ability to engage in abstract processing ascribed to right hemispheric brain networks (Chapman et al., 2006b). Integration of the two significant findings from this study suggests that some apparent brain changes since injury may be adaptive, reflecting plasticity towards positive recovery outcomes, whereas other changes may be maladaptive. Greater perfusion should not be assumed to be better or worse than less perfusion. In fact, brain changes in the children with TBI show dramatic hyperperfusion that indicates plasticity as regions are reconnected after injury — however, discourse performance indicates inefficient function and maladaptive plasticity. For example, in the children with worse discourse abstraction ability, certain brain areas (left frontal lobe and visual cortex) were hyperperfused when compared to normally developing controls and perfusion in these same areas were negatively correlated with discourse scores. In these particular scenarios, the brain perfusion patterns may signal poor compensatory mechanisms within a complex neural network.
183
Furthermore, other children with TBI in this study appeared to successfully compensate for their injury, at least on our discourse outcomes and expected brain patterns. A natural follow-up question to this study is, what internal or external factors are driving the stronger recovery and what appears to be beneficial, adaptive plasticity? Can plasticity be influenced by specialized intervention after brain injury? Preliminary results of an fMRI pre and postintervention study demonstrated that discourse outcomes can be improved and brain function can be altered to engage right frontal regions in children with severe TBI given short-term intensive intervention to abstract core meaning of discourse (Chapman et al., 2001a). In summary, this is one of the first studies to elucidate brain changes and plasticity in children evident 3 years after severe brain injury using complex language measures and functional brain imaging. Our study found regions of hyperperfusion relative to normally developing children. This is in contrast to a pattern of hypoperfusion identified in adults years after TBI (Ichise et al., 1994; Kant et al., 1997; Bonne et al., 2003). Furthermore, children with good and poor discourse recovery showed disparate brain perfusion profiles. These results motivate the emerging emphasis in the field of neurocognitive rehabilitation on employing functional brain imaging. Imaging may help elucidate the brain’s capacity to reorganize on its own and in response to therapies, particularly at later stages postinjury. Studies are needed to determine whether stage-by-stage discourse intervention, in particular, can help mitigate the detrimental and later-emerging effects of disrupted frontal function (Giza et al., 2005). Evidence from neuroscience reveals that brain plasticity remains favorable given intensive therapy at appropriate levels of complexity throughout the life span (Kilgard et al., 2002). Approaches to pediatric brain injury must meld science and treatment in an effort to develop more appropriate interventions to prevent later emerging deficits after brain injury in childhood. Further investigation of the potential to impact brain plasticity in the developing yet injured brain will help fulfill the growing mandate for translational research.
Abbreviations rCBF SPECT TBI
relative cerebral blood perfusion; single photon emission computed tomography; traumatic brain injury.
Acknowledgments This work is supported by a grant from the Kent Waldrep National Paralysis Foundation. The authors are grateful for scientific insight and practical support from Harvey S. Levin, Nicholas Furl, William Cooper, Thomas Harris, Dianne Altuna, Jack Scott, Emily Tobey, Lyn McKinnon, and Zerrin Yetkin. We also express our appreciation for our important collaborations with Dr. Frank McDonald and Our Children’s House at Baylor hospital in Dallas. References Anderson, C.V. and Bigler, E.D. (1995) Ventricular dilation, cortical atrophy, and neuropsychological outcome following traumatic brain injury. J. Neuropsychiatry Clin. Neurosci., 7: 42–48. Bergsneider, M., Hovda, D.A., McArthur, D.L., Etchepare, M., Huang, S.C., Sehati, N., Satz, P., Phelps, M.E. and Becker, D.P. (2001) Metabolic recovery following human traumatic brain injury based on FDG-PET: time course and relationship to neurological disability. J. Head Trauma Rehabil., 16: 135–148. Biddle, K.R., McCabe, A. and Bliss, L.S. (1996) Narrative skills following traumatic brain injury in children and adults. J. Commun. Disord., 29: 447–469. Bigler, E.D. (1996) Brain imaging and behavioral outcome in traumatic brain injury. J. Learn Disabil., 29: 515–530. Bigler, E.D. (1999) Neuroimaging in pediatric traumatic head injury: diagnostic considerations and relationships to neurobehavioral outcome. J. Head Trauma Rehabil., 14: 406–423. Bonne, O., Gilboa, A., Louzoun, Y., Kempf-Sherf, O., Katz, M., Fishman, Y., Ben Nahum, Z., Krausz, Y., Bocher, M., Lester, H., Chisin, R. and Lerer, B. (2003) Cerebral blood flow in chronic symptomatic mild traumatic brain injury. Psychiatry Res.-Neuroimaging, 124: 141–152. Brookshire, B.L., Chapman, S.B., Song, J. and Levin, H.S. (2000) Cognitive and linguistic correlates of children’s discourse after closed head injury: a three-year follow-up. J. Int. Neuropsychol. Soc., 6: 741–751.
184 Bunge, S.A., Dudukovic, N.M., Homason, M.E., Vaidya, C.J. and Gabrieli, J. (2002) Immature frontal lobe contributions to cognitive control in children: evidence from fMRI. Neuron, 33: 301–311. Casey, B.J., Tottenham, N., Liston, C. and Durston, S. (2005) Imaging the developing brain: what have we learned about cognitive development? Trends Cogn. Sci., 9: 104–110. Chang, L.T. (1978) A method for attenuation correction in radionuclide computed tomography. IEEE Trans. Nucl. Sci., 25: 638–643. Chapman, S.B., Bonte, F.J., Wong, S.B.C., Zientz, J.N., Hynan, L.S., Harris, T.S., Gorman, A.R., Roney, C.A. and Lipton, A.M. (2005) Convergence of connected language and SPECT in variants of frontotemporal lobar degeneration. Alzheimer Dis. Assoc. Disord., 19: 202–213. Chapman, S.B., Gamino, J.F., Cook, L.G., Hanten, G. and Levin, H.S. (2006a) Discourse macrolevel production and memory processes after pediatric traumatic brain injury in children. Brain Lang., 97: 178–188. Chapman, S.B., Gamino, J.F., Cook, L.G., Hanten, G., Li, X. and Levin, H.S. (2006b) Impaired discourse gist and working memory in children after brain injury. Brain Lang., 97: 178–188. Chapman, S.B., Levin, H.S. and Lawyer, S.L. (1999) Communication problems resulting from brain injury in children: special issues of assessment and management. In: McDonald, S., Togher, L. and Code, C. (Eds.), Communication Disorders Following Traumatic Brain Injury (Brain Damage, Behaviour and Cognition Series). Taylor & Francis, Hove, England, pp. 235–270. Chapman, S.B., Levin, H.S., Wanek, A., Weyrauch, J. and Kufera, J.A. (1998) Discourse after closed head injury in young children. Brain Lang., 61: 420–449. Chapman, S.B., Max, J.E., Gamino, J.F., McGlothlin, J.H. and Cliff, S.N. (2003) Discourse plasticity in children after stroke: age at injury and lesion effects. Pediatr. Neurol., 29: 34–41. Chapman, S.B., McKinnon, L., Levin, H.S., Chiu, S.B., Sparks, G., Yetkin, Z. and Devous, M.DSr (2001a) Brain reorganization in traumatic brain injury. Presentation at the annual American Association for Advancement of Science meeting, San Francisco. Chapman, S.B., McKinnon, L., Levin, H.S., Song, J., Meier, M.C. and Chiu, S.B. (2001b) Longitudinal outcome of verbal discourse in children with traumatic brain injury: three-year follow-up. J. Head Trauma Rehabil., 16: 441–455. Chapman, S.B., Sparks, G., Levin, H.S., Dennis, M., Roncadin, C., Zhang, L. and Song, J. (2004) Discourse macrolevel processing after severe pediatric traumatic brain injury. Dev. Neuropsychol., 25: 37–60. Chapman, S.B., Watkins, R., Gustafson, C., Moore, S., Levin, H.S. and Kufera, J.A. (1997) Narrative discourse in children with closed head injury, children with language impairment, and typically developing children. Am. J. Speech-Lang. Pathol., 6: 66–75. Christensen, J.R. (2001) What is traumatic brain injury? In: Schoenbrodt, L. (Ed.), Children with Traumatic Brain Injury: A Parent’s Guide. Woodbine House, New York.
Chugani, H.T., Hovda, D.A., Villablanca, J.R., Phelps, M.E. and Xu, W.F. (1991) Metabolic maturation of the brain — a study of local cerebral glucose-utilization in the developing cat. J. Cereb. Blood Flow Metab., 11: 35–47. Chugani, H.T., Muller, R.A. and Chugani, D.C. (1996) Functional brain reorganization in children. Brain Dev., 18: 347–356. Chugani, H.T., Phelps, M.E. and Mazziotta, J.C. (1987) Positron emission tomography study of human-brain functionaldevelopment. Ann. Neurol., 22: 487–497. Cook, L.G., Chapman, S.B. and Gamino, J.F. (2006) Discourse gist in pediatric brain injury: missing the forest for the trees (in press). In: Cain K. and Oakhill J. (Eds.), Cognitive Bases of Children’s Language Comprehension Difficulties., Guilford Publications, Inc., New York. Dapretto, M., Lee, S.S. and Caplan, R. (2005) A functional magnetic resonance imaging study of discourse coherence in typically developing children. Neuroreport, 16: 1661–1665. Delis, D.C., Kramer, J.H., Friedlund, A.J. and Kaplan, E. (1990) A cognitive science approach to neuropsychological assessment. In: McReynolds, P., Rosen, J.C. and Chelune, G.J. (Eds.), Advances in Psychological Assessment. Plunum Press, New York, pp. 101–132. Ewing-Cobbs, L., Levin, H.S., Eisenberg, H.M. and Fletcher, J.M. (1987) Language functions following closed-head injury in children and adolescents. J. Clin. Exp. Neuropsychol., 9: 575–592. Fletcher, J.M., Miner, M.E. and Ewing-Cobbs, L. (1987) Age and recovery from head injury in children: developmental issues. In: Levin, H.S., Grafman, J. and Eisenberg, H.M. (Eds.), Neurobehavioral Recovery from Head Injury. Oxford University Press, New York, pp. 279–292. Friston, K.J., Holmes, A.P., Worsley, K.J., Poline, J.P., Frith, C. and Frackowiack, R. (1995) Statistical parametric maps in functional imaging: a general linear approach. Hum. Brain Mapp., 2: 189–210. Giedd, J.N., Blumenthal, J., Jeffries, N.O., Castellanos, F.X., Liu, H., Zijdenbos, A., Paus, T., Evans, A.C. and Rapoport, J.L. (1999) Brain development during childhood and adolescence: a longitudinal MRI study. Nat. Neurosci., 2: 861–863. Giza, C.C., Griesbach, G.S. and Hovda, D.A. (2005) Experience-dependent behavioral plasticity is disturbed following traumatic injury to the immature brain. Behav. Brain Res., 157: 11–22. Goldenberg, G., Oder, W., Spatt, K. and Podreka, I. (1992) Cerebral correlates of disturbed executive function and memory in survivors of severe closed head injury: a SPECT study. J. Neurol., Neurosurg. Psychiatry, 55: 362–368. Graham, D.I. (1996) Neuropathology of head injury. In: Narayan, R., Wilberger, J.J. and Povlishock, J.T. (Eds.), Neurotrauma. McGraw-Hill, New York, pp. 43–59. Gross, H., Kling, A., Henry, G., Herndon, C. and Lavretsky, H. (1996) Local cerebral glucose metabolism in patients with long-term behavioral and cognitive deficits following mild traumatic brain injury. J. Neuropsychiatry Clin. Neurosci., 8: 324–334.
185 Gullion, C.M., Devous, M.D. and Rush, A.J. (1996) Effects of four normalizing methods on data analytic results in functional brain imaging. Biol. Psychiatry, 40: 1106–1121. Ichise, M., Chung, D.G., Wang, P., Wortzman, G., Gray, B.G. and Franks, W. (1994) Technetium-99M-Hmpao SPECT, CT and MRI in the evaluation of patients with chronic traumatic brain injury — a correlation with neuropsychological performance. J. Nucl. Med., 35: 217–226. Jacobs, A., Put, E., Ingels, M. and Bossuyt, A. (1994) Prospective evaluation of Technetium-99M-Hmpao Spect in mild and moderate traumatic brain injury. J. Nucl. Med., 35: 942–947. Jaffe, K.M., Polissar, N.L., Fay, G.C. and Liao, S.Q. (1995) Recovery trends over three years following pediatric traumatic brain injury. Arch. Phys. Med. Rehabil., 76: 17–26. Johnson, N.S. (1983) What do you do if you can tell the whole story? The development of summarization skills. In: Nelson, K.E. (Ed.), Children’s Language. Gardner Press, New York, pp. 314–383. Kant, R., Smith-Seemiller, L., Isaac, G. and Duffy, J. (1997) Tc-HMPAO SPECT in persistent postconcussion syndrome after mild head injury: comparison with MRI/CT. Brain Injury, 11: 115–124. Kaufmann, P.M., Fletcher, J.M., Levin, H.S., Miner, M.E. and Ewing-Cobbs, L. (1993) Attentional disturbance after pediatric closed-head injury. J. Child Neurol., 8: 348–353. Kennard, M.A. (1938) Reorganization of motor function in the cerebral cortex of monkeys deprived of motor and premotor areas in infancy. J. Neurophysiol., 1: 477–496. Kilgard, M.P., Pandya, P.K., Engineer, N.D. and Moucha, R. (2002) Cortical network reorganization guided by sensory input features. Biol. Cybern., 87: 333–343. Kurth, S.M., Bigler, E.D. and Blatter, D.D. (1994) Neuropsychological outcome and quantitative image-analysis of acute hemorrhage in traumatic brain injury — preliminary findings. Brain Injury, 8: 489–500. Langlois, J.A. (2001) Traumatic brain injury in the United States: assessing outcomes in children. National Center for Injury. Atlanta, Prevention and Control of the Centers for Disease Control and Prevention. Pamphlet. Levin, H.S., Culhane, K.A., Mendelsohn, D., Lilly, M.A., Bruce, D., Fletcher, J.M., Chapman, S.B., Harward, H. and Eisenberg, H.M. (1993) Cognitive function in relation to MRI findings following closed-head injury in children and adolescents. J. Clin. Exp. Neuropsychol., 15: 21. Max, J.E., Lansing, A.E., Koele, S.L., Castillo, C.S., Bokura, H., Schachar, R., Collings, N. and Williams, K.E. (2004) Attention deficit hyperactivity disorder in children and adolescents following traumatic brain injury. Dev. Neuropsychol., 25: 159–177.
Mendelsohn, D., Levin, H.S., Bruce, D., Lilly, M.A., Harward, H., Culhane, K.A. and Eisenberg, H.M. (1992) Late MRI after head injury in children: relationship to clinical features and outcome. Child’s Nervous System, 8: 445–452. Montes, L.G.A. (2005) Anatomic and functional alterations in attention — deficit/hyperactivity disorder. Salud Mental, 28: 1–12. Moses, P., Roe, K., Buxton, R.B., Wong, E.C., Frank, L.R. and Stiles, J. (2002) Functional MRI of global and local processing in children. Neuroimage, 16: 415–424. Nichelli, P., Grafman, J., Pietrini, P., Clark, K., Lee, K.Y. and Miletech, R. (1995) Where the brain appreciates the moral of a story. Neuroreport, 6: 2309–2313. Postman, W.A., Solomon, J., Maisog, J., Chapman, S.B., Tuttle, S., Christian, M.R., Milosky, L. and Braun, A. (2006) Deconstructing discourse: a PET study of narrative production. Presentation at CUNY Sentence Processing Conference, New York. Reilly, J.S., Bates, E.A. and Marchman, V.A. (1998) Narrative discourse in children with early focal brain injury. Brain Lang., 51: 335–375. Robertson, D.A., Gernsbacher, M.A., Guidotti, S.J., Robertson, R.R.W., Irwin, W., Mock, B.J. and Campana, M.E. (2000) Functional neuroanatomy of the cognitive process of mapping during discourse comprehension. Psychol. Sci., 11: 255–260. Schulz, K.P., Fan, J., Tang, C.Y., Newcorn, J.H., Buchsbaum, M.S., Cheung, A.M. and Halperin, J.M. (2004) Response inhibition in adolescents diagnosed with attention deficit hyperactivity disorder during childhood: an event-related fMRI study. Am. J. Psychiatry, 161: 1650–1657. Stein, B.L. and Kirby, J.R. (1992) The effects of text absent and text present conditions on summarization and recall of text. J. Reading Behav., 24: 217–232. Talairach, J. and Tournoux, P. (1988) Co-Planar Stereotaxic Atlas of the Human Brain. Thieme Medical Publishers, New York. Turkington, T.G., Jaszczak, R.J., Pelizzari, C.A., Harris, C.C., MacFall, J.R., Hoffman, J.M. and Coleman, R.E. (1993) Accuracy of registration of PET, SPECT and MR images of a brain phantom. J. Nucl. Med., 34: 1587–1594. Umile, E.M., Plotkin, R.C. and Sandel, M.E. (1998) Functional assessment of mild traumatic brain injury using SPECT and neuropsychological testing. Brain Injury, 12: 577–594. Yorkston, K.M., Jaffe, K.M., Polissar, N.L., Liao, S.Q. and Fay, G.C. (1997) Written language production and neuropsychological function in children with traumatic brain injury. Arch. Phys. Med. Rehabil., 78: 1096–1102.
Plate 11.1. T-test results showing regions of increased rCBF in injured group relative to normally developing controls, superimposed onto MRI transverse slices.
Plate 11.3. Regions showing a positive covariance between rCBF and discourse abstraction performance.
Plate 11.5. Regions showing a negative covariance between rCBF and discourse abstraction performance.
Plate 12.1. Example of different stage progenitors in the DG of the mature mouse hippocampus. Letter A labels a GFP (green) and GFAP (blue)-expressing neural stem cell in the subgranular zone of a nestin-GFP transgenic mouse (Type I cell) (Yu et al., 2005). Letter B marks a GFP+, GFAP–, DCX– progenitor (Type II) and C labels DCX+(red), GFP-, GFAP- maturing progenitor (Type III). Arrow marks GFAP positive process from GFP-expressing cell. GL ¼ granular layer; SGL ¼ subgranular layer. Scale bar ¼ 35 mM.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 12
Activation of neural stem and progenitor cells after brain injury Darry K. Miles and Steven G. Kernie Department of Pediatrics and Center for Developmental Biology, U.T. Southwestern Medical Center, Dallas, TX 75390-9133, USA
Abstract: Neural stem and progenitor cells in the mammalian brain persist and are functional well into adulthood. Reservoirs for these cells are found in both the subventricular zone and the dentate gyrus of the hippocampus. It is still unclear what role these cells may play in humans during normal brain maturation. In addition, there is currently tremendous speculation regarding the potential role of these cells in providing a substrate for recovery and repair after injury. This review provides an overview of the existing data regarding how neural stem and progenitor cells respond to various types of brain injury. In particular, we focus upon their role in the dentate gyrus since this brain area provides a compelling and tractable model of how the brain may use its ability for endogenous regeneration to recover from a variety of injuries. Keywords: neurogenesis; brain injury; hippocampus; dentate gyrus; subventricular zone; transgenic mice; nestin are an area of intense investigation and may include changes in dendritic arborization, spine density, and synaptogenesis (Hoff, 1986; Stroemer et al., 1998; Keyvani and Schallert, 2002). Indeed, regeneration of synaptic connections in hippocampal CA1 pyramidal neurons after experimental traumatic brain injury (TBI) can occur 10–14 days after injury and persist for several months (Scheff et al., 2005). Moreover, the formation of new synapses may also be associated with functional recovery (Ding et al., 2002). In addition to the modification of existing circuitry that occurs through expression of neural plasticity, we now realize that distinct populations of neural stem and progenitor cells exist and continually add functional neurons to the brain throughout the lifespans of humans, birds, and rodents (Alvarez-Buylla and Nottebohm, 1988; Gage, 1998; Garcia-Verdugo et al., 1998; Johansson et al., 1999; Carlen et al., 2002). The generation of new neurons in the adult brain may
Brain injury and plasticity The mammalian nervous system has the unique task of functioning within a setting of relatively static circuitry while maintaining its ability to be plastic, learn, and adapt in response to its environment. Therefore, both the developing and the mature brain contain inherent mechanisms that allow for the continual modification of neural connections. The brain’s endogenous reparative mechanisms after brain injury may utilize similar plastic abilities to restore some of its impaired function. It is well known that patients who suffer neurologic injury have the ability to partially restore function after injury (Ewing-Cobbs et al., 1997; Anderson et al., 2000; Catroppa and Anderson, 2005). The mechanisms that underlie this recovery Corresponding author. Tel.: +1 214 648 4183; Fax: +1 214 648 1960; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57012-8
187
188
comprise a novel mechanism of brain plasticity and this process may offer a method by which the brain repairs itself.
Neural stem and progenitor cells in the adult brain The discovery of multipotent neural progenitor cells has raised recent debate as to the nature and definition of a brain stem cell (Seaberg and van der Kooy, 2003). Adult neural stem cells constitute a true stem cell in that they maintain the ability for unlimited self-renewal and multipotentiality forming neurons, astrocytes and oligodendrocytes (Gage, 1998; Doetsch et al., 1999; Kukekov et al., 1999). The original observations of this phenomenon utilized 3H-thymidine to label dividing cells which demonstrated that precursors cells proliferate in vivo and give rise to mature neurons in the rodent hippocampus and olfactory bulb (Altman and Das, 1965; Kaplan and Hinds, 1977; Bayer et al., 1982). This was later substantiated by in vitro isolation of sphere-forming cells from the adult subventricular zone (SVZ) that could be propagated with epidermal growth factor (EGF) and fibroblast growth factor (FGF) and undergo differentiation into the major cell types in the brain (Reynolds and Weiss, 1992; Richards et al., 1992; Morshead et al., 1994). Additional studies support these initial observations and it is now generally well accepted that the adult brain contains a discrete reservoir of stem cells and that neurogenesis occurs throughout the life of many species including humans (Eriksson et al., 1998; Kempermann et al., 1998; Gage, 2000; Kornack and Rakic, 2001; Alvarez-Buylla et al., 2002; Rakic, 2002; van Praag et al., 2002). Collections of these cells occur in two locations: the subventricular zone (SVZ) and the dentate gyrus (DG). The subventricular zone Two germinal layers, the ventricular zone and subventricular zone (SVZ) give rise to the majority of neurons and glia in the mammalian forebrain (Boulder Committee, 1970). During embryogenesis the ventricular zone involutes while the subventricular zone persists into adulthood
comprising the first several cell layers underneath the ependymal cell lining of the ventricular cavity (Garcia-Verdugo et al., 1998). Early in postnatal life, cells from the SVZ or subependymal layer provide glial progenitor cells that migrate and form oligodendrocytes and astrocytes in the subcortical white matter, striatum, and neocortex (Hardy and Reynolds, 1991; Levison and Goldman, 1993). In addition to providing glial progenitors, stem cells within the anterior portion of the SVZ migrate rostrally to form granular and periglomerular cell interneurons in the olfactory bulb in the rodent but not in humans (Luskin, 1993; Lois and Alvarez-Buylla, 1994; Menezes et al., 1995; Carleton et al., 2003). The SVZ contains slowly dividing astrocyte-like type B cells that express glial fibrillary acidic protein (GFAP) and the intermediate filament nestin which give rise to transition amplifier type C cells and migratory neuroblast type A cells (GarciaVerdugo et al., 1998; Doetsch et al., 1999). Type E cells represent the ependymal layer and maintain close contact with type B cells but do not proliferate in vivo. Supportive evidence that type B cells are the stem cells within the SVZ came from the observation that, after Ara-C ablation of proliferating type A and type C cells, residual type B cells can reconstitute all cells types within the SVZ (Doetsch et al., 1999). Furthermore, in vivo and in vitro ablation of GFAP-expressing cells using a ganciclovir-mediated inducible transgenic mouse model stops the formation of neural stem cellforming neurospheres (Imura et al., 2003; Morshead et al., 2003). Type A neuroblast cells express developmental markers involved in neuronal migration such as doublecortin (DCX) and polysialyated neural cell adhesion molecule (PSA-NCAM). After their generation in the SVZ, immature neuroblasts undergo chainlike migration ensheathed by astrocytes along a restricted pathway termed the rostral migratory stream to the olfactory bulb (Lois et al., 1996; Alvarez-Buylla et al., 2002). Upon arrival to the olfactory bulb, 50–60% of newly born cells will undergo apoptosis over the first 2 months (Winner et al., 2002). The majority of surviving cells integrate into the existing circuitry, forming GABAergic granular and periglomerular neurons
189
(Carlen et al., 2002). Although it is unclear whether this stereotypic sequence of events occurs in humans, cells cultured from the human SVZ are capable of forming astrocytes, oligodendrocytes, and neurons in vitro (Kornack and Rakic, 2001; Sanai et al., 2004; Quinones-Hinojosa et al., 2006;). Moreover, the ability to expand and maintain SVZ stem cells in culture may facilitate usage in transplantation to augment recovery in the damaged or degenerating brain. A recent review discusses the role of the SVZ in cell replacement and brain repair (Romanko et al., 2004a). The dentate gyrus Neurogenesis occurs in the hippocampus of rodents as well as primates and humans (Altman and Das, 1965; Kaplan and Bell, 1984; Kuhn et al., 1996; Eriksson et al., 1998; Gage, 1998; Kornack and Rakic, 1999). The dentate gyrus (DG) receives the primary inputs to the hippocampus via the enterorhinal cortex and it consists of three layers: the principal or granule cell layer (GCL), the largely acellular molecular layer, and the polymorphic cell layer or hilus (Johnston and Amaral, 1998). In the adult hippocampus, precursor cells reside in the subgranular layer (SGL), the border between the granular cell layer and the hilus. In rodents, neurogenesis in the hippocampus occurs primarily postnatally with approximately 85% of its neurons produced during the first 2 weeks after birth (Bayer, 1980). The rate of neurogenesis in adulthood is relatively stable but likely decreases in senescence and may be influenced by many factors (Kuhn et al., 1996; Kempermann et al., 1998; Kornack and Rakic, 1999; Seki, 2002). It is estimated that the adult rodent produces approximately 8,000–12,000 new cells in the SGL everyday, however 450% of new immature precursor cells will die in the first month resulting in the addition of an estimated 2,000–4,000 new neurons daily to the GCL (Kempermann et al., 1997a; Cameron and McKay, 2001, Dayer et al., 2003). It is unclear what role these newly generated neurons play in hippocampal function. One possibility is that new granular cell neurons represent a mechanism for plasticity that may be associated with learning and memory, a function long believed
to reside in the hippocampus (Scoville and Milner, 1957; Zola-Morgan et al., 1986). The process of neurogenesis is highly organized and includes multiple regulatory steps to direct proliferation, migration, differentiation, and survival (Abrous et al., 2005). Similar to the SVZ, the subgranular layer of the DG contains a population of GFAPpositive astrocyte-like cells that express nestin and are believed to be neural stem cells (Imura et al., 2003). These GFAP-expressing cells are thought to undergo asymmetric division and give rise to early neural progenitor cells that early after birth express DCX and PSA-NCAM (Brown et al., 2003; Rao and Shetty, 2004). Although there is overlap in the expression of immature markers, by 2–3 weeks cells begin to express the mature neuronal markers microtubule associated protein (MAP-2) and calbindin (Cameron and McKay, 2001). Mature glutaminergic granular neurons extend dendritic projections that synapse in the molecular layer and send their axonal projections to the CA3 region via mossy fibers (Stanfield and Trice, 1988; Markakis and Gage, 1999; Carlen et al., 2002). Recently, experiments with transgenic mice have assisted in the description of the stem and progenitor cell population within the DG. A common feature of neural precursor cells is the expression of the intermediate neurofilament nestin. Transgenic mice that express green fluorescent protein (GFP) under the control of the neural specific form of the nestin promoter allow visualization of neural progenitor population in developing and adult animals (Yamaguchi et al., 2000; Kawaguchi et al., 2001; Yu et al., 2005). Characterization of GFP expression has yielded distinct populations within the SGL classified as type 1, type 2, and type 3 cells. Type 1 cells express high levels of nestin-GFP, exhibit glial properties, and express GFAP (Doetsch et al., 1999). Type 2 cells remain GFP-positive but lack GFAP expression and can be subdivided into 2a cells that have not acquired expression of immature neuronal markers, and type 2b cells that express DCX and PSA-NCAM (Kronenberg et al., 2003). Type 3 cells are nestin-GFP negative and DCX or PSA-NCAM positive (Fig. 1). Interestingly, type 1 cells have both the electrophysiologic properties and the vascular end feet that are typical of astrocytes. In addition, they
190
Fig. 1. Example of different stage progenitors in the DG of the mature mouse hippocampus. Letter A labels a GFP (green) and GFAP (blue)-expressing neural stem cell in the subgranular zone of a nestin-GFP transgenic mouse (Type I cell) (Yu et al., 2005). Letter B marks a GFP+, GFAP–, DCX– progenitor (Type II) and C labels DCX+(red), GFP-, GFAP- maturing progenitor (Type III). Arrow marks GFAP positive process from GFP-expressing cell. GL ¼ granular layer; SGL ¼ subgranular layer. Scale bar ¼ 35 mM. See Plate 12.1 in Colour Plate Section.
extend long radial glia-like projections outward throughout the granular layer toward the molecular layer (Filippov et al., 2003; Fukuda et al., 2003). Type 2 cells demonstrate sodium current properties indicative of immature neurons. Recently, GABAergic input from the enterorhinal cortex was shown to induce the expression of NeuroD, a bHLH transcription factor involved in the differentiation of neural progenitor cells (Tozuka et al., 2005).
Tracking neural stem and progenitor cells in vivo Limitations in studying neural stem cells include the elusivity of a definitive stem cell marker and the difficulty of fate mapping studies. Therefore, the majority of research on in vivo stem cell populations relies on the use of the halogenated-thymidine analog 50 -bromo-20 -deoxyuridine (BrdU), which is incorporated into dividing cells during S
phaseProgeny cells can then be assessed with immunohistochemical markers for mature cell phenotypes to determine cell fate (del Rio and Soriano, 1989). However, several concerns regarding this approach have been raised in studying neurogenesis (Cooper-Kuhn and Kuhn, 2002). When given during embryogenesis, BrdU has teratogenic effects including cerebellar and cortical dysgenesis, Purkinje cell migration defects, a reduction in litter size, poor weight gain, and decreased spatial learning tasks (Kolb et al., 1999; Sekerkova et al., 2004). BrdU may also be potentially incorporated into damaged postmitotic neurons that either re-enter the cell cycle or undergo DNA repair after injury. Although this has been shown to occur in some hippocampal neurons after hypoxic-ischemic injury, damaged cells that incorporate BrdU do not seem to survive for more than 4 weeks (Kuan et al., 2004). Recently, specific markers of cell proliferation such as Ki-67 and proliferating cell nuclear antigen (PCNA) have become available and can be used in combination with BrdU to evaluate cell mitosis (Cooper-Kuhn and Kuhn, 2002; Kee et al., 2002). An alternative approach used to study fate determination is retroviral labeling that selectively labels dividing cells and allows for long-term fate determination that is not subject to the dilutional effects attributed to BrdU (Goings et al., 2004). Unfortunately, viruses necessitate direct injection into the brain thus creating local trauma that may itself influence cell proliferation and differentiation. It therefore seems appropriate when evaluating neurogenesis to attempt to combine several methods to assist in confirming the presence of cell mitosis and accurately determining fate selection. There is still considerable disparity among studies on many aspects of injury-induced neural stem cell activation that include the age and species of animal analyzed, the method of brain injury, the BrdU dosing paradigm, specific phenotypic markers, and the length of fate determination.
Activation of neural stem and progenitor cells after injury Previous work has uncovered many physiologic factors that influence the production and survival
191
of new neurons in the DG. These factors include diverse stimuli such as age, exercise, environment, growth factors, and adrenal steroids (Kempermann et al., 1997b; Cameron et al., 1998). Many types of brain injury are now thought to stimulate neural precursor cells as well. A major area of recent debate and investigation is whether and to what extent neural progenitor cells participate in brain remodeling and regeneration after injury. Adult animal models of global and focal cerebral ischemia induce a 4–10-fold increase in proliferation in GCL cells in the DG 5–10 days after injury (Liu et al., 1998; Jin et al., 2001; Kee et al., 2001; Yagita et al., 2001; Tureyen et al., 2004). These studies report that approximately 25–50% of newly generated cells survive for up to 1 month and that 50–70% of surviving cells acquire mature neuronal markers. At least one study has proposed that injury also induces accelerated maturation with increased percentages of newly generating cells expressing immature neuronal markers (Kee et al., 2001). Several studies demonstrate that injury-induced new granular cell neurons appropriately extend axons to the CA3 region of the hippocampus after ischemia and TBI (Kee et al., 2001; Emery et al., 2005). In models of TBI a fivefold increase in proliferation is seen in the ipsilateral injured DG with 60% of BrdU-positive cells appearing to undergo normal maturation by cadlbindin expression and persistence for up to 2 months after injury (Dash et al., 2001; Kernie et al., 2001). In various models of unilateral brain injury the contralateral DG and SVZ appear to increase proliferation as well although to a lesser extent. (Dash et al., 2001; Jin et al., 2001; Kernie et al., 2001; Rice et al., 2003). Although many types of injury may stimulate proliferation in the DG, the function and consequence of this remains unclear. One possibility is that neural precursors are dividing to replace injured progenitors or mature granular neurons. However, injury to progenitors is not the only injury stimulus for neurogenesis. In fact, increased proliferation is seen in the subgranular layer (SGL) in response to injury with little or no evidence of cell death within the hippocampus or DG (Liu et al., 1998; Dash et al., 2001; Jin et al., 2001; Arvidsson et al., 2002). However, progenitor cells may retain the ability to regenerate adjacent cell
layers. Recently, extensive regeneration of granular cell neurons after timethyltin-mediated ablation was demonstrated to induce a return of GCL neurons and was associated with restoration of memory retention tasks (Ogita et al., 2005). In addition to neurogenesis being reported in the hippocampus after injury, activation of neural precursor cells within the SVZ has also been reported 1–2 weeks after brain injury (Zhang et al., 2001; Parent et al., 2002; Jin et al., 2003; Rice et al., 2003; Plane et al., 2004; Collin et al., 2005). There also appears to be proliferation of BrdUDCX cells in the striatum ipsilateral to the induced stroke. One possibility is that doublecortin is expressed in injured neurons although a majority of striatal-DCX positive cells appear to be newly generated (Arvidsson et al., 2002). Examination by electron microscopy suggests that doublecortinpositive cells comprise the majority of proliferating cells after injury, and migrating chains of neuroblasts can be seen in the striatum after middle cerebral artery occlusion (MCAO) (Zhang et al., 2004). Most of the newly generated cells formed during the first 2 weeks appear to undergo apoptosis and only 20% of newly generated cells in the striatum survive at 6 weeks (Collin et al., 2005). However, surviving cells appear to differentiate into mature neurons, express regionally specific markers for spiny neurons of the striatum, and persist 2–5 months after injury (Parent et al., 2002; Collin et al., 2005). In contrast to studies in adult mice, progenitors in young mice may be more vulnerable to injury (Brazel et al., 2004; Romanko et al., 2004b). After neonatal hypoxic-ischemic (HI) injury progenitor cells in the DG undergo proliferation but only 10–20% mature to neurons (Bartley et al., 2005). Additionally, newly generated cells are seen in the penumbral region after HI and express mature and immature neuronal markers but fail to either migrate or survive in the cortex or striatum after several weeks (Plane et al., 2004; Ikeda et al., 2005). Along with the increase in DCX-expressing cells there is a concomitant decrease in oligodendrocytes precursors 1 week following HI suggesting that there may be signals that instruct progenitor cells to astrocytic or neuronal fates following disruption of the normal myelination
192
Fig. 2. Contralateral (A) and ipsilateral (B) dentate gyrus (DG) 3 days following hypoxic-ischemic injury in 30-day-old nestin-GFP transgenic mice (Yu et al., 2005). On uninjured contralateral side Type I and Type II cells that express GFP (green) are limited mainly to the SGL. In the injured ipsilateral side (A), GFP-expressing cells increase and are found throughout the DG while DCX-expressing (red) Type III progenitors are diminished. Mol ¼ molecular layer; GL ¼ granular layer; SGL ¼ subgranular layer. Scale bar ¼ 100 mM. See Plate 12.2 in Colour Plate Section.
process that occurs during this time period (Hayashi et al., 2005). Fig. 2 demonstrates an example of activation of nestin-GFP stem and early progenitor cells and the vulnerability of later progenitor cells in the DG following HI. Although in the adult SVZ precursors appear to migrate into the striatum after injury, they may form glial rather than neuronal elements. Examination using lipophilic dyes and retroviral labeling suggest that 4 days after cortical injury increased percentages of SVZ cells may bypass normal RMS migration and move toward the injury in the striatum, corpus callosum, and cortex, with some of these cells having a glial morphology (Goings et al., 2004; Salman et al., 2004).
While known neurogenic areas may respond to injury with stimulation in neurogenesis, there are also data that suggest that neuronal regeneration can occur outside the SVZ and DG. Newly generated immature neurons were found in the hippocampus after selective degeneration of pyramidal cells in young rats by kainic acid treatment or after global ischemia (Nakatomi et al., 2002; Schmidt and Reymann, 2002; Dong et al., 2003; Daval et al., 2004). The cell type responsible for this is unknown but it has been suggested that neural precursor cells within the posterior lateral ventricle may reconstitute the pyramidal cells that are lost (Nakatomi et al., 2002; Daval et al., 2004). However, whether injuryinduced neurogenesis exists outside neurgenic areas
Table 1. Neurogenesis in response to brain injury Author
Injury
Region
Proliferation
Marker
Survival – Markera
Liu (1998) Tuyuren (2004) Kee (2001) Scheepens (2003) Dash (2001) Kernie (2001) Dong (2003) Bartley (2005) Takagi, 1999 Yagita (2001) Jin (2001) Duval (2004)
2 V.O MCAO 2 V.O Perinatal asphxy CCI CCI Kianic acid Neonatal HI 2 V.O 4 V.O MCAO Neonatal hypoxiasvz
DG DG DG HC DG DG CA3 DG DG DG DG/SVZ
m m m m m m m m m m m m
BrdU/PCNA BrdU BrdU 3 H Thy BrdU
DAB 28 NeuN 60% DAB 18 DCX 50%NeuN 30%, DAB 28 Calbindin 65%
Schmidt (2002) Ogita (2005) Nakatomi (2002)
2 V.O Timethyltin Global ischemia
Parent (2002)
MCAO
Collin, (2005) Rice (2003) Zhang (2001) Thored (2005)
Quinolinic acid LFP MCAO MCAO
Zhang (2004) Goings (2003) Plane (2004) Hayashi (2005)
MCAO Ctx aspiration Neonatal HI Neonatal HI
Ardvisson (2002) Magavi (2000) Gu (2000) Jin (2003) Tonchev (2003)
MCAO Ctx apoptosis Stroke MCAO Global ischemia
CA1 CA1 DG CA1 Ctx SVZ Str SVZ/str SVZ/SGZ SVZ/str Str SVZ Ctx, cc SVZ/str Str Ctx Str Ctx Ctx Str/ctx DG Ctx SVZ
DAI DAI DAI DAI DAI DAI DAI DAI DAI DAI DAI DAI
1 7 9, 13 5 3 1 33, 53 3 7, 10 8 8 20
BrdU BrdU BrdU BrdU/PCNA BrdU
m DAI m DAI 2–5 m DAI 2–4
BrdU BrdU/PCNA BrdU, DiI
m DAI 14
BrdU, cdc2 BrdU BrdU BrdU BrdU BrdU
m m m m
DAI DAI DAI DAI
14 2 7 14
m DAI 7 retrovirus m DAI 14 BrdU m DAI 7–21 m DAI 7–21 m DAI 14
m DAI 9 m DAI 9 m DAI 9
Brdu DAI 21 DAB 18 NeuN 0%, TUJ1 0% BrdU BrdU BrdU BrdU BrdU BrdU, DiI, BrdU, Ki-67
DAB DAB DAB DAB
30 Calbindin 65%, GFAP 5% 60 Calbindin 3 TuJ1 28%, 12% NeuN 32 NeuN 22%
DAB 28 NeuN 70% DAB 2 DCX m(DAPI,NSE) DAI20 DAB 28 NeuN DAB 5 DCX: DAB 1–7 NeuN DAB 26 NeuN 1%1, NeuN 53%2 DAB 26 NeuN, Hu, TuJ1 DAB 7 DCX, B tubulin III DAB 24 NeuN 48%, DARP-32 70% DAB 42 NeuN (DARP-32 6% PARV 5% NPY 1%) DAB 32 mPSA-NCAM DAB 14 DCX DAB 42+84 NeuN m Migration glial morph DAB 2 m DCX,m GFAP, k NG2 DAB 2 m DCX k NG2 DAB 32 m DCX mHu mMeis2 26 weeks NeuN, Hu DAB 99 NeuN 5% DAB 3 DCX, Neuro D, Hu retrovirus DAB 5 (DCX, CRMP4, NeuN)1 –3% DAB 5 (NeuN, Hu) 1–3% DAB 5 (DCX, CRMP4, NeuN)1-3%
Note: 1 — % without FGF infusion; 2 — % with FGF infusion on DAI 12–1. a
193
IHC marker is co-localized with BrdU, if available % of total BrdU cells is given. DAI ¼ days after injury; DAB ¼ days after first BrdU dose; DG ¼ dentate gyrus; CCI ¼ controlled cortical impact; LFP ¼ lateral fluid percussion injury; HI ¼ hypoxia-ischemia; SVZ ¼ subventricular zone; SGZ ¼ subgranular zone; ctx ¼ cortex; str ¼ striatum; cc ¼ corpus callosum; hc ¼ hippocampus; 2.V.O. ¼ 2 vessel occlusion; 4 V.O. ¼ 4 vessel occlusion; MCAO ¼ middle cerebral artery occlusion.
194
remains controversial. Several authors have reported no evidence of cortical neurogenesis after injury (Zhang et al., 2001; Arvidsson et al., 2002; Parent et al., 2002; Collin et al., 2005; Ikeda et al., 2005; Salman et al., 2004). One study, however, using targeted apoptotic death of cortical neurons, suggests that new cortical neurons are formed in response to injury and that they can extend long distant projections (Magavi et al., 2000). Cortical neurogenesis has also been reported by two studies after induction of cortical stroke (Gu et al., 2000; Jiang et al., 2001). Studies in nonhuman primates are limited. One investigation induced global ischemia in macaques monkeys and BrdU was injected daily for 5 days prior to analysis. Marked increases were noted in BrdU labeling in the DG, SVZ, and temporal cortex 9 days after injury where approximately 3% of BrdU+ cells co-expressed the immature neuronal markers DCX, HU, and CRMP4 as well as the mature neuronal marker NeuN (Tonchev et al., 2003). Many BrdU+ cells in the DG, SVZ, and temporal cortex co-expressed nestin and Musashi-1 and did not express GFAP at a similar percentage as was seen in controls. In contrast to previous experiments, these authors did not find any BrdU colabeling with neuronal markers in the CA1 region. This demonstrates that neural precursors may respond to brain injury in a similar fashion in nonhuman primates although the overall response may not be as robust as is seen with rodents. Table 1 summarizes the data demonstrating that neurogenesis is induced in a variety of injury paradigms. Molecules implicated in injury-induced neurogenesis The signaling mechanisms that control postnatal neurogenesis are largely unknown and even less is known about the factors that regulate neurogenesis after injury. The inflammatory response is a common feature of many forms of brain injury and may be inhibitory for neurogenesis (Vallieres et al., 2002; Ekdahl et al., 2003). Fibroblast growth factor (FGF) and insulin-like growth factor (IGF-1) are two well-known growth factors that are associated with increased neurogenesis after injury (Aberg et al., 2000; Nakatomi et al., 2002; Yoshimura et al., 2003; Abrous et al., 2005). Brain-derived growth factor (BDNF), a pro-survival neural
growth hormone, is another growth factor whose expression increases after injury but has been shown to decrease neural differentiation (Gustafsson et al., 2003). As summarized in Table 1, there are considerable data that describe the response of neural precursors in brain injury, however, little data exist on how these cells perform long-term, what relevant cell types express factors that affect progenitor activation, and whether injury-induced neurogenesis underlies functional improvement. Many factors may influence the neurogenic response to injury and include the developmental period at which the injury occurs, the location of the injury, and the type of injury. Only until we gain a further understanding of what normal physiologic roles neural stem and progenitors play can we then investigate in a mechanistic manner what benefit these cells may have in brain repair and restoration of function. Abbreviations BrdU CNS DCX DG EGF FGF GCL GFAP GFP HI IGF-1 MCAO PSA-NCAM SGL SVZ TBI
50 -bromo-20 -deoxyuridine central nervous system doublecortin dentate gyrus epidermal growth factor fibroblast growth factor granule cell layer glial fibrillary acidic protein green fluorescent protein hypoxic-ischemic insulin-like growth factor -1 middle cerebral artery occlusion polysialyated neural cell adhesion molecule subgranular layer subventricular zone traumatic brain injury
Acknowledgments We thank the Perot Family Center for the Care of Brain and Nerve Injuries at Children’s Medical Center Dallas for its support. SGK is also supported by the NICHD (1K08 HD01470-01).
195
References Aberg, M.A., Aberg, N.D., Hedbacker, H., Oscarsson, J. and Eriksson, P.S. (2000) Peripheral infusion of IGF-I selectively induces neurogenesis in the adult rat hippocampus. J. Neurosci., 20: 2896–2903. Abrous, D.N., Koehl, M. and Le Moal, M. (2005) Adult neurogenesis: from precursors to network and physiology. Physiol. Rev., 85: 523–569. Altman, J. and Das, G.D. (1965) Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats. J. Comp. Neurol., 124: 319–335. Alvarez-Buylla, A. and Nottebohm, F. (1988) Migration of young neurons in adult avian brain. Nature, 335: 353–354. Alvarez-Buylla, A., Seri, B. and Doetsch, F. (2002) Identification of neural stem cells in the adult vertebrate brain. Brain Res. Bull., 57: 751–758. Anderson, V., Catroppa, C., Morse, S., Haritou, F. and Rosenfeld, J. (2000) Recovery of intellectual ability following traumatic brain injury in childhood: impact of injury severity and age at injury. Pediatr. Neurosurg., 32: 282–290. Arvidsson, A., Collin, T., Kirik, D., Kokaia, Z. and Lindvall, O. (2002) Neuronal replacement from endogenous precursors in the adult brain after stroke. Nat. Med., 8: 963–970. Bartley, J., Soltau, T., Wimborne, H., Kim, S., Martin-Studdard, A., Hess, D., Hill, W., Waller, J. and Carroll, J. (2005) BrdU-positive cells in the neonatal mouse hippocampus following hypoxic-ischemic brain injury. BMC Neurosci., 6: 15. Bayer, S.A. (1980) Development of the hippocampal region in the rat. II. Morphogenesis during embryonic and early postnatal life. J. Comp. Neurol., 190: 115–134. Bayer, S.A., Yackel, J.W. and Puri, P.S. (1982) Neurons in the rat dentate gyrus granular layer substantially increase during juvenile and adult life. Science, 216: 890–892. Brazel, C.Y., Rosti, R.T., Boyce, S., Rothstein, R.P. and Levison, S.W. (2004) Perinatal hypoxia/ischemia damages and depletes progenitors from the mouse subventricular zone. Dev. Neurosci., 26: 266–274. Brown, J.P., Couillard-Despres, S., Cooper-Kuhn, C.M., Winkler, J., Aigner, L. and Kuhn, H.G. (2003) Transient expression of doublecortin during adult neurogenesis. J. Comp. Neurol., 467: 1–10. Cameron, H.A., Hazel, T.G. and Mckay, R.D. (1998) Regulation of neurogenesis by growth factors and neurotransmitters. J. Neurobiol., 36: 287–306. Cameron, H.A. and Mckay, R.D. (2001) Adult neurogenesis produces a large pool of new granule cells in the dentate gyrus. J. Comp. Neurol., 435: 406–417. Carlen, M., Cassidy, R.M., Brismar, H., Smith, G.A., Enquist, L.W. and Frisen, J. (2002) Functional integration of adultborn neurons. Curr. Biol., 12: 606–608. Carleton, A., Petreanu, L.T., Lansford, R., Alvarez-Buylla, A. and Lledo, P.M. (2003) Becoming a new neuron in the adult olfactory bulb. Nat. Neurosci., 6: 507–518. Catroppa, C. and Anderson, V. (2005) A prospective study of the recovery of attention from acute to 2 years following pediatric traumatic brain injury. J. Int. Neuropsychol. Soc., 11: 84–98.
Collin, T., Arvidsson, A., Kokaia, Z. and Lindvall, O. (2005) Quantitative analysis of the generation of different striatal neuronal subtypes in the adult brain following excitotoxic injury. Exp. Neurol., 195: 71–80. Committee, T.B. (1970) Embryonic vertebrate central nervous system: revised terminology. The Boulder Committee. Anat. Rec., 166: 257–261. Cooper-Kuhn, C.M. and Kuhn, H.G. (2002) Is it all DNA repair? Methodological considerations for detecting neurogenesis in the adult brain. Brain Res. Dev. Brain Res., 134: 13–21. Dash, P., Mach, S. and Moore, A. (2001) Enhanced neurogenesis in the rodent hippocampus following traumatic brain injury. J. Neurosci. Res., 63: 313–319. Daval, J.L., Pourie, G., Grojean, S., Lievre, V., Strazielle, C., Blaise, S. and Vert, P. (2004) Neonatal hypoxia triggers transient apoptosis followed by neurogenesis in the rat CA1 hippocampus. Pediatr. Res., 55: 561–567. Dayer, A.G., Ford, A.A., Cleaver, K.M., Yassaee, M. and Cameron, H.A. (2003) Short-term and long-term survival of new neurons in the rat dentate gyrus. J. Comp. Neurol., 460: 563–572. Del Rio, J.A. and Soriano, E. (1989) Immunocytochemical detection of 5’-bromodeoxyuridine incorporation in the central nervous system of the mouse. Brain Res. Dev. Brain Res., 49: 311–317. Ding, Y., Li, J., Lai, Q., Azam, S., Rafols, J.A. and Diaz, F.G. (2002) Functional improvement after motor training is correlated with synaptic plasticity in rat thalamus. Neurol. Res., 24: 829–836. Doetsch, F., Caille, I., Lim, D.A., Garcia-Verdugo, J.M., Lim, D.A. and Alvarez-Buylla, A. (1999) Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell, 97: 703–716. Dong, H., Csernansky, C.A., Goico, B. and Csernansky, J.G. (2003) Hippocampal neurogenesis follows kainic acid-induced apoptosis in neonatal rats. J. Neurosci., 23: 1742–1749. Ekdahl, C.T., Claasen, J.H., Bonde, S., Kokaia, Z. and Lindvall, O. (2003) Inflammation is detrimental for neurogenesis in adult brain. Proc. Natl. Acad. Sci. USA,, 100: 13632–13637. Emery, D.L., Fulp, C.T., Saatman, K.E., Schutz, C., Neugebauer, E. and Mcintosh, T.K. (2005) Newly born granule cells in the dentate gyrus rapidly extend axons into the hippocampal CA3 region following experimental brain injury. J. Neurotrauma., 22: 978–988. Eriksson, P.S., Perfilieva, E., Bjork-Eriksson, T., Alborn, A.M., Nordborg, C., Peterson, D.A. and Gage, F.H. (1998) Neurogenesis in the adult human hippocampus. Nat. Med., 4: 1313–1317. Ewing-Cobbs, L., Fletcher, J.M., Levin, H.S., Francis, D.J., Davidson, K. and Miner, M.E. (1997) Longitudinal neuropsychological outcome in infants and preschoolers with traumatic brain injury. J. Int. Neuropsychol. Soc., 3: 581–591. Filippov, V., Kronenberg, G., Pivneva, T., Reuter, K., Steiner, B., Wang, L.P., Yamaguchi, M., Kettenmann, H. and Kempermann, G. (2003) Subpopulation of nestin-expressing progenitor cells in the adult murine hippocampus shows electrophysiological and morphological characteristics of astrocytes. Mol Cell. Neurosci., 23: 373–382.
196 Fukuda, S., Kato, F., Tozuka, Y., Yamaguchi, M., Miyamoto, Y. and Hisatsune, T. (2003) Two distinct subpopulations of nestin-positive cells in adult mouse dentate gyrus. J. Neurosci., 23: 9357–9366. Gage, F.H. (1998) Stem cells of the central nervous system. Curr. Opin. Neurobiol., 8: 671–676. Gage, F.H. (2000) Mammalian neural stem cells. Science, 287: 1433–1438. Garcia-Verdugo, J.M., Doetsch, F., Wichterle, H., Lim, D.A. and Alvarez-Buylla, A. (1998) Architecture and cell types of the adult subventricular zone: in search of the stem cells. J. Neurobiol., 36: 234–248. Goings, G.E., Sahni, V. and Szele, F.G. (2004) Migration patterns of subventricular zone cells in adult mice change after cerebral cortex injury. Brain Res., 996: 213–226. Gu, W., Brannstrom, T. and Wester, P. (2000) Cortical neurogenesis in adult rats after reversible photothrombotic stroke. J. Cereb. Blood Flow Metab., 20: 1166–1173. Gustafsson, E., Lindvall, O. and Kokaia, Z. (2003) Intraventricular infusion of TrkB-Fc fusion protein promotes ischemia-induced neurogenesis in adult rat dentate gyrus. Stroke, 34: 2710–2715. Hardy, R. and Reynolds, R. (1991) Proliferation and differentiation potential of rat forebrain oligodendroglial progenitors both in vitro and in vivo. Development, 111: 1061–1080. Hayashi, T., Iwai, M., Ikeda, T., Jin, G., Deguchi, K., Nagotani, S., Zhang, H., Sehara, Y., Nagano, I., Shoji, M., Ikenoue, T. and Abe, K. (2005) Neural precursor cells division and migration in neonatal rat brain after ischemic/ hypoxic injury. Brain Res., 1038: 41–49. Hoff, S.F. (1986) Lesion-induced transneuronal plasticity in the adult rat hippocampus. Neuroscience, 19: 1227–1233. Ikeda, T., Iwai, M., Hayashi, T., Nagano, I., Shogi, M., Ikenoue, T. and Abe, K. (2005) Limited differentiation to neurons and astroglia from neural stem cells in the cortex and striatum after ischemia/hypoxia in the neonatal rat brain. Am. J. Obstet. Gynecol., 193: 849–856. Imura, T., Kornblum, H.I. and Sofroniew, M.V. (2003) The predominant neural stem cell isolated from postnatal and adult forebrain but not early embryonic forebrain expresses GFAP. J. Neurosci., 23: 2824–2832. Jiang, W., Gu, W., Brannstrom, T., Rosqvist, R. and Wester, P. (2001) Cortical neurogenesis in adult rats after transient middle cerebral artery occlusion. Stroke, 32: 1201–1217. Jin, K., Minami, M., Lan, J.Q., Mao, X.O., Batteur, S., Simon, R.P. and Greenberg, D.A. (2001) Neurogenesis in dentate subgranular zone and rostral subventricular zone after focal cerebral ischemia in the rat. Proc. Natl. Acad. Sci. USA,, 98: 4710–4715. Jin, K., Sun, Y., Xie, L., Peel, A., Mao, X.O., Batteur, S. and Greenberg, D.A. (2003) Directed migration of neuronal precursors into the ischemic cerebral cortex and striatum. Mol. Cell Neurosci., 24: 171–189. Johnston, D. and Amaral, D.G. (1998) Hippocampus. Oxford University Press, New York. Johansson, C.B., Momma, S., Clarke, D.L., Risling, M., Lendahl, U. and Frisen, J. (1999) Identification of a neural stem cell in the adult mammalian central nervous system. Cell, 96: 25–34.
Kaplan, M.S. and Bell, D.H. (1984) Mitotic neuroblasts in the 9-day-old and 11-month-old rodent hippocampus. J. Neurosci., 4: 1429–1441. Kaplan, M.S. and Hinds, J.W. (1977) Neurogenesis in the adult rat: electron microscopic analysis of light radioautographs. Science, 197: 1092–1094. Kawaguchi, A., Miyata, T., Sawamoto, K., Takashita, N., Murayama, A., Akamatsu, W., Ogawa, M., Okabe, M., Tano, Y., Goldman, S.A. and Okano, H. (2001) Nestin-EGFP transgenic mice: visualization of the self-renewal and multipotency of CNS stem cells. Mol Cell Neurosci., 17: 259–273. Kee, N.J., Preston, E. and Wojtowicz, J.M. (2001) Enhanced neurogenesis after transient global ischemia in the dentate gyrus of the rat. Exp. Brain Res., 136: 313–320. Kee, N., Sivalingam, S., Boonstra, R. and Wojtowicz, J.M. (2002) The utility of Ki-67 and BrdU as proliferative markers of adult neurogenesis. J. Neurosci. Methods, 115: 97–105. Kempermann, G., Kuhn, H.G. and Gage, F.H. (1997a) Genetic influence on neurogenesis in the dentate gyrus of adult mice. Proc. Natl. Acad. Sci. USA, 94: 10409–10414. Kempermann, G., Kuhn, H.G. and Gage, F.H. (1997b) More hippocampal neurons in adult mice living in an enriched environment. Nature, 386: 493–495. Kempermann, G., Kuhn, H.G. and Gage, F.H. (1998) Experience-induced neurogenesis in the senescent dentate gyrus. J. Neurosci., 18: 3206–3212. Kernie, S.G., Erwin, T.M. and Parada, L.F. (2001) Brain remodeling due to neuronal and astrocytic proliferation after controlled cortical injury in mice. J. Neurosci. Res., 66: 317–326. Keyvani, K. and Schallert, T. (2002) Plasticity-associated molecular and structural events in the injured brain. J. Neuropathol. Exp. Neurol., 61: 831–840. Kolb, B., Pedersen, B., Ballermann, M., Gibb, R. and Whishaw, I.Q. (1999) Embryonic and postnatal injections of bromodeoxyuridine produce age-dependent morphological and behavioral abnormalities. J. Neurosci., 19: 2337–2346. Kornack, D.R. and Rakic, P. (1999) Continuation of neurogenesis in the hippocampus of the adult macaque monkey. Proc. Natl. Acad. Sci. USA, 96: 5768–5773. Kornack, D.R. and Rakic, P. (2001) The generation, migration, and differentiation of olfactory neurons in the adult primate brain. Proc. Natl. Acad. Sci. USA, 98: 4752–4757. Kronenberg, G., Reuter, K., Steiner, B., Brandt, M.D., Jessberger, S., Yamaguchi, M. and Kempermann, G. (2003) Subpopulations of proliferating cells of the adult hippocampus respond differently to physiologic neurogenic stimuli. J. Comp. Neurol., 467: 455–463. Kuan, C.Y., Schloemer, A.J., Lu, A., Burns, K.A., Weng, W.L., Williams, M.T., Strauss, K.I., Vorhees, C.V., Flavell, R.A., Davis, R.J., Sharp, F.R. and Rakic, P. (2004) Hypoxiaischemia induces DNA synthesis without cell proliferation in dying neurons in adult rodent brain. J. Neurosci., 24: 10763–10772. Kuhn, H.G., Dickinson-Anson, H. and Gage, F.H. (1996) Neurogenesis in the dentate gyrus of the adult rat: age-related decrease of neuronal progenitor proliferation. J. Neurosci., 16: 2027–2033.
197 Kukekov, V.G., Laywell, E.D., Suslov, O., Davies, K., Scheffler, B., Thomas, L.B., O’brien, T.F., Kusakabe, M. and Steindler, D.A. (1999) Multipotent stem/progenitor cells with similar properties arise from two neurogenic regions of adult human brain. Exp. Neurol., 156: 333–344. Levison, S.W. and Goldman, J.E. (1993) Both oligodendrocytes and astrocytes develop from progenitors in the subventricular zone of postnatal rat forebrain. Neuron, 10: 201–212. Liu, J., Solway, K., Messing, R.O. and Sharp, F.R. (1998) Increased neurogenesis in the dentate gyrus after transient global ischemia in gerbils. J. Neurosci., 18: 7768–7778. Lois, C. and Alvarez-Buylla, A. (1994) Long-distance neuronal migration in the adult mammalian brain. Science, 264: 1145–1148. Lois, C., Garcia-Verdugo, J.M. and Alvarez-Buylla, A. (1996) Chain migration of neuronal precursors. Science, 271: 978–981. Luskin, M.B. (1993) Restricted proliferation and migration of postnatally generated neurons derived from the forebrain subventricular zone. Neuron, 11: 173–189. Magavi, S.S., Leavitt, B.R. and Macklis, J.D. (2000) Induction of neurogenesis in the neocortex of adult mice. Nature, 405: 951–955. Markakis, E.A. and Gage, F.H. (1999) Adult-generated neurons in the dentate gyrus send axonal projections to field CA3 and are surrounded by synaptic vesicles. J. Comp. Neurol., 406: 449–460. Menezes, J.R., Smith, C.M., Nelson, K.C. and Luskin, M.B. (1995) The division of neuronal progenitor cells during migration in the neonatal mammalian forebrain. Mol. Cell. Neurosci., 6: 496–508. Morshead, C.M., Garcia, A.D., Sofroniew, M.V. and Van Der Kooy, D. (2003) The ablation of glial fibrillary acidic proteinpositive cells from the adult central nervous system results in the loss of forebrain neural stem cells but not retinal stem cells. Eur. J. Neurosci., 18: 76–84. Morshead, C.M., Reynolds, B.A., Craig, C.G., Mcburney, M.W., Staines, W.A., Morassutti, D., Weiss, S. and Van Der Kooy, D. (1994) Neural stem cells in the adult mammalian forebrain: a relatively quiescent subpopulation of subependymal cells. Neuron, 13: 1071–1082. Nakatomi, H., Kuriu, T., Okabe, S., Yamamoto, S., Hatano, O., Kawahara, N., Tamura, A., Kirino, T. and Nakafuku, M. (2002) Regeneration of hippocampal pyramidal neurons after ischemic brain injury by recruitment of endogenous neural progenitors. Cell, 110: 429–441. Ogita, K., Nishiyama, N., Sugiyama, C., Higuchi, K., Yoneyama, M. and Yoneda, Y. (2005) Regeneration of granule neurons after lesioning of hippocampal dentate gyrus: evaluation using adult mice treated with trimethyltin chloride as a model. J. Neurosci. Res., 82: 609–621. Parent, J.M., Vexler, Z.S., Gong, C., Derugin, N. and Ferriero, D.M. (2002) Rat forebrain neurogenesis and striatal neuron replacement after focal stroke. Ann. Neurol., 52: 802–813. Plane, J.M., Liu, R., Wang, T.W., Silverstein, F.S. and Parent, J.M. (2004) Neonatal hypoxic-ischemic injury increases forebrain subventricular zone neurogenesis in the mouse. Neurobiol. Dis., 16: 585–595.
Quinones-Hinojosa, A., Sanai, N., Soriano-Navarro, M., Gonzalez-Perez, O., Mirzadeh, Z., Gil-Perotin, S., RomeroRodriguez, R., Berger, M.S., Garcia-Verdugo, J.M. and Alvarez-Buylla, A. (2006) Cellular composition and cytoarchitecture of the adult human subventricular zone: A niche of neural stem cells. J. Comp. Neurol., 494: 415–434. Rakic, P. (2002) Neurogenesis in adult primate neocortex: an evaluation of the evidence. Nat. Rev. Neurosci., 3: 65–71. Rao, M.S. and Shetty, A.K. (2004) Efficacy of doublecortin as a marker to analyse the absolute number and dendritic growth of newly generated neurons in the adult dentate gyrus. Eur. J. Neurosci., 19: 234–246. Reynolds, B.A. and Weiss, S. (1992) Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science, 255: 1707–1710. Rice, A.C., Khaldi, A., Harvey, H.B., Salman, N.J., White, F., Fillmore, H. and Bullock, M.R. (2003) Proliferation and neuronal differentiation of mitotically active cells following traumatic brain injury. Exp. Neurol., 183: 406–417. Richards, L.J., Kilpatrick, T.J. and Bartlett, P.F. (1992) De novo generation of neuronal cells from the adult mouse brain. Proc. Natl. Acad. Sci. USA, 89: 8591–8595. Romanko, M.J., Rola, R., Fike, J.R., Szele, F.G., Dizon, M.L., Felling, R.J., Brazel, C.Y. and Levison, S.W. (2004a) Roles of the mammalian subventricular zone in cell replacement after brain injury. Prog. Neurobiol., 74: 77–99. Romanko, M.J., Rothstein, R.P. and Levison, S.W. (2004b) Neural stem cells in the subventricular zone are resilient to hypoxia/ischemia whereas progenitors are vulnerable. J. Cereb. Blood Flow Metab., 24: 814–825. Salman, H., Ghosh, P. and Kernie, S.G. (2004) Subventricular zone neural stem cells remodel the brain following traumatic injury in adult mice. J. Neurotrauma., 21: 283–292. Sanai, N., Tramontin, A.D., Quinones-Hinojosa, A., Barbaro, N.M., Gupta, N., Kunwar, S., Lawton, M.T., Mcdermott, M.W., Parsa, A.T., Manuel-Garcia Verdugo, J., Berger, M.S. and Alvarez-Buylla, A. (2004) Unique astrocyte ribbon in adult human brain contains neural stem cells but lacks chain migration. Nature, 427: 740–744. Scheff, S.W., Price, D.A., Hicks, R.R., Baldwin, S.A., Robinson, S. and Brackney, C. (2005) Synaptogenesis in the hippocampal CA1 field following traumatic brain injury. J. Neurotrauma, 22: 719–732. Schmidt, W. and Reymann, K.G. (2002) Proliferating cells differentiate into neurons in the hippocampal CA1 region of gerbils after global cerebral ischemia. Neurosci. Lett., 334: 153–156. Scoville, W.B. and Milner, B. (1957) Loss of recent memory after bilateral hippocampal lesions. J. Neurol. Neurosurg. Psychiatry, 20: 11–21. Seaberg, R.M. and Van Der Kooy, D. (2003) Stem and progenitor cells: the premature desertion of rigorous definitions. Trends Neurosci., 26: 125–131. Sekerkova, G., Ilijic, E. and Mugnaini, E. (2004) Bromodeoxyuridine administered during neurogenesis of the projection neurons causes cerebellar defects in rat. J. Comp. Neurol., 470: 221–239.
198 Seki, T. (2002) Expression patterns of immature neuronal markers PSA-NCAM, CRMP-4 and NeuroD in the hippocampus of young adult and aged rodents. J. Neurosci. Res., 70: 327–334. Stanfield, B.B. and Trice, J.E. (1988) Evidence that granule cells generated in the dentate gyrus of adult rats extend axonal projections. Exp. Brain Res., 72: 399–406. Stroemer, R.P., Kent, T.A. and Hulsebosch, C.E. (1998). Enhanced neocortical neural sprouting, synaptogenesis, and behavioral recovery with D-amphetamine therapy after neocortical infarction in rats. Stroke, 29 2381–2393; Discussion 2393–2395. Tonchev, A.B., Yamashima, T., Zhao, L., Okano, H.J. and Okano, H. (2003) Proliferation of neural and neuronal progenitors after global brain ischemia in young adult macaque monkeys. Mol. Cell. Neurosci., 23: 292–301. Tozuka, Y., Fukuda, S., Namba, T., Seki, T. and Hisatsune, T. (2005) GABAergic excitation promotes neuronal differentiation in adult hippocampal progenitor cells. Neuron, 47: 803–815. Tureyen, K., Vemuganti, R., Sailor, K.A., Bowen, K.K. and Dempsey, R.J. (2004) Transient focal cerebral ischemia-induced neurogenesis in the dentate gyrus of the adult mouse. J. Neurosurg., 101: 799–805. Vallieres, L., Campbell, I.L., Gage, F.H. and Sawchenko, P.E. (2002) Reduced hippocampal neurogenesis in adult transgenic mice with chronic astrocytic production of interleukin6. J. Neurosci., 22: 486–492. Van Praag, H., Schinder, A.F., Christie, B.R., Toni, N., Palmer, T.D. and Gage, F.H. (2002) Functional neurogenesis in the adult hippocampus. Nature, 415: 1030–1034. Winner, B., Cooper-Kuhn, C.M., Aigner, R., Winkler, J. and Kuhn, H.G. (2002) Long-term survival and cell death of newly generated neurons in the adult rat olfactory bulb. Eur. J. Neurosci., 16: 1681–1689.
Yagita, Y., Kitagawa, K., Ohtsuki, T., Takasawa, K., Miyata, T., Okano, H., Hori, M. and Matsumoto, M. (2001) Neurogenesis by progenitor cells in the ischemic adult rat hippocampus. Stroke, 32: 1890–1896. Yamaguchi, M., Saito, H., Suzuki, M. and Mori, K. (2000) Visualization of neurogenesis in the central nervous system using nestin promoter-GFP transgenic mice. Neuroreport, 11: 1991–1996. Yoshimura, S., Teramoto, T., Whalen, M.J., Irizarry, M.C., Takagi, Y., Qiu, J., Harada, J., Waeber, C., Breakefield, X.O. and Moskowitz, M.A. (2003) FGF-2 regulates neurogenesis and degeneration in the dentate gyrus after traumatic brain injury in mice. J. Clin. Invest., 112: 1202–1210. Yu, T.S., Dandekar, M., Monteggia, L.M., Parada, L.F. and Kernie, S.G. (2005) Temporally regulated expression of Cre recombinase in neural stem cells. Genesis, 41: 147–153. Zhang, R., Zhang, Z., Wang, L., Wang, Y., Gousev, A., Zhang, L., Ho, K.L., Morshead, C. and Chopp, M. (2004) Activated neural stem cells contribute to stroke-induced neurogenesis and neuroblast migration toward the infarct boundary in adult rats. J. Cereb. Blood Flow Metab., 24: 441–448. Zhang, R.L., Zhang, Z.G., Zhang, L. and Chopp, M. (2001) Proliferation and differentiation of progenitor cells in the cortex and the subventricular zone in the adult rat after focal cerebral ischemia. Neuroscience, 105: 33–41. Zola-Morgan, S., Squire, L.R. and Amaral, D.G. (1986) Human amnesia and the medial temporal region: enduring memory impairment following a bilateral lesion limited to field CA1 of the hippocampus. J. Neurosci., 6: 2950– 2967.
Plate 11.5. Regions showing a negative covariance between rCBF and discourse abstraction performance.
Plate 12.1. Example of different stage progenitors in the DG of the mature mouse hippocampus. Letter A labels a GFP (green) and GFAP (blue)-expressing neural stem cell in the subgranular zone of a nestin-GFP transgenic mouse (Type I cell) (Yu et al., 2005). Letter B marks a GFP+, GFAP–, DCX– progenitor (Type II) and C labels DCX+(red), GFP-, GFAP- maturing progenitor (Type III). Arrow marks GFAP positive process from GFP-expressing cell. GL ¼ granular layer; SGL ¼ subgranular layer. Scale bar ¼ 35 mM.
Plate 12.2. Contralateral (A) and ipsilateral (B) dentate gyrus (DG) 3 days following hypoxic-ischemic injury in 30-day-old nestinGFP transgenic mice (Yu et al., 2005). On uninjured contralateral side Type I and Type II cells that express GFP (green) are limited mainly to the SGL. In the injured ipsilateral side (A), GFP-expressing cells increase and are found throughout the DG while DCXexpressing (red) Type III progenitors are diminished. Mol ¼ molecular layer; GL ¼ granular layer; SGL ¼ subgranular layer. Scale bar ¼ 100 mM.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 13
Cognitive neural plasticity during learning and recovery from brain damage Vanessa Raymont1,2 and Jordan Grafman2, 2
1 Vietnam Head Injury Study, National Naval Medical Center, Bethesda, MD, USA Cognitive Neuroscience Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA
Abstract: The process of neuroplasticity is the ability of the brain to change, either in response to experience or injury. It is a vital process both during normal development and for the recovery after brain injury. Recent research has emphasized that this takes place via both local restitution as well as reorganization and compensatory reassignment. The fact that the brain can undergo such plastic changes has provided evidence for what underlies developmental brain disorders, as well as the variable response to injury at different points in the lifespan. The factors affecting plasticity and its long-term consequences may have increasing importance in exposing the pattern of changes that occur in the normal brain with aging. Keywords: neural plasticity; brain injury; learning; brain disorders; aging; cognitive ability Introduction
the injured and their families given that they are disproportionately young. It is well known that the human brain is functionally altered through experience because of its plasticity (Luciana, 2003). Expression of neural plasticity causes structural and functional changes in the brain at the genetic, molecular, neuronal, system, and behavioral levels, and is an inherent feature of brain function throughout the lifespan. Such changes reflect the brain’s ability to learn, remember, and forget experiences as well as its capacity to reorganize and recover from injury (Buonomano and Merzenich, 1998). Neural plasticity has been described at the molecular (Keyvani and Schallert, 2002), structural, and system (Kolb et al., 2003) levels in animals. Current thinking about systems’ neural plasticity in humans suggests that at least four kinds of plastic changes operate at the representational level after brain damage: (1) homologous area adaptation, (2) cross-modal reassignment, (3) potentiation of topographic representations, and (4) compensatory
This chapter focuses on expression of neural plasticity after brain injury, but the processes described in this chapter are relevant to all forms of brain damage. Traumatic brain injury (TBI) is an insult to the brain caused by an external force, and is surprisingly common. An estimated 5–10% victims of TBI die, but most sustain only minor injuries and do not come to medical attention. Moderate head injury (as categorized by a Glasgow Coma Scale score of 9–12 out of a maximum of 15) affects 60,000–75,000 Americans each year. Of those, two-thirds are moderately to severely disabled three months after the injury (Rimel et al., 1982). Thus this is a phenomenon that not only affects many people each year, but also has considerable long-term implications for
Corresponding author. Tel.: +1 301 496 0220; Fax: +1 301 480 2909; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57013-X
199
200
masquerade. These processes appear to occur via two types of reorganization – rapidly occurring alteration in synaptic sensitivity, possibly related to unmasking of existing connections through change in the inhibitory dynamics, and structural changes that occur over days and weeks. But functionally the effects on the central nervous system (CNS) appear quite similar (Taub, 2004). All these concepts have replaced the traditional view that the CNS has little capacity to reorganize and repair itself in response to injury. This theory extended back to the 19th century, influenced initially by Broca’s studies of localization of function within the brain (Taub, 2004). However, this model never fully explained the phenomenon of spontaneous recovery of function, mainly because the techniques required to explore such processes had not been developed at that time. Beginning in the 1970s, research from many laboratories showed that the mammalian nervous system does indeed have the capacity to reorganize itself functionally after injury (Merzenich et al., 1984), and that environmental manipulation could impact on brain structure and chemistry (Rosenzweig, 1999). The Russian psychologist Luria previously argued for a compensatory process of functional reorganization, so that surviving neural circuits reorganize to provide recovery, and indeed changes in cortical organization have been demonstrated following various types of training and alterations in sensory inputs (Schallert et al., 2000). However, more recent advances in knowledge about the plasticity of the CNS as well as developments in connectionist models of neuropsychological function require that theories must now allow for partial restitution as well as compensation. Additionally, over the past decade both animal and human studies have provided increasing evidence that the adult-cerebral cortex has significant functional plasticity, and that post-injury behavioral experience is a major modulator of subsequent physiological and anatomical changes (Nudo et al., 2001).
Development and learning Many studies have demonstrated multiple sensory and motor representations of a single-brain region
and overlap of representation (Anton et al., 1996: Nudo et al., 2000), and this may help explain the neural underpinnings of plastic changes with learning. Thus, such multiplexing is the diverse use of neurons and fibers so that a particular set of neurons can participate in various functions. There is evidence that expression of neural plasticity occurs during normal human postnatal development. However, this appears to be constrained by many factors. Hypotheses have been presented to explain how postnatal cortical functioning is advanced: (1) that increasing experience of a stimulus or task results in increased localization of functions within the cortex and (2) localized cortical activation within an area becomes more selective for that function. But it has also been suggested that cortical circuits, whose specialization is not yet fully developed, become partially activated in certain task contexts without actually influencing behavior (Johnson, 1999). Thus the structure–function relations observed in the human adult brain appear to be subjected to multiple constraints, both intrinsic and extrinsic, and the relations between brain development and experience thereby become bi-directional (Johnson, 1997). Basic mechanisms that support the expression of this form of plasticity during development include persistence of neurogenesis in some parts of the brain, elimination of neurons through apoptosis (programmed cell death), postnatal proliferation and pruning of synapses, and activity-dependent refinement of neuronal connections (Johnston, 2004). Overproduction of synapses during the postnatal period results in the 2-year old having twice as many synapses in the cerebral cortex as adults, and pruning of excessive synapses progresses until approximately 16 years of age (Johnston et al., 2001). However, clinical disorders that are related to brain plasticity, such as neurofibromatosis, tuberous sclerosis, Fragile X syndrome, and cerebral palsy, are not uncommon in childhood. Thus expression of brain plasticity in children can be divided into four types: (1) adaptive plasticity where changes in neuronal circuitry enhance normal skill development or recovery from brain injury; (2) impaired plasticity which occurs when molecular plasticity pathways are disrupted in
201
genetic or acquired disorders that lead to cognitive impairment; (3) excessive expression of plasticity leading to maladaptive brain circuits as occurs in partial seizures following mesial temporal sclerosis; and (4) plasticity that makes the brain vulnerable to injury, such as occurs in status epilepticus when excitatory mechanisms designed to cause expression of plasticity become over-stimulated, resulting in excitotoxic neuronal damage (Johnston, 2004). Children appear to have a remarkable ability to recover from early TBI, as demonstrated by their ability to recover receptive language after left hemispherectomy performed for epilepsy as late as the second decade of life (Boatman et al., 1999). The publication of case studies of patients exhibiting impressive recovery following focal vascular lesions during the perinatal period or early infancy led to the view that neural plasticity confers an advantage for the developing brain (Levin, 2003). These findings have confirmed the original theories (Kennard, 1936) that the young brain was more capable of reorganization than in maturity (Levin, 2003). However, the child and adult studies often did not compare similar lesions after comparable post-injury intervals, and, in fact more recent, longitudinal findings have noted continued subtle changes in levels of functioning after early TBI (Bates et al., 1997). It has also been suggested that inter-hemispheric transfer of language produced a ‘crowding’ effect on the right-hemisphere functions, leading to compromised visual-spatial functions, based on case studies of early injuries. An alternative view put forward by the early connectionist, Hebb, suggested that the adverse repercussions of brain damage in early life may actually occur because it prevents the normal neural organization and behavioral development from occurring (Hebb, 1947). The site of injury may be important too. For example, the prefrontal cortex (PFC) is immature in children, and therefore the deficits observed secondary to early damage might be a function of immaturity rather than lack of expression of neural plasticity. It appears that in children their primitive executive functions may be mediated by the caudate nucleus, and they become more complex as they are transferred to the PFC in adulthood. Thus in children, subcortical damage may in fact result in executive
dysfunction. Damage to the PFC can only be determined by observing behavior in adulthood (Luciana, 2003). Recent longitudinal studies have shown that hemispheric specialization patterns depend on the site of injury and age of testing, with evidence for an initial delay followed by more typical development (Bates et al., 1997). Similarly, children with focal right-hemisphere injuries before 5 years of age display well-developed spatial ability after an initial period of deficit (Vicari et al., 2000) suggesting that reorganization of function is more extensive prior to this age (Levin, 2003). Most studies of early brain injury have been cross-sectional case studies (Luciana, 2003), and it has been assumed that injuries during the period of cell migration are particularly detrimental. However, such outcomes must be assessed longitudinally, as the apparent recovery in childhood has been noted to potentially reverse in later life, and one function may recover at the expense of others. Preterm birth is particularly associated with high rates of neuro-developmental disability, first and foremost as a secondary effect to hypoxic-ischemic events. Through school age, preterm children exhibit diminished levels of global intellectual function. Several developmental courses are possible, and all have been observed in animal studies (Kolb, 1995). In the first, damage is so severe that development is severely compromised and does not reach a normal state. In the second, development proceeds normally. In the third, development may be hampered early on, but improves with increasing age, suggesting recovery of function. The fourth possibility is one in which evidence of cognitive dysfunction may be subtle or even absent in infancy, but becomes more evident with increased age. This has been termed a ‘sleeper effect’ (McGrath et al., 2000). Animal studies have also suggested that early-injured brains may be vulnerable to degenerative processes (Kolb, 1995). Recent studies suggest that a behaviorally relevant degree of plasticity is retained in the adult cortex, even within early, low-level representations in sensory and motor processing streams. Dancause et al. (2005) used an anterograde tract tracer in animal studies, and found evidence of axonal sprouting after injury to the adult ventral
202
premotor cortex, with establishment of novel connections within a distant target. Recent studies of both humans and animals also showed that new cells can be produced in the hippocampus in the longer term (Eriksson et al., 1998). It appears that this process is at least partly experience-dependent; animals kept in enriched environment show more cell genesis in the hippocampus than animals kept in environments that provided less sensory input (Gould et al., 1999). Indeed, several studies of the sensorimotor cortex suggest that functional activity can be altered in humans by chronic experience of stimuli (Nudo et al., 2001). In somatosensory cortex, the representation of the digits of the skilled hand is expanded in string musicians and blind Braille readers compared to the unskilled hand (Pascal-Leone and Torres, 1993; Elbert et al., 1994). There are many studies in humans that demonstrate that functional reorganization associated with motor learning can also occur over a very short time. On the whole, these studies indicate that motor cortex has the potential for rapid and largescale functional changes in response to motor skill learning. Maps of motor outputs from the primary motor cortex (M1) determined using transcranial magnetic stimulation (TMS) have been shown to change after brief periods of motor training (Nudo et al., 2001). An animal study by Blake et al. (2005) showed that stimuli presented across a long-time window still produce co-representation within the cortex, and that increased excitability accompanies new task learning. Several studies using positron-emission tomography in humans have demonstrated activity changes in motor-cortical structures, and in particular in M1, during the acquisition of new motor skills (Kawashima et al., 1994; Schlaug et al., 1994). Grafton et al. (1992) studied individuals as they learned to track a moving target with their hand and found that as accuracy increased and smooth pursuit movements developed, the activation increased beyond the levels normally associated with such movements. It has been suggested that modification of the strength of horizontal connections in the cortex may mediate such functional changes. The most widely studied model of synaptic mechanisms underlying learning is long-
term potentiation (LTP) and long-term depression (LTD) in the rat hippocampus or cerebral cortex (Nudo et al. 2001). Additionally, functional imaging studies have provided evidence for recruitment of homotopic areas of the right hemisphere in recovery of language in at least a subgroup of adult aphasics (Papanicolaou et al., 1988). A recent PET study also suggests that such reorganization may be usedependent. In a study by Christodoulou et al. (2001), the distribution of brain activation in a working memory task in those who had sustained a TBI 9 months previously was more widely dispersed in the middle frontal and temporal lobes, and may reflect recruitment of additional cerebral resources. Thus it may be possible to follow response to rehabilitation via functional imaging (Levin, 2003).
Recovery from TBI Recovery from motor disorders, language disorders, perceptual deficits, unilateral neglect, attention deficits, and tactile discrimination deficits probably occur because of a combination of compensatory processes and plastic reorganization. Recovery is thought not to continue past 6 months after injury such as from TBI or strokes (Duncan and Lai, 1997); and the ultimate outcome seems to be largely a function of the initial severity of the deficits. The recovery processes can be viewed as natural extensions of normal learning and experience-dependent processes. There is also evidence for similar processes in recovery from deficits regarding attention, memory, language, and executive functions (Mateer and Kerns, 2000). Several techniques have been used in humans to examine the effects of cortical injury on the function of intact cortical tissue. Participants in such studies are typically individuals with cortical lesions (either ischemic or hemorrhagic insults). Although the location of the injury is frequently unknown or uncontrolled, these studies have consistently shown that functional changes occur in several cortical areas after stroke or other brain damage, supporting results from animal experiments (Cramer and Bastings, 2000).
203
The theoretical framework for understanding recovery of function is still evolving. According to one hypothesis presented by von Monakow at the beginning of the twentieth century (Nudo et al., 2001), the function of remote cortical tissue is temporarily suppressed after focal-cortical injury in a process known as ‘diaschisis.’ Recovery is thought to result from the gradual reversal of diaschisis. However, as more injury-induced events at distant sites are examined, it is becoming evident that diaschisis may persist after recovery (Infeld et al., 1995), and that remote effects are complex and include disinhibition and hyperexcitability (Andrews, 1991). Although resolution of diaschisis and behavioral compensation are assumed to play major roles in the phenomenon of motor recovery, it has also been suggested that other cortical areas may compensate for the function of the damaged area (Xerri et al., 1998). It is still not entirely clear as to what degree the reorganization observed in spared tissue represents mechanisms related to restitution of the original function or whether behavioral compensation causes the improvement of function, or if both of these mechanisms are involved. In a study using transcranial magnetic stimulation, it has been shown that, shortly after stroke, the excitability of the motor cortex is reduced, and the cortical representation of the affected muscles is decreased (Traversa et al., 1997). It has been suggested that this effect occurs from a combination of diaschisislike effects and disuse of the affected limb (Nudo et al., 2001). But studies in the motor cortex indicate that the intracortical connections devoted to a processing function may become temporarily enhanced or even enlarged with skill acquisition or frequent exposure to a stimulus. Rehabilitation can lead to an enlargement of the motor map in the injured hemisphere relative to the initial postinjury map (Traversa et al., 1997). It has also been shown that movement of the recovered hand is associated with increased bilateral activation of remote brain areas, such as the cerebellum, and premotor cortex, as well as the sensorimotor cortex, often in the sensorimotor cortex of the uninjured hemisphere, suggesting that the uninjured hemisphere plays a role in recovery (Nelles et al., 1999). As mentioned above, adaptation in
homologous area primarily occurs before puberty and underlies the observation that damage to a particular brain region can be compensated for by shifting its functions to other brain areas (Chugani et al., 1996). This form of neuroplastic change is reported less often in adults, possibly as a consequence of greater local-functional commitment. In cross-modal reassignment new sensory inputs are introduced to a brain region that has been deprived of its main sensory input. For example, PET and fMRI studies of tactile discrimination have shown that adults who became blind early in childhood have somatosensory input to area V1 of the occipital cortex, whereas sighted controls do not (Sadato et al., 1996). Potentiation of topographic representations involves changes within brain regions devoted to a particular kind of process. Compensatory masquerade involves the novel use of an established and intact cognitive process to perform a task previously dependent on an impaired cognitive process. This may be a subtle change detectable only with fine-grained cognitive tasks. Roberson and Murre (1999) have proposed a triage of post-lesion states: (1) small loss of connectivity that leads to autonomous recovery; (2) major loss of connectivity that leads to permanent loss of function, requiring compensatory approach; (3) potentially-rescuable, lesioned circuits, in which case recovery may depend on providing targeted bottom-up and top-down inputs, while maintaining adequate levels of arousal and avoiding activation of competitor circuits. There also appears to be a ‘serial lesion effect,’ when competition occurs between regions in a single lesion, whereas if the same areas are affected in consecutive insults, there is less-functional impact (deCastro and Zrull, 1988).
Long-term implications of plasticity It is likely that environmental factors can predict the long-term repercussions of plastic change in the brain. A more complex environment has been found to be more conducive to brain restoration, whilst stress inhibits neurogenesis (Kozorovitskiy and Gould, 2003). Internal body conditions such
204
as hemodynamics, intracranial pressure variations, glutamate cascades, and other processes that can follow brain lesions may effect the path of recovery and remodeling (Robertson and Murre, 1999). There is increasing evidence suggesting that TBI is a risk factor for Alzheimer’s Disease (AD) later in life (Albensi and Janigro, 2003). However, the underlying mechanisms are still unclear and highly controversial hypotheses have been presented. Mesulam (2000) has hypothesized that TBI induces a heightened state of expression of neural plasticity, which can trigger the neuropathological changes seen in AD, and that AD-promoting factors create a setting where neurons must work harder to meet plasticity demands at their terminals. Only recently have long-term changes in specific neurotransmitter systems been investigated in chronic cortical injury. These studies are important because they may lead to new intervention strategies and potential pharmacologic treatment of chronic stroke. Changes in two neurotransmitter systems, GABA and glutamate, have been implicated in behavioral deficits following stroke, and alterations in the activity of each may play a role in functional recovery (Nudo et al., 2001). There is a growing use of antiepileptic drugs for neuroprotection. Established drugs like phenytoin, phenobarbitol, and carbamazepine have shown neuroprotective activity in the ischemic/hypoxic model of neuronal injury. Animal studies have also shown that newer drugs such as topiramate, levetiracetam, and zonisamide may have similar properties (Wilmore, 2005). However, clinical studies suggest that drugs such as d-amphetamine have the potential to enhance recovery, whilst others such as neuroleptics, benzodiazepines, and anticonvulsants such as phenytoin and phenobarbital may be detrimental. Anecdotal reports suggest that other drugs such as methyphenidate, amantadine, and levodopa have shown some response in TBI (Goldstein, 2003). The notion that level of use affects outcome has important implications for the rehabilitation of individuals with brain injuries. Faced with someone who cannot move his or her arm, do you devote limited rehabilitation resources on physical therapy targeting that arm or help them to develop compensatory strategies? Should someone with
executive deficits be given a structured environment to support more organized behavior or be retrained in deficient attentional-control skills? It seems that the extent of recovery of hemiparesis is proportional to the number of targeted physical rehabilitation that the patient receives (Kwakkel et al., 1999). This is also seen in psychiatric disorders (Wykes et al., 1999) and strokes affecting the left hemisphere (Musso et al., 1999). Constraint-induced movement therapy is used in the rehabilitation of stroke victims and involves the restraint of the less-affected limb because it is assumed to induce use-dependent cortical reorganization (Taub, 2004). The hypothesis that sensorimotor learning and cortical injury interact to remodel structure and function of undamaged parts of the brain during recovery from brain damage via sensorimotor experiences in the weeks and months following injury (Nudo, 2003), has vital implications for rehabilitation, as does the evidence of how pharmaceutical products may alter that process. Finally, increased knowledge about normal variation in gene expression and epigentics that can lead to individual differences in plasticity during normal learning will facilitate recovery of function after brain injury. Thus it is apparent that the cerebral cortex can undergo many forms of plastic changes during learning and after injury. But whilst expression of neural plasticity produces functional improvement after brain injury, it may also produce unwanted effects, such as spasticity and certain forms of epilepsy through the kindling effect (Bach-y-Rita, 2003), subtle cognitive dysfunctions with increasing age, and it may predispose to long-term conditions such as Alzheimer’s disease. Our ability to learn how to exert some significant control over these plastic changes via behavioral strategies, drugs, and other interventions will no doubt improve our ability to facilitate the recovery from the disabling effects of brain damage.
Acknowledgments JG is supported by the National Institute of Neurological Disorders and Stroke intramural research program. VR is supported by a project
205
grant from the United States Army Medical Research and Material Command and administered by the Henry M. Jackson Foundation (‘Vietnam Head Injury Study Phase III: A 30 Year Post-Injury Follow-Up Study,’: Grant number: DAMD17-01-1-0675). References Albensi, B.C. and Janigro, D. (2003) Traumatic brain injury and its effects on synaptic plasticity. Brain Inj., 17: 653–663. Andrews, R.J. (1991) Transhemispheric diaschisis A review and comment. Stroke, 22: 943–949. Anton, J.L., Benali, H., Guigon, E., Di Paola, M., Bittoun, J., Jolivet, O. and Burnod, Y. (1996) Functional MR imaging of the human sensorimotor cortex during haptic discrimination. Neuroreport, 7(18): 2849–2852. Bach-y-Rita, P. (2003) Theoretical basis for brain plasticity after a TBI. Brain Inj., 17: 643–651. Bates, E., Thal, D., Trauner, D., Fenson, J., Aram, D., Eisele, J. and Ness, R. (1997) From first words to grammar in children with focal brain injury. Dev. Neuropsychol., 13: 275–344. Blake, D.T., Strata, F., Kempter, R. and Merzenich, M.M. (2005) Experience-dependent plasticity in SI caused by noncoincident inputs. J. Neurophysiol., 94: 2239–2250. Boatman, D., Freeman, J., Vining, E., Pulsifer, M., Miglioretti, D., Minahan, R., Carson, B., Brandt, J. and McKhann, G. (1999) Language recovery after left hemispherectomy in children with late-onset seizures. Ann. Neurol., 46: 579–586. Buonomano, D.V. and Merzenich, M.M. (1998) Cortical plasticity: from synapses to maps. Annu. Rev. Neurosci., 21: 149–186. Christodoulou, C., DeLuca, J., Ricker, J.H., Madigan, N.K., Bly, B.M., Lange, G., Kalnin, A.J., Liu, W.C., Steffener, J., Diamond, B.J. and Ni, A.C. (2001) Functional magnetic resonance imaging of working memory impairment after traumatic brain injury. J. Neurol., Neurosurg. Psychiatry, 71: 161–168. Chugani, H.T., Muller, R.A. and Chugani, D.C. (1996) Functional brain reorganization in children. Brain Dev., 18: 347–356. Cramer, S.C. and Bastings, E.P. (2000) Mapping clinically relevant plasticity after stroke. Neuropharmacol., 39: 842–851. Dancause, N., Barbay, S., Frost, S.B., Plautz, E.J., Chen, D., Zoubina, E.V., Stowe, A.M. and Nudo, R.J. (2005) Extensive cortical rewiring after brain injury. J. Neurosci., 2: 10167–10179. DeCastro, J.M. and Zrull, M.C. (1988) Recovery of sensorimotor function after frontal cortex damage in rats: evidence that the serial lesion effect is due to serial recovery. Behav. Neurosci., 102: 843–851. Duncan, P.W. and Lai, S.M. (1997) Stroke recovery. Topics in Stroke Rehabil., 4: 51–58. Elbert, T., Flor, H., Birbaumer, N., Knecht, S., Hampson, S., Larbig, W. and Taub, E. (1994) Extensive reorganization of the somatosensory cortex in adult humans after nervous system injury. Neuroreport, 5: 2593–2597.
Eriksson, P.S., Perfilieva, E., Bjork-Eriksson, T., Alborn, A.M., Nordborg, C., Peterson, D.A. and Gage, F.H. (1998) Neurogenesis in the adult human hippocampus. Nat. Med., 4: 1313–1317. Goldstein, L.B. (2003) Neuropharmacology of TBI-induced plasticity. Brain Inj., 17: 685–694. Gould, E., Reeves, A.J., Graziano, M.S. and Gross, C.G. (1999) Neurogenesis in the neocortex of adult of adult primates. Science, 286: 548–552. Hebb, D.O. (1947) The effects of early experience on problem solving at maturity. Am. Psychol., 2: 737–745. Infeld, B., Davis, S.M., Lichtenstein, M., Mitchell, P.J. and Hopper, J.L. (1995) Crossed cerebellar diaschsis and brain recovery after stroke. Stroke, 26: 90–95. Johnson, M.H. (1997) Developmental cognitive neuroscience: an introduction. Blackwell, Oxford. Johnson, M.H. (1999) Cortical plasticity in normal and abnormal cognitive development: evidence and working hypotheses. Dev. Psychopathol., 11: 419–437. Johnston, M.V. (2004) Clinical disorders of brain plasticity. Brain Dev., 26: 73–80. Johnston, M.V., Nishimura, A., Harum, K., Pekar, J. and Blue, M.E. (2001) Sculpting the developing brain. Adv. Pediatr., 48: 1–38. Kawashima, R., Roland, P.E. and O’Sullivan, B.T. (1994) Fields in human motor areas involved in preparation for reaching, actual reaching, and visuomotor learning: a positron emission tomography study. J. Neurosci., 14: 3462–3474. Kennard, M.A. (1936) Age and other factors in motor recovery from precentral lesions in monkeys. Am. J. Physiol., 115: 138–146. Keyvani, K. and Schallert, T. (2002) Plasticity-associated molecular and structural events in the injured brain. J. Neuropathol. Exp. Neurol., 61: 831–840. Kolb, B. (1995) Brain plasticity and behavior. Erlbaum, Mahwah NJ. Kolb, B., Gibb, R. and Robinson, T.E. (2003) Brain plasticity and behavior. Curr. Dir. Psychol. Sci., 12: 1–5. Kozorovitskiy, Y. and Gould, E. (2003) Adult neurogenesis: a mechanism for brain repair? J. Clin. Exp. Neuropsychol., 25(5): 721–732. Kwakkel, G., Wagenaar, R.C., Twisk, J.W., Lankhorst, G.J. and Koetsier, J.C. (1999) Intensity of leg and arm training after primary middle-cerebral artery stroke: a randomized trial. Lancet, 354: 191–196. Levin, H.S. (2003) Neuroplasticity following non-penetrating traumatic brain injury. Brain Inj., 17: 665–674. Luciana, M. (2003) Cognitive development in children born preterm: implications for theories of brain plasticity following early injury. Dev. Psychopathol., 15: 1017–1047. Mateer, C.A. and Kerns, K.A. (2000) Capitalizing on neuroplasticity. Brain and Cogn., 42: 106–109. McGrath, M.M., Sullivan, M.C., Lester, B.M. and Oh, W. (2000) Longitudinal neurologic follow-up in neonatal intensive care unit survivors with various neonatal morbidities. Pediatrics, 106: 1397–1405.
206 Merzenich, M.M., Nelson, R.J., Stryker, M.P., Cynader, M.S., Schoppman, A. and Zook, J.M. (1984) Somatosensory cortical map changes following digital amputation in adult monkeys. J. Comp. Neurol., 24: 591–605. Mesulam, M.M. (2000) A plasticity-based theory of the pathogenesis of Alzheimer’s disease. Ann. N. Y. Acad. Sci., 924: 42–52. Musso, M., Weiller, C., Kiebel, S., Muller, S.P., Bulau, P. and Rijntjes, M. (1999) Training-induced brain plasticity in aphasia. Brain, 122: 1781–1790. Nelles, G., Spiekramann, G., Jueptner, M., Leonhardt, G., Muller, S., Gerhard, H. and Diener, H.C. (1999) Evolution of functional reorganization in hemiplegic stroke: a serial positron emission tomographic activation study. Ann. Neurol., 46: 901–909. Nudo, R.J. (2003) Adaptive plasticity in motor cortex: implications for rehabilitation after brain injury. J. Rehabil. Med., 41(Suppl): 7–10. Nudo, R.J., Barbay, S. and Kleim, J.A. (2000) Role of neuroplasticity in functional recovery after stroke. In: Levin, H.m. and Grafman, J. (Eds.), Neuroplasticity and Reorganization of Function After Brain Injury. Oxford University Press, New York, pp. 168–197. Nudo, R.J., Plautz, E.J. and Frost, S.B. (2001) Role of adaptive plasticity in recovery of function after damage to motor cortex. Muscle Nerve, 24: 1000–1019. Papanicolaou, A.C., Moore, B.D., Deutsch, G., Levin, H.S. and Eisenberg, H.M. (1988) Evidence for right-hemisphere involvement in recovery from aphasia. Arch. Neurol., 45: 1025–1029. Pascal-Leone, A. and Torres, F. (1993) Plasticity of the sensorimotor cortex representation of the reading finger in Braille readers. Brain, 116: 39–52. Rimel, R.W., Giordani, M.A., Barth, T.J. and Jane, J.A. (1982) Moderate head injury: completing the clinical spectrum of brain trauma. Neurosurgery, 11: 344–351. Robertson, I.H. and Murre, J.M.J. (1999) Rehabilitation of brain damage: brain plasticity and principles of guided recovery. Psychol., Bull., 125(5): 544–575.
Rosenzweig, M.R. (1999) Effects of differential experience on brain and cognition throughout the life span. In: Broomanand, S.H. and Fletcher, J.M. (Eds.), The Changing Nervous System: Neurobehavioral Consequences of Early Brain Disorders. Oxford University Press, New York, pp. 25–56. Sadato, N., Pascual-Leone, A., Grafman, J., Ibanez, V., Deiber, M.P., Dold, G. and Hallett, M. (1996) Activation of the primary visual cortex by Braille reading in blind subjects [see comments]. Nature, 380: 526–528. Schallert, T., Bland, D.T. and Leasure, J.L. (2000) Motor rehabilitation, use-related neural events, and reorganization of the brain after injury. In: Levin, H.S. and Grafman, J. (Eds.), Cerebral Reorganization of Function after Brain Damage. Oxford University Press, New York, pp. 145–167. Schlaug, G., Knorr, U. and Seitz, R.J. (1994) Inter-subject variability of cerebral activations in acquiring a motor skill: a study with positron emission tomography. Exp. Brain Res., 98: 523–534. Taub, E. (2004) Harnessing brain plasticity through behavioral techniques to produce new treatments in neurorehabilitation. Am. Psychol., 59: 692–704. Traversa, R., Cininelli, P., Bassi, A., Rossini, P.M. and Bernardi, G. (1997) Mapping of motor cortical reorganization after stroke. A brain stimulation study with focal magnetic pulses. Stroke, 28: 110–117. Vicari, S., Albertoni, A., Chilosi, A.M., Cipriani, P., Cioni, G. and Bates, E. (2000) Plasticity and reorganization during language development in children with early brain injury. Cortex, 36: 31–46. Wilmore, L.J., 2005. Antiepileptic drugs and neuroprotection: current status and future roles. Epilepsy Behav., S25–S28. Wykes, T., Reeder, C., Corner, J., Williams, C. and Everitt, B. (1999) The effects of neurocognitive remediation on executive processing in patients with schizophrenia. Schizophr. Bull., 25: 291–307. Xerri, C., Merzenich, M.M., Peterson, B.E. and Jenkins, W. (1998) Plasticity of primary somatosenosory cortex paralleling sensorimotor skill recovery from stroke in adult monkeys. J. Neurophysiol., 79: 2119–2148.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 14
Novel cell therapy approaches for brain repair Svitlana Garbuzova-Davis1,, Alison E. Willing1, Samuel Saporta1, Paula C. Bickford1, Carmelina Gemma1, Ning Chen1, Cyndy D. Sanberg3, Stephen K. Klasko2, Cesario V. Borlongan4 and Paul R. Sanberg1 1
Center of Excellence for Aging and Brain Repair, Department of Neurosurgery, University of South Florida, College of Medicine, MDC 78, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA 2 University of South Florida Health, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA 3 Saneron CCEL Therapeutics Inc., 3802 Spectrum Blvd., Suite 145, Tampa, FL 33612, USA 4 Department of Neurology, Medical College of Georgia, 15th Street, Augusta, GA 30912, USA
Abstract: Numerous reports elucidate that tissue-specific stem cells are phenotypically plastic and their differentiation pathways are not strictly delineated. Although the identity of all the epigenetic factors which may trigger stem cells to make a lineage selection are still unknown, the plasticity of adult stem cells opens new approaches for their application in the treatment of various disorders. There is increasing researcher interest in hematopoietic stem cells for treatment of not only blood-related diseases but also various unrelated disorders including neurodegenerative diseases. Human umbilical cord blood (hUCB) cells, due to their primitive nature and ability to develop into nonhematopoietic cells of various tissue lineages, including neural cells, may be useful as an alternative cell source for cell-based therapies requiring either the replacement of individual cell types and/or substitution of missing substances. Here we focus on recent findings showing the robustness of adult stem cells derived from hUCB and their potential as a source of transplant cells for the treatment of diseased or injured brains and spinal cords. Depending upon the pathological microenvironment in which the hUCB cells are introduced, neuroprotective and/or trophic effects of these cells, from release of various growth or anti-inflammatory factors to moderation of immuneinflammatory effectors, may be more likely than neural replacement. These protective effects may prove essential to maintaining restored tissue integrity over the course of various diseases or injuries. Keywords: neurodegenerative diseases; brain and spinal cord injury; umbilical cord blood cells; transplantation; alternative treatment nervous system (CNS) of all mammalians, including humans, was discovered (reviewed in Cameron and McKay, 1998; Gage, 2000; Gritti et al., 2002; Lie et al., 2004), the traditional view of the inability of mature nervous tissues to renew and reconstruct themselves has been largely debunked. New neurons are constantly generated from neural stem cells throughout life in restricted brain regions which actually contain adult stem cells. These cells can give rise to differentiated progeny comprising the major cell types of the CNS,
Introduction The brain is a complex organ containing billions of neurons and other cells, specialized by structure and function. A unique cytoarchitecture and neuronal network of rigorous complexity compose the brain, the source of all the qualities that define our humanity. Since neurogenesis in the adult central Corresponding author. Tel.: (813) 974-3189; Fax: (813) 9743078; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57014-1
207
208
neurons and glia, assumed to be responsible for nervous tissue homeostasis and repair throughout adulthood to aging. However, a decline of newly generated neurons has been observed with aging (reviewed in Bernal and Peterson, 2004). Although the mechanisms leading to a reduction of endogenous neurogenesis with increasing age are still unclear, functional decline in the ‘‘normal’’ aging processes is mainly associated with the loss of synaptic densities in many brain structures and impairment of cellular metabolism due to increased oxidative damage to DNA and proteins (reviewed in Limke and Rao, 2002; Brazel and Rao, 2004). Such alterations could be even more critical in age-related neurodegenerative disorders such as Alzheimer’s disease (AD), Parkinson’s disease (PD), and amyotrophic lateral sclerosis (ALS) which afflict more than 6.5 million people in the United States alone. Since current therapies for these devastating disorders merely treat the symptoms but do not provide cures, it is important to develop new therapeutic strategies to replace cells lost in neurodegenerative diseases through neural implantation. In cell-based therapeutics, neural stem cells offer hope for repair of the diseased or injured brain by their endogenous or exogenous (i.e., transplantation) activation. The rise of this neural stem cell concept has raised many questions in these cells therapeutic applications. It has been shown that the fate of adult neural stem cells is under specific environmental control and their proliferation and differentiation biased by different epigenetic signals, such as cytokines and various growth factors (reviewed in Gritti et al., 2002; Baizabal et al., 2003). The changes in microenvironment during disease or injury could influence either endogenous or exogenous repair. For example, Parent (2003) showed increased neurogenesis in the persistent germinative zones (subventricular zone and hippocampal dentate gyrus) using experimental epilepsy and stroke in the adult rodent. Moreover, as recently reported, neurogenesis was elevated in brains of patients with either Huntington’s disease (Curtis et al., 2003) or AD (Jin et al., 2004). These results indicate the great regenerative potential of the human brain, but this compensatory mechanism of selfrepair as a response to brain insult does not
provide complete healing. Limke and Rao (2002) point out, ‘‘As with any disease, the development of new therapies relies heavily on a thorough knowledge of the biology of the system being studied, and the ramifications of alterations of that system during disease.’’ The complexities and specificity of CNS diseases make cell replacement therapy a challenging but potentially very rewarding area for research. Similar to primitive stem cells, adult neural stem cells have capacity for self-renewal and can generate cells other than themselves through asymmetric cell division. Recent findings indicate that neural stem cells display broader than expected multipotentiality. It has been shown that they can differentiate into non-CNS derivates, such as blood cells (Bjornson et al., 1999) or skeletal muscle (Galli et al., 2000). Controversially, bone marrow stem cells can give rise to muscle (Ferrari et al., 1998), hepatic (Petersen et al., 1999), endothelial cells, osteoblasts (Dennis and Charbord, 2002), and neural cells (Sanchez-Ramos et al., 2000; Woodbury et al., 2000). Additionally, the muscle precursors can turn into blood cells (Jackson et al., 1999). The ability of tissue-specific stem cells (neural stem cells and somatic stem cells) to give rise to unrelated cell types may define these cells as pluripotent in nature. In this context, Slack’s (2000) overview of studies on stem cells in epithelial tissues suggested that ‘‘all types of stem cell are the same.’’ Although which epigenetic factors may trigger stem cells to make a lineage selection are still unknown, the plasticity of adult stem cells opens new approaches for their application in treatment of various neurodegenerative disorders. However, in practical view, the use of human neural stem cells as well as embryonic stem cells for brain therapy (i.e., transplantation) is currently obstructed due to ethical and other issues (Vescovi et al., 1999; Henon, 2003; Daar et al, 2004; Sanberg, 2005), in addition to difficulties in ensuring purity of neural cultures and harvesting needed populations for transplantation (Riaz et al., 2002; Linazasoro, 2003; Sathananthan and Trounson, 2005). Owing to these limitations, exploration of new sources of stem cells for therapeutic perspectives is necessary. It seems that tissue-specific stem cells are phenotypically plastic
209
and their differentiation pathways are not strictly delineated. Increasing interest of many researchers in plasticity of hematopoietic stem cells in treatment of not only blood-related diseases but also various unrelated disorders, encourages investigations into the possibilities of using these cells to treat neurodegenerative diseases. Here we focus on recent findings, including ours, showing the robustness of adult stem cells derived from hematopoietic tissue, particularly, human umbilical cord blood (hUCB), in identifying their specific potential as a transplantable cell source for the treatment of diseased or injured brain and spinal cord.
Umbilical cord blood cells: new vision or novel assets Numerous reports elucidate the many advantages of hUCB cells for cellular therapies. These cells are easily accessible in unlimited supply, avoiding ethical and other issues, and hUCB cells may be preferable to other potential cell sources (Sanberg et al., 2001, 2002). Hematopoietic progenitors from hUCB are rich in the most primitive stem cells and are capable of long-term repopulation of blood lineages (Broxmeyer et al., 1989, 1992; Mayani and Lansdorp, 1998; Todaro et al., 2000; Nayar et al., 2002). The number of myeloid progenitor cells in hUCB is similar to the number in bone marrow (Broxmeyer et al., 1992), however, hUCB cells have a greater colony-forming ability (Nakahata and Ogawa, 1982) and can be expanded in long-term cultures in vitro using different growth factors and have longer telomeres than adult cells (Vaziri et al., 1994). Cord blood transplants have already been used to reconstitute bone marrow and blood cell lineages in children with various hematological malignant and nonmalignant diseases (Lu et al., 1996; Sirchia and Rebulla, 1999). Since 1988, when the first transplant was performed in a patient with Fanconi anemia (Gluckman et al., 1989), cord blood transplantations have increased; more than 3000 cases have been reported worldwide (Gluckman, 2000; Broxmeyer, 2004). Moreover, hUCB transplants from unrelated donors (vs. autologous transplantation) have been successfully used for
children and adult patients (Gluckman et al., 1997, 1999; Rubinstein et al., 1998; Ooi et al., 2002; Wagner et al., 2002). Recently, it has been shown that hUCB transplants, versus adult bone marrow stem cells, better restore the host hematopoietic progenitor cell reservoir (Frassoni et al., 2003). Some studies indicate that a single hUCB sample supplies enough hematopoietic stem cells to provide both short- and long-term engraftment (Lu et al., 1996; Sirchia and Rebulla, 1999). This advantage is due to the immune-immaturity of the hUCB cells, which reduces the risk and severity of graft-versus-host disease (GVHD) after transplantation (Madrigal et al., 1997; Gluckman, 2000; Thomson et al., 2000). In the first electron microscopic comparison of cord blood, peripheral blood and bone marrow cells, hUCB cells had a more immature morphology of the myelo-monocytic cells with small numbers of mature neutrophils and unique ultrastructure elements, such as nuclear pockets in the neutrophils, which accelerated the transport of RNA to the cytoplasm (Mikami et al., 2002). The immaturity of immunological properties in hUCB cells is believed to cause a prolonged immunodeficient state after hUCB transplantation (Roncarolo et al., 1994; Garderet et al., 1998; Thomson et al., 2000). However, it has been shown that hUCB cells contain a fully constituted T-cell repertoire (Garderet et al., 1998). When comparing the characteristics of B-cell differentiation in vitro from CD34+ cord blood cells with those of peripheral blood, it was found that B-cell precursors differentiated from cord blood are more immature (Hirose et al., 2001); this may cause the delay in mature B-cell production. Moreover, cord blood lymphocytes expressed cytokine receptor profiles (IL-2, IL-4, IL-6, IL-7, TNF-a, and interferon-g) at lower levels than in adult blood cells (Zola et al., 1995) and produced great amounts of the antiinflammatory cytokine IL-10 (Rainsford and Reen, 2002). Recently, it has been demonstrated that human CD34+ cord blood cells can completely reconstitute the immune system in NOD/SCID mice with functionally competent cells, as indicated by IgM, IgA, and IgG expression (Hiramatsu et al., 2003). These human lymphocytes also formed organized structures in mouse spleens and thymi.
210
Current interest in the transplantation field for clinical applications focuses on the ex vivo expansion of cord blood precursor/progenitor/stem cells to provide a sufficient amount of stem cells for adult transplantation (Conrad and Emerson, 1998; Liesveld, 2003). Major questions arise on selection of an optimal stem cell population for expansion and definition of desired characteristics of the expanded stem cells to be used for engraftment (Shih et al., 2000). Human AC133 (CD133) antigen has been identified as a hematopoietic stem cell marker that may provide an alternative to CD34 for the selection and expansion of hematopoietic cells for transplantation (Yin et al., 1997; Kobari et al., 2001). It has been shown that about 80% of CD34+ cells express CD133 and more than 97% of CD133+ cells are CD133+CD34+ in fresh cord blood (Hao et al., 2003). Although CD133+ cells comprised 0.67% of the total mononuclear hUCB cells (Ma et al., 2002), expansion of CD133+ and CD133+CD34+ cells was significantly higher than those from the CD34+ cells (Hao et al., 2003). These findings suggest that CD133+ may be more primitive hematopoietic progenitor/stem cells than CD34+. In the first clinical trial of autologous transplantation of CD133 selected progenitors in a pediatric patient with relapsed leukemia, complete remission was reported at follow-up, 11 months after transplantation (Koehl et al., 2002). To advance the usefulness of the hUCB cells in treating neurodegenerative diseases, an in vitro study was first conducted by our research group (Sanchez-Ramos et al., 2001). We showed that mononuclear hUCB (MNC hUCB) cells treated with retinoic acid (RA) and nerve growth factor (NGF) expressed molecular markers usually associated with neurons and glia, as determined by immunocytochemistry, Western blot, and DNA microarray. The MNC hUCB cells under neuralization-inducing (RA+NGF) media expressed specific markers for early neural precursors (musashi-1, nestin, TuJ1), mature neurons (NeuN, MAP2), and astrocytes (GFAP). Moreover, cells exposed to RA+NGF treatment increased TuJ1 and GFAP expressions by approximately two times. Similar to this study, we have demonstrated that in standard growth (i.e., DMEM) media, MNC hUCB cells express neural markers, such as nestin, TuJ1, MAP2,
and GFAP (Garbuzova-Davis et al., 2003). Colocalization of nestin and MAP2 and various cytoplasmic expressions of TuJ1 by cells at 2 weeks after plating were also observed; findings that we suggest may depend upon cell cycle or cell development. Additionally, the increased number of cells expressing CD133 antigen, a marker of primitive hematopoietic progenitor/stem cells, in 7 days cultured cells probably gives rise to cells which show immature and mature neural characteristics at the same time. This novel benefit of hUCB cells, that they can express antigens typical of neural lineages within the CNS, has been confirmed by other researchers. Bicknese et al. (2002) and Buzanska et al. (2002) demonstrated that these cells could be induced to express class III b-tubulin, GFAP, and GalC (a marker of oligodendrocytes). Similarly, Ha et al. (2001) showed that these cells can express the neurofilament microtubule associated protein 2 (MAP2) using immunohistochemistry and RT-PCR. In a study reported by Goodwin et al. (2001), a subset of cells from MNC hUCB, which had been maintained in continuous culture for more than 6 months, was described without antigen expression for hematopoietic differentiation. When these cells were exposed to osteogenic, adipogenic agents or basic fibroblast and epidermal growth factors, they expressed bone, fat and neural markers, respectively. According to these data, hUCB contains a cell population, which is capable of expressing antigens of multiple lineages, demonstrating the plasticity of these cells that is very important for cellular therapeutics with the goal of system repair. However, the authors raise critical questions about the nature of this cell population. They cannot conclude that these cells are ‘‘a stem cell population, multiple differentiated progenitors, or cells with transdifferentiation capacity’’ (Goodwin et al., 2001). More investigation needs to be performed using various testing of these cells, including in vivo studies. Confirming the presence of a cell population in hUCB with multipotent ability, McGuckin’s research team developed a negative immunomagnetic selection method that depletes hUCB from hematopoietic lineage marker-expressing cells, therefore isolating a discrete lineage negative (LinNeg) stem cell
211
population (0.1% of MNC hUCB) (Forraz et al., 2004; McGuckin et al., 2004). These selected LinNeg hUCB cells expanded primitive nonadherent hematopoietic progenitors (up to 47-fold) and simultaneously produced adherent cells with neuroglial progenitor cell morphology over 8 weeks. Gene expression analysis showed upregulation of primitive neuroglial progenitor cell markers for GFAP, nestin, musashi-1, and necdin. Recently, we characterized in vitro two different subpopulations of MNC hUCB cells – adherent and floating (Chen et al., 2005). We found that there were a significant number of progenitor/stem and neural cell antigen expressions on cells in the floating population. The adherent cell population mainly contained lymphocytes (over 50%) expressing hematopoietic antigens. These results suggest that a nonhematopoietic subpopulation of cells exists within MNC hUCB cells and seems to have the potential to become neural cells. Although this study is continuing, it is clear that hUCB has a primitive stem cell population that may give rise to both hematopoietic and neural cells. Thus, hUCB cells due to their primitive nature, with ability to transdifferentiate or become nonhematopoietic cells of various tissue lineages, including neural cells, may be useful for cell-based therapies requiring either the replacement of individual cell types and/or substitution of missing substances. Using hUCB cells as an alternative cell source in the treatment of neurodegenerative disorders is therefore an attractive approach.
Umbilical cord blood cells in treatment of diseased brain and spinal cord Stroke Stroke is the most common age-related cerebrovascular disease, which is the third leading cause of morbidity and mortality in the United States (Gorelick et al., 1999). Care for this disease is problematic due to certain disease attributes. Stroke can involve multiple anatomical brain structures, affecting different neuronal cell populations, and disrupting various neuroanatomical pathways. Ischemic injury may be an ongoing
process increasing cell and/or tissue damage; so timing to begin therapeutic course is critical. Currently, the only effective treatment for stroke (tissue plasminogen activator or TPA) must be delivered within a restricted time frame from the initiation of the stroke. These disease outcomes should be taken under consideration in developing any therapeutic intervention, especially, in cellbased therapy for stroke. A recent review (Savitz et al., 2004) discussed preclinical and clinical studies on potential cell therapy for stroke. In on-going clinical trials, patients with intrastriatal neuronal transplantation of the immortalized cell line NT2N (LBS neurons) showed a trend toward improvement in functional outcomes on several scales, compared with baseline measurements before transplantation. Some transplanted patients improved on a test of memory 6 months after transplantation. Another pilot study on intrastriatal transplantation of fetal cells from the pig into patients with basal ganglia infarcts demonstrated that the patients developed no new neurological deficits in the acute setting. But at 2 years, no patients showed improvement on the modified Rankin scale and the study was terminated by the FDA. Although clinical trials on neuronal cell transplantation are feasible, using multipotent stem cells from a different source might be more beneficial. A series of published reports by Chopp and colleagues (Chen et al., 2001a, 2003; Chen et al., 2002; Li et al., 2002) on using bone marrow stromal cells as a source of transplantable cells for treatment of stroke demonstrated functional improvement in rats after focal cerebral ischemia. Transplanted cells migrated to areas of ischemic infarcts and differentiated into neuronal and glial cell types. Authors suggested that recovery mechanisms are likely due to trophic factors released by cells which may promote endogenous neurogenesis and angiogenesis rather than a result of neuronal replacement. The hUCB is another source of multipotential stem cells that has shown promising effects in preclinical studies for treatment of stroke. The intravenous administration of MNC hUCB cells (3 106) at 24 h or 7 days after middle cerebral artery occlusion (MCAO) in a rat model of stroke significantly improved neurological function (Chen
212
et al., 2001b). Upon histological examination of the brains, MNC hUCB cells were observed mainly in the cortex and striatum of the injured hemisphere in the ischemic boundary zone. Few cells were found in the contralateral hemisphere. Using immunohistochemistry, it was determined that some of these MNC hUCB cells were immunoreactive for neuronal markers NeuN (2%) and MAP2 (3%), the astrocytic marker GFAP (6%), and the endothelial cell marker FVIII (8%). The transplanted cells were also detected outside the brain in bone marrow (3%), spleen (1%), muscle, heart, lung, and liver (0.01%–0.5%). Recently, in our study (Willing et al., 2003), we compared the effect of intravenous versus intrastriatal injection of MNC hUCB cells to assess which produced the greatest behavioral recovery in rats with permanent MCAO. It was found that spontaneous activity was significantly less when cells were transplanted 24 h after stroke compared with nontreated stroke animals. Furthermore, behavioral recovery was similar with both cell delivery routes. However, in one functional test (step test) at 2 months after transplant, significant improvements were found only after intravenous delivery of the MNC hUCB cells. Also, in the passive avoidance test, transplanted animals learned the task faster than nontransplanted rats. These results suggest that intravenous delivery of MNC hUCB cells may be more effective than direct striatal delivery in producing long-term functional benefits to the stroke animal. In continuing studies of our research team lead by Dr. Willing, the effect of increasing doses of MNC hUCB cells after MCAO on the behavioral recovery and stroke infarct volume in rats (Vendrame et al., 2004) was examined. Twenty-four hours after induced stroke, rats were intravenously infused with 104–3–5 107 MNC hUCB cells. Results showed that, at 4 weeks after infusion, there was a significant recovery in behavioral performance (spontaneous activity, step test, elevated body swing test) when 106 or more MNC hUCB cells were delivered. Infarct volume measurements revealed an inverse relationship between cell dose and damage volume, which reached significance at the higher doses of MNC hUCB cells (107 cells, po0.01; 3–5 107 cells, po0.05). Moreover,
transplanted cells were localized by immunofluorescence for human nuclei antigen expression and PCR analysis only in the injured brain hemisphere and spleen. These results extend previous observations of MNC hUCB cell infusion in the MCAO rat stroke model by demonstrating a dose relationship between introduced transplanted cells, behavioral improvement, and neuronal sparing. Lately, we showed (Vendrame et al., 2005) that in the brain of rats with permanent MCAO, there was an increase in the number of CD45+/CD11b+ cells (lymphocytes). After MNC hUCB cell transplantation, the number of CD45+/CD11b+ cells returned to the level observed in the normal brain. There was a decrease in the number of CD45+/ CD11b+ cells (resting microglia) in the MCAO brain that was reversed by hUCB transplantation. After stroke, there was also a large increase in the cell population labeled with CD45+/CD11b+, which returned toward normal values in the transplanted animals. This population has been previously characterized as mainly granulocytes, macrophage, and activated microglia. These cellular changes were accompanied by decreases in mRNA and protein expression of pro-inflammatory cytokines and in nuclear factor kappaB (NF-kappaB) DNA binding activity in the brain of stroke animals treated with MNC hUCB cells. In addition to modulating the inflammatory response, we demonstrated that the transplanted cells increased neuronal survival through nonimmune mechanisms. Once thought of as ‘‘cell replacement therapy,’’ we now propose that cord blood treatment in stroke reduces inflammation and provides neuroprotection. Also, we recently showed that MNC hUCB cells could benefit in stroke by providing protective trophic neuronal support through secretion of glial-derived neurotrophic factor (GDNF) or other growth factors (Sanberg et al., 2004). Finally, the therapeutic window for treatment of individuals after stroke is narrow, regardless of the treatment regime; extension of this window would provide a major therapeutic advance. Based on our findings that significant improvement occurred in the behavior of rats receiving MNC hUCB cells 24 h after MCAO, another study from our research group (Newman et al., 2005) attempted to determine the optimal time to administer these
213
cells after stroke. Using ischemic tissue extracts, the migration capability of MNC hUCB cells was investigated. We demonstrated increased migratory activity of MNC hUCB cells toward the extracts harvested at 24–72 h after stroke. The extracts possessed increased levels of certain cytokines and chemokines, suggesting participation of these substances in the cell migration. The results from this study are promising in that the current 3 h therapeutic window for the treatment of stroke victims, using approved anticoagulant treatment, may be extended with the use of MNC hUCB cell therapy to 24–72 h post stroke event. Also, the chemokines present in the supernatant could provide a sound starting point for examining the mechanisms responsible for the in vivo migration of MNC hUCB cells after stroke induction. However, as concluded by Savitz et al. (2004) ‘‘Transplantation is unlikely to succeed if there is a severe arterial occlusion without collateral circulation; inadequate blood supply would not support graft survival.’’ On this point, an other advantage for using hUCB cells for treatment of stroke is potential restoration of vascularity since cord blood contains endothelial progenitor cells which may be of use in proangiogenic neovascularization therapy (Delorme et al., 2005; Shin et al., 2005).
Amyotrophic lateral sclerosis (ALS) Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder characterized by a loss of motor neurons throughout the neural axis that clinically manifests as a progressive muscular weakness leading to paralysis and death, usually within 3–5 years of diagnosis. The diffuse degeneration includes damage to upper motor neurons in the motor cortex, lower motor neurons in the spinal cord, and some brainstem nuclei. New therapeutic strategies, including cell replacement, for this disease are difficult to develop due to multifocal and multicausal motor neuron death (Silani and Leigh, 2003). In agreement with Bruijn’s (2002) comment concerning neural stem cell replacement therapies for ALS, ‘‘it is hard to imagine that transplanted motor neurons would form appropriate
connections with target muscle,’’ there are severe limitations to using this treatment in ALS. Since numerous hypotheses about the etiopathology of ALS have been proposed (reviewed in Cleveland and Rothstein, 2001; Bruijn et al., 2004), increasing evidence points to immune system involvement in the pathogenesis of ALS. If ALS is an autoimmune disease, as some have hypothesized (Kawamata et al., 1992; Alexianu, 1995; Niebroj-Dobosz et al., 1999; Alexianu et al., 2001; Mohamed et al., 2002), hUCB cells may improve disease outcome through immune modulation. Supporting this hypothesis is evidence showing that intravenous administration of a large dose (35 106) of MNC hUCB cells into irradiated G93A mice substantially increased lifespan of mice (Ende et al., 2000; Chen and Ende, 2002). While the survival data was impressive, the investigators did not examine motor function in these animals or determine the underlying mechanism(s). Authors suggested that MNC hUCB cells possibly ‘‘provide enhanced hematopoietic reconstitution of the irradiated hosts own stem cells’’ (Ende et al., 2000). In our study, a single low dose (106 cells) of MNC hUCB administered into the systemic circulation of the presymptomatic G93A mice delayed disease progression at least 2–3 weeks, as determined by testing motor function, and modestly prolonged the lifespan (Garbuzova-Davis et al., 2003). Transplanted cells survived long-term (10–12 weeks) post-transplantation and were found widely distributed in the brain, spinal cord, and other organs. Although most cells were associated with blood vessels, some cells migrated into the parenchyma within the brain and spinal cord and expressed early neural (nestin), neuronal (TuJ-1), and astrocytic (GFAP) markers. However, only morphological evidence was found for differentiated astrocytes. The other finding in our study was that a large number of MNC hUCB cells were present in the spleens of treated mice. As this was the first study demonstrating that MNC hUCB cells altered the peripheral immune system, we examined the immune phenotypes (T- and Bcell antigen expression) of intravenously transplanted cells found in the spleen (Desjarlais et al., 2003). Many of the cells expressed CD8 and CD19
214
while few expressed CD4. Only a few cells were found positive for CD28 and CD80, indicating activation of T- and B-cells. When we analyzed Tand B-cell antigen expression of MNC hUCB cells, which had migrated to the spinal cord after their administration into systemic circulation, most cells expressed CD4 and CD8; fewer expressed CD19, CD28, and CD80. These results suggest that the MNC hUCB cells can not only differentiate into immune cells, but are actively involved in the immune response. It is possible that the beneficial effect of MNC hUCB cells may occur through peripheral immunomodulation of immune effectors due to MNC hUCB development into cells with immune regulatory function. An additional advantage of MNC hUCB cells in treatment of ALS could be through altered function of the lymphoid system which critically influences the immune system. Significant lymphopenia and an increase of the CD4/CD8 ratio have been noted in ALS patients even at an early disease stage (Provinciali et al., 1988). This was confirmed by our recent findings that severe lymphopenia accompanied by spontaneous autorosette formation was determined in G93A mice modeling ALS at the terminal stage of disease (Kuzmenok et al., 2006). It is possible that restoration of the lymphoid system by hUCB cell transplantation may elevate ‘‘defense’’ for motor neurons.
Umbilical cord blood cells in treatment of injured brain and spinal cord Traumatic brain injury (TBI) and spinal cord injury (SCI) result from external physical insult and are associated with high morbidity and mortality. There are currently no sufficient treatments. Brain neurotrauma is characterized not only by focal abnormalities, but rather by multifocal, or even global structural and functional disturbances of the brain network. In both TBI and SCI, the impact initially causes necrotic cell death and then apoptotic cell death in the underlying and surrounding tissues due to multiple subsequent events, such as ischemia and excitotoxicity. Stem cells might participate in reconstructing the molecular and cellular milieu of traumatized brains or spinal cords
(Brodhum et al., 2004; Kulbatski et al., 2005). However, there is no definitive answer about the ideal cell type for transplantation. A recent retrospective analysis of 70 cases of brain trauma or paraplegia treated with neural stem cell transplantation concluded that the transplantation promoted functional recovery (Zhou et al., 2004). In another clinical trial, Rabinovich et al. (2003) reported that cells from fetal nervous and hemopoietic tissues (gestational age 16–22 weeks) had been subarachnoidally implanted into 15 patients with severe SCI at cervical or thoracic spine level. Each patient received from one to four cell grafts at various time intervals. With cell treatment, six patients improved their neurological status exhibiting incomplete restoration of both motor and sensory function. The status of five other cell-treated patients became consistent and was characterized by appearance of contracting activity in some muscles and incomplete restoration of sensitivity. The remaining four patients did not exhibit any clinical improvements. No serious complications of cell transplantation were noted. These results suggest the clinical relevance of the cell-based approach to treating severe consequences of SCI. In our study, we showed that when MNC hUCB cells (2 106) were delivered to the tail vein of rats with TBI, neurological deficits were reduced (Lu et al., 2002). Wide distribution of administered cells in the brain and peripheral organs was detected. The cells which migrated into the parenchyma of the injured brain expressed neuronal markers NeuN and MAP2 and the astrocytic marker GFAP. In another of our studies, intravenously delivered MNC hUCB cells (106), at 1 day or 5 days postinjury in rats with compression injury of the spinal cord, increased the rate of behavioral recovery (Kim et al., 2002; Saporta et al., 2003). However, rats which received the cells 5 days after injury showed significantly improved recovery of motor function compared to those that received cells on the 1st day postinjury, and both were significantly better than animals upon which only laminectomy was performed. The transplanted cells were observed in injured areas of rat spinal cords and were never seen in corresponding areas of spinal cord of noninjured animals. The results
215
of our studies are consistent with the hypothesis that cord blood-derived stem cells participate in the healing of neurological defects caused by not only disease but also by traumatic assault. It is interesting to note that even direct injection of hUCB-derived AC133 (hematopoietic progenitor/stem) cells into a demyelinated lesion of rat spinal cord showed extensive axon remyelination by light and electron microscopic examination (Honmou et al., 2002). Thus, our studies demonstrated that hUCB cells delivered into the systemic circulation were able to migrate into injured/damaged areas of the brain or spinal cord and express neural-like markers without pre-exposure to specific factors. Although the mechanism of cell migration remains unclear, Lu et al. (2002) suggest that hUCB cells ‘‘may enter the brain from blood–brain barrier disruption or in response to signals from cytokines and cell surface receptors and antigens’’ and ‘‘the microenvironment of the brain [spinal cord] after injury may drive hUCB cells into a neural cell phenotype.’’
Umbilical cord blood cells in treatment of metabolic diseases The core strategy in developing a therapy for mucopolysaccharidosis (MPS) is replacement or delivery of the missing enzyme. Cell therapy may show promise as a new treatment for this disease. Recently, cord blood transplants from unrelated donors were shown to improve neurocognitive performance and decrease somatic features in patients with Hurler’s syndrome (MPS type I) (Staba et al., 2004). We investigated the prospects of MNC hUCB as a potential cell source for treatment of Sanfilippo syndrome type B (MPS III B). MPS III B is an autosomal recessive disorder caused by a deficiency of a-N-acetylglucosaminidase enzyme (Naglu). The lack of Naglu enzyme leads to accumulation of heparan sulfate, a glycosaminoglycan (GAG), within lysosomes. Clinical symptoms appear after 2 years of normal development and then progressive cerebral and systemic multiple organ abnormalities are seen.
Our examination of Naglu enzyme activity of MNC hUCB cells in vitro showed that cells contain and extracellularly release the Naglu enzyme, making them a suitable vehicle for use in enzyme replacement therapy (Garbuzova-Davis et al., 2005). When we administered MNC hUCB cells into the cerebral ventricle of Naglu mice modeling MPS III B at 1 month of age, transplanted cells survived long-term (7 months), migrated into the parenchyma of the brain and peripheral organs, expressed neural antigens (nestin, NeuN, GFAP), and exhibited neuron and astrocyte-like morphology. Transplant benefits were also demonstrated by stable cytoarchitecture in the hippocampus and cerebellum, and by reduced GAGs in the livers of treated mutant mice. Although all these results indicate the beneficial effects of MNC hUCB cells, mechanisms of cell migration, as well as mechanisms promoting cell integration and differentiation in the host environment are still unknown. Many questions remain concerning the ability of MNC hUCB cells to treat multifaceted diseases, such as MPS III B, with complex factors underlying the pathogenesis. However, these are the first results supporting enzyme replacement by administered MNC hUCB cells and may lead to new strategies for delivery of the missing enzyme. We are currently investigating systemic administration of hUCB cells into Naglu mice, which may prove even more effective. In one of our more recent studies (GarbuzovaDavis et al., 2006), we attempted to take advantage of the passage of maternal cells into the fetus during pregnancy for prenatal delivery of Naglu enzyme into the enzyme-deficient mouse model of MPS III B. Enzymatically sufficient MNC hUCB cells were intravenously administered into heterozygote females modeling MPS III B on the 5th day of pregnancy during blastocyst implantation. We found that administered MNC hUCB cells transmigrated and diffused into the embryos (E12.5) and some cells expressed CD34 and CD117 antigens. Additionally, transmigrated cells were found in both the maternal and embryonic parts of placentas and also were extensively distributed in the organs and the blood of heterozygote mothers at 1 week after transplantation. More importantly, MNC hUCB cells corrected Naglu enzyme
216
activity in all embryos. Thus, our results indicate that prenatal delivery of Naglu enzyme by MNC hUCB cell administration into mothers of enzymedeficient embryonic mice is possible and may present a significant opportunity for new biotechnologies to treat many inherited disorders.
Alternative and/or combined treatment of diseased or injured brain and spinal cord Recent work in our laboratory lead by Dr. Bickford and others has shown the potential beneficial actions of nutritional approaches to the treatment of aging and neurodegenerative diseases (Joseph et al., 1998; Bickford et al., 1999, 2000; Ferrante et al., 2001; Gemma et al., 2002). The use of fruits or vegetables has the benefit of providing a cocktail of numerous phytochemicals with multiple actions including antioxidant and anti-inflammatory effects and is one reason that many fruits and vegetables have been extensively studied in the field of cancer biology. For example, blueberries are known to contain many phenolic compounds such as anthocyanins that are potent antioxidants; the phenolic content changes with different berry varieties (Zheng and Wang, 2003). Spirulina, a blue green algae used for thousands of years as a food source by the Aztecs, is known to contain large amounts of b-carotene (Annapurna et al., 1991) and several phycocyanins (Bhat and Madyastha, 2001), all with potent antioxidant effects; phycocyanin is also known to have potent COX-2 inhibitory actions (Reddy et al., 2000, 2003) and as COX-2 is increased in ALS (Maihofner et al., 2003) and after stroke (Yokota et al., 2004), foods containing phycocyanin may have dual benefits. In fact, we have shown that aged rats fed diets supplemented with spirulina or spinach demonstrate improved motor learning in either a rodrunning motor learning task or classical eyeblink conditioning (Bickford et al., 2000; Cartford et al., 2002) and show decreases in both markers of oxidative damage and markers of inflammation (Cartford et al., 2002; Gemma et al., 2002). In an animal model of PD, the blueberry or spirulina diet was neuroprotective in that the size of the 6-hyrdoxydopamine (6-OHDA) induced
dopaminergic terminal loss was significantly smaller at 30 days postlesion (Stromberg et al., 2005). At this time, there was a significant reduction in MHC class II positive microglia, suggesting that part of the beneficial mechanism may be via reducing inflammation. At 4 weeks after the dopamine depletion by 6-OHDA injection, a significant increase in GFAPpositive profiles was found in lesioned animals given the control diet, while blueberry- and spirulinatreated animals showed no changes compared to sham-injected rats or to the 1 week time point. In a following study, we showed the protective effects of three antioxidant diets: blueberry, spinach and spirulina against stroke in rats (Wang et al., 2005). Our data indicate that these diets have different effects in reducing ischemia-induced caspase-3 activity and cerebral infarction. Animals were put on a control, blueberry, spinach, or spirulina diet for 4 weeks prior to the insult. We used a 60 min occlusion of the middle cerebral artery and at 24 h examined the size of the infarct using TTC staining. We found a 70% protection in infarct size in the spirulina treated rats and a 50% protection in both the blueberry and spinach treated rats. In these same animals, we observed a significant decrease in caspase-3 activity and in the number of TUNEL positive cells, indicating that a reduction of apoptosis was achieved. All groups also showed significant improvement on horizontal and vertical activity measures when compared with controls. Thus, our results suggest that nutritional supplements (spirulina, blueberries, and spinach) show a broad spectrum of neuroprotection in multiple models of neurodegeneration and can be used as alternative and/or combined treatment of diseased or injured brain and spinal cord. As studies continue using different model of neurodegenerative disease such as ALS, it will be interesting to note changes in the diseased microenvironment by diet supplementations prior to cell transplantation.
Conclusion Numerous reports by us and others have demonstrated the plasticity of MNC hUCB cells in both
217
in vivo and in vitro studies. These cells differentiate into blood, immune, or neural cell lineages and seem a promising novel cell source for treatment of various neurodegenerative diseases and the injured brain or spinal cord. The MNC hUCB cells may modulate host hematopoietic and immune systems or even replace dying cells depending upon the pathological microenvironment in which the MNC hUCB cells are placed. Currently, neuroprotective and/or trophic effects of these cells, from release of various growth or anti-inflammatory factors and/ or by moderation of immune-inflammatory effectors, seems more likely than neural replacement. These protective effects may be essential to maintaining/restoring tissue integrity over the course of various diseases or injuries. As evidence of MNC hUCB cell benefits continues to increase, so should enthusiasm mount for the use of these cells in new and existing therapies. The brain is the source of intelligence and memory, responsible for a variety of crucial interconnected systems. Given our present limited understanding of brain processes and interactions, it would be naı¨ ve to expect simple solutions to brain diseases and injuries with their inevitably complex pathologies. We expect that adult stem cell and nutritional therapies will increasingly be used in conjunction with other therapies in the treatment of brain injuries and diseases.
Conflict of interest disclosure S. Garbuzova-Davis, A.E. Willing, S. Saporta, P.C. Bickford, and C.V. Borlongan are consultants and P.R. Sanberg is Co-Founder and Chairman of Saneron CCEL Therapeutics, Inc. (SCTI, Tampa, FL). SCTI is a start-up company from the University of South Florida which is developing cord blood derived treatments for neurodegenerative and cardiovascular disorders.
References Alexianu, M.E. (1995) The role of immune processes in amyotrophic lateral sclerosis pathogenesis. Rom. J. Neurol. Psychiatry, 33: 215–227.
Alexianu, M.E., Kozovska, M. and Appel, S.H. (2001) Immune reactivity in a mouse model of familial ALS correlates with disease progression. Neurology, 57: 1282–1289. Annapurna, V., Deosthale, Y.G. and Bamji, M.S. (1991) Spirulina as a source of vitamin A. Plant Foods Hum. Nutr., 41: 125–134. Baizabal, J.M., Furlan-Magaril, M., Santa-Olalla, J. and Covarrubias, L. (2003) Neural stem cells in development and regenerative medicine. Arch. Med. Res., 34: 572–588. Bernal, G.M. and Peterson, D.A. (2004) Neural stem cells as therapeutic agents for age-related brain repair. Aging Cell, 3: 345–351. Bhat, V. and Madyastha, K.M. (2001) Scavenging of peroxynitrite by phycocyanin and phycocyanobilin from Spirulina platensis: protection against oxidative damage to DNA. Biochem. Biophys. Res. Commun., 285: 262–266. Bickford, P., Gould, T., Briederick, L., Chadman, K., Pollock, A., Young, D., Shukitt-Hale, B. and Joseph, J. (2000) Antioxidant-rich diets improve cerebellar physiology and motor learning in aged rats. Brain Res. Rev., 866: 211–217. Bickford, P., Shukitt-Hale, B. and Joseph, J. (1999) Effects of aging on cerebellar noradrenergic function and motor learning: nutritional interventions. Mech. Ageing Dev., 111: 141–154. Bicknese, A.R., Goodwin, H.S., Quinn, C.O., Henderson, V.C., Chien, S.N. and Wall, D.A. (2002) Human umbilical cord blood cells can be induced to express markers for neurons and glia. Cell Transp., 11: 261–264. Bjornson, C.R.R., Rietze, R.L., Reynolds, B.A., Magli, M.C. and Vescovi, A.L. (1999) Turning brain into blood: a hematopoietic fate adopted by adult neural stem cells in vivo. Science, 283: 534–537. Brazel, C.Y. and Rao, M.S. (2004) Aging and neuronal replacement. Ageing Res. Rev., 3: 465–483. Brodhum, M., Bauer, R. and Patt, S. (2004) Potential stem cell therapy and application in neurotrauma. Exp. Toxicol. Pathol., 56: 103–112. Broxmeyer, H.E. (2004) Proliferative, self-renewal, and survival characteristics of cord blood hematopoietic stem and progenitor cells. In: Broxmeyer, H.E. (Ed.), Cord Blood: Biology, Immunology, Banking, and Clinical Transplantation. AABB Press, Bethesda, Maryland, pp. 1–21. Broxmeyer, H.E., Douglas, G.W., Hangoc, G., Cooper, S., Bard, J., English, D., Arny, M., Thomas, L. and Boyse, E.A. (1989) Human umbilical cord blood as a potential source of transplantable hematopoietic stem/progenitor cells. Proc. Natl. Acad. Sci. USA, 86: 3828–3832. Broxmeyer, H.E., Hangoc, G., Cooper, S., Ribeiro, R.C., Graves, V., Yoder, M., Wagner, J., Vadhan-Raj, S., Benninger, L., Rubinstein, P. and Broun, E.R. (1992) Growth characteristics and expansion of human umbilical cord blood and estimation of its potential for transplantation in adults. Proc. Natl. Acad. Sci. USA, 89: 4109–4113. Bruijn, L.I. (2002) Amyotrophic lateral sclerosis: fromdisease mechanisms to therapies. BioTechniques, 32: 1112–11121. Bruijn, L.I., Miller, T.M. and Cleveland, D.W. (2004) Unraveling the mechanisms involved in motor neuron degeneration in ALS. Annu. Rev. Neurosci., 27: 723–749.
218 Buzanska, L., Machaj, E.K., Zablocka, B., Pojda, Z. and Domanska-Janik, K. (2002) Human cord blood-derived cells attain neuronal and glial features in vitro. J. Cell Sci., 115: 2131–2138. Cameron, H.A. and McKay, R. (1998) Stem cells and neurogenesis in the adult brain. Curr. Opin. Neurobiol., 8: 677–680. Cartford, M., Gemma, C. and Bickford, P.C. (2002) Eighteenmonth-old Fischer 344 rats fed a spinach-enriched diet show improved delay classical eyeblink conditioning and reduced expression of tumor necrosis factor alpha (TNFalpha) and TNFbeta in the cerebellum. J. Neurosci., 22: 5813–5816. Chen, N., Hudson, J.E., Walczak, P., Misiuta, I., GarbuzovaDavis, S., Jiang, L., Sanchez-Ramos, J., Sanberg, P.R., Zigova, T. and Willing, A.E. (2005) Human umbilical cord blood progenitors: the potential of these hematopoietic cells to become neural. Stem Cells, 23: 1560–1570. Chen, J., Li, Y., Wang, L., Zhang, Z., Lu, D., Lu, M. and Chopp, M. (2001a) Therapeutic benefit of intravenous administration of bone marrow stromal cells after cerebral ischemia in rats. Stroke, 32: 1005–1011. Chen, J., Sanberg, P.R., Li, Y., Wang, L., Lu, M., Willing, A.E., Sanchez-Ramos, J. and Chopp, M. (2001b) Intravenous administration of human umbilical cord blood reduces behavioral deficit after stroke in rats. Stroke, 32: 2682–2688. Chen, J., Zhang, Z.G., Li, Y., Wang, L., Xu, Y., Gautam, S.C., Lu, M., Zhu, Z. and Chopp, M. (2003) Intravenous administration of human bone marrow stromal cells induces angiogenesis in the ischemic boundary zone after stroke in rats. Circ. Res., 92: 692–699. Chen, R. and Ende, N. (2002) The potential for the use of mononuclear cells from human umbilical cord blood in the treatment of amyotrophic lateral sclerosis in SOD1 mice. J. Med., 31: 21–30. Chen, X., Li, Y., Wang, L., Katakowski, M., Zhang, L., Chen, J., Xu, Y., Gautam, S.C. and Chopp, M. (2002) Ischemic rat brain extracts induce human marrow stromal cell growth factor production. Neuropathology, 22: 275–279. Cleveland, D.W. and Rothstein, J.D. (2001) From Charcot to Lou Gehrig: deciphering selective motor neuron death in ALS. Nat. Rev. Neurosci., 2: 806–819. Conrad, P.D. and Emerson, S.G. (1998) Ex vivo expansion of hematopoietic cells from umbilical cord blood for clinical transplantation. J. Leukoc. Biol., 64: 147–155. Curtis, M.A., Penney, E.B., Pearson, A.G., van Roon-Mom, W.M., Butterworth, N.J., Dragunow, M., Connor, B. and Faull, R.L. (2003) Increased cell proliferation and neurogenesis in the adult human Huntington’s disease brain. Proc. Natl. Acad. Sci. USA, 100: 9023–9027. Daar, A.S., Bhatt, A., Court, E. and Singer, P.A. (2004) Stem cell research and transplantation: science leading ethics. Transplant. Proc., 36: 2504–2506. Delorme, B., Basire, A., Gentile, C., Sabatier, F., Monsonis, F., Desouches, C., Blot-Chabaund, M., Uzan, G., Sampol, J. and Dignat-George, F. (2005) Presence of endothelial progenitor cells, distinct from mature endothelial cells, within human CD146+ blood cells. Thromb. Haemost., 4: 1270–1279.
Dennis, J.E. and Charbord, P. (2002) Origin and differentiation of human and murine stroma. Stem Cells, 20: 205–214. Desjarlais, T., Garbuzova-Davis, S., Willing, A.E., Zigova, T., Saporta, S., Davis, C. and Sanberg, P.R. (2003) Putative immunomodulation of human umbilical cord blood cells in a mouse model of ALS. Exp. Neurol., 181: 90. Ende, N., Weistein, F., Chen, R. and Ende, M. (2000) Human umbilical cord blood effect on SOD mice (amyotrophic lateral sclerosis). Life Sci., 67: 53–59. Ferrante, R.J., Klein, A.M., Dedeoglu, A. and Beal, M.F. (2001) Therapeutic efficacy of EGb761 (Gingko biloba extract) in a transgenic mouse model of amyotrophic lateral sclerosis. J. Mol. Neurosci., 17: 89–96. Ferrari, G., Cusella-De Angelis, G., Coletta, M., Paolucci, E., Stornaiuolo, A., Cossu, G. and Mavilio, F. (1998) Muscle regeneration by bone marrow-derived myogenic progenitors. Science, 279: 1528–1530. Forraz, N., Pettengell, R. and McGuckin, C.P. (2004) Characterization of a lineage-negative stem-progenitor cell population optimized for ex-vivo expansion and enriched for LTCIC. Stem Cells, 22: 100–108. Frassoni, F., Podesta, M., Maccario, R., Giorgiani, G., Rossi, G., Zecca, M., Bacigalupo, A., Piaggio, G. and Locatelli, F. (2003) Cord blood transplantation provides better reconstitution of hematopoietic reservoir compared with bone marrow transplantation. Blood, 102: 1138–1141. Gage, F.H. (2000) Mammalian neural stem cells. Science, 287: 1433–1438. Galli, R., Borello, U., Gritti, A., Minasi, M.G., Bjornson, C., Coletta, M., Mora, M., De Angelis, M.G., Fiocco, R., Cossu, G. and Vescovi, A.L. (2000) Skeletal myogenic potential of human and mouse neural stem cells. Nat. Neurosci., 3: 986–991. Garbuzova-Davis, S., Gografe, S.J., Davis Sanberg, C., Willing, A.E., Saporta, S., Cameron, D.F., Desjarlais, T., Daily, J., Kuzmin-Nichols, N., Chamizo, W., Klasko, S.K. and Sanberg, P.R. (2006) Maternal transplantation of human umbilical cord blood cells provides prenatal therapy in Sanfilippo type B mouse model. FASEB, 20: 485–487. Garbuzova-Davis, S., Willing, A.E., Desjarlais, T., Davis Sanberg, C. and Sanberg, P.R. (2005) Transplantation of human umbilical cord blood cells benefits an animal model of Sanfilippo syndrome type B. Stem Cells Dev., 14: 384–394. Garbuzova-Davis, S., Willing, A.E., Zigova, T., Saporta, S., Justen, E.B., Lane, J.C., Hudson, J.E., Chen, N., Davis, C.D. and Sanberg, P.R. (2003) Intravenous administration of human umbilical cord blood cells in a mouse model of amyotrophic lateral sclerosis: distribution, migration, and differentiation. J. Hematother. Stem Cell Res., 12: 255–270. Garderet, L., Dulphy, N., Douay, C., Chalumeau, N., Schaeffer, V., Zilber, M.T., Lim, A., Even, J., Mooney, N., Gelin, C., Gluckman, E., Charron, D. and Toubert, A. (1998) The umbilical cord blood alphabeta T-cell repertoire: characteristics of a polyclonal and naive but completely formed repertoire. Blood, 91: 340–346.
219 Gemma, C., Mesches, M.H., Sepesi, B., Choo, K., Holmes, D.B. and Bickford, P.C. (2002) Diets enriched in foods with high antioxidant activity reverse age-induced decreases in cerebellar beta-adrenergic function and increases in proinflammatory cytokines. J. Neurosci., 22: 6114–6120. Gluckman, E. (2000) Current status of umbilical cord blood hematopoietic stem cell transplantation. Exp. Hematol., 28: 1197–1205. Gluckman, E., Broxmeyer, H.A., Auerbach, A.D., Friedman, H.S., Douglas, G.W., Devergie, A., Esperou, H., Thierry, D., Socie, G. and Lehn, P. (1989) Hematopoietic reconstitution in a patient with Fanconi’s anemia by means of umbilical-cord blood from an HLA-identical sibling. N. Engl. J. Med., 321: 1174–1178. Gluckman, E., Rocha, V., Boyer-Chammard, A., Locatelli, F., Arcese, W., Pasquini, R., Ortega, J., Souillet, G., Ferreira, E., Laporte, J.P., Fernandez, M. and Chastang, C. (1997) Outcome of cord-blood transplantation from related and unrelated donors. Eurocord Transplant Group and the European Blood and Marrow Transplantation Group. N. Engl. J. Med., 337: 373–381. Gluckman, E., Rocha, V. and Chastang, C. (1999) Peripheral stem cells in bone marrow transplantation. Cord blood stem cell transplantation. Baillieres Best Pract. Res. Clin. Haematol., 12: 279–292. Goodwin, H.S., Bicknese, A.R., Chien, S.N., Bogucki, B.D., Oliver, D.A., Quinn, C.O. and Wall, D.A. (2001) Multilineage differentiation activity by cells isolated from umbilical cord blood: expression of bone, fat, and neural markers. Biol. Blood Marrow Transpl., 7: 581–588. Gorelick, P.B., Sacco, R.L., Smith, D.B., Alberts, M., Mustone-Alexander, L., Rader, D., Ross, J.L., Raps, E., Ozer, M.N., Brass, L.M., Malone, M.E., Goldberg, S., Booss, J., Hanley, D.F., Toole, J.F., Greengold, N.L. and Rhew, D.C. (1999) Prevention of a first stroke: a review of guidelines and a multidisciplinary consensus statement from the National Stroke Association. JAMA, 281: 1112–1120. Gritti, A., Vescovi, A.L. and Galli, R. (2002) Adult neural stem cells: plasticity and developmental potential. J. Physiol. Paris., 96: 81–90. Ha, Y., Choi, J.U., Yoon, D.H., Yeon, D.S., Lee, J.J., Kim, H.O. and Cho, Y.E. (2001) Neural phenotype expression of cultured human cord blood cells in vitro. Neuroreport, 12: 3523–3527. Hao, S.G., Sun, G.L., Wu, W.L. and Wu, Y.L. (2003) Studies on the dynamics of biological characteristics of CD133+ cells from human umbilical cord blood during short-term culture. Zhongguo Shi Yan Xue Ye Xue Za Zhi, 11: 569–575. Henon, P.R. (2003) Human embryonic or adult stem cells: an overview on ethics and perspectives for tissue engineering. Adv. Exp. Med. Biol., 534: 27–45. Hiramatsu, H., Nishikomori, R., Heike, T., Ito, M., Kobayashi, K., Katamura, K. and Nakahata, T. (2003) Complete reconstitution of human lymphocytes from cord blood CD34+ cells using the NOD/SCID/{gamma}cnull mice model. Blood, 102: 873–880.
Hirose, Y., Kiyoi, H., Itoh, K., Kato, K., Saito, H. and Naoe, T. (2001) B-cell precursors differentiated from cord blood CD34+ cells are more immature than those derived from granulocyte colony-stimulating factor-mobilized peripheral blood CD34+ cells. Immunology, 104: 410–417. Honmou, O., Oka, S., Houkin, K., Hashi, K. and Kocsis, J.D. (2002) AC133 – positive cells derived from human cord blood remyelinate demyelinated axons in the adult rat spinal cord. Program No. 34.2, 2002 Abstracts Viewer/Itinerary Planner. Washington DC, Society for Neuroscience. Jackson, K.A., Mi, T. and Goodell, M.A. (1999) Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc. Natl. Acad. Sci. USA, 96: 14482–14486. Jin, K., Peel, A.L., Mao, X.O., Xie, L., Cottrell, B.A., Henshall, D.C. and Greenberg, D.A. (2004) Increased hippocampal neurogenesis in Alzheimer’s disease. Proc. Natl. Acad. Sci. USA, 101: 343–347. Joseph, J., Denisova, N., Fisher, D., Bickford, P., Prior, R. and Cao, G. (1998) Age-related neurodegeneration and oxidative stress: putative nutritional intervention. Neurol. Clin., 16: 747–755. Kawamata, T., Akiyama, H., Yamada, T. and McGeer, P.L. (1992) Immunologic reactions in amyotrophic lateral sclerosis brain and spinal cord tissue. Am. J. Pathol., 140: 691–707. Kim, J.J., Saporta, S., Willing, A.E., Fu, E.S., Colina, L.O. and Sanberg, P.R. (2002) Behavioral recovery following intravenous administration of human umbilical cord blood cells in rats with spinal cord injury. Exp. Neurol., 175: 443. Kobari, L., Giarratana, M.C., Pflumio, F., Izac, B., Coulombel, L. and Douay, L. (2001) CD133+ cell selection is an alternative to CD34+ cell selection for ex vivo expansion of hematopoietic stem cells. J. Hematother. Stem Cell Res., 10: 273–281. Koehl, U., Zimmermann, S., Esser, R., Sorensen, J., Gruttner, H.P., Duchscherer, M., Seifried, E., Klingebiel, T. and Schwabe, D. (2002) Autologous transplantation of CD133 selected hematopoietic progenitor cells in a pediatric patient with relapsed leukemia. Bone Marrow Transpl., 29: 927–930. Kulbatski, I., Mothe, A.J., Nomura, H. and Tator, C.H. (2005) Endogenous and exogenous CNS derived stem/progenitor cell approaches for neurotrauma. Curr. Drug Targets, 6: 111–126. Kuzmenok, O.I., Sanberg, P.R., Desjarlais, T., Bennett, S.P. and Garbuzova-Davis, S.N. (2006) Lymphopenia and spontaneous autorosette formation in SOD1 mouse model of ALS. J. Neuroimmunol., 172: 132–136. Li, Y., Chen, J., Chen, X.G., Wang, L., Gautam, S.C., Xu, Y., Katakowski, M., Zhang, L.J., Lu, M., Janakiraman, N. and Chopp, M. (2002) Human marrow stromal cell therapy for stroke in rat: neurotrophins and functional recovery. Neurology, 59: 514–523. Lie, D.C., Song, H., Colamarino, S.A., Ming, G.L. and Gage, F.H. (2004) Neurogenesis in the adult brain: new strategies for central nervous system diseases. Annu. Rev. Pharmacol. Toxicol., 44: 399–421.
220 Liesveld, J.L. (2003) When what you have is not enough: optimizing cord blood transplantation in adults. Leuk. Res., 27: 197–199. Limke, T.L. and Rao, M.S. (2002) Neural stem cells in aging and disease. Cell Mol. Med., 6: 475–496. Linazasoro, G. (2003) Stem cells: solution to the problem of transplants in Parkinson’s disease? Neurologia, 18: 74–100. Lu, D., Sanberg, P.R., Mahmood, A., Li, Y., Wang, L., Sanchez-Ramos, J. and Chopp, M. (2002) Intravenous administration of human umbilical cord blood reduces neurological deficit in the rat after traumatic brain injury. Cell Transpl., 11: 275–285. Lu, L., Shen, R.N. and Broxmeyer, H.E. (1996) Stem cells from bone marrow, umbilical cord blood and peripheral blood for clinical application: current status and future application. Crit. Rev. Oncol. Hematol., 22: 61–78. Ma, Y., Zou, P., Xiao, J. and Huang, S. (2002) The expression and functional characteristics of AC133 antigen in cord blood hematopoietic cells. Zhonghua Nei Ke Za Zhi, 41: 798–800. Madrigal, J.A., Cohen, S.B., Gluckman, E. and Charron, D.J. (1997) Does cord blood transplantation result in lower graftversus-host disease? It takes more than two to tango. Hum. Immunol., 56: 1–5. Maihofner, C., Probst-Cousin, S., Bergmann, M., Neuhuber, W., Neundorfer, B. and Heuss, D. (2003) Expression and localization of cyclooxygenase-1 and -2 in human sporadic amyotrophic lateral sclerosis. Eur. J. Neurosci., 18: 1527–1534. Mayani, H. and Lansdorp, P.M. (1998) Biology of human umbilical cord blood-derived hematopoietic stem/progenitor cells. Stem Cells, 16: 153–165. McGuckin, C.P., Forraz, N., Allouard, Q. and Pettengell, R. (2004) Umbilical cord blood stem cells can expand hematopoietic and neuroglial progenitors in vitro. Exp. Cell Res., 295: 350–359. Mikami, T., Eguchi, M., Kurosawa, H., Sato, Y., Sugita, K., Suzumura, H., Tadokoro, N., Watanabe, H. and Inaba, N. (2002) Ultrastructural and cytochemical characterization of human cord blood cells. Med. Electron Microsc., 35: 96–101. Mohamed, H.A., Mosier, D.R., Zou, L.L., Siklos, L., Alexianu, M.E., Engelhardt, J.I., Beers, D.R., Le, W.D. and Appel, S.H. (2002) Immunoglobulin Fc gamma receptor promotes immunoglobulin uptake, immunoglobulin-mediated calcium increase, and neurotransmitter release in motor neurons. J. Neurosci. Res., 69: 110–116. Nakahata, T. and Ogawa, M. (1982) Hemopoietic colonyforming cells in umbilical cord blood with extensive capability to generate mono- and multipotential hemopoietic progenitors. J. Clin. Invest., 70: 1324–1328. Nayar, B., Raju, G.M. and Deka, D. (2002) Hematopoietic stem/progenitor cell harvesting from umbilical cord blood. Int. J. Gynaecol. Obstet., 79: 31–32. Newman, M.B., Willing, A.E., Manresa, J.J., Davis-Sanberg, C. and Sanberg, P.R. (2005) Stroke-induced migration of human umbilical cord blood cells: time course and cytokines. Stem Cells Dev., 14: 576–586.
Niebroj-Dobosz, I., Jamrozik, Z., Janik, P., Hausmanowa-Petrusewicz, I. and Kwiecinski, H. (1999) Anti-neural antibodies in serum and cerebrospinal fluid of amyotrophic lateral sclerosis (ALS) patients. Acta Neurol. Scand., 100: 238–243. Ooi, J., Iseki, T., Takahashi, S., Tomonari, A., Nagayama, H., Ishii, K., Ito, K., Sato, H., Takahashi, T., Shindo, M., Sekine, R., Ohno, N., Uchimaru, K., Nagamura, F., Shirafuji, N., Tojo, A., Tani, K. and Asano, S. (2002) A clinical comparison of unrelated cord blood transplantation and unrelated bone marrow transplantation for adult patients with acute leukaemia in complete remission. Br. J. Haematol., 118: 140–143. Parent, J.M. (2003) Injury-induced neurogenesis in the adult mammalian brain. Neuroscientist, 9: 261–272. Petersen, B.E., Bowen, W.C., Patrene, K.D., Mars, W.M., Sullivan, A.K., Murase, N., Boggs, S.S., Greenberger, J.S. and Goff, J.P. (1999) Bone marrow as a potential source of hepatic oval cells. Science, 284: 1168–1170. Provinciali, L., Laurenzi, M.A., Vesprini, L., Giovagnoli, A.R., Bartocci, C., Montroni, M., Bagnarelli, P., Clementi, M. and Varaldo, P.E. (1988) Immunity assessment in the early state of amyotropfic lateral sclerosis: a study of virus antibodies and lymphocyte subsets. Acta Neurol. Scand., 78: 449–454. Rabinovich, S.S., Seledtsov, V.I., Poveschenko, O.V., Senuykov, V.V., Taraban, V.Y., Yarochno, V.I., Kolosov, N.G., Savchenko, S.A. and Kozlov, V.A. (2003) Transplantation treatment of spinal cord injury patients. Biomed. Pharmacother., 57: 428–433. Rainsford, E. and Reen, D.J. (2002) Interleukin 10, produced in abundance by human newborn T cell, may be the regulator of increased tolerance associated with cord blood stem cell transplantation. Br. J. Haematol., 116: 702–709. Reddy, C., Bhat, V.B., Kiranmai, G., Reddy, M.N., Reddanna, P. and Madyastha, K.M. (2000) Selective inhibition of cyclooxygenase-2 by C-phycocyanin, a biliprotein from Spirulina platensis. Biochem. Biophys. Res. Commun., 277: 599–603. Reddy, M., Subhashini, J., Mahipal, S.V., Bhat, V.B., Srinivas, R.P., Kiranmai, G., Madyastha, K.M. and Reddanna, P. (2003) C-Phycocyanin, a selective cyclooxygenase-2 inhibitor, induces apoptosis in lipopolysaccharide-stimulated RAW 264.7 macrophages. Biochem. Biophys. Res. Commun., 304: 385–392. Riaz, S.S., Jauniaux, E., Stern, G.M. and Bradford, H.F. (2002) The controlled conversion of human neural progenitor cells derived from foetal ventral mesencephalon into dopaminergic neurons in vitro. Brain Res. Dev. Brain Res., 136: 27–34. Roncarolo, M.G., Bigler, M., Ciuti, E., Martino, S. and Tovo, P.A. (1994) Immune responses by cord blood cells. Blood Cells, 20: 573–585. Rubinstein, P., Carrier, C., Scaradavou, A., Kurtzberg, J., Adamson, J., Migliaccio, A.R., Berkowitz, R.L., Cabbad, M., Dobrila, N.L., Taylor, P.E., Rosenfield, R.E. and Stevens, C.E. (1998) Outcomes among 562 recipients of placental-blood transplants from unrelated donors. N. Engl. J. Med., 339: 1565–1577.
221 Sanberg, P.R. (2005) Stem cell research’s reversal of fortune. Scientist, 19: 12. Sanberg, P.R., Willing, A.E. and Cahill, D.W. (2001) Neurosurgery for the 21st century. What does neurobiology justify for repairing neurodegenerative disorders? Clin. Neurosurg., 48: 113–126. Sanberg, P.R., Willing, A.E. and Cahill, D.W. (2002) Novel cellular approaches to repair of neurodegenerative disease: from Sertoli cells to umbilical cord blood stem cells. Neurotox. Res., 4: 95–101. Sanberg, P.R., Willing, A.E., Davis Sanberg, C., Kazi, Z. and Borlongan, C.V. (2004) Different functional recovery and neurostructural repair of the stroke brain produced by volume-reduced and mononuclear fractioned human umbilical cord blood. Program No. 617.4 Abstract Viewer/Itinerary Planner. Society for Neuroscience, Washington, DC. Sanchez-Ramos, J., Song, S., Cardozo-Pelaez, F., Hazzi, C., Stedeford, T., Willing, A., Freeman, T.B., Saporta, S., Janssen, W., Patel, N., Cooper, D.R. and Sanberg, P.R. (2000) Adult bone marrow stromal cells differentiate into neural cells in vitro. Exp. Neurol., 164: 247–256. Sanchez-Ramos, J.R., Song, S., Kamath, S.G., Zigova, T., Willing, A., Cardozo-Pelaez, F., Stedeford, T., Chopp, M. and Sanberg, P.R. (2001) Expression of neural markers in human umbilical cord blood. Exp. Neurol., 171: 109–115. Saporta, S., Kim, J.J., Willing, A.E., Fu, E.S., Davis, C.D. and Sanberg, P.R. (2003) Human umbilical cord blood stem cells infusion in spinal cord injury: engraftment and beneficial influence on behavior. J. Hematother. Stem Cell Res., 12: 271–278. Sathananthan, A.H. and Trounson, A. (2005) Human embryonic stem cells and their spontaneous differentiation. Ital. J. Anat. Embryol., 110: 151–157. Savitz, S.I., Dinsmore, J.N., Wechsler, L.R., Rosenbaum, D.M. and Caplan, L.R. (2004) Cell therapy for stroke. NeuroRx., 1: 406–414. Shih, C.C., DiGiusto, D. and Forman, S.J. (2000) Ex vivo expansion of transplantable human hematopoietic stem cells: where do we stand in the year 2000? J. Hematother. Stem. Cell Res., 9: 621–628. Shin, J.W., Lee, D.W., Kim, M.J., Song, K.S., Kim, H.S. and Kim, H.O. (2005) Isolation of endothelial progenitor cells from cord blood and induction of differentiation by ex vivo expansion. Yonsei. Med. J., 46: 260–267. Silani, V. and Leigh, N. (2003) Stem therapy for ALS: hope and reality. Amyotrop. Lateral Scler. Other Motor Neuron. Disord., 4: 8–10. Sirchia, G. and Rebulla, P. (1999) Placental/umbilical cord blood transplantation. Haematologica, 84: 738–747. Slack, J.M.W. (2000) Stem cells in epithelial tissues. Science, 287: 1431–1433. Staba, S.L., Escolar, M.L., Poe, M., Kim, Y., Martin, P.L., Szabolcs, P., Allison-Thacker, J., Wood, S., Wenger, D.A., Rubinstein, P., Hopwood, J.J., Krivit, W. and Kurtzberg, J. (2004) Cord-blood transplants from unrelated donors in patients with Hurler’s syndrome. N. Engl. J. Med., 350: 1960–1969.
Stromberg, I., Gemma, C., Vila, J. and Bickford, P.C. (2005) Blueberry- and spirulina-enriched diets enhance striatal dopamine recovery and induce a rapid, transient microglia activation after injury of the rat nigrostriatal dopamine system. Exp. Neurol., 196: 298–307. Thomson, B.G., Robertson, K.A., Gowan, D., Heilman, D., Broxmeyer, H.E., Emanuel, D., Kotylo, P., Brahmi, Z. and Smith, F.O. (2000) Analysis of engraftment, graft-versus-host disease, and immune recovery following unrelated donor cord blood transplantation. Blood, 96: 2703–2711. Todaro, A.M., Pafumi, C., Pernicone, G., Munda, S., Pilastro, M.R., Russo, A., Farina, M. and Messina, A. (2000) Haematopoietic progenitors from umbilical cord blood. Blood Purif., 18: 144–147. Vaziri, H., Dragowska, W., Allsopp, R.C., Thomas, T.E., Harley, C.B. and Lansdorp, P.M. (1994) Evidence for a mitotic clock in human hematopoietic stem cells: loss of telomeric DNA with age. Proc. Natl. Acad. Sci. USA, 91: 9857–9860. Vendrame, M., Cassady, J., Newcomb, J., Butler, T., Pennypacker, K.R., Zigova, T., Sanberg, C.D., Sanberg, P.R. and Willing, A.E. (2004) Infusion of human umbilical cord blood cells in a rat model of stroke dose-dependently rescues behavioral deficits and reduces infarct volume. Stroke, 35: 2390–2395. Vendrame, M., Gemma, C., de Mesquita, D., Collier, L., Bickford, P.C., Sanberg, C.D., Sanberg, P.R., Pennypacker, K.R. and Willing, A.E. (2005) Anti-inflammatory effects of human cord blood cells in a rat model of stroke. Stem Cells Dev., 14: 595–604. Vescovi, A.L., Parati, E.A., Gritti, A., Poulin, P., Ferrario, M., Wanke, E., Frolichsthal-Schoeller, P., Cova, L., ArcellanaPanlilio, M., Colombo, A. and Galli, R. (1999) Isolation and cloning of multipotential stem cells from the embryonic human CNS and establishment of transplantable human neural stem cell lines by epigenetic stimulation. Exp. Neurol., 156: 71–83. Wagner, J.E., Barker, J.N., DeFor, T.E., Baker, K.S., Blazar, B.R., Eide, C., Goldman, A., Kersey, J., Krivit, W., MacMillan, M.L., Orchard, P.J., Peters, C., Weisdorf, D.J., Ramsay, N.K. and Davies, S.M. (2002) Transplantation of unrelated donor umbilical cord blood in 102 patients with malignant and nonmalignant diseases: influence of CD34 cell dose and HLA disparity on treatment-related mortality and survival. Blood, 100: 1611–1618. Wang, Y., Chang, C.F., Chou, J., Chen, H.L., Deng, X., Harvey, B.K., Cadet, J.L. and Bickford, P.C. (2005) Dietary supplementation with blueberries, spinach, or spirulina reduces ischemic brain damage. Exp. Neurol., 193: 75–84. Willing, A.E., Lixian, J., Milliken, M., Poulos, S., Zigova, T., Song, S., Hart, C., Sanchez-Ramos, J. and Sanberg, P.R. (2003) Intravenous versus intrastriatal cord blood administration in a rodent model of stroke. J. Neurosci. Res., 73: 296–307. Woodbury, D., Schwarz, E.J., Prockop, D.J. and Black, I.B. (2000) Adult rat and human bone marrow stromal cells differentiate into neurons. J. Neurosci. Res., 61: 364–370.
222 Yin, A.H., Miraglia, S., Zanjani, E.D., Almeida-Porada, G., Ogawa, M., Leary, A.G., Olweus, J., Kearney, J. and Buck, D.W. (1997) AC133, a novel marker for human hematopoietic stem and progenitor cells. Blood, 90: 5002–5012. Yokota, C., Kaji, T., Kuge, Y., Inoue, H., Tamaki, N. and Minematsu, K. (2004) Temporal and topographic profiles of cyclooxygenase-2 expression during 24 h of focal brain ishemia in rats. Neurosci. Lett., 357: 219–222. Zheng, W. and Wang, S.Y. (2003) Oxygen radical absorbing capacity of phenolics in blueberries, cranberries, choke-
berries, and lingonberries. J. Agric. Food Chem., 51: 502–509. Zhou, Q., Zhang, S.Z., Xu, R.X. and Xu, K. (2004) Neural stem cell transplantation and postoperative management: report of 70 cases. Di Yi Jun Yi Da Xue Xue Bao, 24: 1207–1209. Zola, H., Fusco, M., Macardle, P.J., Flego, L. and Roberton, D. (1995) Expression of cytokine receptors by human cord blood lymphocytes: comparison with adult blood lymphocytes. Pediatr. Res., 38: 397–403.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 15
Recovery of motor function after stroke Jeffrey A. Brown Department of Neurological Surgery, Wayne State University School of Medicine, Detroit, MI, USA
Abstract: Improvement of motor activity may occur after stroke. It may be because of recovery of marginally functional neurons. It may also occur by relearning, a process that strengthens existing pathways and may lead to new functional or structural changes- neuroplasticity. Clinical investigation into the treatment of chronic pain after thalamic infarction has shown improvement in motor function when pain relief is achieved with motor cortex stimulation. More recently, laboratory studies in rats and primates demonstrate significant improvement in forelimb reaching tasks in rats and primates after induced ischemic cortical infarction when rehabilitation is paired with stimulation of the injured cortex and cortical margin at low frequency (50 Hz). Structural changes have also been observed. Dendritic density in layer V of the cortex near the lesion increases after cortical stimulation, consistent with a restorative cortical plasticity. Also, stimulation combined with rehabilitation increases the area of the injured cortex from which movements can be evoked in response to stimulation of the injured cortex in rats. Unilateral cortical stimulation reduces secondary cortical hyperexcitability in the impaired hemisphere after stroke. These findings form the basis for the first clinical study motor cortex stimulation after chronic stroke in humans. A prospective, randomized multicenter study of subthreshold motor cortical electrical stimulation during rehabilitation in patients has been completed. The eight patients entered into this study had weakness from a stroke that occurred at least four months before enrollment. Results demonstrate that the treatment is safe. In addition, there was significant improvement in upper extremity function. These improvements persisted through the 12-week follow-up assessment period after completion of stimulation and rehabilitation. Recently, non-invasive transcranial magnetic stimulation of the motor cortex demonstrates improvements in hand function that persist after stimulation for at least 25 minutes. Such work represents a paradigm shift in the approach towards rehabilitation of the stroke-injured brain away from pharmacologic flooding of neuronal receptors, instead towards targeted physiologic stimulation. Keywords: cerebral cortex; electrical stimulation; stroke; neuropathic pain; neuroplasticity stroke survivors is motor weakness (Gresham et al., 1995). For these individuals, all that has been available to improve these residual motor deficits has been physiotherapy (Gresham et al., 1995). Spontaneous motor recovery does occur after stroke, possibly because of the recovery of marginally affected neurons. Alternatively, there may be reorganization of neural function wherein adjacent brain regions re-establish the function of stroke-damaged areas. This concept is known as neuroplasticity.
Introduction Stroke is the most common cause of disability and the third leading cause of death in the United States. Approximately 700,000 strokes occur annually in the United States. About 2136% of stroke survivors are permanently disabled. The most common disability among the over 5 million Corresponding author. Tel.: +1 516 478 0008; Fax: +1 516 478 0013; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57015-3
223
224
The subject of neuroplasticity after stroke has been extensively reviewed (Teasell et al., 2005). This chapter will be limited to a review of the current status of the preclinical and clinical effects of motor cortex stimulation to enhance the recovery of motor function following the acute phase of an ischemic stroke. After acute injury, marginally functional neurons may recover, leading to early complete recovery of some motor functions, or function may remain marginal. Later recovery processes involve relearning. These processes will lead first to strengthening of existing neural pathways, and, second to new functional or structural changes and thus expression of neuroplasticity (Pascual-Leone et al., 2005). The term plastic derives from the Greek word plastos that means molded. Plasticity of the nervous systems refers to behavioral changes, originally conceived by William James, and to structural changes as conceived by Cajal (Pascual-Leone et al., 2005). The results of electrical stimulation of injured cortical and subcortical motor structures may involve both neuroplastic alterations in function and activation of previously less effective neuronal pathways. Studies of motor cortical stimulation provide an opportunity for better understanding of the mechanism of recovery of motor function after stroke as well as offering a new paradigm for the treatment of the motor effects of stroke.
Background The groundwork for laboratory investigation into the effect of stimulating the motor system on the recovery of motor function after ischemic injury began with work in the field of pain treatment. early laboratory investigation of the effects of frontal lobe stimulation showed that electrical stimulation of the prefrontal cortex could abolish the normal response of midbrain neurons to noxious stimulation. (Hardy and Haigler, 1985). Bipolar stimulation of the medial prefrontal cortex causes a significant increase in the latency of nociceptive responses as measured using hot plate and tail flick techniques. Voluntary motor activity
is not affected nor does such stimulation cause seizures. thus, stimulation of the prefrontal cortex can cause analgesia (Hardy, 1985). Electrical stimulation of the motor cortex in cats, which were deafferentated by sectioning of the spinothalamic tract, inhibits observed thalamic burst hyperactivity observed secondary to the deafferentation (Tsubokawa et al., 1991). These laboratory studies demonstrate the influence of the motor system on sensory activity. many clinical studies using motor cortex stimulation to treat otherwise intractable thalamic pain and other central pain syndromes—trigeminal and other peripheral neuropathic pain syndromes—have been published (Brown and Barbaro, 2003). Some of the earliest publications on the results of motor cortex stimulation for pain control also mentioned that stimulation of the motor cortex resulted in improvement in associated dystonia and spasticity (Tsubokawa et al., 1991, 1993; Franzini et al., 2000) In one clinical series (Katayama et al., 1998), satisfactory control of pain was achieved by electrical stimulation of the motor cortex in 70% of the patients with thalamic pain after stroke when the stimulation elicited muscle contractions in the painful area, whereas it was only possible to alleviate pain in 9% of the patients when the electrical stimulation did not elicit muscle contractions. This finding suggests that pain control requires an intact corticospinal system originating from the motor cortex (Katayama et al., 1998). In a more recent study of motor cortex stimulation for neuropathic pain in 31 patients, preoperative motor status was not confirmed to be a significant factor. (Nuti et al., 2005). In the course of treating patients for central pain after thalamic infarction with motor cortex stimulation, physicians have observed resolution of associated dystonia and intentional myoclonus. (Franzini et al., 2000). Positron emission tomography of the brain during motor cortex stimulation shows increased regional cerebral blood flow in thalamic nuclei connected with motor and premotor cortices, including the medial thalamus and upper brainstem (GarciaLarrea et al., 1999). Four patients with thalamic pain experienced pain control and reduced intentional myoclonus (Franzini et al., 2003).
225
Animal studies on the effect of motor cortex stimulation Laboratory investigations into the effect of stimulating the motor system on recovery after ischemic cortical injury have confirmed the observations of enhanced motor recovery brought about by cortical stimulation. (Adkins-Muir and Jones, 2003; Kleim et al., 2003;Teskey et al., 2003). Studies of cortical ischemic infarction in a primate model confirmed other findings that electrical stimulation of the motor cortex enhanced performance of training after experimentally induced strokes (Plautz et al., 2003). In these studies, the proximal forelimb motor cortex (M1) was mapped using intracortical microstimulation techniques. An infarct was created using bipolar electrocoagulation over the neurophysiologically identified M1 hand representation regions of the squirrel monkeys. After 3.5–5 months, spontaneous motor recovery had stabilized but significant motor impairments remained. When cortical stimulation is combined with rehabilitative training for 2–4 weeks, pellet retrieval from progressively smaller wells shows statistically significant gains, although the performance did not reach preinfarct levels. Mapped hand representational area significantly increases and a new cortical area representing the hand becomes apparent adjacent to the infarct as well as at a considerable distance from the infarct. This study is important because it demonstrates that poststroke motor gains can be achieved beyond the subacute poststroke period. Studies in rats have shown that subdural electrical motor cortex stimulation near a cortical lesion after ischemic cortical injury significantly improves results of a skilled forelimb retrieval task (Adkins-Muir and Jones, 2003). In this study by Adkins-Muir and Jones, rats were trained to reach for food pellets placed on a staircase that was designed to present serially more difficult reaching tasks (Adkins-Muir and Jones, 2003). Endothelin1, a vasoconstrictor, was then applied to the cortex to cause ischemic cortical injury. Two weeks later, a recording electrode was inserted close to the area of the lesion and rehabilitation training ensued for 10 days. Intermittent cortical stimulation at 50 Hz during poststroke
rehabilitation significantly improved the results of forelimb retrieval. Stimulation at a higher frequency (250 Hz) did not improve function. The level of performance achieved persisted 2 days after termination of the stimulation. Significant structural changes were also observed after electrical stimulation. Dendritic density in layer V of the cortex at or near the lesion increased as demonstrated by the presence of more microtubuleassociated protein 2 immunoreactivity. There was no increase in dendritic density in the normal cortex. Thus, after a focal ischemic cortical injury, electrical stimulation of the motor cortex combined with rehabilitation creates a ‘‘restorative cortical plasticity’’(Adkins-Muir and Jones, 2003). Most importantly, the study demonstrated structural changes developed in addition to recovery of motor function. Another animal study investigated the hypothesis that motor cortex stimulation combined with rehabilitation expands contralateral cortical representation after ischemic cortical infarction. In this study, forelimb cortical representation was mapped in rats using microstimulation techniques. Rats were trained for 2 weeks in a task to retrieve food pellets through a slot in a Plexiglas cage. A cortical ischemic lesion was then created using bipolar coagulation in the distal forelimb region, and cortical stimulating electrodes were implanted over the ischemic cortex and on the remaining intact forelimb cortical region. This study showed that a combination of stimulation and rehabilitation increased the area of the injured cortex from which movements could be evoked in response to microstimulation of the injured cortex compared to that of rats who did not undergo cortical stimulation during rehabilitation (Kleim et al., 2003). The authors concluded that increased motor representation from cortical stimulation after stroke is a consequence of enhanced synaptic function and restoration of cortical circuitry. In yet another study, an ischemic cortical injury was induced by surgical removal of pia mater from the caudal forelimb area of the sensorimotor cortex on the side opposite the paw preferred by rats for pasta retrieval tasks. Recording electrodes were implanted to measure frontal evoked potentials. After recovery from the induced stroke, rats were
226
stimulated at 50% of motor movement threshold at 100 Hz while the rats performed pasta retrieval tasks. Those rats receiving stimulation increasingly began using their preferred limb so that they were able to return to their preinfarct reaching retrieval levels when stimulation at frequencies of 50 or 100 Hz were used. The polysynaptic component of the recorded evoked potentials was enhanced, and the amount of current required to elicit movement could be reduced over time. This study showed that cortical stimulation reduced the secondary cortical hyperexcitability that developed in the impaired hemisphere after stroke (Teskey et al., 2003). The observed large polysynaptic potentials in the evoked potentials recorded from rats undergoing cortical stimulation indicate that synaptic efficacy has strengthened. Long-term cortical synaptic potentiation occurred in cortical layer V. The author’s hypothesis is that this synaptic potentiation represents an example of anatomic cortical reorganization. The findings of these animal experiments in stroke models support the hypothesis that increased motor cortical representation is derived from enhanced synaptic function and restoration of cortical circuitry brought about by expression of neural plasticity initiated by cortical stimulation together with the effect of rehabilitation. These studies in animals confirm the results of clinical studies showing that enhanced motor function occurs during cortical stimulation for central pain, and provide the basis for clinical investigations regarding the safety of using cortical stimulation to enhance outcome after nonhemorrhagic stroke in humans. (Brown and Pilitsis, 2005; Brown et al., 2006).
Clinical investigations in motor cortex stimulation We have earlier tested the hypothesis that subthreshold cortical stimulation of primary motor cortex could facilitate rehabilitation of motor function and produce residual motion in an impaired limb in a patient who had suffered a nonhemorrhagic cortical or subcortical infarction (Brown et al., 2003). In this case, the report discussed a 65-year-old man with a subcortical
ischemic infarct and right spastic hemiparesis occurring 19 months before treatment, who underwent subthreshold epidural motor cortex stimulation delivered concurrently with 3 weeks of structured rehabilitation. Before stimulation, the patient’s affected arm rested in a flexion posture without the ability to flex or extend the fingers; after stimulation and rehabilitation, he was able to grasp a pen and write letters. The FuglMeyer motor-scale score of the patients (FuglMeyer et al., 1975), a measurement of hand/arm function based on a 100-point scale, improved from 36 to 46. This improvement was sustained at the last assessment performed 4 weeks after conclusion of the rehabilitation therapy (Brown et al., 2003). A prospective, randomized, multicenter study of safety of subthreshold motor cortical electrical stimulation confirmed that the procedure is safe, and that it can facilitate motor recovery after strokes (Brown et al., 2006). The study was performed in patients with motor deficit resulting from a stroke that occurred at least 4 months prior to enrollment, and has been completed (Brown et al., 2006). Patients were randomized into a treatment and a control group consisting of X and Y patients: (1) The patients in the treatment group underwent electrode implantation and subsequent epidural electrical stimulation (at 50 Hz, 50% of the current needed to evoke gross motor movement) concurrent with 3 weeks of rehabilitation. (2) The stroke patients in the control group received the same 3 weeks of rehabilitation but did not undergo implantation. The site for hand function was identified on a functional magnetic resonance imaging (fMRI) scan before the implantation, and this activation site was integrated into a neuronavigational system used during implantation. After completing a circular 4 cm surgical craniotomy over the targeted site, transdural electrical stimulation was undertaken to identify electromyographic activation of the hand/finger electrodes. An electrode grid was sutured to the dura with the center of the grid over the point deemed to be the center of the fMRI ‘‘hot spot.’’ The electrode lead was tunneled to a supraclavicular exit site and the bone flap replaced. Rehabilitation
227
began 1 week later. Control patients did not undergo device implantation, but they completed the same rehabilitation protocol as the treatment patients. Cortical stimulation was initiated at strength of 50% of the movement threshold or 6.5 mA if no movement could be evoked. The threshold for evoking gross movement in the contralateral hand was determined before the first session of each week of treatment. The mean time since stroke was 18+/18 (9–33) months in the surgical group and 38+/(15–68) months in the control group. Of the eight patients completing surgical implantation and/or rehabilitation, the stimulation plus rehabilitation group improved significantly better than controls in the upper extremity FuglMeyer score (P ¼ 0.003, overall) and the hand function score of the stroke impact scale (P ¼ 0.001, overall). Two patients dropped out of the study due to infection. The improvements persisted through the 12-week follow-up assessment (study week 16). The observed improvement through cortical stimulation may be caused by reorganization of the cortex through expression of neural plasticity. Such reorganization may include enhancement of inhibition in regions that have developed hyperactivity after the stroke. Thalamic hyperactivity has been observed in animal experiments after motor cortex stimulation in a cat deafferentiation model wherein the spinothalamic tract had been sectioned (Tsubokawa et al., 1991). The hypothesis that the thalamus is involved in neuropathic pain is supported by the presence of positron emission tomography demonstrated increased blood flow during motor cortex stimulation of patients with central pain syndromes (Garcia-Larrea et al., 1999). The most significant increase in regional cerebral blood flow was seen in the ventral lateral thalamus. This may occur because of corticothalamic projections from motor areas (GarciaLarrea et al., 1999). Functional enhancement persists after withdrawal of cortical stimulation, as seen in both animal studies and studies in humans indicating that the motor improvement is a result of expression of neural plasticity and not (only) caused by direct enhancement of surrounding marginally
functional cortical neurons. The combination of rehabilitation and stimulation may cause expression of plasticity of marginally effective circuits, leading to improved voluntary function. Noninvasive approaches to cortical stimulation for stroke recovery enhancement have also been investigated. (Hummel and Cohen, 2005; Hummel et al., 2005; Pascual-Leone et al., 2005). Transcranial direct current stimulation applied in a doubleblind protocol to motor regions of the affected hemisphere causes improvements in pinch force, hand function as measured by such activities of daily living as turning over cards or stacking checkers, and simple reaction times in the paretic hand. These improvements outlast the stimulation period for at least 25 minutes. Changes are accompanied by increased corticomotor excitability and reduced intracortical inhibition to transcranial magnetic stimulation. Such findings raise the possibility of using repeated noninvasive transcranial cortical stimulation to improve motor function after stroke. This work represents a paradigm shift in our approach towards rehabilitation of the stroke-injured brain. Rather than pharmacologic flooding of neuronal receptors, it is now possible to use targeted electrical stimulation to reprogram and enhance brain function. Our future ability to accomplish this task depends on additional research into our ability to reprogram brain function.
Acknowledgments The author acknowledges the assistance of Brookes C. Brown in reviewing this manuscript.
References Adkins-Muir, D.L. and Jones, T.A. (2003) Cortical electrical stimulation combined with rehabilitative training: enhanced functional recovery and dendritic plasticity following focal cortical ischemia in rats. Neurol. Res., 25: 780–788. Brown, J.A. and Barbaro, N.M. (2003) Motor cortex stimulation for central and neuropathic pain: current status. Pain, 104: 431–435. Brown, J.A., Lutsep, H., Cramer, S.C. and Weinand, M. (2003) Motor cortex stimulation for enhancement of recovery after stroke: case report. Neurol. Res., 25: 815–818.
228 Brown, J.A., Lutsep, H.L., Weinand, M. and Cramer, S.C. (2006) Motor cortex stimulation for the enhancement of recovery from stroke: a prospective multicenter safety study. Neurosurgery, 58(3): 464–473. Brown, J.A. and Pilitsis, J.G. (2005) Motor cortex stimulation for central and neuropathic facial pain: a prospective study of 10 patients and observations of enhanced sensory and motor function during stimulation. Neurosurgery, 56: 290–297. Franzini, A., Ferroli, P., Dones, I., Marras, C. and Broggi, G. (2003) Chronic motor cortex stimulation for movement disorders: a promising perspective. Neurol. Res., 25: 123–126. Franzini, A., Ferroli, P., Servello, D. and Broggi, G. (2000) Reversal of thalamic hand syndrome by long-term motor cortex stimulation. J. Neurosurg., 93: 873–875. Fugl-Meyer, A.R., Jaasko, L., Leyman, I., Olsson, S. and Steglind, S. (1975) The post-stroke hemiplegic patient. 1. A method for evaluation of physical performance. Scand. J. Rehabil. Med., 7: 13–31. Garcia-Larrea, L., Peyron, R., Mertens, P., Gregoire, M.C., Lavenne, F., Le, B.D., Convers, P., Mauguiere, F., Sindou, M. and Laurent, B. (1999) Electrical stimulation of motor cortex for pain control: a combined PET-scan and electrophysiological study. Pain, 83: 259–273. Gresham, G., Duncan, P., Stason, W., Adams, H., Adelman, A., Alexander, D., Bishop, D., Diller, L., Donaldson, N., Granger, C., Holland, A., Kelly-Hayes, M., McDowell, F., Myers, L., Phipps, M., Roth, E., Siebens, H., Tarvin, G., Trombley, C., 1995. Post-Stroke Rehabilitation. Rockville, MD, U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. Hardy, S.G. (1985) Analgesia elicited by prefrontal stimulation. Brain Res., 339: 281–284. Hardy, S.G. and Haigler, H.J. (1985) Prefrontal influences upon the midbrain: a possible route for pain modulation. Brain Res., 339: 285–293. Hummel, F., Celnik, P., Giraux, P., Floel, A., Wu, W.H., Gerloff, C. and Cohen, L.G. (2005) Effects of non-invasive cortical stimulation on skilled motor function in chronic stroke. Brain, 128: 490–499.
Hummel, F. and Cohen, L.G. (2005) Improvement of motor function with noninvasive cortical stimulation in a patient with chronic stroke. Neurorehabil. Neural Repair, 19: 14–19. Katayama, Y., Fukaya, C. and Yamamoto, T. (1998) Poststroke pain control by chronic motor cortex stimulation: neurological characteristics predicting a favorable response. J. Neurosurg., 89: 585–591. Kleim, J.A., Bruneau, R., VandenBerg, P., MacDonald, E., Mulrooney, R. and Pocock, D. (2003) Motor cortex stimulation enhances motor recovery and reduces peri-infarct dysfunction following ischemic insult. Neurol. Res., 25: 789–793. Nuti, C., Peyron, R., Garcia-Larrea, L., Brunon, J., Laurent, B., Sindou, M. and Mertens, P. (2005) Motor cortex stimulation for refractory neuropathic pain: four year outcome and predictors of efficacy. Pain, 118: 43–52. Pascual-Leone, A., Amedi, A., Fregni, F. and Merabet, L.B. (2005) The plastic human brain cortex. Ann. Rev. Neurosci., 28: 377–401. Plautz, E.J., Barbay, S., Frost, S.B., Friel, K.M., Dancause, N., Zoubina, E.V., Stowe, A.M., Quaney, B.M. and Nudo, R.J. (2003) Post-infarct cortical plasticity and behavioral recovery using concurrent cortical stimulation and rehabilitative training: a feasibility study in primates. Neurol. Res., 25: 801–810. Teasell, R., Bayona, N.A. and Bitensky, J. (2005) Plasticity and reorganization of the brain post stroke. Top. Stroke Rehabil., 12: 11–26. Teskey, G.C., Flynn, C., Goertzen, C.D., Monfils, M.H. and Young, N.A. (2003) Cortical stimulation improves skilled forelimb use following a focal ischemic infarct in the rat. Neurol. Res., 25: 794–800. Tsubokawa, T., Katayama, Y., Yamamoto, T., Hirayama, T. and Koyama, S. (1991) Treatment of thalamic pain by chronic motor cortex stimulation. Pacing Clin. Electrophysiol., 14: 131–134. Tsubokawa, T., Katayama, Y., Yamamoto, T., Hirayama, T. and Koyama, S. (1993) Chronic motor cortex stimulation in patients with thalamic pain. J. Neurosurg., 78: 393–401.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 16
Functional plasticity following spinal cord lesions Alain Frigon and Serge Rossignol Center and Group for Neurological Sciences, CIHR Group in Neurological Sciences, CIHR Regenerative Medicine and Nanomedicine Team, Faculty of Medicine, Universite´ de Montre´al, Montreal, QC, Canada
Abstract: Spinal cord injury results in marked modification and reorganization of several reflex pathways caudal to the injury. The sudden loss or disruption of descending input engenders substantial changes at the level of primary afferents, interneurons, and motoneurons thus dramatically influencing sensorimotor interactions in the spinal cord. As a general rule reflexes are initially depressed following spinal cord injury due to severe reductions in motoneuron excitability but recover and in some instances become exaggerated. It is thought that modified inhibitory connections and/or altered transmission in some of these reflex pathways after spinal injury as well as the recovery and enhancement of membrane properties in motoneurons underlie several symptoms such as spasticity and may explain some characteristics of spinal locomotion observed in spinally transected animals. Indeed, after partial or complete spinal lesions at the last thoracic vertebra cats recover locomotion when the hindlimbs are placed on a treadmill. Although some deficits in spinal locomotion are related to lesion of specific descending motor pathways, other characteristics can also be explained by changes in the excitability of reflex pathways mentioned above. Consequently it may be the case that to reestablish a stable walking pattern that modified afferent inflow to the spinal cord incurred after injury must be normalized to enable a more normal re-expression of locomotor rhythm generating networks. Indeed, recent evidence demonstrates that step training, which has extensively been shown to facilitate and ameliorate locomotor recovery in spinal animals, directly influences transmission in simple reflex pathways after complete spinal lesions. Keywords: spinal reflexes; neuroplasticity; spinal locomotion; reflex pathways; spinal cord injury; spasticity the absence of afferent feedback in curarised cats (Grillner and Zangger, 1979; Grillner, 1981; Rossignol et al., 1990) (see Fig. 1). The underlying mechanisms for this recovery probably implicate interrelated anatomical, neurochemical, and physiological changes in the spinal cord below the lesion. As such, post spinal excitability changes in reflex pathways and the evolution in the expression of locomotion may in part be interrelated (Muir, 1999; Cote et al., 2003; Cote and Gossard, 2004). In order to better understand how these neuroplastic changes in the cord relate to functional recovery after injury, this review will concentrate firstly on observable modifications at the level of simple circuits such as spinal reflexes and secondly
General introduction After spinal lesions, there is gradual recovery below the lesion of some motor functions, including locomotion (Barbeau and Rossignol, 1987; Be´langer et al., 1996; Rossignol et al., 2000, 2002). Such recovery undoubtedly results from the reexpression of a genetically-determined intrinsic spinal circuitry called a Central Pattern Generator (CPG) that can manifest itself even when spinalization is performed before having learned to walk (Forssberg et al., 1980a, b) and operates even in Corresponding author. Tel.: (514) 343-6111 ext 3305; Fax: (514) 343-7972; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57016-5
231
232
at the level of more complex locomotor patterngenerating networks. In a final section, changes in reflex excitability will be discussed and related to characteristics of spinal locomotion. Changes in reflex pathways after spinal cord injury It has been well established that following lesions of the spinal cord neural circuitry caudal to the injury undergoes substantial modifications in segmental connections due to the dramatic loss of supraspinal and propriospinal inputs (Sherrington, 1899; Mendell, 1984). As such, exteroceptive and proprioceptive reflex pathways display rapid, perdurable, and evolving plasticity after spinal injury. Initially, spinal reflexes are substantially depressed due to the severe decrease in motoneuron excitability as synaptic input from supralesional centers is diminished or lost in complete or partially transected spinal cords (Kuhn, 1950; Barnes et al., 1962; Murray and Goldberger, 1974; Walmsley and Tracey, 1983; Conway et al., 1988; Bennett et al., 1999, 2001b, 2004; Hultborn, 2003). This period, defined as spinal shock, is characterized by muscular paralysis, hypotonus, and the abolition of spinal reflexes (Ashby and Verrier, 1975; Brown, 1994; Ditunno et al., 2004). It appears that afferent transmission in reflex pathways increases immediately
Fig. 1. General scheme of reflex pathways and spinal locomotor control. This scheme is subdivided in three parts. The supraspinal level includes various descending pathways from the telencephalon and brain stem involved in activating, stopping, or modulating characteristics of the spinal central pattern generator (CPG) for locomotion as well as the excitability of transmission in reflex pathways at motoneuronal or premotoneuronal (presynaptic and/or interneuronal) levels. The large arrow emerging from the Supraspinal level encompasses all these functions without any further details. The spinal cord level includes the CPG with a generally reciprocal activity between the Flexor (F) and Extensor (E) sides. These two antagonist phases of the CPG circuitry are separated to indicate that each part may exert a function on other spinal mechanisms (represented by 3 output neurons emerging from each part of the CPG) as well as interact between each other (inhibitory connections between F and E). The Interneurons are
represented by two large pink and blue interneurons [IN] which are interposed between afferents and motoneurons in disynaptic pathways as well as other more specific inhibitory interneurons (in black) representing disynaptic inhibitory pathways (such as Ib inhibitory interneurons) which can also be inhibited by other interneurons in certain tasks such as locomotion. Finally, motoneuron pools include both a-motoneurons projecting to extrafusal muscle fibers and g-motoneurons projecting to intrafusal muscle fibers. Recurrent inhibition s through Renshaw cells inhibits a-motoneurons (represented) and g-motoneurons (not represented) and Ia interneurons responsible for reciprocal inhibition between a-motoneurons. In the periphery, one ankle flexor muscle (pink) and one extensor muscle (blue) are represented with a spindle in both. Group Ia and II represent sensory fibers from spindles and are responsible for indicating rate and amount of muscle stretch, respectively. The stimulation symbol on the Ia fiber from the extensor illustrates direct stimulation of Ia afferents as performed during H-Reflex studies. Ib fibers originate from Golgi tendon organs, which measure the force output of the muscle. Connectivity of the various afferents is partial and is largely based on that established in Rossignol et al. (2006). See Plate 16.1 in Colour Plate Section.
233
after spinal lesions (Nelson et al., 1979; Crenna et al., 1982; Lundberg, 1982; Cook and Woolf, 1985; Baker and Chandler, 1987b; Li et al., 2004b;) since raising motoneuronal excitability to pre injury levels by applying an excitatory conditioning stimulus to the motor pool produces enhanced reflexes (Bennett et al., 2004). However, without artificially augmenting motoneuron excitability, the loss of persistent inward currents (PICs) and plateau potentials obscures any increase in reflex amplitude. Gradually, as motoneuron excitability recovers and the ability to generate PICs and plateau potentials is reestablished (Bennett et al., 2001a) another phase ensues highlighted by exaggerated reflex responsiveness, clonus, and hypertonus (Dimitrijevic and Nathan, 1967b; Brown, 1994; Young, 1994) collectively defined as spasticity. In man, the clinical hyper-reflexia resulting from partial or complete spinal lesions has been extensively described (Dimitrijevic and Nathan, 1967a, b, 1968, 1970, 1971, 1973) and the celerity of changes in input–output properties of spinal reflexes is attributed to the massive loss of control from descending and propriospinal sources (Liddell, 1934; Dimitrijevic and Nathan, 1967b; Engberg et al., 1968; Bennett et al., 2004). As such, several investigators have developed animal models to study changes in reflex pathways after lesions to the spinal cord. For example, animals with partial spinal lesions (Murray and Goldberger, 1974; Aoki et al., 1976; Decima and Morales, 1983; Hultborn and Malmsten, 1983a, b; Carter et al., 1991), or complete transections (Bailey et al., 1980; Smith et al., 1983; Bennett et al., 1999, 2004) display enhanced monosynaptic and/or polysynaptic reflex responses at segmental levels caudal to the lesion. Plastic changes in a reflex arc can occur at primary afferents, interneurons, and motoneurons and it appears that disrupting or interrupting descending pathways results in marked reorganization at all three levels. Therefore, following injury to the spinal cord reflex circuits undergo anatomical and physiological changes and activating or modulating these pathways potentially could enhance functional motor recovery (Muir and Steeves, 1997; Barbeau et al., 1999; Pearson, 2001). It is imperative to understand the reorganization of
spinal pathways after spinal cord injury (SCI) to design and improve training protocols and pharmacological treatments aimed at ameliorating motor functions. The following sections describe changes in simple reflex pathways resulting from altered or abolished descending input after SCI in humans and animal models. Stretch reflexes The stretch reflex, mediated by group Ia and II afferents has both dynamic and static components and has received particular attention following SCI. Fig. 1 represents pathways from group Ia and II fibers originating from different receptors in muscle spindles responsible for detecting changes in muscle length (II) and rate of this change (Ia). Changes in stretch reflex pathways have been ascribed a causal link for spasticity since SCI patients display increased resistance to muscle lengthening, especially to higher stretch velocity. Consequently, the stretch reflex has been used to demonstrate neurophysiological changes after SCI in humans. Immediately following SCI, during spinal shock, stretch reflexes are absent or substantially depressed (Leis et al., 1996a) but recover progressively with time and as spasticity ensues they become exaggerated (Hiersemenzel et al., 2000). Indeed, it is generally reported that stretch reflexes are enhanced in chronic SCI patients (Dimitrijevic and Nathan, 1967a; Taylor et al., 1984; Hiersemenzel et al., 2000; Calancie et al., 2004; Nakazawa et al., 2006b). For example, stretch reflexes of intact individuals were compared to those of spastic patients who had suffered a SCI (Dimitrijevic and Nathan, 1967a). It was shown that although latency was unchanged the amplitude and duration of stretch reflexes were increased after SCI. In addition, following a brief muscle stretch there was a silent period before the muscle became tonically active with an after-discharge that could spread to several muscles of both limbs. This widespread excitation is a hallmark of spasticity. Moreover, changes in stretch reflex pathways are dependent on lesion extent. For example, stretch reflexes, evoked by patellar and Achilles tendon taps, were compared in motor-complete
234
and incomplete SCI individuals within the first 5 days post injury (Calancie et al., 2004). It was found that motor incomplete, compared with motor-complete patients had significantly larger stretch reflexes. As time after injury progressed, stretch reflexes gradually increased in both groups and although the relative increase was larger in motor-complete individuals, motor-incomplete patients still displayed much larger stretch reflexes. That stretch reflexes are larger in incomplete rather than complete SCI has recently been confirmed for chronic patients (Nakazawa et al., 2006a). Spasticity generally results from asymmetric descending influences since animal models using partial lesions, in contrast to complete transections, produce exaggerated reflex responses ipsilateral to the injury (Hultborn, 2003). In humans, in whom clinical signs of spasticity are frequent after SCI, complete loss of descending input is rare and extant projections from some propriospinal and supraspinal sources remain, which may explain why spasticity is a frequent consequence. Since stretch reflex pathways are reorganized after SCI in humans several animal models have replicated these changes to elucidate the mechanisms subserving this plasticity. Using partial lesions of the cat spinal cord, Murray and Goldberger evaluated stretch reflexes in both hindlimbs before and after injury (Murray and Goldberger, 1974). Contrary to other reflexes, stretch reflexes showed little depression on the lesioned side immediately after injury. As time elapsed, however, induction threshold decreased and stretch reflex amplitude increased in the injured side. Reflex responses on the intact side were similar to pre injury levels and thus acted as an appropriate reference. Furthermore, it was reported that increased reflex activity paralleled ameliorations in locomotor performance. In other studies, a model of stretch hyperreflexia was developed in cats by unilaterally sectioning the dorsolateral column in the spinal cord at T13–L3 levels (Taylor et al., 1997, 1999). Ramp and hold flexion of both ankle joints generated stretch reflexes in triceps surae muscles before and after injury in awake cats. To determine whether the lesion influenced stretch reflex threshold and gain the force applied by the animal before the stretch
was monitored as an indicator of motoneuronal pool excitability and matched prior to and after injury. Measures of force such as threshold angle (joint excursion producing a 50 g force) and incremental dynamic stiffness (slope of relationship between dynamic force and joint angle) were used to infer changes in the length–tension relationship and stretch reflex gain. It was shown that incremental dynamic stiffness was increased and threshold angle reduced after spinal lesions at matched pre stretch background force levels on the side ipsilateral to the lesion only, indicating that at a given joint excursion the stretch-evoked response generated more force. The EMG response to stretch in soleus was also greater after injury on the injured side. Intrinsic muscle stiffness was ruled out since ketamine administration, which severely depresses reflex activity, completely abolished the increased stiffness associated with the lesion. Thus, the authors concluded that increased stiffness was due to enhanced reflex activity. Recently, a model of spasticity was developed using a sacral SCI in the adult rat, which does not interfere with bladder, bowel, or hindlimb locomotor function (Bennett et al., 1999). The spinalization was performed at S2 and therefore only influenced cutaneous and muscular regions of the tail. The rat was then inserted inside a hollow plexiglass tube with the tail freely hanging out at the back. This model enabled changes in various reflex pathways of the tail to be investigated, in awake rats, before and after complete sacral transections. A rotating manipulator was developed to generate rapid downward or upward rotations, which induced stretch reflexes to extensor or flexor muscles, respectively. In intact rats stretch applied to either flexors or extensors generated little or no stretch reflex response, indicating that under normal circumstances supralesional centers powerfully inhibit these pathways. However, following spinal shock after SCI in which the tail remained flaccid and non responsive, the tail subsequently became hyperreflexive to muscle stretch approximately 14 days after sacral transection. In these rats with a spastic tail, stretches identical to those performed pre injury produced large and sustained muscle responses resulting from phasic and tonic stretch reflexes. Therefore, similar to humans, in
235
animal models of spasticity stretch reflexes appear to have lower induction thresholds and greatly exaggerated responses after SCI. H-reflexes The H-reflex, which bypasses muscle spindles and the fusimotor drive provided by gamma motoneurons by directly activating Ia afferents (see stimulation symbol on the 1a afferent on the extensor side of Fig. 1), has proven very useful in evaluating, both in clinical and research settings, the central component of the stretch reflex pathway during different conditions and pathologies (Schieppati, 1987; Zehr, 2002; Misiaszek, 2003). In humans, H-reflexes, primarily evoked by stimulating the tibial nerve in the popliteal fossa, and recording electromyographic (EMG) activity in soleus have been studied in acute and/or chronic SCI to evaluate changes in transmission to motoneurons from group Ia afferents (Ashby et al., 1974; Ashby and Verrier 1975; Cadilhac et al., 1977; Taylor et al., 1984; Leis et al., 1996a, b; Hiersemenzel et al., 2000). In acute SCI, during spinal shock, despite the loss of stretch reflexes, Hreflexes are generally present albeit reduced (Weaver et al., 1963; Diamantopoulos and Zander, 1967; Ashby et al., 1974; Cadilhac et al., 1977; Little and Halar, 1985; Leis et al., 1996a). Contrary to stretch reflexes, H-reflexes recover rapidly after SCI and as spasticity ensues H-reflex amplitude increases further (Hiersemenzel et al., 2000). Recently, changes in H-reflex excitability were evaluated in complete spinal adult rats to study reorganization of reflex pathways after SCI (Valero-Cabre et al., 2004). Intramuscular recordings of multiple hindlimb muscles were made before and after a T9 spinal section in rats. H-reflexes were evoked in tibialis anterior, gastrocnemius, and plantar muscles by stimulating the sciatic nerve. Although M-wave amplitude transiently decreased after SCI and recovered toward pre operative values, H-reflex amplitude, expressed as the H:M ratio, was significantly enhanced in all three muscles studied immediately after spinal transection and remained elevated for the remainder of the study (60 days). In addition, H-reflex recruitment curves were facilitated, induction threshold
was reduced, and onset of responses was delayed for all three muscles post spinalization. Although these studies point to increased excitability of H-reflex pathways after SCI, some studies in humans (Ashby et al., 1974; Taylor et al., 1984; Schindler-Ivens and Shields, 2000, 2004) and animals (Thompson et al., 1992; Chen et al., 2001; Lee et al., 2005) show that H-reflex amplitude (H:M ratio) and threshold (Boorman et al., 1996) are not altered after SCI. For example, Schindler-Ivens and Shields (2004) demonstrated that soleus H-reflex threshold, gain, and amplitude were unchanged in clinically complete chronic SCI patients exhibiting signs of spasticity compared to intact individuals. A possible explanation for these discrepant findings is that the degree of injury influences Hreflexes. This does not seem to be the case since no difference in H-reflex amplitude was found in incomplete and complete SCI patients (Nakazawa et al., 2006c). Moreover, severity of injury (mild, moderate, and complete) on plantar muscle Hreflexes, was assessed in adult rats at varying times after SCI injury under chloral hydrate anesthesia, which similar to ketamine has negligible effects on H-reflexes (Lee et al., 2005). The amplitude of baseline H-reflexes tested at 0.1 Hz did not change consistently after SCI. However, similar to other studies (Thompson et al., 1992; Skinner et al., 1996) rate-sensitive depression at high stimulation frequencies was altered after SCI. Indeed, the frequency of stimulation appears to be a critical factor when assessing modifications in H-reflexes after SCI. It is well established that increasing the frequency of stimulation evokes ratesensitive depression of H-reflexes, attributed to neurotransmitter depletion at Ia afferent terminals (homosynaptic depression, HD) (Hultborn et al., 1996; Kohn et al., 1997) or to increased presynaptic inhibition (PSI) (Thompson et al., 1992; Calancie et al., 1993; Schindler-Ivens and Shields, 2000). However, this ‘normal’ rate-sensitive depression is altered after SCI in rats (Thompson et al., 1992; Skinner et al., 1996) and humans (Ishikawa et al., 1966; Calancie et al., 1993; Schindler-Ivens and Shields, 2000), which may lead to specious conclusions concerning changes in H-reflex amplitude or gain. For example, Thompson et al. (1992) using a spinal contusion injury in
236
rats measured the rate-sensitive depression of Hreflexes at different stimulation frequencies. In intact rats and within 6 days of SCI, H-reflexes were attenuated at frequencies greater than 0.3 Hz and as stimulus frequency increased so did H-reflex depression. However, 28 days postinjury the ratesensitive depression of H-reflexes was severely diminished and in some animals reflexes were potentiated with increasing stimulus frequency. This impairment in rate-sensitive depression after SCI was maintained for the duration of the study. In another study, rate-sensitive depression was assessed after SCI in humans (Schindler-Ivens and Shields, 2000). Soleus H-reflexes were evoked at different stimulation frequencies in intact, acute, and chronic SCI individuals. Although H-reflex amplitude was diminished in all three groups with increasing stimulation frequency, the H-reflex depression was considerably less in chronic SCI patients. Moreover, rate-sensitive depression was gradually reduced in an acute SCI individual tested for several weeks reaching levels similar to chronic SCI. These results in rats and humans indicate that impaired rate-sensitive depression does not occur immediately post injury, but evolves and persists over time. Consequently, changes in Hreflex amplitude after SCI should be interpreted with caution since putative increases in H-reflex amplitude or gain might simply result from altered rate-sensitive depression if reflexes are evoked at stimulation frequencies sensitive to this effect. To avoid effects from rate-sensitive depression, which is altered after SCI, H-reflexes should be elicited at stimulation frequencies no greater than 0.1 Hz. H-reflex excitability has also been evaluated during walking in intact and SCI patients. In intact humans soleus H-reflex excitability exhibits phasedependency during the step cycle progressively increasing from early to late stance and becoming quiescent during the swing phase (Capaday and Stein, 1986; Crenna and Frigo, 1987; Simonsen and Dyhre-Poulsen, 1999; Rossignol et al., 2006). This pattern of modulation persists in incomplete SCI patients showing weak signs of spasticity but changes dramatically in more severe cases of spasticity (Fung and Barbeau, 1994). In the latter, modulation during stance and inhibition during swing are abolished. Furthermore, conversely to
intact and SCI individuals with mild spasticity, which show marked attenuation of soleus H-reflexes during walking compared to standing, severe spastic SCI patients have ambulatory soleus Hreflexes of greater or similar value relative to standing. Therefore, phase- and task-dependent modulation of H-reflexes is impaired after SCI and alleviating or reducing spasticity appears to restore this modulation. Intracellular recordings and group Ia afferents Intracellular motoneuron recordings following peripheral nerve stimulation have been investigated in a few studies after SCI in animal models (Nelson and Mendell, 1979; Mayer et al., 1984; Munson et al., 1986; Hochman and McCrea, 1994a). Acute experiments performed under anesthesia, which abolish some motoneuronal properties such as plateau potentials (Guertin and Hounsgaard, 1999; Hultborn, 2003), might mask alterations in reflexes but enable determination of changes at premotoneuronal levels. For example, Hochman and McCrea (1994a) compared Ia EPSPs of lumbar ankle extensor motoneurons, evoked by peripheral nerve stimulation, in non lesioned and chronic cats spinalized at L1–L2 6 weeks prior to the acute experiment. It was found that Ia EPSPs in lateral gastrocnemius (LG) and soleus motoneurons after stimulating the homonymous nerve (LGS) were significantly greater in spinal cats compared to non lesioned animals and had reduced rise times. However, medial gastrocnemius (MG) Ia EPSP amplitude and rise time were largely unchanged in response to MG nerve stimulation, corroborating earlier reports (Nelson and Mendell, 1979; Mayer et al., 1984; Munson et al., 1986), which indicates that not all motoneuronal pools are influenced similarly by spinal cord transection. Furthermore, heteronymous rather than homonymous monosynaptic connections from Ia afferents became stronger following spinalization. For example, stimulation of LGS and MG nerves with recordings in MG and LG motoneurons, respectively, were twice as large in chronic spinal cats compared to non lesioned cats. This indicates a reorganization of synaptic connections between synergists after spinalization, and not an alteration
237
in passive membrane properties since a general change in motoneuron excitability would influence all responses similarly (Hochman and McCrea, 1994b). The authors concluded that larger homonymous and heteronymous Ia EPSPs could in part mediate increased ankle extensor stretch reflexes in chronic spinal cats. Mechanisms subserving changes in reflexes from group Ia afferents Changes in reflexes from group Ia afferents can result from altered gamma-motoneuron activity, PSI, or HD at Ia terminals, and/or changes in motoneuron excitability (see Fig. 1). Modified alpha-motoneuron excitability is discussed later on since it impacts all reflex pathways. There is evidence to suggest that diminished stretch reflexes during spinal shock result from depressed gamma motoneuron activity. For example, several studies report that H-reflexes, which are not influenced directly by fusimotor drive, are present during spinal shock (Weaver et al., 1963; Diamantopoulos and Zander, 1967; Ashby et al., 1974; Cadilhac et al., 1977; Little and Halar, 1985; Leis et al., 1996a), whereas stretch reflexes are absent or severely depressed (Leis et al., 1996a), indicating that gamma motoneuron activity is reduced immediately after SCI. Although fusimotor drive recovers, enhanced activity does not appear to contribute to increased stretch reflexes after SCI since primary muscle spindle responses are depressed or unchanged in chronic spinal cats (Bailey et al., 1980). In this study, intact cats were compared with acute and chronic spinal cats at various epochs post spinalization. Muscle spindle activity was recorded in MG Ia fibers in dorsal roots in response to static stretch. It was shown that acute spinal cats (3 days post spinalization) were hypotonic and slightly hyperreflexic and that spindle responses were significantly depressed. In chronic spinal cats the limb was hypertonic and hyperreflexic, reaching a peak 30 days post spinalization, and although muscle spindle activity recovered it was still lower than control levels. These results indicate that depressed motoneuron excitability observed in spinal shock could in part be mediated by decreased facilitatory inputs from
muscle spindles but that exaggerated stretch reflexes in spasticity are not attributed to increased fusimotor activity. Altered PSI and/or HD at Ia terminals are attractive candidates since these mechanisms are under control from descending pathways (Rudomin and Schmidt, 1999). In acute stages of injury PSI or HD appear to be greatly increased, which may contribute to depressed motoneuronal excitability (Ashby et al., 1974; Ashby and Verrier, 1975; Calancie et al., 1993). For example, soleus Hreflexes were evoked with or without homonymous tendon vibration to assess changes in PSI and/or HD in acute and chronic SCI (Ashby et al., 1974; Ashby and Verrier, 1975). It was found that conversely to intact individuals, vibration completely abolished H-reflexes in acute SCI, indicating that PSI and/or HD are heightened shortly after SCI. However, contrary to acute stages, PSI and/or HD become considerably diminished in chronic SCI patients (Taylor et al., 1984; Levin and Chapman, 1987; Calancie et al., 1993; Nielsen et al., 1993; Faist et al., 1994). HD, assessed at different stimulation frequencies or by applying a preceding muscle stretch, is thought to result from reduced transmitter release due to prior activation of Ia afferents and is modified with changes in descending input in humans (Ishikawa et al., 1966; Calancie et al., 1993; Nielsen et al., 1993, 1995; Schindler-Ivens and Shields, 2000) and rats (Thompson et al., 1992; Skinner et al., 1996). Indeed, activation history influences soleus Ia afferent pathway excitability since stretching the soleus has considerably less effect in reducing H-reflexes in chronic SCI patients compared to intact participants (Nielsen et al., 1993). In other studies, Hreflexes evoked at different stimulation frequencies were used to assess HD. For example, H-reflexes in intrinsic foot muscles, elicited by stimulating the tibial nerve, were evaluated before and after a contusion injury at the 8th thoracic level at different frequencies of stimulation at various times after SCI (Thompson et al., 1992). At 6 days post injury no rate-dependent depression, as a function of intact controls, was found. However, by 28 days H-reflexes evoked at stimulation rates above 1 Hz were considerably greater than controls and this reduction in HD persisted for the duration of the
238
study. Similar findings have recently been demonstrated in humans (Schindler-Ivens and Shields, 2004). These results indicate that diminished HD evolves gradually after SCI and persists with time. To solely assess changes in PSI at Ia afferent terminals soleus H-reflexes were conditioned by stimulating the femoral nerve, which facilitates soleus H-reflexes by reducing PSI at Ia afferent terminals (Hultborn et al., 1987) in intact and chronic SCI participants (Faist et al., 1994). It was found that facilitation of soleus H-reflexes was considerably greater in SCI patients compared to intact participants suggesting that presynaptic inhibitory control is impaired with spinal lesions. In another study soleus H-reflexes were conditioned by stimulating the common peroneal nerve at C–T intervals known to evoke PSI of the soleus Ia afferent pathway in intact and complete chronic SCI individuals (Levin and Chapman, 1987). In intact individuals CP conditioning elicited profound attenuation of soleus H-reflexes, whereas, in SCI patients this depression was significantly reduced, indicating that PSI of the soleus H-reflex pathway is altered after SCI. Therefore, these two studies provide evidence that PSI at Ia afferent terminals is impaired and diminished following SCI. Recurrent inhibition Renshaw cells, which mediate classic recurrent inhibition of alpha-motoneurons (Renshaw, 1941), also strongly project to gamma-motoneurons, Ia inhibitory interneurons (see Fig. 1) and other Renshaw cells. Since segmental and descending pathways project to Renshaw cells (McCrea et al., 1980; Pratt and Jordan, 1980; Katz and PierrotDeseilligny, 1998) their activity may be modified following SCI promoting widespread changes in interneuronal and motoneuronal activity. Using a conditioned H-reflex technique (PierrotDeseilligny and Bussel, 1975) it was demonstrated that recurrent inhibition is consistently increased in SCI patients (Shefner et al., 1992). In this study, stimulation of the tibial nerve in the popliteal fossa, to evoke soleus H-reflexes was followed by a supramaximal pulse 10 ms after, allowing investigation of recurrent inhibitory pathways (see Shefner et al. for full methods). No significant
differences in H:M ratios between intact and SCI individuals were found. However, marked differences were found for the conditioned H-reflex (H0 ), which negatively correlates with the level of recurrent inhibition. In all intact individuals the H0 response was clearly present, whereas in most SCI patients this response was absent or very small. This indicated that recurrent inhibition, which is normally under inhibitory control from brainstem centers (Fung et al., 1987), is augmented following interruption or disruption of descending inputs to Renshaw cells. Interestingly, with multiple testing sessions the appearance of H0 responses, or reduced recurrent inhibition, in SCI patients was associated with clinical improvements, particularly a decrease in stiffness. Increased recurrent inhibition has also been demonstrated on the lesioned side of hemisected cats (Hultborn and Malmsten, 1983b). In this study, stimulating cut dorsal roots evoked test reflexes with discharges recorded in intact ventral roots. Recurrent inhibition was evoked by conditioning test reflexes with antidromic volleys in peripheral motor nerves, which reduce the excitability of alpha-motoneurons. It was demonstrated that recurrent inhibition was predominantly larger on the hemisected side compared to the non lesioned side. These studies in humans and cats indicate that Renshaw cell activity, and thus recurrent inhibition, is heightened after SCI. Although increased recurrent inhibition would depress motoneuronal excitability and therefore prevent hyperreflexia their powerful connections to other interneurons, especially Ia inhibitory interneurons should be considered. Reciprocal inhibition Reciprocal inhibition, mediated by disynaptic connections from homonymous Ia afferents to antagonist motoneurons (see Fig. 1), via the Ia inhibitory interneuron, normally prevents simultaneous contraction of agonist and antagonist during movement (Baldissera et al., 1981). Ia reciprocal inhibition is generally reduced (Dimitrijevic and Nathan, 1967b; Yanagisawa and Tanaka, 1978; Boorman et al., 1996; Okuma et al., 2002; Crone et al., 2003; Xia and Rymer,
239
2005) in complete and incomplete SCI humans, resulting in abnormal motor behavior characterized by increased coactivation of agonists and antagonists. For example, it was reported that spontaneous lower limb movements in SCI humans were delineated by simultaneous activation of agonists and antagonists (Dimitrijevic and Nathan, 1967b). Furthermore, this coactivation could persist for several minutes indicating that normal reciprocal inhibition between antagonistic muscle groups is severely impaired after SCI. Since these early studies, others have shown that instead of normal reciprocal inhibition between agonist and antagonist a reciprocal excitation or facilitation in spinal spasticity potentially mediated by an oligosynaptic excitatory pathway emerges (Okuma et al., 2002; Crone et al., 2003; Xia and Rymer, 2005). For one study, soleus H-reflexes were conditioned by stimulating the common peroneal nerve which, in normal individuals, evokes reciprocal inhibition of the group Ia afferent pathway of the soleus (Crone et al., 2003). At a condition-test (CT) interval of 2–4 ms common peroneal nerve stimulation elicited short-latency inhibition of soleus Hreflexes in intact individuals, which is mediated by a disynaptic reciprocal Ia inhibitory pathway (Crone et al., 1987, 1994; Crone and Nielsen, 1989). However, in SCI patients, this inhibition is replaced by marked facilitation of soleus H-reflexes. The authors concluded that this facilitation was mediated by an oligosynaptic group I afferent pathway, which is normally under inhibitory control from supraspinal structures. Moreover, the time-course of this facilitation was associated with the emergence of hyperactive stretch reflexes, which indicates that changes in reciprocal pathways may be one of the underlying causes of spasticity. Recently, reciprocal pathways were evaluated by evoking stretch reflexes in ankle flexors and extensors of incomplete and complete SCI patients by applying a tap to the tibialis anterior and Achilles tendon, respectively (Xia and Rymer, 2005). In both groups there was evidence of reciprocal facilitation. In other words, responses in antagonists were greater than the muscle being stretched. Moreover, the incidence of reciprocal facilitation was considerably greater in incomplete spinal individuals. These two studies provide strong evidence
that reciprocal pathways are reorganized after SCI and that varying degrees of change in Ia reciprocal pathways depend on the extent of the lesion since the prevalence of reciprocal facilitation is greater in incomplete SCI (Xia and Rymer, 2005). This indicates that remaining descending connections have more disruptive effects on the control of reciprocal inhibition than complete lesions. In another study using cases of asymmetric spinal spasticity, where one side exhibits more clinical signs, the side displaying better recovery and less spasticity was associated with more pronounced Ia reciprocal inhibition (Okuma et al., 2002). For this, soleus H-reflexes were preceded (0.5–10 ms) by a conditioning stimulation consisting of single pulses to the common peroneal nerve to induce reciprocal inhibition of the soleus group Ia afferent pathway. This was done in both legs to determine if there was asymmetric reciprocal inhibition. Compared to intact individuals, SCI patients had much greater H:M ratios but no differences were seen between the two limbs in the injured group. However, significant differences in the level of reciprocal inhibition were found between the legs of spinal patients, whereas intact individuals showed no side-to-side differences. In all patients studied, reciprocal inhibition was greater on the less affected side and in some instances a reciprocal facilitation was observed on the spastic side. The authors thus suggested that the level of reciprocal inhibition could be a good indicator of the clinical severity of spasticity and that restoring reciprocal inhibition benefits functional recovery. Besides the emergence of an oligosynaptic pathway, another potential mechanism for reciprocal facilitation is increased Renshaw cell activity (Shefner et al., 1992), which exerts inhibitory actions on Ia inhibitory interneurons (Hultborn et al., 1971; Fedina and Hultborn, 1972). As such, increased recurrent inhibition from Renshaw cells on Ia inhibitory interneurons after SCI would lead to decreased reciprocal inhibition and potentially to an increased likelihood of cocontraction (see the third section and Fig. 2). Therefore, absence, impairment, or asymmetries of descending inputs to Ia inhibitory interneurons result in reorganization of reciprocal reflex pathways between antagonistic motor pools.
240
Fig. 2. (A) Comparison of locomotion in the intact and spinal states in the same cat. (B) Stick figures of the swing and stance phases reconstructed from videos taken at 60 frames/s. Each frame is de-interlaced to extract each video field and light reflecting markers placed on the various joints are software detected and used to reconstruct the movement as stick figures for each joint as indicated. The angles measured are indicated and the arrows show the direction of limb movement. The stick figures are displaced from one another by a value corresponding to the foot displacement in the X direction so that each stick is clearly individualized. The length calibration found in B is therefore different in the X and Y axes. (C) Same as in A but 41 days after spinalization with daily locomotor training on the treadmill. Note that in this case the limb is somewhat extended throughout the cycle and that overall step length is reduced compared to A. The calibration is explained in A. (D) Joint angular displacements are illustrated for the intact state (black, 25 cycles) and spinal state (red, 16 cycles). Note the shift of hip in extension (shift of the hip trace) and of the knee and ankle extension during stance. The cycle is normalized to unity. Electromyographic (EMG) recordings of the same sections as in C. Ipsilateral (i) refers to the side of the video camera while (co) is contralateral. Semitendinosus (St) is mainly a knee flexor but also acts as a hip extensor; Tibialis Anterior (TA) is an ankle flexor; Vastus Lateralis (VL) is a knee extensor and Gastrocnemius Lateralis (GL) an ankle extensor; Flexor Digitorum Longus plantarflexes the digits and is active during stance. EMGs are largely superimposed in both conditions although some changes are distinguishable and discussed in the text. EMGs have also been normalized to unity. See Plate 16.2 in Colour Plate Section.
Nonreciprocal group I pathway Ib afferents from Golgi tendon organs mediate disynaptic inhibition of homonymous muscles via Ib interneurons (see Fig. 1) and transmit information related to contraction-generated tension
(Jami, 1992). However, other segmental afferents, such as group Ia and cutaneous afferents, project to Ib interneurons (see Fig. 1) and as such nonreciprocal group I pathway is considered a more appropriate term (Jankowska, 1992). The group I pathway exhibits a reflex reversal (e.g. switches to
241
a disynaptic excitatory connection) under certain circumstances, such as during locomotion (Gossard and Hultborn, 1991; Pearson and Collins, 1993; Gossard et al., 1994; McCrea et al., 1995; Prochazka, 1996). This is represented in Fig. 1 by the inhibitory interneurons inhibiting the black inhibitory interneurons. Thus, the nonreciprocal group I pathway exerts powerful actions on locomotor networks. Cortico- and reticulospinal tracts have strong effects on Ib interneurons (Schomburg, 1990; Jankowska, 1992) and it would thus be anticipated that disruption of these pathways engenders a reorganization in nonreciprocal group I pathways. However, in humans interrupting or disrupting these pathways does not seem to influence transmission from group Ib afferents (Downes et al., 1995). To demonstrate this, soleus H-reflexes were conditioned by stimulating the MG nerve at a condition-test interval of 6 ms in intact and SCI individuals, which has previously been shown to evoke group Ib afferent disynaptic inhibition in humans (Pierrot-Deseilligny and Morin, 1979; Pierrot-Deseilligny et al., 1981; Delwaide et al., 1991). As there were no changes in Ib-mediated inhibition between the two groups it was concluded that reflex effects from Ib afferents are unchanged after SCI in humans (Downes et al., 1995). However, since transmission from other afferent inputs, which converge on Ib interneurons, is altered following SCI it is probable that considerable modifications occur in group I nonreciprocal pathways. Alternatively, that the nonreciprocal group I pathway is largely immutable following SCI may highlight its significance for functional motor recovery. Group II pathway Group II afferents mediate nonmonosynaptic stretch reflexes, flexor reflexes, and some postural reflexes and an increase in the gain of group II reflex pathways is thought to contribute to spasticity following SCI (Jankowska, 1992; Eriksson et al., 1996; Jankowska and Hammar, 2002). Jankowska and Hammar (2002) posit that group II reflexes exert such potent effects in spasticity that interneuronal contribution from group Ia
reciprocal and Ia/Ib nonreciprocal pathways are comparatively negligible. The evidence for reorganization of group II pathways following SCI is sparse due to the greater difficulty in selectively activating these pathways. However, pharmacological agents that alleviate spastic symptoms are thought to selectively target group II afferent pathways (Bras et al., 1990; Jankowska et al., 2000; Jankowska and Hammar, 2002). For example, systemic or intrathecal administration of the noradrenergic (NA) agonists clonidine and tizanidine, or the NA precursor L-DOPA, abolishes activation of interneurons in group II reflex pathways and reduces spastic signs. Thus, in cats the loss of NA-releasing descending tract neurons, which exert powerful inhibitory control of interneurons intercalated in group II pathways may promote the characteristic exaggeration in reflex responsiveness (Jankowska and Hammar, 2002). Therefore, hyperreflexia in spinal spasticity may partly be due to modifications in descending inhibitory control of group II afferent pathways. Although changes in group II pathways has not been investigated with SCI, there is evidence that modified descending input to spinal centres alters transmission in these pathways in humans (Marque et al., 2001; Maupas et al., 2004). In one study, transmission in group II pathways was assessed by conditioning the quadriceps H-reflex with a stimulus to the common peroneal nerve in hemiplegic post stroke patients with clinical signs of spasticity (Marque et al., 2001). In intact individuals this conditioning stimulus has been shown to evoke an early (10–12 ms C–T interval) and a late (15–20 ms C–T interval) facilitation of the quadriceps H-reflexes via group I and group II afferent activation, respectively (Marque et al., 1996; Simonetta-Moreau et al., 1999). It was shown that the spastic limb had considerably greater group I and II mediated facilitation compared to the non spastic side and matched controls. There were no significant differences between the non spastic side and control participants. The authors attributed the greater facilitation to enhanced transmission in group I and II pathways resulting from changes at premotoneuronal levels due to altered descending control. In a subsequent study, the same group used tizanidine, an alpha-2 noradrenergic agonist
242
known to selectively depress transmission in group II pathways, to evaluate its effects on the facilitation of quadriceps H-reflexes by common peroneal nerve conditioning in hemiplegic spastic poststroke patients and intact individuals (Maupas et al., 2004). Oral administration of tizanidine considerably reduced the group II-, and to a lesser extent, group I-mediated facilitation of quadriceps H-reflexes so that no significant differences were apparent between spastic and unaffected sides. Therefore, these results show that altered descending input modifies group II afferent transmission and that administration of NA agonists in part reduces spasticity by selectively targeting these pathways. Cutaneous pathways Cutaneous reflexes, via polysynaptic pathways, predominantly modify ongoing movement and participate in positioning the foot during locomotion (Zehr and Stein, 1999; Bouyer and Rossignol, 2003a, b; Rossignol et al., 2006). For example, stimulation of the dorsal surface of the hindpaw during stepping generates a coordinated reflex response of the hindlimb musculature allowing the perturbed limb safe passage over an impeding obstacle (Forssberg, 1979). These responses are preserved after spinalization (Forssberg et al., 1975). Since spinal interneuronal networks mediating cutaneous reflexes receive strong projections from descending sources (Schomburg, 1990) lesions of the spinal cord greatly impact their behavior. This section will predominantly discuss changes in cutaneous reflex pathways originating from low-threshold cutaneous afferents. High-threshold cutaneous and/ or nociceptive afferents are discussed later on. Changes in cutaneous pathways projecting to motoneuronal pools has been investigated by stimulating low threshold cutaneous afferents and observing modulatory effects on soleus H-reflexes in intact and SCI individuals (Lebizec et al., 1983; Levin and Chapman, 1987; Fung and Barbeau, 1994; Knikou and Conway, 2001). What becomes evident is that not all cutaneous pathways are similarly influenced by altered descending input. For example, in intact individuals, a conditioning stimulus to the sural nerve has been shown to
facilitate the soleus H-reflex (Hugon and Delwaide, 1969) and has been attributed to supraspinal influences (Delwaide et al., 1981). To test whether supraspinal sources were responsible for this facilitation, Lebizec et al. (1983) studied sural nerve conditioning of soleus H-reflexes in complete SCI patients. It was found that facilitation persisted in SCI individuals and that the magnitude of increase was similar to previously studied intact individuals. What differed, however, was that progressive increases in conditioning intensity did not inhibit soleus H-reflexes, which normally occurs in intact individuals (Delwaide et al., 1981). In other studies, the conditioning effects of medial plantar nerve (MPN) stimulation, which innervates the sole of the foot, on soleus H-reflex excitability was assessed by electrical stimulation (Fung and Barbeau, 1994) and by mechanical loading of the foot (Knikou and Conway, 2001) in intact and SCI individuals. It was found that MPN conditioning of the soleus H-reflex, which produces a clear inhibition of the H-reflex, did not differ in intact and SCI individuals under resting conditions. Moreover, Fung and Barbeau (1994) further tested MPN conditioning during locomotion. In intact individuals, MPN conditioning modulated soleus H-reflex excitability phase-dependently with the inhibition being greater during early stance and swing phases. Although severe spastic SCI patients did not exhibit phase-dependent modulation of soleus H-reflexes during locomotion the conditioning stimulation from MPN was modulated similarly to intact participants. The authors thus suggested that deficient Ia modulatory mechanisms during walking could be restored with conditioning stimuli mimicking foot contact. In another study the effects from the superficial peroneal nerve (SPN) onto the soleus H-reflex pathway were investigated in intact and SCI individuals (Levin and Chapman, 1987). In intact individuals, a conditioning stimulus to the SPN, which activates primarily low threshold Ab cutaneous afferents, at C–T intervals ranging from 30–190 ms generally facilitates the soleus H-reflex. Conversely to sural nerve stimulation (Lebizec et al., 1983), no facilitation from the SPN was observed in SCI patients regardless of C–T intervals and moreover, in roughly half the SCI individuals
243
tested, SPN conditioning produced an attenuation of soleus H-reflex amplitude rather than facilitation. In intact individuals, inhibitory actions from cutaneous afferents are typically associated with noxious stimulation (Delwaide et al., 1981). These results indicate that facilitatory actions from the SPN projecting to the soleus H-reflex pathway are altered following SCI. These studies in human SCI patients indicate that not all cutaneous pathways are similarly impacted following injury. Thus, although supraspinal influences are not required for sural-induced facilitation or MPN-induced inhibition of soleus H-reflexes, it does appear that the loss of supraspinal control generates some changes in cutaneous pathways and that activating preserved pathways might benefit functional motor recovery. This is in agreement with a recent study showing that step training in spinal cats differentially influences transmission in several cutaneous pathways (Cote and Gossard, 2004). Thus, plastic changes in reflex pathways either resulting from SCI or training are not uniformly distributed amongst cutaneous pathways. Changes in transmission of cutaneous pathways have been shown to occur after SCI in animal models (Afelt, 1970; Baker and Chandler, 1987b; Bennett et al., 1999, 2004; Valero-Cabre et al., 2004). For example, to study changes in the transmission of cutaneous afferents after complete spinal transection intracellular recordings of hindlimb triceps surae motoneurons were made at various epochs following injury in response to sural-nerve stimulation in cats (Baker and Chandler, 1987b). Similar to Lebizec et al. (1983), there were no changes in short or long latency post-synaptic potential patterns or latencies, although response amplitudes were markedly greater in chronic compared to acute spinal cats. This confirmed that cutaneous pathways from the sural nerve after spinalization persist over time after SCI but that transmission was enhanced. It was established that augmented reflex responses were the result of reorganized synaptic connections at interneuronal levels since no changes in motoneuron membrane properties were observed (Baker and Chandler, 1987a), although this was done under anesthesia which abolishes certain properties
such as plateau potentials. However, whether or not responses in chronic spinal cats were greater than pre injury levels was not evaluated. To address this point, Valero-Cabre et al. (2004) stimulated the tibial nerve at the ankle at various stimulus intensities to selectively activate different afferent populations before and following T9 spinal transections in rats. Based on onset latencies it was possible to distinguish three components mediated by different afferents (C1: Aab; C2: Ad; C3: C fibers) and to determine their changes after SCI. Ipsilateral and contralateral responses were recorded in biceps femoris and tibialis anterior muscles, respectively. It was found that immediately after injury the ipsilateral C1 component was largely unaffected whereas C2 and C3 components were completely abolished. At 14 days post injury, C2 and C3 reappeared but remained diminished for the rest of the study, while the C1 component was markedly increased (peak 300% increase) compared to pre transection levels maintaining elevated values for the duration of the study. The crossed spinal reflex (stimulation of the tibial nerve on the left side recorded in the tibialis anterior of the right side) was abolished for all three components 1 h post injury. Whereas C1 and C3 showed some recovery, the C2 component reappeared in a few rats during the 60 days post-transection. In another study, cutaneous responses to light touch were small or absent in intact rats but following sacral transection (414 days postinjury) these responses were greatly enhanced in both amplitude and duration. To quantify these responses electrical stimulation of the distal part on the dorsal surface of the tail was used to evoke a predominantly pure activation of low-threshold cutaneous afferents (Bennett et al., 1999, 2004). In both intact and acute spinal rats this stimulation induced a short latency inhibition, which lasted up to 500 ms, followed by an excitation, which could not be potentiated with repetitive stimuli. However, in chronic spinal rats, the same stimulation produced a pure short latency excitation and repetitive stimulation evoked progressively larger responses. Thus, after sacral SCI the short latency inhibitory cutaneous reflex reverses to an excitatory response and due to the lack of inhibition repetitive stimuli can generate increased and sustained muscular
244
activity resulting from altered interneuronal connections and recovered motoneuron properties such as plateau potentials. These results in humans and animal models show that cutaneous transmission is altered following spinal transections and that reflexes mediated by distinct afferent populations are modified and recover differently after SCI. Moreover, crossed reflex responses show marked differences in reorganization compared to ipsilateral pathways. Since various reflex pathways are differently influenced a general change in motoneuronal excitability cannot constitute the primary mechanism. Thus alterations at interneuronal levels are undoubtedly occurring. Interlimb reflexes coupling upper and lower limbs Interlimb reflexes evoked by cutaneous and/or muscle-nerve stimulation of the lower limbs and recorded in upper limb musculature, or vice-versa, is postulated to participate in the coordination of the four limbs during locomotion in intact humans and animals (Gassel and Ott, 1973; Kearney and Chan, 1979, 1981; Piesiur-Strehlow and Meinck, 1980; Meinck and Piesiur-Strehlow, 1981; Zehr et al., 2001). Following cervical SCI interlimb reflexes are more prevalent with consistently earlier onset latencies (Calancie, 1991; Calancie et al., 1996, 2002, 2005) compared to the intact state (Zehr et al., 2001). For example, interlimb responses recorded in upper limb muscles evoked by stimulating the tibial nerve in the popliteal fossa do not emerge before 6 months after cervical SCI (Calancie et al., 2002) and responses become increasingly prevalent in individuals as time elapses subsequent to the lesion (Calancie et al., 2005). Furthermore, interlimb reflexes, which are more prominent in proximal muscles in intact individuals (Zehr et al., 2001), are rare in proximal motor pools after SCI, becoming more evident in distal musculature (Calancie et al., 2002). As such, few interlimb responses are evoked in proximal upperarm musculature compared to muscles of the forearm and hand after SCI (Calancie et al., 2005). Interestingly, in intact individuals, interlimb reflex activity is characterized by an initial inhibitory response followed by an excitatory one (Zehr et al.,
2001) whereas, after SCI interlimb responses are almost exclusively excitatory (Calancie et al., 1996). Thus, taken together, these observations provide strong evidence that interlimb reflexes, mediated by propriospinal connections coupling cervical and lumbosacral motor pools, are deeply reorganized and are constantly evolving following SCI. High-threshold afferents High-threshold afferents, via polysynaptic connections from muscle, cutaneous, or nociceptive receptors, are involved with posture and locomotion (Lundberg, 1979; Grillner, 1981), and mediate several motor behaviors such as the classic withdrawal reflex (Sherrington, 1910). Since supraspinal centres exert powerful actions on these pathways (Schomburg, 1990) the loss of descending control induces marked reorganization. In intact humans, activation of high threshold afferents consistently evokes a short latency flexion response (Pedersen, 1954; Hugon and Delwaide, 1969), which becomes infrequent and/or absent after motor-complete SCI (Roby-Brami and Bussel, 1987; Knikou et al., 2005). However, in the latter group a longer latency reflex arises conducted by low-threshold cutaneous and/or muscle afferent pathways (Dimitrijevic and Nathan, 1968; Shahani and Young, 1971; Roby-Brami and Bussel, 1987; Andersen et al., 2004; Knikou et al., 2005). In one study, Roby-Brami and Bussel (1987) electrically stimulated tibial and sural nerves to evoke flexion reflexes in complete SCI patients. It was found that early latency reflexes were infrequent in SCI patients whereas longer latency reflexes (4120 ms) were consistently observed. The onset of late reflexes become progressively delayed with increasing stimulus intensity and occurs independently of the early response. Afferents mediating early responses are thought to inhibit the pathway responsible for the late response since longer latency reflexes are more prominent and/or prevalent if the early component is suppressed or weakened (Anden et al., 1966a, b; Jankowska et al., 1967a, b). Indeed, in the acute or chronic spinal cat the late reflex surfaces with DOPA injection, suppressing the early response in the process (Anden et al., 1966a, b; Jankowska et al., 1967a, b). It had previously been
245
observed in spinal cats that late reflexes appeared provided the early flexion reflex was weak (Anden et al., 1966b), which might explain why in man the late flexion reflex occurs without drug administration (Roby-Brami and Bussel, 1987). The emergence of a polysynaptic pathway has also been demonstrated in spinal rats (Bennett et al., 2004). In this study stimulating the tip of the tail evoked polysynaptic latency responses, which became significantly more common and amplified after chronic spinal transection (Bennett et al., 2004). The longer latency component has been implicated in the generation of spinal locomotion in the cat (Jankowska et al., 1967a; Schomburg et al., 1998) and may therefore represent a pathway released from inhibition by the loss of descending signals since it is absent in intact individuals (Roby-Brami and Bussel, 1987). In more recent studies, electrical stimulation of afferents or manipulation of the hip joint has been shown to modulate the late flexion reflex in SCI humans (Knikou and Conway, 2005; Knikou et al., 2006). Flexion reflexes were evoked by electrically stimulating the sural nerve below the lateral malleolus at high intensities. Like previous reports (Roby-Brami and Bussel, 1987) short-latency reflexes were rare after SCI, whereas long latency flexion reflexes were consistently evoked. It was demonstrated that hip flexion depressed the late flexion reflex, corroborating earlier findings (Dimitrijevic and Nathan, 1968), while extension of the hip produced facilitation. A similar depression of the late flexion reflex was observed with functional electrical stimulation of the rectus femoris, a hip flexor (Knikou and Conway, 2005). These effects highlight the importance of sensory feedback in modulating spinal interneuronal systems, which may prove useful for rehabilitative strategies (Knikou and Conway, 2005; Knikou et al., 2006). Other studies have shown that other parameters of high-threshold afferent pathways are altered following SCI. For example, spinal lesions modify receptive fields of nociceptive withdrawal reflexes in rats (Schouenborg et al., 1992) and humans (Andersen et al., 2004). In these studies nociceptive withdrawal reflex receptive fields are expanded after SCI and reflex threshold is decreased. For example, electrical stimulation at various sites on the
plantar surface of the foot, just above threshold for tibialis anterior responses, was used to elicit flexion reflexes in intact and complete SCI individuals (Andersen et al., 2004). Response amplitude in tibialis anterior and soleus were generally augmented in the SCI group compared to intact individuals at virtually every site tested, indicating that receptive fields for these muscles had expanded into previously absent regions. Moreover, ankle flexion in SCI patients following electrical stimulation was considerably greater than intact individuals. These results provided quantitative evidence that nociceptive reflex receptive fields are expanded in agonist and antagonist muscles after SCI and that reflex amplitude is enhanced. Another novel property of high threshold afferent pathways after SCI is that successively applied stimuli generate windup, a considerably potentiated response (Bennett et al., 1999; Hornby et al., 2003). For example, a suprathreshold stimulus to the plantar skin of the foot followed by successively applied stimuli at intervals less than 3 s generate windup of flexion reflexes in SCI patients (Hornby et al., 2003). A similar phenomenon is found in sacral spinal rats following light brushing of tail hairs, which induces a powerful flexor response after SCI no seen in the intact state, which can be greatly potentiated by applying three consecutive stimuli (Bennett et al., 1999). This windup of flexion reflexes in rats has been linked to the development of unusually long (200–500 ms) polysynaptic excitatory (pEPSPs) inputs to motoneurons after SCI (Li et al., 2004a). These pEPSPs in conjunction with the loss of inhibitory control of PICs and plateau potentials in turn produce slow maintained motoneuronal depolarizations, which cause spastic reflexes (Bennett et al., 2004). Furthermore, under normal circumstances, as described by Sherrington (1910) and termed local sign, contraction of a particular muscle withdraws the skin region receiving the nociceptive stimuli (Schouenborg and Weng, 1994; Andersen et al., 1999). However, following SCI this property is altered. For example, local sign withdrawal, a reflex directing the limb away from noxious stimuli, was evaluated in intact and complete SCI patients (Schmit et al., 2003). Noxious cutaneous stimuli applied over various sites in the lower limb
246
produced local sign withdrawal in intact individuals but SCI patients had flexion responses irrespective of stimulus location. Flexion was observed at the hip and ankle regardless of whether stimuli were applied to regions of the foot, ankle, or lower leg. The authors concluded that interneuronal processing of cutaneous stimuli is altered following SCI either directly by the loss of descending input or indirectly via plastic changes in afferent pathways. Therefore, along with the emergence of novel or latent polysynaptic pathways these various changes imply that high threshold or nociceptive reflex pathways are modified following injury to the spinal cord. Mechanisms mediating plastic changes in reflex circuitry Although alterations in reflex pathways after SCI are well-known phenomena, there is a paucity of information with regards to precise mechanisms subserving this plasticity. Changes in reflex pathways after SCI can transpire in numerous ways. The first one implicates changes in motoneuron excitability either by pre- or postsynaptic mechanisms, which have recently been reviewed after SCI (Heckmann et al., 2005) and are briefly summarized here. PICs, which normally generate sustained depolarizations called ‘plateau potentials’, are sharply reduced with the loss of descending monoaminergic inputs during acute SCI (Conway et al., 1988; Hounsgaard et al., 1988; Bennett et al., 2001b), resulting in marked decreases in motoneuronal and reflex excitability. Subsequent increases in reflexes are made possible by the recovery of motoneuron excitability and the ability to generate PICs and plateau potentials (Bennett et al., 1999, 2001a, b; Li and Bennett, 2003), which are suggested to reappear because of a supersensitivity to remnant monoamines. Although PICs with chronic SCI recover to levels similar to those recorded in intact motoneurons (Bennett et al., 1998; Lee and Heckman, 1998; Li and Bennett, 2003) the loss of descending control that normally turn off motoneurons could account for abnormal prolonged motoneuronal firing, as
observed in spasticity (Li et al., 2004a). Moreover, although the focus has been on changes in PICs influencing motoneuronal excitability, little is known about similar phenomena impacting interneurons. It may be the case that PICs throughout the interneuronal relay of a reflex pathway are modified following SCI. Therefore changes in PICs and hence plateau potentials appear to considerably influence reflex behavior after SCI and are thought to largely contribute to the development of spasticity. Altered afferent transmission, which influences motoneuron excitability, can result from changes in PSI and/or HD. As previously mentioned, early after injury PSI and/or HD of the Ia afferent pathway is increased, which may contribute to depressed motoneuronal excitability and hyporeflexia (Ashby et al., 1974; Ashby and Verrier, 1975; Calancie et al., 1993). However, PSI and/or HD gradually diminishes with chronic SCI, which may partly subserve enhanced reflexes (Taylor et al., 1984; Levin and Chapman, 1987; Calancie et al., 1993; Faist et al., 1994). Changes in PSI and synaptic transmission after SCI are certainly not limited to group Ia afferents and are undoubtedly present for other primary sensory afferents and interneurons interposed in these pathways. Furthermore, modifying synaptic input from sensory afferents has been shown to influence plateau potential thresholds (Bennett et al., 1998). Therefore, a combination of pre and postsynaptic changes at the motoneuronal level influences excitability of the motor pool. Plasticity in reflex pathways can also occur through the emergence of new connections either by activating latent pathways released from inhibition or by establishing new synapses through collateral sprouting. Due to the rapid nature of the plasticity, and the abrupt emergence of new reflex pathways after spinalization (Bennett et al., 2004), activation of latent connections is a likely candidate (Wall, 1975). Moreover, collateral sprouting of dorsal horn afferents, spinal interneurons, and extant descending projections has been implicated in reorganizing reflex pathways (Murray and Goldberger, 1974; Helgren and Goldberger, 1993; Raineteau and Schwab, 2001). Indeed, SCI has been shown to trigger sprouting of primary
247
afferents in the spinal cord (Krenz and Weaver, 1998) and studies have attempted to delineate molecules involved in initiating or promoting this phenomenon (Schnell et al., 1994; Reier, 2004; Buchli and Schwab, 2005; Li et al., 2005). However, to what extent these new connections are functional is unclear. Another means of inducing plastic changes in spinal reflexes is by strengthening extant or newly formed connections. Modified synaptic transmission through long-term potentiation or depression-like phenomena has been shown to occur at various sites in the central nervous system, including the spinal cord (Pockett and Figurov, 1993; Randic et al., 1993; Liu and Sandkuhler, 1995; Lozier and Kendig, 1995; Pockett, 1995). Thus, altered synaptic transmission through LTP or LTD could enhance or depress excitability of reflex pathways after SCI and/or training. Furthermore, biochemical properties of the spinal cord change and evolve over time after SCI and with step training (reviewed in Edgerton et al., 2004). For example, it has been established that certain pharmacological agents are more effective at specific stages after a lesion, indicating that the sensitivity of the spinal cord to certain drugs is altered and continues to evolve following SCI (Chau et al., 2002; Giroux et al., 2003). Indeed, there is an upregulation of certain receptor types in the injured spinal cord, which return to control values thereafter (Roudet et al., 1996; Giroux et al., 1999). Moreover, step training influences the biochemical milieu of the spinal cord. For example, studies have shown that strychnine or bicuculline, a glycinergic agonist and GABAA receptor antagonist respectively, which were ineffective in improving stepping in trained spinal cats, substantially ameliorated stepping ability of nontrained spinal cats (Edgerton et al., 1997; de Leon et al., 1999b). These studies indicated that training modulates inhibitory connections within the spinal cord thus facilitating the expression of spinal locomotion. Although considerable more work is required to elucidate the precise mechanisms subserving changes in reflex pathways after SCI it is evident that modifications in spinal reflexes occur at several loci and at different levels, which evolve
temporally and with various interventions, such as locomotor training.
Locomotor recovery after spinal cord lesions The above sections described changes in more or less simple reflex pathways after spinalization. One of the major findings however has been that, after a complete spinal transection, complex motor patterns such as locomotion can recover, involving the reactivation of more elaborate spinal circuits. This second part first summarizes aspects of this functional locomotor recovery after partial or complete spinal sections in cats, then attempts to explain characteristics of spinal locomotion as a function of post spinal excitability changes in simple reflex pathways described in the first section. Recovery of locomotion after partial spinal sections at the last thoracic segment in cats Fig. 1 only indicates that supraspinal structures in the brain stem and telencephalon exert an influence on spinal mechanisms generating locomotion (more details are found in Rossignol et al., 2006). Obviously, supraspinal descending pathways exert powerful effects on spinal mechanisms and are capable of initiating or stopping locomotion and modifying it to reach a target or avoid obstacles. It is of interest to determine if some descending pathways play a sine qua non role in evoking voluntary quadrupedal locomotion. As will be seen, probably none of the descending pathways are uniquely necessary since cats recover voluntary locomotion after various lesions of dorsal or ventral spinal tracts, albeit some deficits may persist in relation to the specific pathways or groups of pathways lesioned. This undoubtedly results from the remarkable compensation from remaining descending pathways. As an example, previous work suggested that medial and medio-lateral pathways are essential for locomotion (Eidelberg, 1981) since sparing a small part of one ventrolateral quadrant allowed recovery of locomotion in chronically lesioned cats (Afelt, 1974; Eidelberg et al., 1981a, b; Contamin, 1983;). However, we and others have indicated
248
that cats (Gorska et al., 1990, 1993a, b; Zmyslowski et al., 1993; Bem et al., 1995; Brustein and Rossignol, 1998, 1999; Rossignol et al., 1999;) and even monkeys (Vilensky et al., 1992) can walk with all four limbs after large lesions of medial and mediolateral pathways that carry vestibular and reticulospinal tracts. Our experiments (Brustein and Rossignol, 1998, 1999) showed that after large ventral and ventrolateral lesions at T13 cats initially behaved as complete spinal cats for 3–6 weeks but all cats eventually walked voluntarily with all limbs. Coupling between the hindlimbs remained stable but cats often walked with a more crouched posture and coordination between the hindlimbs and forelimbs was at times irregular. HRP was injected below the spinal lesion and the number and location of surviving cells with spinal projections were evaluated. In the Pontine Reticular Formation and the Medullary Reticular Formation (MRF), the number of labeled cells amounted to 5–48% of normal values depending on the lesions; there were no vestibulospinal neurons. Rubrospinal cells whose axons course through the dorsolateral funiculi were either normal or decreased in number depending on the amount of encroachment of the lesion on the rubrospinal tract. HRP-labeled cells appeared to be slightly more numerous and distributed somewhat differently in the motor cortex compared to control cats (Rossignol et al., 1999). Therefore, locomotor adaptation may have been achieved by remaining corticospinal pathways or from remaining reticulospinal and rubrospinal cells. Unfortunately, propriospinal neurons were not studied although they could have a role in such compensation (Jordan and Schmidt, 2002). Cats with lesions of dorsal and dorsolateral pathways, destroying among others the corticospinal tracts, can also walk over ground with all four limbs (Gorska et al., 1993b; Zmyslowski et al., 1993; Bem et al., 1995; Jiang and Drew, 1996). After a period of impaired voluntary quadrupedal locomotion for a few days, cats continued to walk with a more crouched position for a few weeks with an increased step cycle duration. Cats had a significant persistent foot drag and could not correctly modify their gait to step over obstacles on a treadmill; instead the dorsum of the foot hit the
obstacle (Drew et al., 1996). Despite these deficits, cats performed remarkably well on the treadmill. Recovery of locomotion after complete spinal section at T13 in cats In numerous animal species hindlimb locomotion on a treadmill can recover after complete spinal cord sections (Sherrington, 1910; Shurrager and Dykman, 1951; Delcomyn, 1980; Grillner, 1981; Rossignol, 1996; Rossignol et al., 1996, 2000). Grillner and colleagues studied quantitatively the kinematics and EMG activity of locomotion after spinalization, especially in kittens (Grillner, 1973; Forssberg et al., 1980a, b) and should be credited for the surging interest in functional capacities of the spinal cord for the last 35 years. Besides illustrating that spinal cats can walk with the hindlimbs, this group established the key concept that spinalized kittens prior to ‘‘learning’’ to walk can express a locomotor pattern, which can be maintained for several months, thus showing that the locomotor circuitry is genetically determined and extant within the spinal cord. Within the context of plasticity, it is important to ask which spinal levels are implicated in the expression of locomotion since these levels might be critical after spinalization. This became apparent with pharmacological experiments using intrathecal cannulae implanted chronically in spinal cats to deliver various drugs (Chau et al., 1998b; Giroux et al., 2001) to evaluate their effects on the initiation or characteristics of locomotion. A postmortem study of the cannulae revealed that drugs acting over restricted portions of the cord (midlumbar L3–L4) had major effects. Further studies showed that intraspinal microinjections at these specific and restricted levels of a noradrenergic agonist such as clonidine and an antagonist, yohimbine, could respectively induce or block spinal locomotion in cats spinalized at T13 1 week before these acute experiments (Marcoux and Rossignol, 2000). Furthermore, we studied the effect of a second more caudal spinal lesion in cats first spinalized at T13 after having recovered locomotion (Langlet and Rossignol, 2002). Whereas a second spinalization at the levels of L2 or rostral L3 did not abolish locomotion, lesions at caudal L3 and L4 prevented locomotion
249
entirely despite several weeks of additional locomotor training. Since these segments are rostral to the motoneurons of the hindlimbs (Vanderhorst and Holstege, 1997), it was concluded that these midlumbar segments were crucial in spinal locomotion. This was in keeping with the idea that important interneurons are located in these pre motoneuronal segments and discharge during fictive locomotion (Edgley et al., 1988; Shefchyk et al., 1990; Jankowska and Edgley, 1993; Davies and Edgley, 1994; Jankowska et al., 2003). Hindlimb locomotion thus recovers after complete spinalization in adult cats (Barbeau and Rossignol, 1987; Be´langer et al., 1996; Rossignol et al., 2000, 2002). After 2–3 weeks of treadmill locomotor training cats can walk with plantar foot contact and support the weight of their hindquarters. Although young animals possess better potential for expressing spinal locomotion (Smith et al., 1982; Bregman and Goldberger, 1983; Bradley and Smith, 1988; Howland et al., 1995a, b), adult cats can also express a full pattern of hindlimb locomotion. We and others (Edgerton et al., 1991, 1997; de Leon et al., 1998a, b; Roy et al., 1998) have clearly established that, when trained several times a week on a treadmill, adult spinal cats gradually recover regular, bilateral, weight-bearing steps. Moreover, we have found that after 2–3 months locomotor performance is remarkably stable from one day to the next when measuring cycle duration at a given fixed treadmill speed, suggesting that even if the animals have regained plantar foot placement and weight support, there are still spinal processes evolving to stabilize the spinal circuitry (Barbeau and Rossignol, 1987).
Reflex pathways and spinal locomotion Effects of post spinal changes in reflex pathways on spinal locomotion In the first section of this review as well as in Rossignol et al. (2006) contributions of reflexes to different aspects of normal locomotion (foot placement, force generation, cycle timing) have been summarized. What is the importance of these sensory inputs to spinal locomotion?
The importance of cutaneous feedback for the full expression of spinal locomotion in cats was recently addressed (Bouyer and Rossignol, 2003a, b). For example, intact and spinalized cats are capable of compensating for the progressive loss of cutaneous nerves in the hindpaw (Bouyer and Rossignol, 2003a, b). Functional deficits, such as inappropriate paw placement, only became apparent once all cutaneous nerves were cut indicating that a modicum of cutaneous information is necessary for normal locomotion. However, if a complete cutaneous denervation was performed prior to spinalization than the spinal animal never recovered proper paw placement or weight bearing despite several weeks of rigorous step training. These studies indicated that some cutaneous information is required for flawless locomotion and that at least some intact cutaneous pathways are critical for expressing spinal locomotion. Muscle proprioceptors also play an important role. Indeed, activation of load pathways appears critical for functional locomotor recovery following spinal transection since they exert powerful actions on rhythm generating networks of several species, including humans (Prochazka, 1996; Harkema et al., 1997; Duysens et al., 2000). Progressive increases in weight bearing improve stepping in chronic spinal cats (Barbeau and Rossignol, 1987; Barbeau et al., 1987) and recent findings indicate that the level of hindlimb loading markedly influences the quantity and quality of locomotion in spinal rats (Timoszyk et al., 2005), indicating a strong contribution of these reflex pathways in the recovery process. This is not surprising since group I afferents reinforce ongoing extensor activity during the stance phase (Conway et al., 1987; Pearson and Collins, 1993; Guertin et al., 1995; Rossignol et al., 2006) and are ideal substrates in assisting force generation. As detailed earlier, significant changes in reflex pathways occur and how these changes may actually be reflected in some aspects of spinal locomotion relative to normal locomotion is of interest. Fig. 2 compares characteristics of locomotion in the same cat before and 41 days after a complete section of the spinal cord at T13. In Fig. 2(A) and (B) are represented, as figurines, the swing and stance phases in intact and spinal states,
250
respectively. After spinalization the overall step length is decreased although the cat walks at the same speed (0.4 m/s) in both cases. Shortening of the step cycle may result from hyperexcitable muscle stretch pathways that provide strong signals to terminate the stance phase. Hyperactivity in muscle stretch pathways might also account for the more ‘‘spastic’’ gait adopted by the spinal cat. Indeed, during stance, the knee and ankle yield vary little at stance onset and remain strongly extended throughout stance in the spinal state. After spinalization the hip joint has a slightly smaller overall excursion but, more remarkably, the hip joint clearly moves in a more extended range. Some of these changes might also be seen in a few muscle discharges (although only a limited sample is shown here). For instance, iVL, a knee extensor, exhibits a gradually increasing activity up to the end of stance and the overall period of activity of coGL, an ankle extensor, is prolonged after spinalization. Another interesting aspect is that following spinalization periods of cocontraction between antagonist muscles become increasingly frequent. Here, clearly the ankle extensor iGL is coactive for a short period with the ankle flexor iTA, which might result from reduced reciprocal inhibition reported in the first part of this review. Obviously some of these considerations are speculative but nevertheless of interest since it could eventually be possible to see how modifying specific reflex pathways (by training, for instance) translates into changes of locomotor characteristics in spinal animals. Normalization of reflex pathways after spinal cord injury with locomotor training As we have seen SCI results in dramatic reorganization of reflex pathways caudal to the injury. Consequently, transmission to and from the spinal cord is markedly altered and may impair functional motor recovery. However, different paradigms, such as operant conditioning (Segal and Wolf, 1994; Chen et al., 1996), step training (Trimble et al., 1998; Cote et al., 2003; Cote and Gossard, 2004), and cycle training (Skinner et al., 1996; Kiser et al., 2005; Reese et al., 2005) have been shown to modify transmission in reflex
pathways after SCI. It may be the case that in order to re-express motor programs, such as locomotion, that modified afferent input to the spinal cord incurred after injury must be shaped or normalized. Locomotor training after spinal transections in chronic adult spinal cats is an established method to improve the rate and quality of stepping recovery. What is unclear, however, are the mechanisms subserving this functional recovery. Several authors have suggested that stepping provides spinal locomotor rhythm generating networks appropriate sensory cues that serve to entrain the locomotor pattern (Lovely et al., 1986; Rossignol, 1996; Harkema, 2001). Although, plastic changes in locomotor rhythm generating networks are undoubtedly occurring recent evidence indicates that transmission in afferent pathways from load and cutaneous receptors are modified by step training after SCI (Cote et al., 2003; Cote and Gossard, 2004). To demonstrate step-trainingdependent changes in afferent transmission after SCI, Cote et al., (2003) compared a group of cats that were spinalized and not trained (shams) with cats that were spinalized and step trained (trained). After a period of 3–4 weeks, an acute experiment was performed to evaluate afferent transmission using intracellular recordings of hindlimb motoneurons in response to peripheral nerve stimulation in shams and trained cats. It was shown that step training reduced the amplitude of monosynaptic excitation and disynaptic Ib inhibition and that these changes were not due to similar modifications in motoneuronal properties since AHP duration was unaltered with 1 month of training (Cote et al., 2003). Decreased monosynaptic reflexes were attributed to increased PSI at Ia terminals and the authors posited that step training potentially reduces spasticity by decreasing Ia afferent transmission. The authors also postulated that a normalization of inhibitory control systems in the spinal cord might be critically important for locomotor recovery. Transmission in these load pathways were also assessed during fictive locomotion since disynaptic Ib inhibition exhibits a reversal to a polysynaptic excitation during locomotion (Gossard and Hultborn, 1991; Gossard et al., 1994; McCrea et al., 1995). It was shown
251
that trained cats possessed greater amplitude polysynaptic excitation than shams, which may be useful for the recovery of weight bearing in spinal locomotion. In a subsequent study (Cote and Gossard, 2004) cutaneous transmission projecting to hindlimb motoneurons was tested in a similar fashion. It was found that training did not change the distribution of the various responses evoked by cutaneous stimulation at rest but that in the majority of cases training decreased mean cutaneous response amplitudes, in particular transmission in the MPN. Therefore, these results demonstrated that step training in chronic adult spinal cats influences transmission in several afferent pathways and that functional locomotor recovery may in part be mediated by a normalization of exaggerated or erroneous reflex information, which occurs after SCI. The authors thus suggested that hyperexcitability of reflexes after spinal transection is counterproductive for the expression of locomotion and that step training provides a ‘normalization’ of sensory feedback. Clonidine, which exerts powerful actions on locomotor circuits in spinalized cats (Barbeau and Rossignol, 1987), similar to step training depresses transmission in cutaneous (Barbeau et al., 1987; Chau et al., 1998a, b) and group I pathways (Cote et al., 2003). Therefore, as previously proposed pharmacological agents and/ or step training may aid locomotor recovery by normalizing activity in reflex pathways after SCI (Naftchi, 1982; Robinson and Goldberger, 1986; de Leon et al., 1999a; Cote et al., 2003; Cote and Gossard, 2004). Using a different protocol it was shown that training spinally transected adult rats modified Hreflex excitability (Skinner et al., 1996). In this study, adult rats were completely spinalized at T10 and 5 days after surgery a 3-month training regimen was initiated. Training consisted of passive cycling with rats suspended and hindfeet strapped onto pedals. H-reflexes in plantar muscles and their sensitivity to high frequency stimulation, evoked by stimulating the tibial nerve were compared in three groups at the conclusion of training. In the control group (no transection), high frequency stimulation resulted in marked depression of H-reflexes, which was considerably reduced in
the transected group, corroborating an earlier report (Thompson et al., 1992). This frequencydependent depression has previously been attributed to PSI or HD at Ia afferent terminals. However, in the trained spinal rats this frequencydependent depression returned to normal values, suggesting that training restores or reorganizes presynaptic inhibitory mechanisms. In a subsequent study, the same group demonstrated that rate-sensitive depression could be increased as early as 15 days after initiating training but that full restoration emerged gradually over time (Reese et al., 2005). Similar to animal models, locomotor training can significantly influence gating in transmission of the H-reflex pathway after SCI in humans (Trimble et al., 1998; Kiser et al., 2005). Using a similar paradigm to Skinner et al. (1996) a motorized bicycle was used to restore normal ratesensitive depression of the H-reflex and thus improve spasticity in a C7 SCI patient (Kiser et al., 2005). Training consisted of passive cycling, five times a week for 13 weeks. Cycle training restored normal rate-sensitive depression of soleus Hreflexes, which paralleled improvements in spasticity. However, 4 weeks after training cessation rate-sensitive depression began to decrease and spastic signs returned indicating that cycle training must be maintained to retain ameliorations. In another case study of an incomplete SCI man, rate-sensitive depression of soleus H-reflexes was assessed before and after 4 months of treadmill training (Trimble et al., 1998). Soleus H-reflex recruitment curves were elicited at 0.1 and 1 Hz to compare the effects of low-frequency depression in the SCI man to intact individuals. Low-frequency depression (1 Hz) of H-reflexes in the SCI patient were considerably less before training compared to intact individuals but significantly increased after training, which paralleled ameliorations in locomotor performance. Although these are only case studies, it does suggest that training, similar to animal models, normalizes afferent input to the spinal cord, which may benefit functional motor recovery. Considerable more research using larger sample sizes is required to elucidate the neurophysiological mechanisms subserving functional motor recovery in humans with SCI.
252
Therefore, although training benefits may stem from other plastic changes in spinal circuitry, locomotor improvements appear to be generated, in part, by normalizing erratic or disorganized sensory feedback incurred after SCI. Concluding remarks The spinal cord after a lesion provides the opportunity to study changes in relatively simple sensorimotor systems. And yet, these changes are complex and involve several mechanisms, which impact the behavior of these ‘‘simple circuits’’. We have attempted to summarize how changes in reflex pathways may account for various symptoms, such as spasticity, incurred after spinal injury. Furthermore, we have tried to relate these changes in reflex excitability to behavioral characteristics in more complex networks underlying locomotion, which utilize in part elements of these simple reflex circuits. This leads to the notion that perhaps rehabilitative techniques to reinstate function after spinal cord injury may rely on or be manifested in the normalization of these reflex pathways. Having a clearer view of the relation between characteristics of simple reflexes and more complex behaviors such as locomotion may generate more focused rehabilitative approaches as well as reflex testing as a tool to assess the progression of therapies or the prognosis of spinal injury. What is clear however is that the spinal cord, like the brain is reprogrammed after spinal cord injury and optimizes remnant functions. This demonstrated spinal plasticity proffers an even clearer basis for influencing and directing this neuroplasticity present after spinal injury through rehabilitative approaches. Acknowledgments The authors thank Janyne Provencher and Daniel Cyr for the iconography. A. Frigon has a scholarship from the Natural Sciences and Engineering Research Council of Canada. S. Rossignol is funded through a Tier 1 Canada Research Chair on the Spinal Cord by the Canadian Institute for Health Research of Canada (CIHR) as well as through a CIHR Group Grant and an individual
Grant and through a CIHR Regenerative Medicine and Nanomedicine Team grant (Multidisciplinary approach to promote and evaluate locomotion after spinal cord lesions and stroke).
References Afelt, Z. (1970) Reflex activity in chronic spinal cats. Acta Neurobiol. Exp. (Wars), 30: 129–144. Afelt, Z. (1974) Functional significance of ventral descending tracts of the spinal cord in the cat. Acta Neurobiol. Exp., 34: 393–407. Anden, N.E., Jukes, M.G.M. and Lundberg, A. (1966a) The effect of DOPA on the spinal cord. 2. A pharmacological analysis. Acta Physiol. Scand., 67: 387–397. Anden, N.E., Jukes, M.G.M., Lundberg, A. and Vyklicky, L. (1966b) The effect of DOPA on the spinal cord. 3. Depolarization evoked in the central terminals of ipsilateral Ia afferents by volleys in the flexor reflex afferents. Acta. Physiol. Scand., 68: 322–336. Andersen, O.K., Finnerup, N.B., Spaich, E.G., Jensen, T.S. and Arendt-Nielsen, L. (2004) Expansion of nociceptive withdrawal reflex receptive fields in spinal cord injured humans. Clin. Neurophysiol., 115: 2798–2810. Andersen, O.K., Sonnenborg, F.A. and Arendt-Nielsen, L. (1999) Modular organization of human leg withdrawal reflexes elicited by electrical stimulation of the foot sole. Muscle Nerve, 22: 1520–1530. Aoki, M., Mori, S. and Fujimori, B. (1976) Exaggeration of knee-jerk following spinal hemisection in monkeys. Brain Res., 107: 471–485. Ashby, P. and Verrier, M. (1975) Neurophysiological changes following spinal cord lesions in man. Can. J. Neurol. Sci., 2: 91–100. Ashby, P., Verrier, M. and Lightfoot, E. (1974) Segmental reflex pathways in spinal shock and spinal spasticity in man. J. Neurol. Neurosurg. Psychiatry, 37: 1352–1360. Bailey, C.S., Lieberman, J.S. and Kitchell, R.L. (1980) Response of muscle spindle primary endings to static stretch in acute and chronic spinal cats. Am. J. Vet. Res., 41: 2030–2036. Baker, L.L. and Chandler, S.H. (1987a) Characterization of hindlimb motoneuron membrane properties in acute and chronic spinal cats. Brain Res., 420: 333–339. Baker, L.L. and Chandler, S.H. (1987b) Characterization of postsynaptic potentials evoked by sural nerve stimulation in hindlimb motoneurons from acute and chronic spinal cats. Brain Res., 420: 340–350. Baldissera, F., Hultborn, H. and Illert, M. (1981) Integration in spinal neuronal systems. In: Brookhart, J.M. and Mountcastle, V.B. (Eds.), Handbook of Physiology. Amer. Physiol. Soc., Maryland, pp. 509–595. Barbeau, H., Julien, C. and Rossignol, S. (1987) The effects of clonidine and yohimbine on locomotion and cutaneous reflexes in the adult chronic spinal cat. Brain Res., 437: 83–96.
253 Barbeau, H., McCrea, D.A., O’Donovan, M.J., Rossignol, S., Grill, W.M. and Lemay, M.A. (1999) Tapping into spinal circuits to restore motor function. Brain Rese. Rev., 30: 27–51. Barbeau, H. and Rossignol, S. (1987) Recovery of locomotion after chronic spinalization in the adult cat. Brain Res., 412: 84–95. Barnes, C.D., Joynt, R.J. and Schottelius, B.A. (1962) Motoneuron resting potentials in spinal shock. Am. J. Physiol., 203: 1113–1116. Be´langer, M., Drew, T., Provencher, J. and Rossignol, S. (1996) A comparison of treadmill locomotion in adult cats before and after spinal transection. J. Neurophysiol., 76: 471–491. Bem, T., Gorska, T., Majczynski, H. and Zmyslowski, W. (1995) Different patterns of fore–hindlimb coordination during overground locomotion in cats with ventral and lateral spinal lesions. Exp. Brain Res., 104: 70–80. Bennett, D.J., Gorassini, M., Fouad, K., Sanelli, L., Han, Y. and Cheng, J. (1999) Spasticity in rats with sacral spinal cord injury. J. Neurotrauma, 16: 69–84. Bennett, D.J., Hultborn, H., Fedirchuk, B. and Gorassini, M. (1998) Synaptic activation of plateaus in hindlimb motoneurons of decerebrate cats. J. Neurophysiol., 80: 2023–2037. Bennett, D.J., Li, Y., Harvey, P.J. and Gorassini, M. (2001a) Evidence for plateau potentials in tail motoneurons of awake chronic spinal rats with spasticity. J. Neurophysiol., 86: 1972–1982. Bennett, D.J., Li, Y. and Siu, M. (2001b) Plateau potentials in sacrocaudal motoneurons of chronic spinal rats, recorded in vitro. J. Neurophysiol., 86: 1955–1971. Bennett, D.J., Sanelli, L., Cooke, C.L., Harvey, P.J. and Gorassini, M.A. (2004) Spastic long-lasting reflexes in the awake rat after sacral spinal cord injury. J. Neurophysiol., 91: 2247–2258. Boorman, G.I., Lee, R.G., Becker, W.J. and Windhorst, U.R. (1996) Impaired ‘‘natural reciprocal inhibition’’ in patients with spasticity due to incomplete spinal cord injury. Electroencephalogr. Clin. Neurophysiol., 101: 84–92. Bouyer, L.J.G. and Rossignol, S. (2003a) Contribution of cutaneous inputs from the hindpaw to the control of locomotion: 1. Intact cats. J. Neurophysiol., 90: 3625–3639. Bouyer, L.J.G. and Rossignol, S. (2003b) Contribution of cutaneous inputs from the hindpaw to the control of locomotion: 2. Spinal cats. J. Neurophysiol., 90: 3640–3653. Bradley, N.S. and Smith, J.L. (1988) Neuromuscular patterns of stereotypic hindlimb behaviors in the first postnatal months. II. Stepping in spinal kittens. Develop Brain Res., 38: 53–67. Bras, H., Jankowska, E., Noga, B. and Skoog, B. (1990) Comparison of effects of various types of NA and 5-HT agonists on transmission from group II muscle afferents in the cat. Eur. J. Neurosci., 211: 1029–1039. Bregman, B.S. and Goldberger, M.E. (1983) Infant lesion effect: I. Development of motor behavior following neonatal spinal cord damage in cats. Develop Brain Res., 9: 103–117. Brown, P. (1994) Pathophysiology of spasticity. J. Neurol. Neurosurg. Psychiatry, 57: 773–777. Brustein, E. and Rossignol, S. (1998) Recovery of locomotion after ventral and ventrolateral spinal lesions in the cat. I. Deficits and adaptive mechanisms. J. Neurophysiol., 80: 1245–1267.
Brustein, E. and Rossignol, S. (1999) Recovery of locomotion after ventral and ventrolateral spinal lesions in the cat. II. Effects of noradrenergic and serotoninergic drugs. J. Neurophysiol., 81: 1513–1530. Buchli, A.D. and Schwab, M.E. (2005) Inhibition of Nogo: a key strategy to increase regeneration, plasticity and functional recovery of the lesioned central nervous system. Ann. Med., 37: 556–567. Cadilhac, J., Georgesco, M., Benezech, J., Duday, H. and Dapres, G. (1977) Somatosensory evoked potentials and Hoffmann reflex in acute spinal cord lesions; physiopathological and prognostic aspects. Electroencephalogr. Clin. Neurophysiol., 43: 160–167. Calancie, B. (1991) Interlimb reflexes following cervical spinal cord injury in man. Exp. Brain Res., 85: 458–469. Calancie, B., Alexeeva, N., Broton, J.G. and Molano, M.R. (2005) Interlimb reflex activity after spinal cord injury in man: strengthening response patterns are consistent with ongoing synaptic plasticity. Clin. Neurophysiol., 116: 75–86. Calancie, B., Broton, J.G., Klose, K.J., Traad, M., Difini, J. and Ayyar, D.R. (1993) Evidence that alterations in presynaptic inhibition contribute to segmental hypo- and hyperexcitability after spinal cord injury in man. Electroencephalogr. Clin. Neurophysiol., 89: 177–186. Calancie, B., Lutton, S. and Broton, J.G. (1996) Central nervous system plasticity after spinal cord injury in man: interlimb reflexes and the influence of cutaneous stimulation. Electroencephalogr. Clin. Neurophysiol., 101: 304–315. Calancie, B., Molano, M.R. and Broton, J.G. (2002) Interlimb reflexes and synaptic plasticity become evident months after human spinal cord injury. Brain, 125: 1150–1161. Calancie, B., Molano, M.R. and Broton, J.G. (2004) Tendon reflexes for predicting movement recovery after acute spinal cord injury in humans. Clin. Neurophysiol., 115: 2350–2363. Capaday, C. and Stein, R.B. (1986) Amplitude modulation of the soleus H-reflex in the human during walking and standing. J. Neurosci., 6: 1308–1313. Carter, R.L., Ritz, L.A., Shank, C.P., Scott, E.W. and Sypert, G.W. (1991) Correlative electrophysiological and behavioral evaluation following L5 lesions in the cat: a model of spasticity. Exp. Neurol., 114: 206–215. Chau, C., Barbeau, H. and Rossignol, S. (1998a) Early locomotor training with clonidine in spinal cats. J. Neurophysiol., 79: 392–409. Chau, C., Barbeau, H. and Rossignol, S. (1998b) Effects of intrathecal a1- and a2-noradrenergic agonists and norepinephrine on locomotion in chronic spinal cats. J. Neurophysiol., 79: 2941–2963. Chau, C., Giroux, N., Barbeau, H., Jordan, L.M. and Rossignol, S. (2002) Effects of intrathecal glutamatergic drugs on locomotion. I. NMDA in short-term spinal cats. J. Neurophysiol., 88: 3032–3045. Chen, X.Y., Feng-Chen, K.C., Chen, L., Stark, D.M. and Wolpaw, J.R. (2001) Short-term and medium-term effects of spinal cord tract transections on soleus H-reflex in freely moving rats. J. Neurotrauma, 18: 313–327.
254 Chen, X.Y., Wolpaw, J.R., Jakeman, L.B. and Stokes, B.T. (1996) Operant conditioning of H-reflex in spinal cord-injured rats. J. Neurotrauma, 13: 755–766. Contamin, F. (1983) Sections me´dullaires incomple`tes et locomotion chez le chat. Bull. Acad. Nat. Med., 167: 727–730. Conway, B.A., Hultborn, H. and Kiehn, O. (1987) Proprioceptive input resets central locomotor rhythm in the spinal cat. Exp. Brain Res., 68: 643–656. Conway, B.A., Hultborn, H., Kiehn, O. and Mintz, I. (1988) Plateau potentials in alpha-motoneurones induced by intravenous injection of L-Dopa and clonidine in the spinal cat. J. Physiol., 405: 369–384. Cook, A.J. and Woolf, C.J. (1985) Cutaneous receptive field and morphological properties of hamstring flexor alphamotoneurones in the rat. J. Physiol., 364: 249–263. Cote, M.P. and Gossard, J.P. (2004) Step training-dependent plasticity in spinal cutaneous pathways. J. Neurosci., 24: 11317–11327. Cote, M.P., Menard, A. and Gossard, J.P. (2003) Spinal cats on the treadmill: changes in load pathways. J. Neurosci., 23: 2789–2796. Crenna, P. and Frigo, C. (1987) Excitability of the soleus Hreflex arc during walking and stepping in man. Exp. Brain Res., 66: 49–60. Crenna, P., Schieppati, M. and de Curtis, M. (1982) Longlatency, nonreciprocal reflex responses of antagonistic hind limb muscles after cutaneous nerve stimulation in the cat. Exp. Neurol., 76: 58–71. Crone, C., Hultborn, H., Jespersen, B. and Nielsen, J. (1987) Reciprocal Ia inhibition between ankle flexors and extensors in man. J. Physiol., 389: 163–185. Crone, C., Johnsen, L.L., Biering-Sorensen, F. and Nielsen, J.B. (2003) Appearance of reciprocal facilitation of ankle extensors from ankle flexors in patients with stroke or spinal cord injury. Brain, 126: 495–507. Crone, C. and Nielsen, J. (1989) Spinal mechanisms in man contributing to reciprocal inhibition during voluntary dorsiflexion of the foot. J. Physiol., 416: 255–272. Crone, C., Nielsen, J., Petersen, N., Ballegaard, M. and Hultborn, H. (1994) Disynaptic reciprocal inhibition of ankle extensors in spastic patients. Brain, 117(Pt 5): 1161–1168. Davies, H.E. and Edgley, S.A. (1994) Inputs to group II activated midlumbar interneurones from descending motor pathways in the cat. J. Physiol., 479: 463–473. de Leon, R.D., Hodgson, J.A., Roy, R.R. and Edgerton, V.R. (1998a) Full weight-bearing hindlimb standing following stand training in the adult spinal cat. J. Neurophysiol., 80: 83–91. de Leon, R.D., Hodgson, J.A., Roy, R.R. and Edgerton, V.R. (1998b) Locomotor capacity attributable to step training versus spontaneous recovery after spinalization in adult cats. J. Neurophysiol., 79: 1329–1340. de Leon, R.D., Hodgson, J.A., Roy, R.R. and Edgerton, V.R. (1999a) Retention of hindlimb stepping ability in adult spinal cats after the cessation of step training. J. Neurophysiol., 81: 85–94. de Leon, R.D., Tamaki, H., Hodgson, J.A., Roy, R.R. and Edgerton, V.R. (1999b) Hindlimb locomotor and postural training modulates glycinergic inhibition in the spinal cord of the adult spinal cat. J. Neurophysiol., 82: 359–369.
Decima, E.E. and Morales, F.R. (1983) Long-lasting facilitation of a monosynaptic pathway as a result of ‘‘partial’’ axotomy of its presynaptic elements. Exp. Neurol., 79: 532–551. Delcomyn, F. (1980) Neural basis of rhythmic behavior in animals. Science, 210: 492–498. Delwaide, P.J., Crenna, P. and Fleron, M.H. (1981) Cutaneous nerve stimulation and motoneuronal excitability: I, soleus and tibialis anterior excitability after ipsilateral and contralateral sural nerve stimulation. J. Neurol. Neurosurg. Psychiatry, 44: 699–707. Delwaide, P.J., Pepin, J.L. and Maertens, d.N. (1991) Shortlatency autogenic inhibition in patients with Parkinsonian rigidity. Ann. Neurol., 30: 83–89. Diamantopoulos, E. and Zander, O.P. (1967) Excitability of motor neurones in spinal shock in man. J. Neurol. Neurosurg. Psychiatry, 30: 427–431. Dimitrijevic, M.R. and Nathan, P.W. (1967a) Studies of spasticity in man. 2. Analysis of stretch reflexes in spasticity. Brain, 90: 333–358. Dimitrijevic, M.R. and Nathan, P.W. (1967b) Studies of spasticity in man. I. Some features of spasticity. Brain, 90: 1–30. Dimitrijevic, M.R. and Nathan, P.W. (1968) Studies of spasticity in man. 3. Analysis of reflex activity evoked by noxious cutaneous stimulation. Brain, 91: 349–368. Dimitrijevic, M.R. and Nathan, P.W. (1970) Studies of spasticity in man. 4. Changes in flexion reflex with repetitive cutaneous stimulation in spinal man. Brain, 93: 743–768. Dimitrijevic, M.R. and Nathan, P.W. (1971) Studies of spasticity in man. 5. Dishabituation of the flexion reflex in spinal man. Brain, 94: 77–90. Dimitrijevic, M.R. and Nathan, P.W. (1973) Studies of spasticity in man. Habituation, dishabituation and sensitization of tenson reflexes in spinal man. Brain, 96: 337–354. Ditunno, J.F., Little, J.W., Tessler, A. and Burns, A.S. (2004) Spinal shock revisited: a four-phase model. Spinal Cord, 42: 383–395. Downes, L., Ashby, P. and Bugaresti, J. (1995) Reflex effects from Golgi tendon organ (Ib) afferents are unchanged after spinal cord lesions in humans. Neurology, 45: 1720–1724. Drew, T., Jiang, W., Kably, B. and Lavoie, S. (1996) Role of the motor cortex in the control of visually triggered gait modifications. Can. J. Physiol. Pharmacol., 74: 426–442. Duysens, J., Clarac, F. and Cruse, H. (2000) Load-regulating mechanisms in gait and posture: comparative aspects. Physiol. Rev., 80: 83–133. Edgerton, V.R., de Guzman, C.P., Gregor, R.J., Roy, R.R., Hodgson, J.A. and Lovely, R.G. (1991) Trainability of the spinal cord to generate hindlimb stepping patterns in adult spinalized cats. In: Shimamura, M., Grillner, S. and Edgerton, V.R. (Eds.), Neurobiological Basis of Human Locomotion. Japan Scientific Societies Press, Tokyo, pp. 411–423. Edgerton, V.R., de Leon, R.D., Tillakaratne, N., Recktenwald, M.R., Hodgson, J.A. and Roy, R.R. (1997) Use-dependent plasticity in spinal stepping and standing. Adv. Neurol., 72: 233–247.
255 Edgerton, V.R., Tillakaratne, N.J., Bigbee, A.J., de Leon, R.D. and Roy, R.R. (2004) Plasticity of the spinal neural circuitry after injury. Annu. Rev. Neurosci., 27: 145–167. Edgley, S.A., Jankowska, E. and Shefchyk, S. (1988) Evidence that mid-lumbar neurones in reflex pathways from group II afferents are involved in locomotion in the cat. J. Physiol., 403: 57–71. Eidelberg, E. (1981) Consequences of spinal cord lesions upon motor function, with special reference to locomotor activity. Prog. Neurobiol., 17: 185–202. Eidelberg, E., Story, J.L., Walden, J.G. and Meyer, B.L. (1981a) Anatomical correlates of return of locomotor function after partial spinal cord lesions in cats. Exp. Brain Res., 42: 81–88. Eidelberg, E., Walden, J.G. and Nguyen, L.H. (1981b) Locomotor control in macaque monkeys. Brain, 104: 647–663. Engberg, I., Lundberg, A. and Ryall, R.W. (1968) Reticulospinal inhibition of interneurones. J. Physiol., 194: 225–236. Eriksson, J., Olausson, B. and Jankowska, E. (1996) Antispastic effects of L-dopa. Exp. Brain Res., 111: 296–304. Faist, M., Mazevet, D., Dietz, V. and Pierrot-Deseilligny, E. (1994) A quantitative assessment of presynaptic inhibition of Ia afferents in spastics. Differences in hemiplegics and paraplegics. Brain, 117(Pt 6): 1449–1455. Fedina, L. and Hultborn, H. (1972) Facilitation from ipsilateral primary afferents of interneuronal transmission in the Ia inhibitory pathway to motoneurones. Acta Physiol. Scand., 86: 59–81. Forssberg, H. (1979) Stumbling corrective reaction: a phasedependent compensatory reaction during locomotion. J. Neurophysiol., 42: 936–953. Forssberg, H., Grillner, S. and Halbertsma, J. (1980a) The locomotion of the low spinal cat. I. Coordination within a hindlimb. Acta Physiol. Scand., 108: 269–281. Forssberg, H., Grillner, S., Halbertsma, J. and Rossignol, S. (1980b) The locomotion of the low spinal cat: II. Interlimb coordination. Acta Physiol Scand., 108: 283–295. Forssberg, H., Grillner, S. and Rossignol, S. (1975) Phase dependent reflex reversal during walking in chronic spinal cats. Brain Res., 85: 103–107. Fung, J. and Barbeau, H. (1994) Effects of conditioning cutaneomuscular stimulation on the soleus H-reflex in normal and spastic paretic subjects during walking and standing. J. Neurophysiol., 72: 2090–2104. Fung, S.J., Pompeiano, O. and Barnes, C.D. (1987) Suppression of the recurrent inhibitory pathway in lumbar cord segments during locus coeruleus stimulation in cats. Brain Res., 402: 351–354. Gassel, M.M. and Ott, K.H. (1973) Patterns of reflex excitability change after widespread cutaneous stimulation in man. J. Neurol. Neurosurg. Psychiatry, 36: 282–287. Giroux, N., Chau, C., Barbeau, H., Reader, T.A. and Rossignol, S. (2003) Effects of intrathecal glutamatergic drugs on locomotion. II. NMDA and AP-5 in intact and late spinal cats. J. Neurophysiol., 90: 1027–1045. Giroux, N., Reader, T.A. and Rossignol, S. (2001) Comparison of the effect of intrathecal administration of clonidine and yohimbine on the locomotion of intact and spinal cats. J. Neurophysiol., 85: 2516–2536.
Giroux, N., Rossignol, S. and Reader, T.A. (1999) Autoradiographic study of a1-, a2-Noradrenergic and Serotonin 1A receptors in the spinal cord of normal and chronically transected cats. J. Comp. Neurol., 406: 402–414. Gorska, T., Bem, T. and Majczynski, H. (1990) Locomotion in cats with ventral spinal lesions: support patterns and duration of support phases during unrestrained walking. Acta Neurobiol. Exp., 50: 191–200. Gorska, T., Bem, T., Majczynski, H. and Zmyslowski, W. (1993a) Unrestrained walking in cats with partial spinal lesions. Brain Res. Bull., 32: 241–249. Gorska, T., Majczynski, H., Bem, T. and Zmyslowski, W. (1993b) Hindlimb swing, stance and step relationships during unrestrained walking in cats with lateral funicular lesion. Acta Neurobiol. Exp., 53: 133–142. Gossard, J.-P., Brownstone, R.M., Barajon, I. and Hultborn, H. (1994) Transmission in a locomotor-related group Ib pathway from hindlimb extensor muscles in the cat. Exp. Brain Res., 98: 213–228. Gossard, J.-P. and Hultborn, H. (1991) The organization of the spinal rhythm generation in locomotion. In: Wernig, A. (Ed.), Restorative Neurology. Elsevier, Amsterdam, pp. 385–404. Grillner, S. (1973) Locomotion in the spinal cat. In: Stein, R.B., Pearson, K.G., Smith, R.S. and Redford, J.B. (Eds.), Control of posture and locomotion. Adv. Behav. Biol. 7. Plenum Press, New York, pp. 515–535. Grillner, S. (1981) Control of locomotion in bipeds, tetrapods, and fish. In: Brookhart, J.M. and Mountcastle, V.B. (Eds.), Handbook of Physiology. The Nervous System II. Amer. Physiol. Soc., Bethesda, Maryland, pp. 1179–1236. Grillner, S. and Zangger, P. (1979) On the central generation of locomotion in the low spinal cat. Exp. Brain Res., 34: 241–261. Guertin, P., Angel, M.J., Perreault, M.-C. and McCrea, D.A. (1995) Ankle extensor group I afferents excite extensors throughout the hindlimb during fictive locomotion in the cat. J. Physiol., 487: 197–209. Guertin, P.A. and Hounsgaard, J. (1999) Non-volatile general anaesthetics reduce spinal activity by suppressing plateau potentials. Neuroscience, 88: 353–358. Harkema, S.J. (2001) Neural plasticity after human spinal cord injury: application of locomotor training to the rehabilitation of walking. Neuroscientist, 7: 455–468. Harkema, S.J., Hurlay, S.L., Patel, U.K., Requejo, P.S., Dobkin, B.H. and Edgerton, V.R. (1997) Human lumbosacral spinal cord interpret loading during stepping. J. Neurophysiol., 77: 797–811. Heckmann, C.J., Gorassini, M.A. and Bennett, D.J. (2005) Persistent inward currents in motoneuron dendrites: implications for motor output. Muscle Nerve, 31: 135–156. Helgren, M.E. and Goldberger, M.E. (1993) The recovery of postural reflexes and locomotion following low thoracic hemisection in adult cats involves compensation by undamaged primary afferent pathways. Exp. Neurol., 123: 17–34. Hiersemenzel, L.P., Curt, A. and Dietz, V. (2000) From spinal shock to spasticity: neuronal adaptations to a spinal cord injury. Neurology, 54: 1574–1582.
256 Hochman, S. and McCrea, D.A. (1994a) Effects of chronic spinalization on ankle extensor motoneurons. I.Composite monosynaptic Ia EPSPs in four motoneuron pools. J. Neurophysiol., 71: 1452–1467. Hochman, S. and McCrea, D.A. (1994b) Effects of chronic spinalization on ankle extensor motoneurons.II.Motoneuron electrical properties. J. Neurophysiol., 71: 1468–1479. Hornby, T.G., Rymer, W.Z., Benz, E.N. and Schmit, B.D. (2003) Windup of flexion reflexes in chronic human spinal cord injury: a marker for neuronal plateau potentials? J. Neurophysiol., 89: 416–426. Hounsgaard, J., Hultborn, H., Jespersen, J. and Kiehn, O. (1988) Bistability of alpha-motoneurones in the decerebrate cat and in the acute spinal cat after intravenous 5-hydroxytryptophan. J. Physiol., 405: 345–367. Howland, D.R., Bregman, B.S. and Goldberger, M.E. (1995a) The development of quadrupedal locomotion in the kitten. Exp. Neurol., 135: 93–107. Howland, D.R., Bregman, B.S., Tessler, A. and Goldberger, M.E. (1995b) Development of locomotor behavior in the spinal kitten. Exp. Neurol., 135: 108–122. Hugon, M. and Delwaide, P.J. (1969) Modifications de l’excitabilite de l’arc myotatique et variations des courbes de recrutement du reflexe de Hoffmann. Arch. Int. Physiol. Bioch., 77: 125–127. Hultborn, H. (2003) Changes in neuronal properties and spinal reflexes during development of spasticity following spinal cord lesions and stroke: studies in animal models and patients. J. Rehabil. Med., Suppl. 41: 46–55. Hultborn, H., Illert, M., Nielsen, J., Paul, A., Ballegaard, M. and Wiese, H. (1996) On the mechanism of the post-activation depression of the H-reflex in human subjects. Exp Brain Res., 108: 450–462. Hultborn, H., Jankowska, E. and Lindstrom, S. (1971) Relative contribution from different nerves to recurrent depression of Ia IPSPs in motoneurones. J. Physiol., 215: 637–664. Hultborn, H. and Malmsten, J. (1983a) Changes in segmental reflexes following chronic spinal cord hemisection in the cat. I. Increased monosynaptic and polysynaptic ventral root discharges. Acta Physiol. Scand., 119: 405–422. Hultborn, H. and Malmsten, J. (1983b) Changes in segmental reflexes following chronic spinal cord hemisection. II. Conditioned monosynaptic test reflexes. in the cat. Acta Physiol. Scand., 119: 423–433. Hultborn, H., Meunier, S., Morin, C. and Pierrot-Deseilligny, E. (1987) Assessing changes in presynaptic inhibition of I a fibres: a study in man and the cat. J. Physiol., 389: 729–756. Ishikawa, K., Ott, K., Porter, R.W. and Stuart, D. (1966) Low frequency depression of the H wave in normal and spinal man. Exp. Neurol., 15: 140–156. Jami, L. (1992) Golgi tendon organs in mammalian skeletal muscle: functional properties and central actions. Physiol. Rev., 72: 623–666. Jankowska, E. (1992) Interneuronal relay in spinal pathways from proprioceptors. Prog. Neurobiol., 38: 335–378.
Jankowska, E. and Edgley, S. (1993) Interaction between pathways controlling posture and gait at the level of spinal interneurones in the cat. Prog. Brain Res., 97: 161–171. Jankowska, E. and Hammar, I. (2002) Spinal interneurones; how can studies in animals contribute to the understanding of spinal interneuronal systems in man? Brain Res. Brain Res. Rev., 40: 19–28. Jankowska, E., Hammar, I., Chojnicka, B. and Heden, C.H. (2000) Effects of monoamines on interneurons in four spinal reflex pathways from group I and/or group II muscle afferents. Eur. J. Neurosci., 12: 701–714. Jankowska, E., Hammar, I., Slawinska, U., Maleszak, K. and Edgley, S.A. (2003) Neuronal basis of crossed actions from the reticular formation on feline hindlimb motoneurons. J. Neurosci., 23: 1867–1878. Jankowska, E., Jukes, M.G.M., Lund, S. and Lundberg, A. (1967a) The effect of DOPA on the spinal cord. 5. Reciprocal organization of pathways transmitting excitatory action to alpha motoneurones of flexors and extensors. Acta Physiol. Scand., 70: 369–388. Jankowska, E., Jukes, M.G.M., Lund, S. and Lundberg, A. (1967b) The effects of DOPA on the spinal cord. 6. Half centre organization of interneurones transmitting effects from the flexor reflex afferents. Acta Physiol. Scand., 70: 389–402. Jiang, W. and Drew, T. (1996) Effects of bilateral lesions of the dorsolateral funiculi and dorsal columns at the level of the low thoracic spinal cord on the control of locomotion in the adult cat: I. Treadmill walking. J. Neurophysiol., 76: 849–866. Jordan, L.M. and Schmidt, B.J. (2002) Propriospinal neurons involved in the control of locomotion: potential targets for repair strategies. In: McKerracher, L., Doucet, G. and Rossignol, S. (Eds.), Spinal Cord Trauma: Neural Repair and Functional Recovery. Elsevier, New York, pp. 125–139. Katz, R. and Pierrot-Deseilligny, E. (1998) Recurrent inhibition in humans. Prog. Neurobiol., 57: 325–355. Kearney, R.E. and Chan, C.W.Y. (1979) Reflex response of human arm muscles to cutaneous stimulation of the foot. Brain Res., 170: 214–217. Kearney, R.E. and Chan, C.W. (1981) Interlimb reflexes evoked in human arm muscles by ankle displacement. Electroencephalogr. Clin. Neurophysiol., 52: 65–71. Kiser, T.S., Reese, N.B., Maresh, T., Hearn, S., Yates, C., Skinner, R.D., Pait, T.G. and Garcia-Rill, E. (2005) Use of a motorized bicycle exercise trainer to normalize frequency-dependent habituation of the H-reflex in spinal cord injury. J. Spinal Cord Med., 28: 241–245. Knikou, M. and Conway, B.A. (2001) Modulation of soleus Hreflex following ipsilateral mechanical loading of the sole of the foot in normal and complete spinal cord injured humans. Neurosci. Lett., 303: 107–110. Knikou, M. and Conway, B.A. (2005) Effects of electrically induced muscle contraction on flexion reflex in human spinal cord injury. Spinal Cord, 43: 640–648. Knikou, M., Kay, E. and Rymer, W.Z. (2006) Modulation of flexion reflex induced by hip angle changes in human spinal cord injury. Exp. Brain Res., 168: 577–586.
257 Kohn, A.F., Floeter, M.K. and Hallett, M. (1997) Presynaptic inhibition compared with homosynaptic depression as an explanation for soleus H-reflex depression in humans. Exp. Brain Res., 116: 375–380. Krenz, N.R. and Weaver, L.C. (1998) Sprouting of primary afferent fibers after spinal cord transection in the rat. Neuroscience, 85: 443–458. Kuhn, R.A. (1950) Functional capacity of the isolated human spinal cord. Brain, LXXIII: 1–51. Langlet, C. and Rossignol, S. (2002) The rostral lumbar spinal segments play a crucial role in the expression of spinal locomotion. Soc. Neurosci. Abstracts 28[67.19]. Ref Type: Abstract Lebizec, J., Demaire, C. and Honore, J. (1983) Spinal increase of the Hoffmann reflex by stimulation of the sural nerve at a low intensity: study of the paraplegic. Electroencephalogr. Clin. Neurophysiol., 55: 162–164. Lee, J.K., Emch, G.S., Johnson, C.S. and Wrathall, J.R. (2005) Effect of spinal cord injury severity on alterations of the Hreflex. Exp. Neurol., 196: 430–440. Lee, R.H. and Heckman, C.J. (1998) Bistability in spinal motoneurons in vivo: systematic variations in rhythmic firing patterns. J. Neurophysiol., 80: 572–582. Leis, A.A., Kronenberg, M.F., Stetkarova, I., Paske, W.C. and Stokic, D.S. (1996a) Spinal motoneuron excitability after acute spinal cord injury in humans. Neurology, 47: 231–237. Leis, A.A., Zhou, H.H., Mehta, M., Harkey III, H.L. and Paske, W.C. (1996b) Behavior of the H-reflex in humans following mechanical perturbation or injury to rostral spinal cord. Muscle Nerve, 19: 1373–1382. Levin, M. and Chapman, C.E. (1987) Inhibitory and facilitatory effects from the peroneal nerve onto the soleus H-reflex in normal and spinal man. Electroenc. Clin. Neurophysiol., 67: 468–478. Li, Y. and Bennett, D.J. (2003) Persistent sodium and calcium currents cause plateau potentials in motoneurons of chronic spinal rats. J. Neurophysiol., 90: 857–869. Li, Y., Gorassini, M.A. and Bennett, D.J. (2004a) Role of persistent sodium and calcium currents in motoneuron firing and spasticity in chronic spinal rats. J. Neurophysiol., 91: 767–783. Li, Y., Harvey, P.J., Li, X. and Bennett, D.J. (2004b) Spastic long-lasting reflexes of the chronic spinal rat studied in vitro. J. Neurophysiol., 91: 2236–2246. Li, S., Kim, J.E., Budel, S., Hampton, T.G. and Strittmatter, S.M. (2005) Transgenic inhibition of Nogo-66 receptor function allows axonal sprouting and improved locomotion after spinal injury. Mol. Cell. Neurosci., 29: 26–39. Liddell, E.G.T. (1934) Spinal shock and some features in isolation-alteration of the spinal cord in cats. Brain, 57: 386–400. Little, J.W. and Halar, E.M. (1985) H-reflex changes following spinal cord injury. Arch. Phys. Med. Rehabil., 66: 19–22. Liu, X.G. and Sandkuhler, J. (1995) Long-term potentiation of C-fiber-evoked potentials in the rat spinal dorsal horn is prevented by spinal N-methyl-D-aspartic acid receptor blockage. Neurosci. Lett., 191: 43–46.
Lovely, R.G., Gregor, R.J., Roy, R.R. and Edgerton, V.R. (1986) Effects of training on the recovery of full-weight-bearing stepping in the adult spinal cat. Exp. Neurol., 92: 421–435. Lozier, A.P. and Kendig, J.J. (1995) Long-term potentiation in an isolated peripheral nerve-spinal cord preparation. J. Neurophysiol., 74: 1001–1009. Lundberg, A. (1979) Multisensory control of spinal reflex pathways. Prog. Brain Res., 50: 11–28. Lundberg, A. (1982) Inhibitory control from the brain stem of transmission from primary afferents to motoneurons primary afferent terminals and ascending pathways. In: Sjolund, B. and Bjorklund, A. (Eds.), Brain Stem Control of Spinal Mechanisms. Elsevier, North-Holland, pp. 179–224. Marcoux, J. and Rossignol, S. (2000) Initiating or blocking locomotion in spinal cats by applying noradrenergic drugs to restricted lumbar spinal segments. J. Neurosci., 20: 8577–8585. Marque, P., Pierrot-Deseilligny, E. and Simonetta-Moreau, M. (1996) Evidence for excitation of the human lower limb motoneurones by group II muscle afferents. Exp. Brain Res., 109: 357–360. Marque, P., Simonetta-Moreau, M., Maupas, E. and Roques, C.F. (2001) Facilitation of transmission in heteronymous group II pathways in spastic hemiplegic patients. J. Neurol. Neurosurg. Psychiatry, 70: 36–42. Maupas, E., Marque, P., Roques, C.F. and Simonetta-Moreau, M. (2004) Modulation of the transmission in group II heteronymous pathways by tizanidine in spastic hemiplegic patients. J. Neurol. Neurosurg. Psychiatry, 75: 130–135. Mayer, R.F., Burke, R.E., Toop, J., Walmsley, B. and Hodgson, J.A. (1984) The effect of spinal cord transection on motor units in cat medial gastrocnemius muscles. Muscle Nerve, 7: 23–31. McCrea, D.A., Pratt, C.A. and Jordan, L.M. (1980) Renshaw cell activity and recurrent effects on motoneurons during fictive locomotion. J. Neurophysiol., 44: 475–488. McCrea, D.A., Shefchyk, S.J., Stephens, M.J. and Pearson, K.G. (1995) Disynaptic group I excitation of ankle extensor motoneurones during fictive locomotion in the cat. J. Physiol., 487: 527–539. Meinck, H.M. and Piesiur-Strehlow, B. (1981) Reflexes evoked in leg muscles from arm afferents: a propriospinal pathway in man? Exp. Brain Res., 43: 78–86. Mendell, L.M. (1984) Modifiability of spinal synapses. Physiol. Rev., 64: 260–324. Misiaszek, J.E. (2003) The H-reflex as a tool in neurophysiology: its limitations and uses in understanding nervous system function. Muscle Nerve, 28: 144–160. Muir, G.D. (1999) Locomotor plasticity after spinal injury in the chick. J. Neurotrauma, 16: 705–711. Muir, G.D. and Steeves, J.D. (1997) Sensorimotor stimulation to improve locomotor recovery after spinal cord injury. TINS, 20: 72–77. Munson, J.B., Foehring, R.C., Lofton, S.A., Zengel, J.E. and Sypert, G.W. (1986) Plasticity of medial gastrocnemius motor units following cordotomy in the cat. J. Neurophysiol., 55: 619–634.
258 Murray, M. and Goldberger, M.E. (1974) Restitution of function and collateral sprouting in the cat spinal cord: the partially hemisected animal. J. Comp. Neurol., 158: 19–36. Naftchi, N.E. (1982) Functional restoration of the traumatically injured spinal cord in cats by clonidine. Science, 217: 1042–1044. Nakazawa, K., Kawashima, N. and Akai, M. (2006a) Enhanced stretch reflex excitability of the soleus muscle in persons with incomplete rather than complete chronic spinal cord injury. Arch. Phys. Med. Rehabil., 87: 71–75. Nakazawa, K., Kawashima, N. and Akai, M. (2006b) Enhanced stretch reflex excitability of the soleus muscle in persons with incomplete rather than complete chronic spinal cord injury. Arch. Phys. Med. Rehabil., 87: 71–75. Nakazawa, K., Kawashima, N. and Akai, M. (2006c) Enhanced stretch reflex excitability of the soleus muscle in persons with incomplete rather than complete chronic spinal cord injury. Arch. Phys. Med. Rehabil., 87: 71–75. Nelson, S.G., Collatos, T.C., Niechaj, A. and Mendell, L.M. (1979) Immediate increase in Ia-motoneuron synaptic transmission caudal to spinal cord transection. J. Neurophysiol., 42: 655–664. Nelson, S.G. and Mendell, L.M. (1979) Enhancement in Iamotoneuron synaptic transmission caudal to chronic spinal cord transection. J. Neurophysiol., 42: 642–654. Nielsen, J., Petersen, N., Ballegaard, M., Biering-Sorensen, F. and Kiehn, O. (1993) H-reflexes are less depressed following muscle stretch in spastic spinal cord injured patients than in healthy subjects. Exp. Brain Res., 97: 173–176. Nielsen, J., Petersen, N. and Crone, C. (1995) Changes in transmission across synapses of Ia afferents in spastic patients. Brain, 118: 995–1004. Okuma, Y., Mizuno, Y. and Lee, R.G. (2002) Reciprocal Ia inhibition in patients with asymmetric spinal spasticity. Clin. Neurophysiol., 113: 292–297. Pearson, K.G. (2001) Could enhanced reflex function contribute to improving locomotion after spinal cord repair? J. Physiol., 533: 75–81. Pearson, K.G. and Collins, D.F. (1993) Reversal of the influence of group Ib afferents from plantaris on activity in medial gastrocnemius muscle during locomotor activity. J. Neurophysiol., 70: 1009–1017. Pedersen, E. (1954) Studies on the central pathway of the flexion reflex in man and animal. Acta Psychiatr. Neurol. Scand., 88: 1–81. Pierrot-Deseilligny, E. and Bussel, B. (1975) Evidence for recurrent inhibition by motoneurons in human subjects. Brain Res., 88: 105–108. Pierrot-Deseilligny, E. and Morin, C. (1979) Evidence for Ib inhibition in human subjects. Brain Res., 166: 176–179. Pierrot-Deseilligny, E., Morin, C., Bergego, C. and Tankov, N. (1981) Pattern of group I fibre projections from ankle flexor and extensor muscles in man. Exp. Brain Res., 42: 337–350. Piesiur-Strehlow, B. and Meinck, H.M. (1980) Response patterns of human lumbo-sacral motoneurone pools to distant somatosensory stimuli. Electroenc. Clin. Neurophysiol., 48: 673–682.
Pockett, S. (1995) Long-term potentiation and depression in the intermediate gray matter of rat spinal cord in vitro. Neuroscience, 67: 791–798. Pockett, S. and Figurov, A. (1993) Long-term potentiation and depression in the ventral horn of rat spinal cord in vitro. Neuroreport, 4: 97–99. Pratt, C.A. and Jordan, L.M. (1980) Recurrent inhibition of motoneurons in decerebrate cats during controlled treadmill locomotion. J. Neurophysiol., 44: 489–500. Prochazka, A. (1996) Proprioceptive feedback and movement regulation. In: Rowell, L.B. and Sheperd, J.T. (Eds.), Handbook of Physiology. Secion 12. Exercise: Regulation and Integration of Multiple Systems. American Physiological Society, New York, pp. 89–127. Raineteau, O. and Schwab, M.E. (2001) Plasticity of motor systems after incomplete spinal cord injury. Nat. Rev. Neurosci., 2: 263–273. Randic, M., Jiang, M.C. and Cerne, R. (1993) Long-term potentiation and long-term depression of primary afferent neurotransmission in the rat spinal cord. J. Neurosci., 13: 5228–5241. Reese, N.B., Skinner, R.D., Mitchell, D., Yates, C., Barnes, C.N., Kiser, T.S. and Garcia-Rill, E. (2005) Restoration of frequency-dependent depression of the H-reflex by passive exercise in spinal rats. Spinal Cord, 44: 28–34. Reier, P.J. (2004) Cellular transplantation strategies for spinal cord injury and translational neurobiology. NeuroRx, 1: 424–451. Renshaw, B. (1941) Influence of discharge of motoneurones upon excitation of neighboring motoneurons. J. Neurophysiol., 4: 167–183. Robinson, G.A. and Goldberger, M.E. (1986) The development and recovery of motor function in spinal cats. II. Pharmacological enhancement of recovery. Exp. Brain Res., 62: 387–400. Roby-Brami, A. and Bussel, B. (1987) Long-latency spinal reflex in man after flexor reflex afferent stimulation. Brain, 110: 707–725. Rossignol, S. (1996) Neural control of stereotypic limb movements. In: Rowell, L.B. and Sheperd, J.T. (Eds.), Handbook of Physiology, Section 12. Exercise: Regulation and Integration of Multiple Systems. Oxford University Press, New York, pp. 173–216. Rossignol, S., Be´langer, M., Chau, C., Giroux, N., Brustein, E., Bouyer, L., Grenier, C.-A., Drew, T., Barbeau, H. and Reader, T. (2000) The spinal cat. In: Kalb, R.G. and Strittmatter, S.M. (Eds.), Neurobiology of Spinal Cord Injury. Humana Press, Totowa, pp. 57–87. Rossignol, S., Chau, C., Brustein, E., Be´langer, M., Barbeau, H. and Drew, T. (1996) Locomotor capacities after complete and partial lesions of the spinal cord. Acta Neurobiol. Exp., 56: 449–463. Rossignol, S., Chau, C., Giroux, N., Brustein, E., Bouyer, L., Marcoux, J., Langlet, C., Barthe´lemy, D., Provencher, J., Leblond, H., Barbeau, H. and Reader, T.A. (2002) The cat model of spinal injury. In: McKerracher, L., Doucet, G. and Rossignol, S. (Eds.), Spinal Cord Trauma: Regeneration, Neural Repair and Functional Recovery. Elsevier, New York, pp. 151–168.
259 Rossignol, S., Drew, T., Brustein, E. and Jiang, W. (1999) Locomotor performance and adaptation after partial or complete spinal cord lesions in the cat. In: Binder, M.D. (Ed.), Peripheral and Spinal Mechanisms in the Neural Control of Movement. Elsevier, Amsterdam, pp. 349–365. Rossignol, S., Dubuc, R. and Gossard, J.P. (2006) Dynamic sensorimotor interactions in locomotion. Physiol. Rev., 86: 89–154. Rossignol, S., Pearson, K.G., Be´langer, M. and Drew, T. (1990) A comparison of locomotion and other rhythmic motor programs in the intact, the spinal and the paralysed cat. Neurosci. Lett. Suppl., 38: S42 (Abstract). Roudet, C., Gimenez, R.M., Privat, A., Feuerstein, C. and Savasta, M. (1996) Regional study of spinal alpha 2-adrenoceptor densities after intraspinal noradrenergic-rich implants on adult rats bearing complete spinal cord transection or selective chemical noradrenergic denervation. Neurosci. Lett., 208: 89–92. Roy, R.R., Talmadge, R.J., Hodgson, J.A., Zhong, H., Baldwin, K.M. and Edgerton, V.R. (1998) Training effects on soleus of cats spinal cord transected (T12-13) as adults. Muscle Nerve, 21: 63–71. Rudomin, P. and Schmidt, R.F. (1999) Presynaptic inhibition in the vertebrate spinal cord revisited. Exp. Brain Res., 129: 1–37. Schieppati, M. (1987) The Hoffmann reflex: a means of assessing spinal reflex excitability and its descending control in man. Prog. Neurobiol., 28: 345–376. Schindler-Ivens, S. and Shields, R.K. (2000) Low frequency depression of H-reflexes in humans with acute and chronic spinal cord injury. Exp. Brain Res., 133: 233–241. Schindler-Ivens, S.M. and Shields, R.K. (2004) Soleus H-reflex recruitment is not altered in persons with chronic spinal cord injury. Arch. Phys. Med. Rehabil., 85: 840–847. Schmit, B.D., Hornby, T.G., Tysseling-Mattiace, V.M. and Benz, E.N. (2003) Absence of local sign withdrawal in chronic human spinal cord injury. J. Neurophysiol., 90: 3232–3241. Schnell, L., Schneider, R., Kolbeck, R., Barde, Y.A. and Schwab, M.E. (1994) Neurotrophin-3 enhances sprouting of corticospinal tract during development and after adult spinal cord lesion. Nature, 367: 170–173. Schomburg, E.D. (1990) Spinal sensorimotor systems and their supraspinal control. Neurosci. Res., 7: 265–340. Schomburg, E.D., Petersen, N., Barajon, I. and Hultborn, H. (1998) Flexor reflex afferents reset the step cycle during fictive locomotion in the cat. Exp. Brain Res., 122: 339–350. Schouenborg, J., Holmberg, H. and Weng, H.R. (1992) Functional organization of the nociceptive withdrawal reflexes. II. Changes of excitability and receptive fields after spinalization in the rat. Exp. Brain Res., 90: 469–478. Schouenborg, J. and Weng, H.R. (1994) Sensorimotor transformation in a spinal motor system. Exp. Brain Res., 100: 170–174. Segal, R.L. and Wolf, S.L. (1994) Operant conditioning of spinal stretch reflexes in patients with spinal cord injuries. Exp. Neurol., 130: 202–213. Shahani, B.T. and Young, R.R. (1971) Human flexor reflexes. J. Neurol. Neurosurg. Psychiatry, 34: 616–627.
Shefchyk, S., McCrea, D.A., Kriellaars, D., Fortier, P. and Jordan, L. (1990) Activity of interneurons within the L4 spinal segment of the cat during brainstem-evoked fictive locomotion. Exp. Brain Res., 80: 290–295. Shefner, J.M., Berman, S.A., Sarkarati, M. and Young, R.R. (1992) Recurrent inhibition is increased in patients with spinal cord injury. Neurology, 42: 2162–2168. Sherrington, C.S. (1899) On the spinal animal. MedicoChirugical Trans., 82: 449–486. Sherrington, C.S. (1910) Flexion-reflex of the limb, crossed extension-reflex, and reflex stepping and standing. J. Physiol., 40: 28–121. Shurrager, P.S. and Dykman, R.A. (1951) Walking spinal carnivores. J. Comp. Physiol. Psychol., 44: 252–262. Simonetta-Moreau, M., Marque, P., Marchand-Pauvert, V. and Pierrot-Deseilligny, E. (1999) The pattern of excitation of human lower limb motoneurones by probable group II muscle afferents. J. Physiol., 517(Pt 1): 287–300. Simonsen, E.B. and Dyhre-Poulsen, P. (1999) Amplitude of the human soleus H reflex during walking and running. J. Physiol., 515: 929–939. Skinner, R.D., Houle, J.D., Reese, N.B., Berry, C.L. and Garcia-Rill, E. (1996) Effects of exercise and fetal spinal cord implants on the H- reflex in chronically spinalized adult rats. Brain Res., 729: 127–131. Smith, J.L., Edgerton, V.R., Eldred, E. and Zernicke, R.F. (1983) The chronic spinalized cat: a model for neuromuscular plasticity. Birth Defects Orig. Artic. Ser., 19: 357–373. Smith, J.L., Smith, L.A., Zernicke, R.F. and Hoy, M. (1982) Locomotion in exercised and non-exercised cats cordotomized at two or twelve weeks of age. Exp. Neurol., 76: 393–413. Taylor, S., Ashby, P. and Verrier, M. (1984) Neurophysiological changes following traumatic spinal lesions in man. J. Neurol. Neurosurg. Psychiatry, 47: 1102–1108. Taylor, J.S., Friedman, R.F., Munson, J.B. and Vierck Jr., C.J. (1997) Stretch hyperreflexia of triceps surae muscles in the conscious cat after dorsolateral spinal lesions. J. Neurosci., 17: 5004–5015. Taylor, J., Munson, J. and Vierck Jr., C. (1999) Effects of dorsolateral spinal lesions on stretch reflex threshold and stiffness in awake cats. Eur. J. Neurosci., 11: 363–368. Thompson, F.J., Reier, P.J., Lucas, C.C. and Parmer, R. (1992) Altered patterns of reflex excitability subsequent to contusion injury of the rat spinal cord. J. Neurophysiol., 68: 1473–1486. Timoszyk, W.K., Nessler, J.A., Acosta, C., Roy, R.R., Edgerton, V.R., Reinkensmeyer, D.J. and de Leon, R. (2005) Hindlimb loading determines stepping quantity and quality following spinal cord transection. Brain Res., 1050: 180–189. Trimble, M.H., Kukulka, C.G. and Behrman, A.L. (1998) The effect of treadmill gait training on low-frequency depression of the soleus H-reflex: comparison of a spinal cord injured man to normal subjects. Neurosci. Lett., 246: 186–188. Valero-Cabre, A., Fores, J. and Navarro, X. (2004) Reorganization of reflex responses mediated by different afferent sensory fibers after spinal cord transection. J. Neurophysiol., 91: 2838–2848.
260 Vanderhorst, V.G.J.M. and Holstege, G. (1997) Organization of lumbosacral motoneuronal cell groups innervating hindlimb, pelvic floor, and axial muscles in the cat. J. Comp. Neurol., 382: 46–76. Vilensky, J.A., Moore, A.M., Eidelberg, E. and Walden, J.G. (1992) Recovery of locomotion in monkeys with spinal cord lesions. J. Motor Behav., 24: 288–296. Wall, P.D. (1975) Signs of plasticity and reconnection in spinal cord damage. Ciba. Foundation Symposium, 35–63. Walmsley, B. and Tracey, D.J. (1983) The effect of transection and cold block of the spinal cord on synaptic transmission between Ia afferents and motoneurones. Neuroscience, 9: 445–451. Weaver, R.A., Landau, W.M. and Higgins, J.F. (1963) Fusimotor function. II. Evidence of fusimotor depression in human spinal shock. Arch. Neurol., 9: 127–132. Xia, R. and Rymer, W.Z. (2005) Reflex reciprocal facilitation of antagonist muscles in spinal cord injury. Spinal Cord, 43: 14–21.
Yanagisawa, N. and Tanaka, R. (1978) Reciprocal Ia inhibition in spastic paralysis in man. Electroencephalogr. Clin. Neurophysiol., Suppl., 34: 521–526. Young, R.R. (1994) Spasticity: a review. Neurology, 44: s12–s20. Zehr, P.E. (2002) Considerations for use of the Hoffmann reflex in exercise studies. Eur. J. Appl. Physiol., 86: 455–468. Zehr, E.P., Collins, D.F. and Chua, R. (2001) Human interlimb reflexes evoked by electrical stimulation of cutaneous nerves innervating the hand and foot. Exp. Brain Res., 140: 495–504. Zehr, E.P. and Stein, R.B. (1999) What functions do reflexes serve during human locomotion? Prog. Neurobiol., 58: 185–205. Zmyslowski, W., Gorska, T., Majczynski, H. and Bem, T. (1993) Hindlimb muscle activity during unrestrained walking in cats with lesions of the lateral funiculi. Acta Neurobiol. Exp., 53: 143–153.
Plate 16.1. General scheme of reflex pathways and spinal locomotor control. This scheme is subdivided in three parts. The supraspinal level includes various descending pathways from the telencephalon and brain stem involved in activating, stopping, or modulating characteristics of the spinal central pattern generator (CPG) for locomotion as well as the excitability of transmission in reflex pathways at motoneuronal or premotoneuronal (presynaptic and/or interneuronal) levels. The large arrow emerging from the Supraspinal level encompasses all these functions without any further details. The spinal cord level includes the CPG with a generally reciprocal activity between the Flexor (F) and Extensor (E) sides. These two antagonist phases of the CPG circuitry are separated to indicate that each part may exert a function on other spinal mechanisms (represented by 3 output neurons emerging from each part of the CPG) as well as interact between each other (inhibitory connections between F and E). The Interneurons are represented by two large pink and blue interneurons [IN] which are interposed between afferents and motoneurons in disynaptic pathways as well as other more specific inhibitory interneurons (in black) representing disynaptic inhibitory pathways (such as Ib inhibitory interneurons) which can also be inhibited by other interneurons in certain tasks such as locomotion. Finally, motoneuron pools include both a-motoneurons projecting to extrafusal muscle fibers and g-motoneurons projecting to intrafusal muscle fibers. Recurrent inhibition s through Renshaw cells inhibits a-motoneurons (represented) and g-motoneurons (not represented) and Ia interneurons responsible for reciprocal inhibition between a-motoneurons. In the periphery, one ankle flexor muscle (pink) and one extensor muscle (blue) are represented with a spindle in both. Group Ia and II represent sensory fibers from spindles and are responsible for indicating rate and amount of muscle stretch, respectively. The stimulation symbol on the Ia fiber from the extensor illustrates direct stimulation of Ia afferents as performed during H-Reflex studies. Ib fibers originate from Golgi tendon organs, which measure the force output of the muscle. Connectivity of the various afferents is partial and is largely based on that established in Rossignol et al. (2006).
Plate 16.2. (A) Comparison of locomotion in the intact and spinal states in the same cat. (B) Stick figures of the swing and stance phases reconstructed from videos taken at 60 frames/s. Each frame is de-interlaced to extract each video field and light reflecting markers placed on the various joints are software detected and used to reconstruct the movement as stick figures for each joint as indicated. The angles measured are indicated and the arrows show the direction of limb movement. The stick figures are displaced from one another by a value corresponding to the foot displacement in the X direction so that each stick is clearly individualized. The length calibration found in B is therefore different in the X and Y axes. (C) Same as in A but 41 days after spinalization with daily locomotor training on the treadmill. Note that in this case the limb is somewhat extended throughout the cycle and that overall step length is reduced compared to A. The calibration is explained in A. (D) Joint angular displacements are illustrated for the intact state (black, 25 cycles) and spinal state (red, 16 cycles). Note the shift of hip in extension (shift of the hip trace) and of the knee and ankle extension during stance. The cycle is normalized to unity. Electromyographic (EMG) recordings of the same sections as in C. Ipsilateral (i) refers to the side of the video camera while (co) is contralateral. Semitendinosus (St) is mainly a knee flexor but also acts as a hip extensor; Tibialis Anterior (TA) is an ankle flexor; Vastus Lateralis (VL) is a knee extensor and Gastrocnemius Lateralis (GL) an ankle extensor; Flexor Digitorum Longus plantarflexes the digits and is active during stance. EMGs are largely superimposed in both conditions although some changes are distinguishable and discussed in the text. EMGs have also been normalized to unity.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 17
The education and re-education of the spinal cord Jonathan R. Wolpaw Wadsworth Center, Laboratory of Nervous System Disorders, New York State Department of Health and State University of New York, Albany, NY 12201, USA
Abstract: In normal life, activity-dependent plasticity occurs in the spinal cord as well as in the brain. Like CNS plasticity elsewhere, this spinal cord plasticity can occur at many neuronal and synaptic sites and by a variety of mechanisms. Spinal cord plasticity is prominent in postnatal development and contributes to acquisition of standard behaviors such as locomotion and rapid withdrawal from pain. Later on in life, spinal cord plasticity contributes to acquisition and maintenance of specialized motor skills, and to compensation for the peripheral and central changes associated with aging, disease, and trauma. Mastery of even the simplest behaviors is accompanied by complex spinal and supraspinal plasticity. This complexity is necessary, to preserve the full roster of behaviors, and is also inevitable, due to the ubiquity of activitydependent plasticity in the CNS. Careful investigation of spinal cord plasticity is essential for understanding motor skills; and, because of the relative simplicity and accessibility of the spinal cord, is a logical and convenient starting point for exploring skill acquisition. Appropriate induction and guidance of activity-dependent plasticity in the spinal cord is likely to be a key part of the realization of effective new rehabilitation methods for spinal cord injuries, cerebral palsy, and other chronic motor disorders. Keywords: spinal cord; spinal cord injury; plasticity; conditioning; learning; memory; behavior Understanding of spinal cord function is now evolving further as part of the growing recognition of the ubiquity of activity-dependent plasticity in the CNS. Appreciation of the multiple mechanisms of synaptic and neuronal plasticity, of their existence in many different regions, and of the frequency with which they are activated, has overturned the traditional view of a hard-wired CNS that stores the effects of past experience by only a few mechanisms at only a few specialized sites. Activity-dependent plasticity, or persistent CNS modification that results from past experience and affects future behavior, is now recognized as a feature of the entire CNS, including the spinal cord. The spinal cord has capacities for activity-dependent plasticity similar to those found elsewhere in the CNS. During development and later in life, it changes in response to input from the brain and from the periphery. Like plasticity elsewhere in the
Introduction The development of neuroscience over the past 200 years has been marked by the gradual and belated elevation in the status of the spinal cord. Originally thought to be simply a big well-protected nerve through which the brain interacts with the world, the spinal cord evolved first into a way station between the brain and the periphery that harbored a few simple reflexes, and then progressed to a repository of highly stereotyped behaviors such as locomotion. Nevertheless, it remained until very recently a hard-wired structure that simply responded, albeit in complex ways, to inputs from the brain and from the periphery.
Corresponding author. Tel.: +1 518 473 3631; Fax: +1 518 486 4910; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57017-7
261
262
CNS, this spinal cord plasticity involves synaptic and neuronal mechanisms (e.g., long-term potentiation (LTP), modifications in neuronal morphology and electrical properties); it is affected by growth factors, and it is associated with gene activation (e.g., Liu and Sandku¨hler, 1997; Eyre et al., 2000; Gibson et al., 2000; Inglis et al., 2000; Mendell et al., 2001; Wolpaw and Tennissen, 2001; Tillakaratne et al., 2002; Dupont-Versteegden et al., 2004; Ding et al., 2005.) Many laboratory and clinical studies have addressed the functional impact of such spinal cord plasticity in health and disease. The main measures of function have been two major classes of spinal cord responses to peripheral inputs: flexion withdrawal reflexes, which are mediated by oligosynaptic nociceptive pathways to spinal motoneurons from unmyelinated C fibers and small myelinated A-delta fibers; and proprioceptive reflexes, which are mediated by mono- and oligosynaptic pathways to spinal motoneurons from larger afferents innervating muscle spindles, Golgi tendon organs, and other receptors that register muscle length and tension and limb position (Matthews, 1972; Baldissera et al., 1981; Burke, 1998; Kandel et al., 2000; Zehr, 2002; Misiaszek, 2003; Sandrini et al., 2005; Voerman et al., 2005). In the isolated spinal cord, flexion withdrawal reflexes display habituation and sensitization in a variety of stimulation protocols (Mendell, 1984). Most studies have involved pain mechanisms and sensitization of spinal cord responses to C and A fiber inputs (see Chapters by Price and by Lenz and other recent reviews (e.g., Herrero et al., 2000; Zimmermann, 2001; Ji et al., 2003; Melzack et al., 2004; Katz and Rothenberg, 2005; Flor and Andrasik, 2006)). In addition, these reflexes can show classical and operant conditioning phenomena (reviewed in Patterson, 1976; Kandel, 1977; Thompson, 2001). Furthermore, activity in proprioceptive afferents can change the spinal cord (e.g., Lloyd, 1949; Kandel, 1977). The spinal locomotor pattern generator (LPG), which is normally activated and influenced by descending supraspinal activity, can function autonomously in the isolated spinal cord. This is most apparent in lower vertebrates, but evident also in higher vertebrates such as the cat (Rossignol, 1996;
Kiehn and Butt, 2003; Clarac et al., 2004; Juvin et al., 2005; Grillner et al., 2005; Rossignol et al., 2006). A spinal LPG exists in humans as well (Holmes, 1915; Kuhn, 1950; Bussel et al., 1988; Calancie et al., 1994; Dietz et al., 1995; Dobkin et al., 1995; Rossignol, 1996, 2000; Kiehn et al., 1998; Dimitrijevic et al., 1998; Orlovsky et al., 1999; Dietz, 2003; Zehr, 2005). Studies begun 50 years ago and resumed in the 1980s demonstrate that the locomotion produced by the isolated spinal cord improves with training and that this improvement derives from spinal cord plasticity (Shurrager and Dykman, 1951; Lovely et al., 1986; Barbeau and Rossignol, 1987; Barbeau et al., 2002; Rossignol et al., 2002, 2004; Tillakaratne et al., 2002; Edgerton et al., 2004). This work (reviewed in Chapters by Frigon and Brown) implies that the operation of a human LPG could be encouraged and guided in those with spinal cord injuries so as to restore locomotion. Comparable methods for inducing and guiding activity-dependent spinal cord plasticity could promote restoration of bladder, bowel, and other autonomic functions after spinal cord injuries. Interruption or impairment of descending spinal cord pathways leads to gradual changes in spinal cord function (Riddoch, 1917; Kuhn, 1950; Mountcastle, 1980; Ronthal, 1998; Hiersemenzel et al., 2000). The end result typically includes increased resistance to passive muscle stretch, especially in antigravity muscles (i.e., leg extensors and arm flexors), hyperactive tendon jerks, and increased flexion withdrawal responses. These functional effects make up the syndrome of spasticity. The progressive development of these signs reflects plasticity in the spinal cord caused by loss or distortion of supraspinal input and by associated changes in peripheral inputs. This spinal cord plasticity includes modifications in motoneuron and motor unit properties, in motoneuron synaptic coverage, in primary afferent EPSPs, and in interneuronal pathways (Nelson and Mendell, 1979; Cope et al., 1986; Munson et al., 1986; Boorman et al., 1991; Thompson et al., 1992; Shefner et al., 1992; Hochman and McCrea, 1994a–c; Tai and Goshgarian, 1996; Tai et al., 1997; Hultborn, 2003). Much more rapid activity-dependent spinal cord plasticity occurring in response to the
263
more focused change in descending input caused by a cerebellar lesion was first described nearly 80 years ago (DiGiorgio, 1929, 1942; Manni, 1950; Gerard, 1961; Chamberlain et al., 1963). This phenomenon, labeled ‘‘spinal fixation,’’ was thought by Gerard and others to be an excellent model for the fixation or consolidation of memory. Similar phenomena occur in the spinal cord in connection with other supraspinal lesions and after manipulation of vestibular sensory inputs (Giulio, 1952; Straka and Dieringer, 1995). In sum, altered descending input that persists for a sufficient time causes spinal cord plasticity that remains after the input ceases. New appreciation of these data and their implications has coincided with and has been encouraged by the energy and optimism now centered on possibilities for restoring spinal cord structure and function after injury (e.g., see Chapter by Merzenich et al., Dobkin and Havton, 2004; Liverman et al., 2005). The expectations for achieving regeneration of damaged pathways and lost neurons inevitably lead to the question of how newly regenerated spinal cord is to become capable of useful function. An effectively functioning adult spinal cord depends on appropriate activity-dependent plasticity during early development and throughout subsequent life. A newly, and probably imperfectly, regenerated spinal cord will almost certainly not support effective function well (Muir and Steeves, 1997); it will probably display diffuse infantile reflexes and/or other disordered and dysfunctional behaviors. Thus, as methods for producing spinal cord regeneration become available, methods for appropriately re-educating the regenerated spinal cord are likely to become essential. This anticipated need is spurring attention to the activity-dependent spinal cord plasticity that shapes the functional properties of spinal cord neurons and pathways to support behaviors as varied as locomotion, urination, ballet, and playing a musical instrument. Furthermore, new appreciation of the inherent capabilities for plasticity of the injured unregenerated spinal cord gives additional reason for investigating activitydependent spinal cord plasticity (e.g., Chapter by Frigon and Grillner and Walle´n, 2004). Understanding spinal cord plasticity is necessary for
understanding both the changes caused by injury and the processes that might be engaged and guided to restore effective function. Attention to activity-dependent spinal cord plasticity is further encouraged by the growing recognition that acquisition and maintenance of normal motor behaviors depend on activity-dependent plasticity at many sites in the CNS, including the spinal cord. Peripheral and descending inputs during practice change the spinal cord, and these changes combine with changes elsewhere in the CNS to support the new behavior. Thus, understanding the mechanisms of spinal cord plasticity and its interactions with plasticity elsewhere is required for understanding normal motor behaviors as well as the complex motor disabilities that accompany spinal cord injuries and other chronic neurological disorders. Activity-dependent spinal cord plasticity during normal life In the normal CNS, descending and peripheral inputs to the spinal cord are a continual barrage of activity in many different pathways. The immediate effects of this activity (e.g., voluntary movements, responses to peripheral disturbances, locomotion, respiration, urination, task-specific adjustments in spinal reflexes) are very obvious. In contrast, the gradual long-term effects of these inputs on the spinal cord are not obvious. Nevertheless, these gradual long-term effects are important: they serve to establish and maintain spinal cord function in a state that supports effective motor behaviors. Gradual activity-dependent plasticity, driven by descending and associated peripheral inputs, shapes spinal cord function during development and continues to adjust it throughout later life. The next sections review the spectrum of activity-dependent plasticity during normal life, and Table 1 lists representative examples. Developmental plasticity In early life, both descending and peripheral inputs have crucial roles in the plasticity that produces a
264 Table 1. Examples of activity-dependent spinal cord plasticity in normal life Development Focusing of proprioceptive reflexes (Myklebust et al., 1986; O’Sullivan et al., 1991) Directionality of nociceptive withdrawal responses (Levinsson et al., 1999; Waldenstrom et al., 2003) Lateralization of descending cortical control (Eyre et al., 2001; Martin et al., 2004) Maturation of bladder reflex pathways (de Groat, 2002) Motoneuron morphology (Inglis et al., 2000) Skill Acquisition and Other Experiences Later in Life Ballet (Nielsen et al., 1993; Koceja et al., 1991; Goode and Van Hoven, 1982) Athletic training (Rochcongar et al., 1979; Casabona et al., 1990) Activity level (Yamanaka et al., 1999; Gomez-Pinilla et al., 2002) Hopping (Voigt et al., 1998) Backward locomotion (Schneider and Capaday, 2003) Limb trajectory maintenance (Meyer-Lohmann et al., 1986) Age-related changes in proprioceptive reflexes (Koceja and Mynark 2000; Morita et al., 1995) Operant conditioning of proprioceptive reflexes (Wolpaw et al., 1983; Chen and Wolpaw, 1995; Segal, 1997)
normally functioning adult spinal cord, a spinal cord that has characteristic adult reflex patterns, supports basic motor skills like locomotion, and also supports more specialized skills such as dancing or playing the piano. Rapid withdrawal from painful stimuli is a critical function of spinal cord pathways, and it is acquired early in life. In the neonatal rat, focal nociceptive stimulation evokes diffuse and frequently inappropriate muscle activations and limb movements. In contrast, in the normal adult this stimulation excites only the correct muscles – the muscles that withdraw the limb from the painful stimulus. Schouenborg’s group has shown the importance of descending input in achieving these correctly focused adult flexion withdrawal reflexes (e.g., Levinsson et al., 1999; Waldenstrom et al., 2003). When spinal cord transection at birth removes descending input, the adult pattern fails to develop; so that nonspecific and inappropriate withdrawal reflexes are still present in adulthood (Fig. 1A). Spinal cord proprioceptive reflexes contribute to locomotion and other basic motor behaviors
(Rossignol, 1996). In human infants, muscle stretch causes short-latency spinally-mediated stretch reflexes in the stretched muscles and in their antagonists as well (Myklebust et al., 1986; O’Sullivan et al., 1998). The antagonist stretch reflexes gradually disappear during childhood, and the adult is left with standard ‘‘knee-jerk’’ reflexes that are limited to the stretched muscles. However, when perinatal supraspinal damage (i.e., cerebral palsy) disrupts activity in descending pathways, this evolution may fail to occur, and antagonist stretch reflexes may last into adulthood and contribute to motor disabilities. Fig. 1B shows agonist and antagonist stretch reflexes from a normal baby, a normal adult, and an adult with cerebral palsy. Like the normal baby and unlike the normal adult, the adult with cerebral palsy shows shortlatency responses in both agonist and antagonist muscles. In this situation, the original damage is supraspinal. Thus, the probable reason for the abnormal persistence of infantile spinal reflexes into adulthood is distortion of the descending input that gradually eliminates these reflexes in the initial years of life. Intact descending activity is also important for development of urinary function. In newborn animals and humans, voiding is easily evoked by peripheral stimuli through a purely spinal reflex pathway. During postnatal development, this reflex voiding is suppressed and brain control of voiding becomes dominant. The change seems to reflect change in the relative strengths of peripheral and descending excitatory inputs to the preganglionic neurons in the spinal cord that mediate micturition (de Groat, 2002; Vizzard, 2006). The decreased response to peripheral input probably results from a presynaptic decrease in glutamic acid release, rather than from a postsynaptic change. Neonatal spinal cord transection prevents this evolution, and infantile reflex voiding persists into adulthood. The primitive pattern can also reemerge in adults after spinal cord injury. In development, corticospinal pathways change so that they produce the normal adult pattern of mainly contralateral innervation of limb muscles (Eyre et al., 2001, Eyre, 2003; ten Donkelaar et al., 2004; Martin, 2005). Perinatal damage to sensorimotor cortex on one side can prevent this normal
265
Fig. 1. Activity-dependent spinal cord plasticity during development. (A) Direction of limb movement produced by flexion withdrawal responses to a nociceptive stimulus in normal adult rats and in adult rats that had undergone spinal cord transection just after birth. Direction is almost always appropriate, i.e., away from the stimulus, in normal adults, but is often inappropriate in transected adults. Neonatal transection prevents normal shaping of flexion withdrawal reflexes by descending input (modified from Levinsson et al., 1999). (B) Short-latency electromyographic (EMG) responses of soleus (solid) and tibialis anterior (dotted) muscles to sudden foot dorsiflexion, which stretches the soleus and shortens the tibialis anterior, in a normal infant, a normal adult, and an adult with cerebral palsy. In the normal infant, spinal stretch reflexes occur in both muscles. In the normal adult, a reflex occurs only in the stretched muscle, i.e., the soleus, and little or no response occurs in the tibialis anterior. In contrast, in the adult with cerebral palsy, in whom perinatal supraspinal injury has impaired descending input, the infantile pattern persists: reflexes occur in both muscles. (From B. Myklebust, unpublished data (Myklebust et al., 1982, 1986 for comparable data)). (C) Ipsilateral (solid) and contralateral (dashed) EMG responses (first dorsal interosseus muscle) to transcranial magnetic stimulation over motor cortex in a normal adult (left) and an adult with cerebral palsy (right). Horizontal scale bar is 20 ms, vertical bar is 200 mV. The large ipsilateral response seen in the subject with cerebral palsy indicates abnormal preservation of the strong ipsilateral corticospinal connections that normally disappear early in life (modified from Eyre et al., 2001). (D) Densities in 8-week-old and adult cats of putative corticospinal tract boutons (varicosities) in cervical spinal cord ipsilateral (dark gray) and contralateral (light gray) to forelimb muscle paralyzed from age 3 weeks to 7 weeks. Muscle paralysis during development reduces corticospinal tract innervation and this deficit persists into adulthood. Asterisks indicate significant differences from the contralateral data (modified from Martin et al., 2004).
266
evolution, and lead to abnormal adult innervation in which the undamaged side has strong ipsilateral as well as contralateral projections to the spinal cord. This abnormal innervation seems to result from the lack of normal activity-dependent competition between ipsilateral and contralateral connections. Fig. 1C, which shows responses to transcranial magnetic stimulation in a normal adult and an adult with cerebral palsy, illustrates the abnormality (Eyre et al., 2001). The normal adult has a large contralateral response and a minimal ipsilateral response, while the adult with cerebral palsy has large responses on both sides. Cerebral palsy can also affect muscle afferent connections in spinal cord, and may in addition affect spinal neuronal properties (e.g., expression of parvalbumin and the early immediate gene c-Jun) (Gibson et al., 2000). The functional effect of the abnormal innervation is especially severe in humans, in whom corticospinal connections become prominent early in development and come to have major roles in movement control (Eyre et al., 2000; Mayer and Esquenazi, 2003). The motor abnormalities caused by perinatal hemispheric damage may be minimal in the infant and only become obvious later on, as complex motor skills fail to develop normally. Both peripheral and descending inputs contribute to the development of a properly functioning adult spinal cord. In rats, the evolution of normal flexion withdrawal responses depends not only on descending pathways (e.g., Fig. 1A) but on peripheral input as well. Peripheral anesthesia during development prevents development of normal adult reflexes (Waldenstrom et al., 2003). Painful input is not required: tactile input alone can support normal development. Abnormal peripheral input during development caused by paralysis of a particular muscle may also lead to adult abnormalities in corticospinal motoneuronal connections (Fig. 1D) and motor control (Martin et al., 2004).
Plasticity with the acquisition and maintenance of motor skills Acquisition of motor skills later on in life is accompanied by modifications in spinal cord circuits.
These modifications have been demonstrated in animals and humans mainly by measuring the spinal stretch reflex (SSR) (produced largely by the monosynaptic pathway comprised of the Ia afferent from the muscle spindle, its synapse on the motoneuron, and the motoneuron), and its electrical analog, the H-reflex, which is evoked by direct electrical stimulation of Ia afferents (Magladery et al., 1951; Matthews, 1972; Henneman and Mendell, 1981; Brown, 1984; Zehr, 2002). The simple spinal pathway of these reflexes contributes to both simple and complex behaviors. As a result, changes in this pathway change many behaviors and/or change the CNS activity responsible for these behaviors. Many studies indicate that these spinal reflexes are affected by the nature, intensity, and duration of past physical activity and by specific training regimens. These spinal reflexes differ between athletes and nonathletes and among different groups of athletes (Rochcongar et al., 1979; Goode and Van Hoven, 1982; Casabona et al., 1990; Koceja et al., 1991, 2004; Nielsen et al., 1993; Auge´ and Morrison, 2000). Nielsen et al. (1993) measured H-reflexes in soleus muscles of people who were sedentary, moderately active, or extremely active and in professional ballet dancers. H-reflexes and disynaptic reciprocal inhibition were larger in moderately active people than in sedentary people, and were even larger in extremely active people. Given that the human soleus muscle is composed largely of slow (i.e., type I) fibers, exercise-induced change in motor unit properties cannot readily explain these reflex increases associated with physical activity. The most notable result, illustrated in Fig. 2A, was that both the H-reflex and disynaptic reciprocal inhibition were smallest in the professional dancers, even though they were much more active than the other groups. The dancers’ reflexes were smaller than those of sedentary people, and far smaller than those of people who were physically active in other ways. Beginning from the knowledge that muscle cocontraction (i.e., simultaneous contraction of agonists and antagonists) is associated with increased presynaptic inhibition and decreased reciprocal inhibition, Nielsen et al. (1993) proposed that the prolonged cocontractions needed for the classical ballet postures led to
267
Fig. 2. Activity-dependent spinal cord plasticity associated with skill acquisition and with aging. (A) Soleus H reflexes are much smaller in professional ballet dancers than in other well-trained athletes (e.g., runners, swimmers, cyclists) (H-reflexes of sedentary subjects fall in between.) (modified from Nielsen et al., 1993). (B) Working for reward, monkeys performed an elbow flexion-extension task on which brief perturbations were randomly superimposed. Biceps EMG and elbow angle (flexion upward) for an unperturbed trial (dotted), a perturbed trial early in training (solid), and a perturbed trial late in training (dashed) are shown. Early in training, perturbation elicits both a spinal stretch reflex (SSR) and a long-latency polysynaptic response (LLR). After intermittent training over several years, the SSR is much larger and the LLR has disappeared. The SSR has gradually taken over the role of opposing the perturbation. This improves performance: the disturbance in the smooth course of elbow flexion is smaller and briefer (modified from Meyer-Lohmann et al., 1986). (C) Change in soleus H-reflex size as a function of time in the backward-walking step cycle as a person masters backward walking over 10 days. Top: Soleus EMG of Day 1 and Day 10 just before and after onset of the soleus burst associated with the stance phase of the step cycle. Dotted line shows soleus EMG for quiet standing. Bottom: H-reflex size (as % of size during quiet standing) versus time in the backward-walking step cycle for Days 1, 4, 7, and 10 of training. Soleus EMG does not change with training. In contrast, the marked increase in H-reflex size prior to the soleus burst that is seen on Day 1 disappears by Day 10. (D) Soleus H-reflexes in prone (black) and standing (gray) positions from a young person and an old person. In old subjects, the H-reflex tends to be smaller and to be less affected by body position (modified from Koceja et al., 1995).
persistent decreases in synaptic transmission at Ia synapses, and thus to weak H-reflexes and weak reciprocal inhibition. The decreased direct peripheral influence on motoneurons that is reflected in the smaller reflexes may increase cortical control and thereby enable more precise movements. In studies of people with different histories of physical activity, it is difficult to rule out the
possible confounding impacts of differences in basic genetic endowments (e.g., between dancers and other people, or between dancers and runners). Carefully controlled investigations of the effects of specialized training regimens do not face this difficulty, and such studies provide further evidence of activity-dependent spinal cord plasticity. In an early experiment, monkeys learned to make
268
smooth repetitive flexion and extension movements about the elbow, and random brief perturbations were superimposed (Meyer-Lohmann et al., 1986). Over months and years, the SSR evoked by the perturbation gradually increased and essentially took over the task of responding to the perturbation, while longer-latency reflex responses gradually disappeared. As Fig. 2B shows, the larger SSR was adaptive, that is, it was associated with faster recovery from the disturbance in trajectory caused by the perturbation. The paper concluded (p. 398) that the results ‘‘demonstrate a growing role for fast segmental mechanisms in the reaction to external disturbances as motor learning progresses.’’ Subsequent studies in humans have described reflex changes occurring over days and weeks due to specialized training regimens (Pe´rot et al., 1991; Voigt et al., 1998; Yamanaka et al., 1999; Schneider and Capaday, 2003). Fifteen min/day of practice in a backward walking task produced a gradual change in the dependence of H-reflex amplitude on the point in the step cycle at which the reflex was elicited. On the first day of backward walking training, the soleus H-reflex was large well before the onset of the soleus burst associated with the stance phase of the cycle. By the tenth day, the H-reflex was not detectable until the onset of the burst. Fig. 2C illustrates this change. The early increase in the H-reflex on the first day may represent compensation for uncertainty as to when the stance phase of the step cycle, with its need for greater soleus activity, will start. The greater sensitivity of the reflex arc helps ensure a rapid excitatory response to foot contact. As training goes on, and the time of contact becomes more predictable, this compensation becomes less important. More evidence for adaptive spinal cord plasticity in response to particular demands in adult life is provided by studies of reflex changes in humans that occur during aging (Sabbahi and Sedgwick, 1982; DeVries et al., 1985; Koceja et al., 1995; Morita et al., 1995; Angulo-Kinzler et al., 1998; Zheng et al., 2000; Koceja and Mynark, 2000; Scaglioni et al., 2003; Kido et al., 2004). The agerelated modifications in reflex strength and taskrelated reflex modulation described in these studies and illustrated in Fig. 2D probably reflect both
direct and indirect effects of aging, i.e., direct effects on the neuronal circuitry of the reflexes and indirect effects secondary to the effects of aging elsewhere in the CNS or in the muscles, joints, and nerves that implement motor acts.
Sites and mechanisms of activity-dependent spinal cord plasticity The phenomena summarized in the last section show that activity-dependent changes in spinal cord function accompany the acquisition and maintenance of motor skills throughout life. However, these phenomena do not differentiate between the respective contributions to modifying spinal cord function of: (1) plasticity in the spinal cord itself; (2) changes in descending input to the spinal cord resulting from supraspinal plasticity; and (3) peripheral neuromuscular changes that alter sensory inputs or the effects of spinal cord output. Progress in understanding how plasticity of particular kinds at particular anatomical sites underlies the acquisition and maintenance of particular behaviors, depends on studies of the same spinal cord reflexes that are used to recognize this plasticity. Simple spinal cord reflexes, such as stretch reflexes and flexion withdrawal reflexes are normally activated as parts of complex behaviors. At the same time, they are in themselves simple behaviors, the simplest of which the mammalian CNS is capable; and adaptive changes in these reflexes are basically simple skills (i.e., ‘‘adaptive behaviors acquired through practice’’ (Compact Oxford Engish Dictionary, 1993)) that may be used as laboratory models of the plasticity underlying skill acquisition. Operant conditioning of the SSR, or its electrical analog the H-reflex, which has been described in monkeys, humans, rats, and mice, provides clear evidence of activity-dependent plasticity at particular sites in the spinal cord, and is elucidating the mechanisms and interactions of the spinal and supraspinal plasticity that underlies these simple skills (Wolpaw et al., 1983; Wolpaw, 1987; Evatt et al., 1989; Wolf and Segal, 1996; Segal, 1997; Wolpaw and Tennissen, 2001; Carp et al., 2005; Chen and Wolpaw, 2005; Wolpaw and Chen, 2006).
269
In the basic paradigm, used in monkeys, humans, rats, and mice, SSR or H-reflex amplitude is measured as electromyographic activity (EMG), and reward is given when the size is greater than (for up-conditioning) or less than (for down-conditioning) a specific criterion. The primary observation is that, after imposition of the reward criterion, the amplitude of the reflex changes appropriately over days and weeks (Fig. 3A). This adaptive change has two phases, a small rapid phase-1 change that occurs in the first few hours or days (believed to reflect rapid appropriate change in descending influence on the spinal reflex pathway), and a much slower phase-2 change that continues to grow for weeks (and appears to reflect gradual spinal cord plasticity caused by the longterm continuation of the descending influence responsible for phase 1) (Wolpaw and O’Keefe, 1984; Chen et al., 2001). This descending influence is conveyed by the corticospinal tract (CST); other major descending pathways are not needed (Chen and Wolpaw, 1997, 2002). Conditioning is possible in humans or rats with partial spinal cord injuries, but does not occur in people with strokes involving sensorimotor cortex or in rats in which contralateral sensorimotor cortex has been ablated or the CST has been transected (Segal and Wolf, 1994; Segal, 1997; Chen et al., 2002, 2006). This spinal cord plasticity includes alterations in motoneuron properties (Carp and Wolpaw, 1994; Halter et al., 1995; Carp et al., 2001a, b). Downconditioning is associated with a positive shift in motoneuron firing threshold and a decrease in axonal conduction velocity. Both of these changes suggest a positive shift in sodium channel activation voltage. As illustrated in Fig. 3B, the threshold change (and an accompanying small decrease in EPSP size) could largely account for the smaller reflex amplitude. Although activity-dependent synaptic plasticity has received the most attention in the past, the occurrence and importance of plasticity in neuronal properties (such as in neuronal voltage-gated ion channels) is now being recognized (Spitzer, 1999; Cantrell and Catterall, 2001; Carr et al., 2003). The shift in motoneuron firing threshold with H-reflex down-conditioning is an example of such plasticity, and indicates its behavioral significance. Other physiological and
anatomical studies suggest that SSR or H-reflex conditioning also changes the Ia afferent-motoneuron synapse, other synaptic terminals on the motoneuron (e.g., Fig. 3C), motor unit properties, and interneurons that carry oligosynaptic Group 1 input to the motoneuron (Carp and Wolpaw, 1995; Feng-Chen and Wolpaw, 1996; Carp et al., 2001b). The most recent studies have shown that the cerebellum is essential for acquisition and longterm maintenance of H-reflex down-conditioning, probably through its connections to sensorimotor cortex (Chen and Wolpaw, 2005; Wolpaw and Chen, 2006), and that the basal ganglia are essential at least for acquisition (Chen et al., 2004). Furthermore, these studies suggest that cortical plasticity induced and maintained by the cerebellum is responsible for the long-term survival of the spinal cord plasticity that is directly responsible for the altered H-reflex. Fig. 3D summarizes present understanding of the hierarchy of spinal and supraspinal plasticity that appears to underlie the acquisition and maintenance of this ostensibly simple skill of H-reflex conditioning.
Principles underlying activity-dependent plasticity and their relations to behavioral change The evidence that activity-dependent plasticity occurs at multiple spinal and supraspinal anatomical sites (with acquisition of even the simple skill of an H-reflex decrease or increase) is consistent with the rapidly increasing evidence that activity-dependent plasticity is ubiquitous in the CNS and occurs at multiple sites with simple learning in invertebrates and vertebrates (Wolpaw and Lee, 1989; Lieb and Frost, 1997; Thompson et al., 1997; Cohen et al., 1997; Carrier et al., 1997; Whelan and Pearson, 1997; Lisberger, 1998; Garcia et al., 1999; Hansel et al., 2001; King et al., 2001; Wolpaw and Tennissen, 2001; Medina et al., 2002; van Alphen and De Zeeuw, 2002; Vaynman and Gomez-Pinilla, 2005). Given that the primary CNS function is to produce appropriate behaviors and that activitydependent plasticity is ubiquitous in the CNS, complex plasticity, especially in the spinal cord, appears to be both necessary and inevitable.
270
Fig. 3. Operant conditioning of the spinal stretch reflex (SSR)/H-reflex pathway and the associated spinal and supraspinal plasticity. (A) The graphs show average poststimulus soleus EMG for representative days before (solid) and after (dotted) soleus H-reflex conditioning from a rat in which the H-reflex has been increased by the up-conditioning mode (left) or decreased by the downconditioning mode (right). The H-reflex is much larger after up-conditioning and much smaller after down-conditioning, while background EMG (indicated here by EMG at zero time) and M responses (i.e., direct muscle responses) are unchanged. (B) Triceps surae motoneurons on the conditioned side of H-reflex down-conditioned (HR) monkeys were found to have more positive firing thresholds and slightly smaller Ia EPSPs. Together, these two findings can explain why the H-reflex became smaller (A and B from Wolpaw, 1997). (C) Effects of successful down-conditioning on GABAergic terminals on soleus motoneurons (assessed by glutamic acid decarboxylase (GAD67)-immunoreactivity (GAD-IR). Top left: Soleus motoneuron labeled by Alexa Fluor-488 conjugated with CTB injected into the muscle. Bottom left: Same motoneuron showing GAD-IR (i.e., GABAergic) terminals (dark) located on the periphery of the motoneuron.(Bar ¼ 10 Fm.). Right: Average (7SEM) values for: down-conditioning successful (DS), down-conditioning failed (DF), and naive control (NC) rat groups for: number of GABAergic terminals per motoneuron; terminal GAD density; and GABAergic terminal coverage of soma (expressed as percent of perimeter). (***Po0.0001; compared to the NC group.) After successful down-conditioning, soleus motoneurons have more GABAergic terminals, and these terminals are more densely labeled and occupy more of the soma (from Wang et al., 2006)). (D) Spinal and supraspinal sites (shaded ovals) of plasticity associated with operant conditioning of the SSR or its electrical analog, the H-reflex. ‘‘MN’’ is the motoneuron, ‘‘CST’’ is the main corticospinal tract, and each ‘‘IN’’ is one or more spinal interneuron types. Open synaptic terminals are excitatory, solid ones are inhibitory, half-open ones could be either, and the subdivided one is a cluster of C terminals. Dashed pathways imply the possibility of intervening spinal interneurons. The monosynaptic and probably oligosynaptic SSR/H-reflex pathway from Ia and Ib afferents to the motoneuron is shown. Definite (heavy shading) or highly probable (light shading) sites of plasticity include: the motoneuron membrane (i.e. firing threshold and axonal conduction velocity), motor unit properties, C terminals on the motoneuron, the Ia afferent synaptic connection, and terminals conveying disynaptic group I inhibition or excitation to the motoneuron. The essential roles of the corticospinal tract (originating in sensorimotor cortex) and of cerebellar output to cortex are indicated (updated from Wolpaw, 1997). See Plate 17.3 in Colour Plate Section.
Along with its homologous brainstem nuclei, the spinal cord is the final assembly point for neuromuscular behaviors, both simple and complex. It is, to use the term Sherrington originally applied
to the motoneuron itself, ‘‘the final common path’’ (Clarke and O’Malley, 1996, p. 375), assembling and executing the end product of activity elsewhere in the CNS. For example, the motoneurons,
271
interneurons, and synapses of the lumbosacral spinal cord produce all the many kinds of locomotion and postural maintenance, withdraw the legs from painful stimuli, contribute to actions involving all four limbs, support bladder and bowel function, and produce numerous specialized actions. The fact that the spinal cord supports these many behaviors, and incorporates new ones throughout life, implies that its neuronal and synaptic elements are continually adjusted and readjusted to serve the current repertoire of behaviors. That such adjustments take place in the short-term as the organism shifts from one behavior to another or cycles through the different phases of a single behavior is indicated by studies like those demonstrating the changes in presynaptic inhibition from standing to walking to running, and the changes in responses to primary afferent input that occur over the step cycle (Capaday and Stein, 1987; Stein, 1995; Faist et al., 1996; Rossignol, 1996; Pearson and Ramirez, 1997; Zehr, 2002). As the data reviewed above indicate, both long-term and short-term adjustments occur. Gradual activity-dependent plasticity, initiated and guided by descending and peripheral inputs, presumably maintains spinal cord circuitry in a state appropriate for its current roster of behaviors. The long-term control – a consensus resulting from the different patterns of activity associated with the different behaviors – seems to operate as a coarse adjustment, setting ranges over which fine adjustments specific to each behavior are made. For example, at a given time, the strength of primary afferent input to soleus motoneurons has a range that includes values appropriate for standing, walking, and running (Zehr, 2002). In this situation, the neural activity that adds a new behavior to the repertoire is likely to induce plasticity that supports the new behavior and, in addition, plasticity that preserves the old behaviors. This is true whether the activity results from daily practice and the behavior is an athletic skill or the activity is produced by peripheral or central damage and the behavior reflects or compensates for a functional abnormality. For example, the increased motoneuron response to primary afferent input that underlies H-reflex up-conditioning (e.g., Fig. 3A (left)) affects other behaviors that involve the same input pathway (Chen et al.,
2005a) (Fig. 4). These effects may lead to additional ‘‘compensatory’’ activity-dependent plasticity that restores these other behaviors. The still mysterious plasticity that maintains a normal contralateral H-reflex in a monkey that has undergone H-reflex down-conditioning (Wolpaw and Lee, 1989; Wolpaw et al., 1993) could be compensatory, serving to preserve normal contralateral function. In addition, because activity-dependent plasticity may occur at many places in the spinal cord, the modifications in activity caused by the plasticity responsible for a new behavior or for maintaining old behaviors are likely to lead to further ‘‘reactive’’ plasticity at other sites. Thus, the smaller stretch reflexes in the ostensibly unaffected arm contralateral to an arm paralyzed by a hemispheric stroke (Thilmann et al., 1990), may result from reactive plasticity caused by change in activity in the spinal pathways that connect the two sides of the spinal cord. These considerations imply that acquisition of a new behavior probably involves three categories of plasticity: primary plasticity that supports the new behavior, compensatory plasticity that preserves old behaviors in spite of the impact of the primary plasticity, and reactive plasticity that is caused by changes in activity due to primary and compensatory plasticity. This etiological categorization explains the multisite plasticity that accompanies even the simplest behavioral change. It indicates that multisite plasticity is necessary to preserve the complete repertoire of behaviors and inevitable due to the ubiquitous capacity of the CNS for activity-dependent plasticity. It also helps to explain why some instances of plasticity (e.g., the contralateral spinal cord plasticity with H-reflex conditioning (Wolpaw and Lee, 1989)) may bear no ostensible relationship to the behavioral changes that they accompany. Recognition of these three etiological categories of plasticity may help define factors affecting the efficacy of new rehabilitation methods (see below). The normal activity-dependent spinal cord plasticity occurring during development, during skill acquisition later in life, or induced in the laboratory, is largely created and guided by descending input from the brain, sometimes accompanied by associated peripheral inputs to the spinal cord.
272
Fig. 4. H-reflexes during the stance phase of locomotion and right soleus bursts during undisturbed locomotion before (solid) and after (dotted) conditioning from a down-conditioned (HRdown) and an up-conditioned (HRup) rat. The stance H-reflexes are each the average of 109–166 trials and the stance bursts are each the average of 131–462 bursts. After conditioning, both the stance H-reflex and the soleus burst are smaller in the HRdown rat and larger in the HRup rat. (from Chen et al., 2005a).
This descending influence is likely to be associated with, and may often depend on, plasticity in cortex, cerebellum, and/or other brain areas. The behavioral effects associated with spinal cord plasticity seem to reflect the complex interactions of plasticity at multiple spinal and supraspinal locations (Carrier et al., 1997; Whelan and Pearson, 1997; Wolpaw and Tennissen, 2001; Wolpaw and Chen, 2006). Theoretical significance of activity-dependent spinal cord plasticity The spinal cord’s impressive capacity for activitydependent plasticity suggests that most motor
skills, particularly those acquired through prolonged practice, depend to some degree on spinal cord plasticity. This inference is consistent with the strong evidence that activity can gradually change the spinal cord. It seems to explain why intense practice over a long time is needed for acquisition and maintenance of athletic skills and other motor skills such as playing the piano and other musical instruments. In fact, such skills probably cannot be fully understood solely on the basis of plasticity in cerebral cortex, cerebellum, or other brain areas. The plasticity in the spinal cord must also be taken into account. The role of spinal cord plasticity has rarely been accorded any, much less adequate, recognition in studies focused on the role
273
of cortical plasticity in accounting for new behaviors. The fact that spinal cord plasticity can, and probably often does, contribute to normal and pathological behavioral changes complicates investigations of such changes. Nevertheless, without adequate recognition of and attention to spinal cord plasticity, efforts to clarify how CNS plasticity explains behavioral changes are likely to yield only incomplete or misleading insights. The fact that motor skills depend on multisite and multilevel (i.e., spinal and supraspinal) plasticity suggests that intellectual skills, like language mastery or mathematical facility, may also depend on widely distributed plasticity. The rapid behavioral changes that traditionally engage most research attention, like single-trial acquisition of a new word, may merely reflect small adjustments in patterns of plasticity slowly created by prolonged practice, adjustments comparable to the change in presynaptic inhibition associated with the transition from standing to running, or the change in descending input that produces phase-1 change in the SSR (see above). As a result, elucidation of many skills may require investigation of gradually acquired activity-dependent plasticity in the brain similar to that found in the spinal cord. Indeed, such investigation might logically start with simple motor skills and with the spinal cord, for its relative simplicity and accessibility and its well-defined connections to the brain facilitate studies of activity-dependent plasticity and of how multiple sites of plasticity conspire to produce behavioral changes.
Possible clinical uses of activity-dependent spinal cord plasticity In developing new rehabilitation methods for long-term neuromuscular disorders such as spinal cord injury or cerebral palsy, the spinal cord’s capacity for activity-dependent plasticity is both a challenge and an opportunity. On the one hand, this capacity may add to the disabilities that follow spinal cord injury and will certainly influence the results of new therapeutic methods that induce regeneration of spinal cord pathways and neurons. On the other hand, it offers the
opportunity to induce and guide restoration of function, and could enable imperfect regeneration to support significant functional recovery. For both reasons, the effective engagement of activitydependent plasticity in the spinal cord is likely to be a key part of new rehabilitation programs for people with spinal cord injuries or other chronic neuromuscular disorders. Therapeutic initiation and guidance of activity-dependent spinal cord plasticity will depend on training protocols that create appropriate patterns of peripheral and descending inputs to the spinal cord. The development of such methods has just begun for locomotion (see Chapter 15 and 16). Other important behaviors, such as urination, have not yet been addressed. Poirrier et al. (2004) evaluated the effect of repetitive transcranial magnetic stimulation (rTMS) on locomotion in rats with high or low thoracic spinal cord compression injuries. rTMS (10 Hz 5 s every 2 min for 20 min) was delivered 5 d/wk for 8 weeks. In rats with low thoracic injury, locomotion improved substantially more in TMS rats than in unstimulated rats. Fig. 5 shows this result. In contrast, rats with high thoracic injury, which normally recovered better than those with low thoracic injury, showed no benefit from TMS. Supported by histological data, the investigators hypothesize that TMS improves locomotion in rats with low thoracic injuries by increasing the strength of descending serotonergic input to the lumbosacral locomotor pattern generator. Another recent study (Chen et al., 2005b) explored the possibility that operant conditioning of a spinal reflex can improve locomotion in spinal cord-injured rats. Mid-thoracic transection of the right lateral column (LC) caused a persistent asymmetry in treadmill locomotion. Rats were then either exposed or not exposed to an H-reflex up-conditioning protocol that greatly increased right soleus motoneuron response to primary afferent input, and locomotion was reevaluated. H-reflex up-conditioning, which increased the right soleus burst, eliminated the locomotor asymmetry. In contrast, the asymmetry persisted in the unconditioned rats. These results suggest that appropriate reflex conditioning protocols could improve function in people with
274
Fig. 5. BBB scores (mean7SEM) (reflecting locomotor function (Basso et al., 1995)) for control (N ¼ 6) and rTMS-treated (N ¼ 6) rats with low thoracic spinal cord lesions. BBB scores were significantly higher for the stimulated rats from the second week postinjury onward (from Poirrier et al., 2004.).
partial spinal cord injuries. They could be especially useful when regeneration becomes possible and precise methods for re-educating the regenerated spinal cord become essential for restoring effective function. These endeavors will be greatly affected by the complexity of the activity-dependent plasticity that accompanies even simple training protocols (e.g., Fig. 3D). They will also be affected by a distinctive feature of activity-dependent spinal cord plasticity as it functions in normal life and in response to disease: the slow rate of its impact on behavior. In spite of the rapidity of activity-dependent processes like LTP (which are known to occur in the spinal cord), the behavioral changes due to activity-dependent spinal cord plasticity develop gradually, probably because each one is the result of multiple activity-dependent processes. Changes in spinal reflexes during development and during the learning of skills like ballet occur gradually over months and years; those caused by H-reflex operant conditioning or other specialized training regimens occur over days and weeks. Although
reflexes like the H-reflex can differ substantially between established behaviors (e.g., standing and running (Zehr, 2002)), or even between different phases of one behavior (e.g., stance and swing phases of walking (Faist et al., 1996)), the specification of the reflex strengths associated with a particular behavior develops slowly. This feature is probably fortunate: rapid, large changes in specific reflexes independent of the contexts of specific behaviors could wreak havoc with motor control and necessitate prodigious supraspinal compensation. At the same time, the characteristically gradual effect of activity-dependent spinal cord plasticity on behavior implies that laboratory studies and clinical applications need to extend over sufficient time periods. Furthermore, the ubiquity of activity-dependent plasticity and the inevitable interactions among primary, compensatory, and reactive plasticity, imply that functional effects are likely to change over time. Early gains will not always evolve into long-term improvements; and, conversely, early deleterious effects may give way to long-term benefits.
275
Acknowledgments Dr. Ann M. Tennissen provided invaluable assistance in the preparation of the manuscript. Work in the author’s laboratory has been supported by NIH (NS22189 and HD36020), The United Cerebral Palsy Research and Educational Foundation, The Paralyzed Veterans of America, The International Spinal Research Trust, and The Christopher Reeve Paralysis Foundation.
References Angulo-Kinzler, R.M., Mynark, R.G. and Koceja, D.M. (1998) Soleus H-reflex gain in elderly and young adults: Modulation due to body position. J. Geron. Series A, Biol. Sci. Med. Sci., 53: M120–M125. Auge´ II, W.K. and Morrison, D.S. (2000) Assessment of the infraspinatus spinal stretch reflex in the normal, athletic, and multidirectionally unstable shoulder. Am. J. Sports Med., 28: 206–213. Baldissera, F., Hultborn, H. and Illert, M. (1981) Integration in spinal neuronal systems. In: Brooks, V.B. (Ed.), Handbook of Physiology, Section I: The Nervous System. Vol. 2, Motor Control, Part I. Williams and Wilkins Company, Baltimore, MD, pp. 509–595. Barbeau, H. and Rossignol, S. (1987) Recovery of locomotion after chronic spinalization in the adult cat. Brain Res., 412: 84–95. Barbeau, H., Ladouceur, M., Mirbagheri, M.M. and Kearney, R.E. (2002) The effect of locomotor training combined with functional electrical stimulation in chronic spinal cord injured subjects: walking and reflex studies. Brain Res. Rev., 40: 274–291. Basso, D.M., Beattie, M.S. and Bresnahan, J.C. (1995) A sensitive and reliable locomotor rating scale for open field testing in rats. J. Neutotrauma, 12: 1–21. Boorman, G., Hulliger, M., Lee, R.G., Tako, K. and Tanaka, R. (1991) Reciprocal Ia inhibition in patients with spinal spasticity. Neurosci. Lett., 127: 57–60. Brown, W.F. (1984) The Physiological and Technical Basis of Electromyography. Butterworths, Boston, MA. Burke, R.E. (1998) Spinal Cord: Ventral Horn. In: Shepherd, G.M. (Ed.) The Synaptic Organization of the Brain, Vol. 3. Oxford University Press, New York, pp. 77–120. Bussel, B., Roby-Brami, A., Azouvi, P., Biraben, A., Yakovleff, A. and Held, J.P. (1988) Myoclonus in a patient with spinal cord transection: possible involvement of the spinal stepping generator. Brain, 111: 1235–1245. Calancie, B., Needham-Shropshire, B., Jacobs, P., Willer, K., Zych, G. and Green, B.A. (1994) Involuntary stepping after chronic spinal cord injury: evidence for a central rhythm generator for locomotion in man. Brain, 117: 1143–1159.
Cantrell, A.R. and Catterall, W.A. (2001) Neuromodulation of Na+ channels: An unexpected form of cellular plasticity. Nat. Rev. Neurosci., 2: 397–407. Capaday, C. and Stein, R.B. (1987) A method for stimulating the reflex output of a motoneuron pool. J. Neurosci. Meth., 21: 91–104. Carp, J.S. and Wolpaw, J.R. (1994) Motoneuron plasticity underlying operantly conditioned decrease in primate H-reflex. J. Neurophysiol., 72: 431–442. Carp, J.S. and Wolpaw, J.R. (1995) Motoneuron properties after operantly conditioned increase in primate H-reflex. J. Neurophysiol., 73: 1365–1373. Carp, J.S., Chen, X.Y., Sheikh, H. and Wolpaw, J.R. (2001a) Operant conditioning of rat H-reflexes affects motoneuronal axonal conduction velocity. Exp. Brain Res., 136: 69–73. Carp, J.S., Chen, X.Y., Sheikh, H. and Wolpaw, J.R. (2001b) Motor unit properties after operant conditioning of rat Hreflex. Exp. Brain Res., 140: 382–386. Carp, J.S., Tennissen, A.M., Chen, X.Y. and Wolpaw, J.R. (2005) Operant conditioning of H-reflex in mice. Program No. 175.2. Abstract Viewer/Itinerary Planner. Washington, DC: Soc. Neurosci., 2005. Online. Carr, D.B., Day, M., Cantrell, A.R., Held, J., Scheuer, T., Catterall, W.A. and Surmeier, D.J. (2003) Transmitter modulation of slow, activity-dependent alterations in sodium channel availability endows neurons with a novel form of cellular plasticity. Neuron, 39: 793–806. Carrier, L., Brustein, E. and Rossignol, S. (1997) Locomotion of the hindlimbs after neurectomy of ankle flexors in intact and spinal cats: model for the study of locomotor plasticity. J. Neurophysiol., 77: 1979–1993. Casabona, A., Polizzi, M.C. and Perciavalle, V. (1990) Differences in H-reflex between athletes trained for explosive contraction and non-trained subjects. Eur. J. Appl. Physiol., 61: 26–32. Chamberlain, T., Halick, P. and Gerard, R.W. (1963) Fixation of experience in the rat spinal cord. J. Neurophysiol., 22: 662–673. Chen, X.Y., Carp, J.S., Chen, L. and Wolpaw, J.R. (2006) Sensorimotor cortex ablation prevents H-reflex up-conditioning and causes a paradoxical response to down-conditioning in rats. J. Neurophysiol., 96: 119–127. Chen, Y., Chen, X.Y., Jakeman, L.B., Schalk, G., Stokes, B.T. and Wolpaw, J.R. (2005a) The interaction of a new motor skill and an old one: H-reflex conditioning and locomotion in rats. J. Neurosci., 25: 6898–6906. Chen, Y., Chen, X.Y., Jakeman, L.B., Stokes, B.T., Liu, R.L. and Wolpaw, J.R. (2005b) Reeducating the injured spinal cord by operant conditioning of a reflex pathway. Program No. 175.4. Abstract Viewer/Itinerary Planner. Washington, DC: Soc. Neurosci., 2005. Online. Chen, L., Chen, Y., Liu, R.L., Chen, X.Y. and Wolpaw, J.R. (2004) Bilateral globus pallidus ablation in rats prevents down-conditioning of H-reflex. Program No. 417.3. Abstract Viewer/Itinerary Planner. San Diego: Soc. Neurosci., Abst. 2004. Online. Chen, X.Y., Chen, L. and Wolpaw, J.R. (2001) Time course of H-reflex conditioning in the rat. Neurosci. Lett., 302: 85–88.
276 Chen, X.Y. and Wolpaw, J.R. (1995) Operant conditioning of H-reflex in freely moving rats. J. Neurophysiol., 73: 411–415. Chen, X.Y. and Wolpaw, J.R. (1997) Dorsal column but not lateral column transection prevents down conditioning of H-reflex in rats. J. Neurophysiol., 78: 1730–1734. Chen, X.Y. and Wolpaw, J.R. (2002) Probable corticospinal tract control of spinal cord plasticity in rats. J. Neurophysiol., 87: 645–652. Chen, X.Y. and Wolpaw, J.R. (2005) Ablation of cerebellar nuclei prevents H-reflex down-conditioning in rats. Learn. Mem., 12: 248–254. Chen, L., Wolpaw, J.R., Smith, B. and Chen, X.Y. (2002) Sensorimotor cortex ablation prevents H-reflex down conditioning. Program No.66.15.2002 Abstract Viewer/Itinerary Planner. Washington, DC: Soc. Neurosci., CD-ROM. Clarac, F., Pearlstein, E., Pflieger, J.F. and Vinay, L. (2004) The in vitro neonatal rat spinal cord preparation: a new insight into mammalian locomotor mechanisms. J. Comp. Physiol. A, Neuroeth. Sensory, Neural, Beh. Physiol., 190: 343–357. Clarke, E. and O’Malley, C.D. (1996) The Human Brain and Spinal Cord. Norman Publishing, San Francisco, CA, p. 375. Cohen, T.E., Kaplan, S.W., Kandel, E.R. and Hawkins, R.D. (1997) A simplified preparation for relating cellular events to behavior: mechanisms contributing to habituation, dishabituation, and sensitization of the Aplysia gill-withdrawal reflex. J. Neurosci., 17: 2886–2899. Compact Oxford English Dictionary (1993) In: Weiner E.S.C. and Simpson J.A. (Eds.), Ed. 2. Oxford University Press, Oxford, p. 1782. Cope, T.C., Bodine, S.C., Fournier, M. and Edgerton, V.R. (1986) Soleus motor units in chronic spinal transected cats: physiological and morphological alterations. J. Neurophysiol., 55: 1202–1220. de Groat, W.C. (2002) Plasticity of bladder reflex pathways during postnatal development. Physiol. Beh., 77: 689–692. DeVries, H.A., Wiswell, R.A., Romero, G.T. and Heckathorne, E. (1985) Changes with age in monosynaptic reflexes elicited by mechanical and electrical stimulation. Am. J. Phys. Med. Rehabil., 64: 71–81. Dietz, V. (2003) Spinal cord pattern generators for locomotion. Clin. Neurophysiol., 114: 1379–1389. Dietz, V., Colombo, G., Jensen, L. and Baumgartner, L. (1995) Locomotor capacity of spinal cord in paraplegic patients. Ann. Neurol., 37: 574–586. DiGiorgio, A.M. (1929) Persistenza nellanimale spinale, di asymmetrie posturali e motorie di origine cerebellare: I, II, III. Archivio di Fisiologia., 27: 518–580. DiGiogio, A.M. (1942) Azione del cervelletto-neocerebellumsul tono posturale degli arti e localizzazioni cerebellari nellanimale rombencefalico. Archivio di Fisiologia., 42: 25–79. Dimitrijevic, M.R., Gerasimenko, Y. and Pinter, M.M. (1998) Evidence for a spinal central pattern generator in humans. Ann. NY Acad. Sci., 860: 360–376. Ding, Y., Kastin, A.J. and Pan, W. (2005) Neural plasticity after spinal cord injury. Curr. Pharma. Des., 11: 1441–1450. Dobkin, B., Harkema, S., Requejo, P.S. and Edgerton, V.R. (1995) Modulation of locomotor-like EMG activity in
subjects with complete and incomplete spinal cord injury. J. Neurol. Rehabil., 9: 183–190. Dobkin, B.H. and Havton, L.A. (2004) Basic advances and new avenues in therapy of spinal cord injury. Ann. Rev. Med., 55: 255–282. Dupont-Versteegden, E.E., Houle´, J.D., Dennis, R.A., Zhang, J., Knox, B.S., Wagoner, G. and Peterson, C.A. (2004) Exercise-induced gene expression in soleus muscle is dependent on time after spinal cord injury in rats. Muscle Nerve, 29: 73–81. Edgerton, V.R., Tillakaratne, N.J.K., Bigbee, A.J., de Leon, R.D. and Roy, R.R. (2004) Plasticity of the spinal neural circuitry after injury. Ann. Rev. Neurosci., 27: 145–167. Evatt, M.L., Wolf, S.L. and Segal, R.L. (1989) Modification of human spinal stretch reflexes: Preliminary studies. Neurosci. Lett., 105: 350–355. Eyre, J.A., Miller, S., Clowry, G.J., Conway, E.A. and Watts, C. (2000) Functional corticospinal projections are established prenatally in the human foetus permitting involvement in the development of spinal motor centres. Brain, 123: 51–64. Eyre, J.A., Taylor, J.P., Villagra, F., Smith, M. and Miller, S. (2001) Evidence of activity-dependent withdrawal of corticospinal projections during human development. Neurol., 57: 1543–1554. Eyre, J.A. (2003) Development and plasticity of the corticospinal system in man. Neural Plast., 10: 93–106. Faist, M., Dietz, V. and Pierrot-Deseilligny, E. (1996) Modulation, probably presynaptic in origin, of monosynaptic Ia excitation during human gait. Exp. Brain Res., 109: 441–449. Feng-Chen, K.C. and Wolpaw, J.R. (1996) Operant conditioning of H-reflex changes synaptic terminals on primate motoneurons. Proc. Natl. Acad. Sci. USA, 93: 9206–9211. Flor, H. and Andrasik, F. (2006) Chronic pain. In: Selzer, M.E., Clarke, S., Cohen, L.G., Duncan, P.W. and Gage, F.H. (Eds.), Textbook of Neural Repair and Rehabilitation. Cambridge University Press, Cambridge, UK, pp. 219–230. Garcia, K.S., Steele, P.M. and Mauk, M.D. (1999) Cerebellar cortex lesions prevent acquisition of conditioned eyelid responses. J. Neurosci., 19: 10940–10947. Gerard, R.W. (1961) The fixation of experience. In: Gerard, R.W. and Konorski, J. (Eds.), Brain Mechanisms and Learning. Blackwell, Oxford, pp. 21–32. Gibson, C.L., Arnott, G.A. and Clowry, G.J. (2000) Plasticity in the rat spinal cord seen in response to lesions to the motor cortex during development but not to lesions in maturity. Exp. Neurobiol., 166: 422–434. Giulio, L. (1952) Sulla funzione del midollo spinale. Persistenza di asimmetrie da eccitamento labirintico. Bollettino—Societa Italiana Biologia Sperimentale., 28: 1651–1652. Gomez-Pinilla, F., Ying, Z., Roy, R.R., Molteni, R. and Edgerton, V.R. (2002) unitary exercise induces a BDNFmediated mechanism that promotes neuroplasticity. J. Neurophysiol., 88: 2187–2195. Goode, D.J. and Van Hoven, J. (1982) Loss of patellar and Achilles tendon reflex in classical ballet dancers. Arch. Neurol., 39: 323.
277 Grillner, S. and Walle´n, P. (2004) Innate versus learned movements—a false dichotomy? Prog. Brain Res., 143: 3–12. Grillner, S., Markram, H., De Schutter, E., Silberberg, G. and LeBeau, F.E. (2005) Microcircuits in action—from CPGs to neocortex. Trends Neurosci., 28: 525–533. Halter, J.A., Carp, J.S. and Wolpaw, J.R. (1995) Operantly conditioned motoneuron plasticity: possible role of sodium channels. J. Neurophysiol., 74: 867–871. Hansel, C., Linden, D.J. and D’Angelo, E. (2001) Beyond parallel fiber LTD: the diversity of synaptic and non-synaptic plasticity in the cerebellum. Nat. Neurosci., 4: 467–475. Henneman, E. and Mendell, L.M. (1981) Functional organization of motoneuron pool and inputs. In: Brooks V.B. (Ed.), Handbook of Physiology, Section I: The Nervous System, Vol. 2, Motor Control, Part I. Williams and Wilkins, Baltimore, MD, pp. 423–507. Herrero, J.F., Laird, J.M. and Lopez-Garcia, J.A. (2000) Windup of spinal cord neurones and pain sensation: much ado about something? Prog. Neurobiol., 61: 169–203. Hiersemenzel, L.P., Curt, A. and Dietz, V. (2000) From spinal shock to spasticity: neuronal adaptations to a spinal cord injury. Neurol., 54: 1574–1582. Hochman, S. and McCrea, D.A. (1994a) Effects of chronic spinalization on ankle extensor motoneurons II. Motoneuron electrical properties. J. Neurophysiol., 71: 1468–1479. Hochman, S. and McCrea, D.A. (1994b) Effects of chronic spinalization on ankle extensor motoneurons I. Composite monosynaptic Ia EPSPs in four motoneuron pools. J. Neurophysiol., 71: 1452–1467. Hochman, S. and McCrea, D.A. (1994c) Effects of chronic spinalization on ankle extensor motoneurons III. Composite Ia EPSPs in motoneurons separated into motor unit types. J. Neurophysiol., 71: 1480–1490. Holmes, G. (1915) Spinal injuries of warfare. Brit. Med. J., 2: 815–821. Hultborn, H. (2003) Changes in neuronal properties and spinal reflexes during development of spasticity following spinal cord lesions and stroke: studies in animal models and patients, J. Rehabil. Med., 41(Suppl): 46–55. Inglis, F.M., Zuckerman, K.E. and Kalb, R.G. (2000) Experience-dependent development of spinal motor neurons. Neuron, 26: 299–305. Ji, R., Kohno, T., Moore, K.A. and Woolf, C.J. (2003) Central sensitization and LTP: Do pain and memory share similar mechanisms? Trends Neurosci., 26: 696–705. Juvin, L., Simmers, J. and Morin, D. (2005) Propriospinal circuitry underlying interlimb coordination in mammalian quadrupedal locomotion. J. Neurosci., 25: 6025–6035. Kandel, E.R. (1977) Neuronal plasticity and the modification of behavior. In: Brookhart J.M. and Mountcastle V.B. (Eds.), Handbook of Physiology, Section I: The Nervous System, Vol. 1, Cellular Biology of Neurons. Williams and Wilkins, Bethesda, MD, pp. 1137–1382. Kandel, E.R., Schwartz, J.H. and Jessell, T.M. (2000) Principles of Neural Science. McGraw-Hill, New York. Katz, W.A. and Rothenberg, R. (2005) Section 3: The nature of pain: pathophysiology. J. Clin. Rheum., 11(2 Suppl): S11–S15.
Kido, A., Tanaka, N. and Stein, R.B. (2004) Spinal excitation and inhibition decrease as humans age. Can. J. Physiol. Pharm., 82: 238–248. Kiehn, O. and Butt, S.J. (2003) Physiological, anatomical and genetic identification of CPG neurons in the developing mammalian spinal cord. Prog. Neurobiol., 70: 347–361. Kiehn, O., Harris-Warrick, R.M., Jordan, L.M., Hultborn, H. and Kudo, N. (Eds.). (1998) Neuronal Mechanisms for generating locomotor activity. New York Academy of Sciences, New York. King, D.A.T., Krupa, D.J., Foy, M.R. and Thompson, R.F. (2001) Mechanisms of neuronal conditioning. Int. Rev. Neurobiol., 45: 313–337. Koceja, D.M., Burke, J.R. and Kamen, G. (1991) Organization of segmental reflexes in trained dancers. Int. J. Sports Med., 12: 285–289. Koceja, D.M., Davison, E. and Robertson, C.T. (2004) Neuromuscular characteristics of endurance- and power-trained athletes. Res. Quart. Exererc. Sport, 75: 23–30. Koceja, D.M., Markus, C.A. and Trimble, M.H. (1995) Postural modulation of the soleus H reflex in young and old subjects. Electroencephalogr. Clin. Neurophysiol., 97: 387–393. Koceja, D.M. and Mynark, R.G. (2000) Comparison of heteronymous monosynaptic Ia facilitation in young and elderly subjects in supine and standing positions. Int. J. Neurosci., 103: 1–17. Kuhn, R.A. (1950) Functional capacity of the isolated human spinal cord. Brain, 1: 1–51. Levinsson, A., Luo, X.L., Holmberg, H. and Schouenborg, J. (1999) Developmental tuning in a spinal nociceptive system: effects of neonatal spinalization. J. Neurosci., 19: 10397–10403. Lieb, J.R. and Frost, W.N. (1997) Realistic simulation of the Aplysia siphon-withdrawal reflex circuit: roles of circuit elements in producing motor output. J. Neurophysiol., 77: 1249–1268. Lisberger, S.G. (1998) Physiologic basis for motor learning in the vestibulo-ocular reflex. Otolaryng. — Head Neck Surg., 119: 43–48. Liu, X. and Sandku¨hler, J. (1997) Characterization of long-term potentiation of C-fiber-evoked potentials in spinal dorsal horn of adult rat: essential role of NK1 and NK2 receptors. J. Neurophysiol., 78: 1973–1982. Liverman, C.T., Altevogt, B.M., Joy, J.E. and Johnson, R.T. (Eds.). (2005) Spinal Cord Injury, Progress, Promise, and Priorities. The National Academies Press, Washington, DC. Lloyd, D.P.C. (1949) Post-tetanic potentiation of response in monosynaptic reflex pathways of the spinal cord. J. Genl. Physiol., 33: 147–170. Lovely, R.G., Gregor, R.J., Roy, R.R. and Edgerton, V.R. (1986) Effects of training on the recovery of full-weightbearing stepping in the adult spinal cat. Exp. Neurol., 92: 421–435. Magladery, J.W., Porter, W.E., Park, A.M. and Teasdall, R.D. (1951) Electrophysiological studies of nerve and reflex
278 activity in normal man. IV. The two-neuron reflex and identification of certain action potentials from spinal roots and cord. Bull. Johns Hopkins Hosp., 88: 499–519. Manni, E. (1950) Localizzazoni cerebellari corticali nella cavia, Nota 1: II ‘‘corpus cerebelli’’. Archivio di Fisiologia, 49: 213–237. Martin, J.H. (2005) The corticospinal system: from development to motor control. Neuroscientist, 11: 161–173. Martin, J.H., Choy, M., Pullman, S. and Meng, Z. (2004) Corticospinal system development depends on motor experience. J. Neurosci., 24: 2122–2132. Matthews, P.B.C. (1972) Mammalian Muscle Receptors and Their Central Actions. Williams and Wilkins, Baltimore, MD. Mayer, N.H. and Esquenazi, A. (2003) Muscle overactivity and movement dysfunction in the upper motoneuron syndrome. Phys. Med. Rehabil. Clin. NA., 14: 855–883. Medina, J.F., Repa, J.C., Mauk, M.D. and LeDoux, J.E. (2002) Parallels between cerebellum- and amygdala-dependent conditioning. Nat. Rev. Neurosci., 3: 122–131. Melzack, R., Coderre, T.J., Katz, J. and Vaccarino, A.L. (2004) Central neuroplasticity and pathological pain. Ann. NY Acad. Sci., 933: 157–174. Mendell, L.M. (1984) Modifiability of spinal synapses. Physiol. Rev., 64: 260–324. Mendell, L.M., Munson, J.B. and Arvanian, V.L. (2001) Neurotrophins and synaptic plasticity in the mammalian spinal cord. J. Physiol., 533: 91–97. Meyer-Lohmann, J., Christakos, C.N. and Wolf, H. (1986) Dominance of the short-latency component in perturbation induced electromyographic responses of long-trained monkeys. Exp. Brain Res., 64: 393–399. Misiaszek, J.E. (2003) The H-reflex as a tool in neurophysiology: its limitations and uses in understanding nervous system function. Muscle Nerve, 28: 144–160. Mountcastle, V.B. (1980) Effects of spinal cord transaction. In: Mountcastle, V.B. (Ed.) Medical Physiology, Vol. 1. Mosby, St. Louis, MO, pp. 781–786. Morita, H., Shindo, M., Yanagawa, S., Yoshida, T., Momoi, H. and Yanagisawa, N. (1995) Progressive decrease in heteronymous monosynaptic la facilitation with human ageing. Exp. Brain Res., 104: 167–170. Muir, G.D. and Steeves, J.D. (1997) Sensorimotor stimulation to improve locomotor recovery after spinal cord injury. Trends Neurosci., 20: 72–77. Munson, J.B., Foehring, R.C., Lofton, S.A., Zengel, J.E. and Sypert, G.W. (1986) Plasticity of medial gastrocnemius motor units following cordotomy in the cat. J. Neurophysiol., 55: 619–634. Myklebust, B.M., Gottlieb, G.L., Penn, R.L. and Agarwal, G.C. (1982) Reciprocal excitation of antagonistic muscles as a differentiating feature in spasticity. Ann. Neurol., 12: 367–374. Myklebust, B.M., Gottlieb, G.L. and Agarwal, G.C. (1986) Stretch reflexes of the normal human infant. Dev. Med. Child Neurol., 28: 440–449.
Nelson, S.G. and Mendell, L.M. (1979) Enhancement in Iamotoneuron synaptic transmission caudal to chronic spinal cord transection. J. Neurophysiol., 42: 642–654. Nielsen, J., Crone, C. and Hultborn, H. (1993) H-reflexes are smaller in dancers from the Royal Danish Ballet than in welltrained athletes. Eur. J. Appl. Physiol., 66: 116–121. Orlovsky, G.N., Deliagina, T.G. and Grillner, S. (1999) Neuronal Control of Locomotion from Mollusc to Man. Oxford University Press, New York. O’Sullivan, M.C., Miller, S., Ramesh, V., Conway, E., Gilfillan, K., McDonough, S. and Eyre, J.A. (1998) Abnormal development of biceps brachii phasic stretch reflex and persistence of short latency heteronymous reflexes from biceps to triceps brachii in spastic cerebral palsy. Brain, 121: 2381–2395. Patterson, M.M. (1976) Mechanisms of classical conditioning and fixation in spinal mammals. Adv. Psychobiol., 3: 381–436. Pearson, K.G. and Ramirez, J.M. (1997) Sensory modulation of pattern-generating circuits. In: Stein, P.S.G., et al. (Eds.), Neurons, Networks and Motor Behavior. MIT Press, Cambridge, MA, pp. 225–235. Pe´rot, C., Goubel, F. and Mora, I. (1991) Quantification of Tand H-responses before and after a period of endurance training. Eur. J. Appl. Physiol., 63: 368–375. Poirrier, A.L., Nyssen, Y., Scholtes, F., Multon, S., Rinkin, C., Weber, G., Bouhy, D., Brook, G., Franzen, R. and Schoenen, J. (2004) Repetitive transcranial magnetic stimulation improves open field locomotor recovery after low but not high thoracic spinal cord compression-injury in adult rats. J. Neurosci. Res., 75: 253–261. Riddoch, G. (1917) The reflex functions of the completely divided spinal cord in man, compared with those associated with less severe lesions. Brain, 40: 264–402. Rochcongar, P., Dassonville, J. and Le Bars, R. (1979) Modifications du reflexe de Hoffmann en fonktion de lentraıˆ nement chez le sportif. Eur. J. Appl. Physiol., 40: 165–170. Ronthal, M. (1998) Spinal Cord Injury in Clinical Neurology. In: Joynt, R.J. and Griggs, R.C. (Eds.) Clinical Neurology, Vol. 47. Lippincott, Williams and Wilkins, Hagerstown, MD, pp. 1–28. Rossignol, S. (1996) Neural control of stereotypic limb movements. In: Rowell, L.B. and Sheperd, J.T. (Eds.), Handbook of Physiology. Oxford University Press, New York, pp. 173–216. Rossignol, S. (2000) Locomotion and its recovery after spinal injury. Curr. Opin. Neurol., 10: 708–716. Rossignol, S., Bouyer, L., Barthe´lemy, D., Langlet, C. and Leblond, H. (2002) Recovery of locomotion in the cat following spinal cord lesions. Brain Res. Rev., 40: 257–266. Rossignol, S., Brustein, E., Bouyer, L., Barthelemy, D., Langlet, C. and Leblond, H. (2004) Adaptive changes of locomotion after central and peripheral lesions. Can. J. Physiol. Pharm., 82: 617–627. Rossignol, S., Dubuc, R. and Gossard, J.P. (2006) Dynamic sensorimotor interactions in locomotion. Physiol. Rev., 86: 89–154.
279 Sabbahi, M.A. and Sedgwick, E.M. (1982) Age-related changes in monosynaptic reflex excitability. J. Gerontol., 37: 24–32. Sandrini, G., Serrao, M., Rossi, P., Romaniello, A., Cruccu, G. and Willer, J.C. (2005) The lower limb flexion reflex in humans. Prog. Neurobiol., 77: 353–395. Scaglioni, G., Narici, M.V., Maffiuletti, N.A., Pensini, M. and Martin, A. (2003) Effect of ageing on the electrical and mechanical properties of human soleus motor units activated by the H reflex and M wave. J. Physiol., 548: 649–661. Schneider, C. and Capaday, C. (2003) Progressive adaptation of the soleus H-reflex with daily training at walking backward. J. Neurophysiol., 89: 648–656. Segal, R.L. (1997) Plasticity in the central nervous system: pperant conditioning of the spinal stretch reflex. Top. Stroke Rehabil., 3: 76–87. Segal, R.L. and Wolf, S.L. (1994) Operant conditioning of spinal stretch reflex in patients with spinal cord injuries. Exp. Neurol., 130: 202–213. Shefner, J.M., Berman, S.A., Sarkarati, M. and Young, R.R. (1992) Recurrent inhibition is increased in patients with spinal cord injury. Neurol, 42: 2162–2168. Shurrager, P.S. and Dykman, R.A. (1951) Walking spinal carnivores. J. Comp. Physiol. Psych., 44: 252–262. Spitzer, N.C. (1999) New dimensions of neuronal plasticity. Nat. Neurosci., 2: 489–491. Stein, R.B. (1995) Presynaptic inhibition in humans. Prog. Neurobiol., 47: 533–544. Straka, H. and Dieringer, N. (1995) Spinal plasticity after hemilabyrinthectomy and its relation to postural recovery in the frog. J. Neurophysiol., 73: 1617–1631. Tai, Q. and Goshgarian, H.G. (1996) Ultrastructural quantitative analysis of glutamatergic and GABAergic synaptic terminals in the phrenic nucleus after spinal cord injury. J. Comp. Neurol., 372: 343–355. Tai, Q., Palazzolo, K.L. and Goshgarian, H.G. (1997) Synaptic plasticity of 5-hydroxytrypamine-immunoreactive terminals in the phrenic nucleus following spinal cord injury: a quantitative electron microscopic analysis. J. Comp Neurol., 386: 613–624. ten Donkelaar, H.J., Lammens, M., Wesseling, P., Hori, A., Keyser, A. and Rotteveel, J. (2004) Development and malformations of the human pyramidal tract. J. Neurol., 251: 1429–1442. Thilmann, A., Fellows, S. and Garms, E. (1990) Pathological stretch reflexes on the ‘‘good’’ side of hemiparetic patients. J. Neurol. Neurosurg. Psych., 53: 208–214. Thompson, R.F. (2001) Spinal plasticity. In: Patterson, M.M. and Grau, J.W. (Eds.), Spinal Cord Plasticity: Alterations in Reflex Function. Kluwer Academic, Boston, MA, pp. 1–11. Thompson, R.F., Bao, S., Chen, L., Cipriano, B.D., Grethe, J.S., Kim, J.J., Thompson, J.K., Tracy, J.A., Weninger, M.S. and Krupa, D.J. (1997) Associative learning. Int. Rev. Neurobiol., 41: 151–189. Thompson, F.J., Reier, P.J., Lucas, C.C. and Parmer, R. (1992) Altered patterns of reflex excitability subsequent to contusion injury of the rat spinal cord. J. Neurophysiol., 68: 1473–1486.
Tillakaratne, N.J.K., de Leon, R.D., Hoang, T.X., Roy, R.R., Edgerton, V.R. and Tobin, A.J. (2002) Use-dependent modulation of inhibitory capacity in the feline lumbar spinal cord. J. Neurosci., 22: 3130–3143. van Alphen, A.M. and De Zeeuw, C.I. (2002) Cerebellar LTD facilitates but is not essential for long-term adaptation of the vestibulo-ocular reflex. Eur. J. Neurosci., 16: 486–490. Vaynman, S. and Gomez-Pinilla, F. (2005) License to run: exercise impacts functional plasticity in the intact and injured central nervous system by using neurotrophins. Neurorehabil. Neural. Repair,, 19: 283–295. Vizzard, M.A. (2006) Neurochemical plasticity and the role of neurotrophic factors in bladder reflex pathways after spinal cord injury. Prog. Brain Res., 152: 97–115. Voerman, G.E., Gregoric, M. and Hermens, H.J. (2005) Neurophysiological methods for the assessment of spasticity: the Hoffmann reflex, the tendon reflex, and the stretch reflex. Disabil. Rehabil., 27: 33–68. Voigt, M., Chelli, F. and Frigo, C. (1998) Changes in the excitability of soleus muscle short latency stretch reflexes during human hopping after 4 weeks of hopping training. Eur. J. Appl. Physiol., 78: 522–532. Waldenstrom, A., Thelin, J., Thimansson, E., Levinsson, A. and Schouenborg, J. (2003) Developmental learning in a pain-related system: Evidence for a cross-modality mechanism. J. Neurosci., 23: 7719–7725. Wang, Y., Pillai, S., Wolpaw, J.R. and Chen, X.Y. (2006) Motor learning changes GABAergic terminals on spinal motoneurons in normal rats. Eur. J. Neurosci., 23: 41–50. Whelan, P.J. and Pearson, K.G. (1997) Plasticity in reflex pathways controlling stepping in the cat. J. Neurophysiol., 78: 1643–1650. Wolf, S.L. and Segal, R.L. (1996) Reducing human biceps brachii spinal stretch reflex magnitude. J. Neurophysiol., 75: 1637–1646. Wolpaw, J.R. (1987) Operant conditioning of primate spinal reflexes: the H-reflex. J. Neurophysiol., 57: 443–459. Wolpaw, J.R. (1997) The complex structure of a simple memory. Trends Neurosci., 20: 588–594. Wolpaw, J.R., Braitman, D.J. and Seegal, R.F. (1983) Adaptive plasticity in the primate spinal stretch reflex: initial development. J. Neurophysiol., 50: 1296–1311. Wolpaw, J.R. and Chen, X.Y. (2006) The cerebellum in maintenance of a motor skill: a hierarchy of supraspinal and spinal cord plasticity underlies H-reflex conditioning. Learn. Mem., 13: 208–215. Wolpaw, J.R., Herchenroder, P.A. and Carp, J.S. (1993) Operant conditioning of the primate H-reflex: factors affecting the magnitude of change. Exp. Brain Res., 97: 31–39. Wolpaw, J.R. and Lee, C.L. (1989) Memory traces in primate spinal cord produced by operant conditioning of H-reflex. J. Neurophysiol., 61: 563–572. Wolpaw, J.R. and O’Keefe, J.A. (1984) Adaptive plasticity in the primate spinal stretch reflex: evidence for a two-phase process. J. Neurosci., 4: 2718–2724.
280 Wolpaw, J.R. and Tennissen, A.M. (2001) Activity-dependent spinal cord plasticity in health and disease. Ann. Rev. Neurosci., 24: 807–843. Yamanaka, K., Yamamoto, S., Nakazawa, K., Yano, H., Suzuki, Y. and Fukunaga, T. (1999) The effects of long-term bed rest on H-reflex and motor evoked potential in the human soleus muscle during standing. Neurosci. Lett., 266: 101–104. Zehr, E.P. (2002) Considerations for use of the Hoffman reflex in exercise studies. Eur. J. Appl. Physiol., 86: 455–468.
Zehr, E.P. (2005) Neural control of rhythmic human movement: the common core hypothesis. Exerc. Sports Sci. Rev., 33: 54–60. Zheng, Z., Gibson, S.J., Khalil, Z., Helme, R. and McMeeken, J.M. (2000) Age-related differences in the time course of capsaicin-induced hyperalgesia. Pain, 85: 51–58. Zimmermann, M. (2001) Pathobiology of neuropathic pain. Eur. J. Pharm., 429: 23–37.
A
B HR threshold
D
0.3 0.2
Naive EPSP
0.1 0.0
0
H
6
M
12
H
18 0 6 TIME (ms)
12
HR EPSP
18 Terminals/ M N
M
C
Coverage of Soma(%)
10 µm
GAD-IRDensity
EMG (mV)
Naive threshold
6.0
***
4.0
2.0 30
*** 25
20 12 10
***
8 6 4
DS
DF NC Rat Group
Plate 17.3. Operant conditioning of the spinal stretch reflex (SSR)/H-reflex pathway and the associated spinal and supraspinal plasticity. (A) The graphs show average poststimulus soleus EMG for representative days before (solid) and after (dotted) soleus Hreflex conditioning from a rat in which the H-reflex has been increased by the up-conditioning mode (left) or decreased by the downconditioning mode (right). The H-reflex is much larger after up-conditioning and much smaller after down-conditioning, while background EMG (indicated here by EMG at zero time) and M responses (i.e., direct muscle responses) are unchanged. (B) Triceps surae motoneurons on the conditioned side of H-reflex down-conditioned (HR) monkeys were found to have more positive firing thresholds and slightly smaller Ia EPSPs. Together, these two findings can explain why the H-reflex became smaller (A and B from Wolpaw, 1997). (C) Effects of successful down-conditioning on GABAergic terminals on soleus motoneurons (assessed by glutamic acid decarboxylase (GAD67)-immunoreactivity (GAD-IR). Top left: Soleus motoneuron labeled by Alexa Fluor-488 conjugated with CTB injected into the muscle. Bottom left: Same motoneuron showing GAD-IR (i.e., GABAergic) terminals (dark) located on the periphery of the motoneuron.(Bar ¼ 10 Fm.). Right: Average (7SEM) values for: down-conditioning successful (DS), down-conditioning failed (DF), and naive control (NC) rat groups for: number of GABAergic terminals per motoneuron; terminal GAD density; and GABAergic terminal coverage of soma (expressed as percent of perimeter). (***Po0.0001; compared to the NC group.) After successful down-conditioning, soleus motoneurons have more GABAergic terminals, and these terminals are more densely labeled and occupy more of the soma (from Wang et al., 2006)). (D) Spinal and supraspinal sites (shaded ovals) of plasticity associated with operant conditioning of the SSR or its electrical analog, the H-reflex. ‘‘MN’’ is the motoneuron, ‘‘CST’’ is the main corticospinal tract, and each ‘‘IN’’ is one or more spinal interneuron types. Open synaptic terminals are excitatory, solid ones are inhibitory, half-open ones could be either, and the subdivided one is a cluster of C terminals. Dashed pathways imply the possibility of intervening spinal interneurons. The monosynaptic and probably oligosynaptic SSR/H-reflex pathway from Ia and Ib afferents to the motoneuron is shown. Definite (heavy shading) or highly probable (light shading) sites of plasticity include: the motoneuron membrane (i.e. firing threshold and axonal conduction velocity), motor unit properties, C terminals on the motoneuron, the Ia afferent synaptic connection, and terminals conveying disynaptic group I inhibition or excitation to the motoneuron. The essential roles of the corticospinal tract (originating in sensorimotor cortex) and of cerebellar output to cortex are indicated (updated from Wolpaw, 1997).
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 18
Cochlear implants: cortical plasticity in congenital deprivation Andrej Kral1,3,, Jochen Tillein1,4, Silvia Heid2, Rainer Klinke2 and Rainer Hartmann2 1
Laboratories of Auditory Neuroscience, Institute of Neurophysiology and Pathophysiology, University of Hamburg School of Medicine, Hamburg, Germany 2 Institute of Sensory Physiology & Neurophysiology, J.W. Goethe University School of Medicine, Frankfurt am Main, Germany 3 School of Behavioral and Brain Sciences, University of Texas at Dallas, Dallas, TX, USA 4 Medel Company, Innsbruck, Austria
Abstract: Congenital auditory deprivation (deafness) leads to a dysfunctional intrinsic cortical microcircuitry. This chapter reviews these deficits with a particular emphasis on layer-specific activity within the primary auditory cortex. Evidence for a delay in activation of supragranular layers and reduction in activity in infragranular layers is discussed. Such deficits indicate the incompetence of the primary auditory cortex to not only properly process thalamic input and generate output within the infragranular layers, but also incorporate top-down modulations from higher order auditory cortex into the processing within primary auditory cortex. Such deficits are the consequence of a misguided postnatal development. Maturation of primary auditory cortex in deaf animals shows evidence of a developmental delay and further alterations in gross synaptic currents, spread of activation, and morphology of local field potentials recorded at the cortical surface. Additionally, degenerative changes can be observed. When hearing is initiated early in life (e.g., by chronic cochlear-implant stimulation), many of these deficits are counterbalanced. However, plasticity of the auditory cortex decreases with increasing age, so that a sensitive period for plastic adaptation can be demonstrated within the second to sixth months of life in the deaf cat. Potential molecular mechanisms of the existence of sensitive period are discussed. Data from animal research may be compared to electroencephalographic data obtained from cochlear-implanted congenitally deaf children. After cochlear implantation in humans, three phases of plastic adaptation can be observed: a fast one, taking place within the first few weeks after implantation, showing no sensitive period; a slower one, taking place within the first months after implantation (a sensitive period up to 4 years of age); and possibly a third, and the longest one, related to increasing activation of higher order cortical areas. Keywords: sensitive period; layer-specific activity; top-down projection; current source density; development; maturation; auditory cortex electrode array placed on a thin silastic carrier (implanted in the cochlea), a subcutaneous receiver (implanted in the skull behind the ear), and a microphone with a sound or speech processor (worn extracorporally). The processor receives the signal from the microphone, preprocesses it using the selected coding strategy, and transmits the
Introduction Cochlear implants (Fig. 1) are the most successful of all neuroprosthetic devices. They consist of an Corresponding author. Tel.: +49 40 42803 7046; Fax: +49 40 42803 7752; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57018-9
283
284
Fig. 1. A commercial cochlear implant device consisting of an extracorporal sound processor with a microphone and an attached transmitter coil (A, courtesy of Cochlear Corp., Melbourne, Australia). This device transmits signals magnetically to a subcutaneously located receiver unit (B, Cochlear Corp., Melbourne, Australia) connected to an indifferent extracochlear electrode and an intracochlear electrode array. The intracochlear electrode arrays are either precurved with a stand (C, Cochlear Corp., Melbourne, Australia) or straight (D, MedEl Company, Innsbruck, Austria).
information through HF transmission via a coil onto the subcutaneous receiver. This then relays the signals to the stimulating electrodes. The stimulation array of the cochlear implant is inserted into the scala tympani through either the round window or a cochleostomy. Cochlear implants are used to treat hearing loss caused by a nonfunctional organ of Corti. Nonetheless, the indication of a cochlear implantation has been expanding over the last decade, now sometimes also including individuals with residual hearing and lack of benefit from conventional high-power hearing aids (von Ilberg et al., 1999; Kiefer et al., 2004), individuals suffering from strong tinnitus (Thedinger et al., 1985; McKerrow et al., 1991; Dauman et al., 1993; Ito and Sakakihara, 1994; Tyler, 1995; Ruckenstein et al., 2001), and individuals with the so-called auditory neuropathy (Miyamoto et al., 1999; Trautwein et al., 2000; Shallop et al., 2001; Sininger and Trautwein, 2002). ‘‘Electric hearing’’ is also a suitable tool to investigate effects of auditory deprivation and developmental plasticity. Functional consequences of auditory deprivation cannot be investigated
as easily as in visual deprivation: simple suturing of the ear canals does not suffice to block all hearing. The thresholds rarely increase by more than 40 dB, and bone conduction is not attenuated at all. Consequently, mechanical manipulations on the external meatus cannot prevent hearing experience. Surgical destruction of the middle ear also does not suffice, as it does not affect bone conduction. Consequently, sounds from swallowing, chewing, breathing, sneezing, coughing, but most importantly own vocalizations are not attenuated by such intervention. Cochlear destructions, like the frequently used cochlear ablation, are irreversible (for detailed comparison on the different deprivation models see Kral et al., 2001; Syka, 2002). The need for a neurophysiological model of congenital auditory deprivation arose with the introduction of cochlear implantation. The first models of auditory deprivation were based on pharmacological destruction of the inner ear; functional data on a deprived auditory system could be gathered using electrical stimulation of the surviving fibers of the auditory nerve. Later, congenitally deaf animal strains have been introduced to this area of research.
285
Electrical stimulation of the auditory nerve Electrical stimulation of the auditory nerve induces a pattern of activity in the afferent auditory system that differs from that induced by acoustic stimulation of a hearing cochlea. Destruction or degeneration of hair cells eliminates spontaneous activity in the auditory nerve (Hartmann et al., 1984). Electrical stimulation is additionally characterized by
Lack of precise spatial information — a ‘‘beam forming’’ is possible in electrical stimulation of the auditory nerve, but the resulting tuning curves are much less sharp than with acoustic stimulation (Kral et al., 1998). Lack of stochasticity — electrically evoked action potentials express much less temporal jitter than the action potentials evoked by acoustic stimuli in a hearing cochlea, resulting in a ‘‘hypersynchronization’’ of the evoked activity to the electrical stimulation. This has several unwanted effects including steep loudness growth. Despite of high-frequency stimulation in commercial cochlear implants where individual stimulation pulses can fall into the refractory period of the fibers, and despite of the attempts to introduce stochasticity in the elicited firing pattern by adding subthreshold electrical noise, the stochasticity of auditory nerve firing pattern is less in cochlear implants patients than in hearing individuals (Hartmann et al., 1984; Rubinstein et al., 1999). Compressed dynamic range — responses of single auditory nerve fibers to electrical stimulation saturate within 3–10 dB above threshold. This represents a substantial collapse of the normal dynamic range of 40–80 dB in auditory nerve fibers. Consequently, the loudness range is substantially compressed in cochlear implant users as compared to normal acoustic hearing (review in Hartmann and Klinke, 1990).
Plasticity, development, and deprivation Cochlear implants provide a way to obtain information about the function of the auditory system that has been deaf for a certain time. The deaf ear can be chronically stimulated electrically by using
cochlear implants making it possible to study the central auditory plasticity. Central reorganizations induced by cochlear implants involve several different types of plasticity: 1. Reorganization of the auditory system can be induced in hearing-experienced individuals (animals and humans) who became deaf as adults and received a cochlear implant after a certain period of deafness. To compensate for deafness-induced degenerative changes in the auditory system, these individuals have to undergo plastic reorganization to adapt to the abnormal characteristics of the neural activity evoked by the cochlear implant. This type of plastic reorganization is similar to other forms of learning-induced plasticity with the difference that the reorganization is preceded by a period of deprivation and is more extensive because the evoked activity becomes different from that evoked by sound during the period of hearing. The ‘‘interpretation’’ of the new stimulation mode has to be newly learned. 2. Previous studies have consistently shown that the development of the central auditory system is shaped by acoustic experience (for review on owls spatial orientation, see Knudsen, 2004; for review on language-related aspects, see Skuse, 1993; Ruben, 1997; Kuhl, 2004; for review on auditory aspects and neurophysiology in mammals, see Kral et al., 2001; Syka, 2002). Since the auditory system in the congenitally deaf individual does not have any input, the shaping of the auditory system through experience does not occur and the normal adaptation to the acoustic environment does not take place (review in Kral et al., 2001; Hartmann and Kral, 2004). Such an acoustically naı¨ ve auditory system has to undergo additional adaptations after cochlear implantation. It has to catch up with the activity-dependent maturation that the auditory system has missed during previous development. Therefore, the plasticity in a naive auditory system cannot be functionally compared to plasticity in an experienced auditory system. The term developmental plasticity will be used for this type of plasticity in the following text.
286
Postnatal development of a naı¨ve auditory system Auditory deprivation induces many changes in the central auditory nervous system. These can be divided into dystrophic changes of the morphology of neurons in different nuclei of the auditory system and changes in the connectivity, eventually leading to deficits in functional properties of the central auditory system. These changes are usually assumed to occur in the ascending auditory pathways including the auditory cerebral cortices, but changes may also occur in descending systems. Degenerative changes in the auditory nervous system have been described for neonatally deafened and congenitally deaf animals. Dystrophic morphological abnormalities in the central auditory system of naı¨ ve animals include changes in the structure of the neuronal body and synaptic endings (for details, see recent review articles, Kral et al., 2001; Hartmann and Kral, 2004; Middlebrooks et al., 2005). The main afferent projections in the auditory system are only marginally affected by congenital auditory deprivation (Heid et al., 1997). The synaptic changes, which have been investigated most extensively in the cochlear nucleus (Hultcrantz et al., 1991; Larsen and Kirchhoff, 1992; Lustig et al., 1994; Saada et al., 1996; Ryugo et al., 1997), include dystrophic and hypertrophic changes. Reductions in synaptic numbers and densities have been demonstrated in the midbrain of neonatally deafened animals (Hardie et al., 1998). Functional changes in the different nuclei of the ascending auditory pathways including the different regions of the cerebral auditory cortex have been studied using electrical stimulation of auditory nerve fibers in the cochlea of neonatally deaf and deafened animals. For experiments in animals, the availability of suitable animal species is limited. Such animals should be completely and congenitally deaf with well-preserved auditory nerve. There are genetically modified rodent strains with hearing loss (reviewed in Kiernan and Steel, 2000), but they are often not completely deaf at birth. Investigations of central deficits from auditory deprivation are now dominated by two models: neonatally deafened animals and congenitally deaf species
such as the deaf white cat (review e.g., in Hartmann and Kral, 2004). Both these models have advantages and disadvantages. In neonatally deafened animals, the destruction of the inner ear is achieved by systemic application of ototoxic substances during the phase of hearing acquisition. The advantage of neonatally deafened animals is the easy availability, and the disadvantage is the pronounced and rapid degeneration of spiral ganglion cells (cell loss from 50–90% of normal counts after several weeks to months of deafness, see Leake-Jones et al., 1982; Leake et al., 1987, 1999; Leake and Hradek, 1988; Dodson, 1997a, b). It is an important advantage of congenitally deaf strains that some of them show a slow degeneration of spiral ganglion cells, comparable to human congenital deafness. The disadvantage of congenitally deaf animals is their lesser availability because their litters are often small. In the congenitally deaf (white) cats (CDCs), all hair cells are lost spontaneously prior to hearing onset (Heid et al., 1998). The cochlea shows the picture of Scheibe dysplasia, with preserved bony structure, with preserved auditory nerve and spiral ganglion, but with dystrophic and degenerative changes in the scala media (including a collapse of the Reissner’s membrane, a retraction of the tectorial membrane, and some other more subtle deficits). Although there is some degeneration in the spiral ganglion, the degeneration progresses much slower than in neonatally deafened animals. Most importantly, in the first halfturn of the cochlea (where a cochlear implant can be inserted in a cat, Kral et al., 1998), there is no significant loss of spiral ganglion cells up to the age of 2 years (Heid et al., 1998). This means that in contrast to neonatal deafening, no degeneration of spiral ganglion cells can be observed at the site of most effective electrical stimulation, thus allowing studies of central auditory evoked responses elicited by cochlear implant stimulation in congenitally deaf cats.
Functional changes in the afferent auditory system were not prominent in neonatally or congenitally deaf animals. In the auditory midbrain, several parameters including poststimulus time
287
histograms, maximum following frequency and entrainment, as well as latencies or amplitudes of the responses were not significantly affected by the absence of auditory experience (Snyder et al., 1990, 1991). Reductions in occurrence of longlatency responses in the midbrain were observed (Snyder et al., 1991). Other investigators also described reductions in the temporal jitter of the responses (Shepherd et al., 1999). Greater abnormalities were observed in the auditory cortex, where auditory deprivation induced numerous and extensive processing deficits. In the adult auditory cortex, the representations of the stimulated auditory partition changed. Although a rudimentary cochleotopy could be demonstrated in congenitally deaf cats (Hartmann et al., 1997), the cochleotopy became progressively more smeared with the duration of the absence of auditory experience in neonatally deafened animals (Raggio and Schreiner, 1999), indicating a degenerative process in the spatial organization of the auditory cortex acting over the time of complete auditory deprivation. It has to be considered that the process of degeneration of the spiral ganglion cells in neonatally deafened animals may contribute to this smearing of the cochleotopic gradient (Dodson, 1997a, b; Leake et al., 1999; Dodson and Mohuiddin, 2000). Other characteristics of
responses to simple electrical stimuli from cells in the primary auditory cortex (field A1), like the dynamic range, latencies of responses, and poststimulus time histograms, were similar in the cortex of deaf and hearing cats (Raggio and Schreiner, 1994). However, spontaneous activity was slightly, but significantly increased in the field A1 of congenitally deaf cats (Fig. 2, Kral et al., 2003). Current source density analysis in deaf auditory cortex Further deafness-induced deficits in the cortical microcircuitry have been observed in congenitally deaf cats using the current source density (CSD) method. The CSD analysis makes it possible to describe the activation of the auditory cortex in a layer-specific manner. The current density analysis is based on Maxwell’s electrical field theory and has been adapted for use in neurophysiology by Walter Pitts and later by Nicholson and Freeman (Pitts, 1952; Nicholson and Freeman, 1975). Applied to the cerebral cortex, the method relies on measurements of local field potentials in different cortical layers with microelectrodes (Fig. 3). From these signals, the second spatial derivative
Fig. 2. Distribution of cortical spontaneous activity in halothane-anesthetized adult hearing controls (dashed) and congenitally deaf cats. Congenitally deaf cats show higher spontaneous activity (median in controls ¼ 8.0 spikes/s, median in deaf cats ¼ 9.8 spikes/s, w2 test, a ¼ 1%, reprinted with permission from Kral et al., 2003).
288
Fig. 3. One-dimensional CSD signals represent the second spatial derivative of the local field potentials recorded during a penetration through the cortex perpendicular to the cortical surface (modified with permission from Kral et al., 2000).
multiplied by a resistivity tensor is computed to obtain the so-called CSD. The CSD signals are the effect of synaptic currents in close proximity of the tips of the recording microelectrodes. For example, once a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) receptors are activated, a sodium channel opens and sodium flows into the cell (the electrical current is physically oriented into the cell, inward current), leaving a current sink in the extracellular space. As the inward current depolarizes the membrane of the neuron (‘‘discharges’’ the capacitor), it sets free the positive ions originally attracted to the extracellular side of the neuronal membrane by the membrane potential. The positive ions become mobile and detach from the membrane, causing a current source in the extracellular space (current oriented out of the cell, outward current). The current source balances the current sink and neutralizes the extracellular space. However, this passive return current is distributed over larger portions of the neuronal membrane near the activated synapse and is therefore locally of smaller magnitude. Depending on the method of recordings and filtering the original signals,
these passive return currents may be less obvious in the computed CSD profiles. Inhibitory synaptic currents, in contrast, are caused by the outflow of positive ions (potassium) or inflow of negative ions (chloride), thus causing a current source in the extracellular space (provided a resting transmembrane potential). One-dimensional CSD analysis reveals only the current sinks and sources that are the consequence of the currents flowing perpendicularly to the electrode penetration. Consequently, the CSD signals obtained from electrode penetrations perpendicular to the cortical surface will in fact reveal the extracellular synaptic currents of cells with elongated structure running in parallel to the electrode penetration. In the cortex, penetrations oriented perpendicularly to the cortical surface will therefore mainly reveal synaptic currents from pyramidal cells that are large, oriented parallel to the electrode penetration and to each other, and are arranged regularly. Smaller cells with globular architecture and less regular arrangement in individual cortical layers (e.g., stellate cells) will be less represented in these signals. Owing to the fact that passive currents are
289 distributed over larger portions of neuronal membranes and in all three space dimensions, they produce smaller one-dimensional CSD signals than ‘‘true’’ (active) synaptic currents that are much more localized. From a theoretical point of view, the local field potentials used for calculation of the CSD should be recorded simultaneously. An advantage of a simultaneous recording is that single-sweep data can be used for its computation, thus allowing a trial-to-trial analysis. With such an approach, the differences in the impedance of individual electrodes have to be considered; otherwise, they would introduce errors in the estimation of the current source densities. Additionally, the electrode shank affects the shape of the electrical field around the electrode (if of different impedance than the surrounding tissue). An often-used alternative is to record local field potentials with an individual electrode successively located at different recording positions and perform the calculation of current source densities off line (e.g., Friauf and Shatz, 1991; Cruikshank et al., 2002). This method only allows CSD computation on averaged local field potentials as the average represents the invariant part of the evoked response, and thus the shift in time between individual recording positions does not invoke a bias into the data. Glass microelectrodes with very narrow shanks can be used for recording, minimizing the tissue trauma and the influence of the electrode on the geometry of the electrical field, which confers an advantage. Using glass microelectrodes, iontophoretic application of dyes is possible and thus exact reconstructions of the electrode penetration within the tissue become available. For technical details of the method compare e.g. Nicholson and Freeman (1975), Mitzdorf (1985), and Somogyvari et al. (2005). In our laboratories both techniques are used, but the data reviewed here rely solely on the single-electrode technique. The CSD method allows an effective assessment of the synaptic activity in different layers of the auditory cortex. In contrast to intracellular recordings, the current CSD method gives information about several hundreds of
synapses at the same time and by that undersampling of the synaptic activity is avoided. 1. Extracellular synaptic currents have a different shape when compared to intracellularly recorded synaptic currents. Spikes, in the first approximation, appear like a temporal derivative of the time function of the spike recorded intracellularly. Synaptic currents are most probably less ‘‘distorted.’’ 2. The sensitivity of the CSD method is lower than that of intracellular recordings or whole-cell patch clamp regarding the number of active synapses. The results of the CSD method represents the activity in hundreds of active synapses mixed together in a single signal. The computation of the CSD therefore represents the average synaptic currents at the recording position. It is evident that the pattern of activity obtained with this method is specific for individual cortical layers (Fig. 4). The method produces well-reproducible results, and the activity in the different layers (like the borders of layer IV) can be distinguished in the waveform of CSD signals. Penetrations with smaller distances between recording positions reveal additional sink and sources that are hidden in the more course penetrations (Kral and Hartmann, unpublished data). Recording steps from 50 mm to 300500 mm have been used in different studies. The finer the steps, the more spatial and temporal details can be obtained from the CSD signals. To be able to determine CSD signals from functionally corresponding cortical positions in different animals, the cortical area has to be mapped first. This can be done using local field potentials recorded from the cortical surface with high-impedance glass microelectrodes. The resulting functional maps can be quantitatively analyzed, and the location with largest responses can be determined (region of interest, ROI, size: 1 mm2). In the studies reviewed here, the signals used for CSD computation and single-unit activity were recorded in such defined ROIs.
Activity in the auditory cortex of congenitally deaf cats was analyzed using the above method and compared to hearing controls stimulated
290
Fig. 4. Neurophysiological approach for determining CSD signals. First, lowest cortical threshold were determined using local field potentials recorded with low-impedance electrodes at a 3 3 recording grid at the field A1 (A). At 10 dB above the lowest thresholds, the primary auditory cortex was mapped using local field potentials recorded at the cortical surface with glass microelectrodes (Z ¼ 6 MO) at 100–170 recording positions (B). At the spot with largest responses, a region of interest (ROI) with 1 mm2 was defined and there the cortex was penetrated at a grid of 2 3 positions (500 mm spaced). Two recording positions were marked by iontophoretic application of horseradish peroxidase and the penetration was histologically reconstructed after the experiment (C). Recorded local field potentials could then be assigned to cortical layers and CSD signals could be computed (sinks are filled; reprinted with permission from Kral et al., 2005).
electrically (Fig. 5). Such comparison revealed numerous deficits in the activation of the auditory cortex of congenitally deaf cats. A significant decrease was shown in the mean amplitude of gross synaptic currents, both expressed in the maximum current of each sink and
in the temporal integrals of the sinks (Kral et al., 2000). These results may be caused by desynchronization of synaptic activity, a reduction of the number of activated synapses, or a reduction in the amplitude of the individual synaptic currents in the primary auditory cortex of congenitally deaf cats.
291
Fig. 5. Representative examples of CSD signals in naı¨ ve adult deaf cats (A) and electrically stimulated controls (B), sinks are filled. The amplitudes of the signals in deaf animals are significantly smaller (WilcoxonMannWhitney test, a ¼ 5%) and concentrated to mainly the supragranular layers. The shortest latencies evaluated in each layer separately were significantly delayed in congenitally deaf cats in supragranular and infragranular layers and no difference was found in layer IV (C, data from Kral et al., 2000).
Desynchronization of neuronal activity has further been documented by a less synchronous activation of the cortical column in congenitally deaf cats (Fig. 5). The supragranular layers in congenitally deaf cats were activated with significant delays not found in hearing controls. A delay in activation of cells in the supragranular layers in relation to cells in layer IV may have the consequence of desynchronization of the excitatory drive on pyramidal cells of layer V (Larkum
et al., 1999). This in turn may prevent an appropriate activation of these cells, and thus impair the function of the intrinsic cortical microcircuitry in cortical ‘‘modules’’ (columns), affecting activity in infragranular layers. Reductions of synaptic activity were also found in the infragranular cortical layers V and VI in congenitally deaf cats (Kral et al., 2000). These layers are targets of descending projections from the higher auditory cortex, and reductions in the activity in layer V and VI further
292
indicate reduced input from descending corticocortical projections. In congenitally deaf animals, activity in field A1 was restricted to a shorter interval after the stimulus. This is considered a consequence of the above-described functional disintegration of the activity in the cortical module. Long-latency activity was also reduced in deaf animals. Comparisons in source amplitudes between deaf and hearing animals revealed a significant reduction in layers III and IV in the field A1 of the deaf animals (Hubka et al., 2004; Kral et al., 2005). Inhibitory synapses are most abundant in layer I, with layers IIIV following (Prieto et al., 1994a, b). This result therefore indicates reductions in the synaptic function within layers III and IV of congenitally deaf cats. Layers III and IV are essential for cortical propagation of activity from thalamus to supragranular layers — according to some authors they perform a gating function (Rozas et al., 2001). This gating develops during postnatal life, shaped by experience. Accordingly in the visual system, neonatal deprivation was shown to increase the sensitivity of layer IV (Maffei et al., 2004). Down-regulation of inhibition in this layer, as indicated by the above data in the auditory cortex of deaf animals, thus corresponds to the findings in the visual system. The deficits that are revealed using the CSD method are not prominently expressed in surfacerecorded local field potentials, which are dominated by supragranular activity. However, the amplitude of the negative wave Nb in the evoked potentials recorded from the surface of the cerebral cortex is significantly reduced in adult congenitally deaf cats. The wave P1 is reduced in amplitude and its latency is prolonged in adult congenitally deaf cats (Klinke et al., 1999; Kral et al., 2005). These processing deficits may have several causes. They may be caused by degenerative processes in the auditory cortex due to inactivity. Alternatively, a misguided developmental sequence may lead to a dysfunctionality of the auditory cortex. Last but not least, recruitment for another function is the last option; this may lead to a reduced possibility for activation of the primary auditory cortex by auditory input. The last mentioned option can, however, be ruled out for the following reasons: If the cells in
the primary auditory cortex were recruited for function other than the normal one, it would be expected that the more active synapses would prevail in competition for synaptic space and those that normally receive input from sources that are inactive (the auditory thalamus) would be suppressed. The most probable consequence of such a condition would be an increase in auditory activation thresholds. However, currently available data show that this is not the case, neither in congenitally deaf nor in neonatally deafened animals (Raggio and Schreiner, 1999; Kral et al., 2005). In congenitally deaf cats, the thresholds for activation of the auditory cortex via the lemniscal pathway (via a cochlear implant) were not higher than those in hearing controls; on the contrary, they were significantly lower (Kral et al., 2005). This ‘‘hypersensitivity’’ to auditory inputs most probably appeared at the central, possibly thalamocortical level (for arguments see Kral et al., 2005; for possible mechanisms see Kotak et al., 2005). Evidences of cross-modal reorganization of the higher order auditory cortex by the visual inputs have been demonstrated in many human studies (e.g.,Nishimura et al., 1999; Petitto et al., 2000; Finney et al., 2001). Reorganization did occur not only for linguistic visual material, but also for nonlinguistic, moving visual stimuli (Finney et al., 2001, 2003). Nonetheless, different cortical areas differ in their receptiveness for cross-modal reorganization. Only a reorganization of higher order auditory areas has been clearly demonstrated, and cross-modal reorganizations of the primary areas have not been unambiguously shown in these studies. Previously, Stewart and Starr investigated visual responses in the primary auditory cortex of awake congenitally deaf cats with visual flashes. They did not obtain any multiunit responses in field A1 (Stewart and Starr, 1970). However, later studies claimed to have found visually evoked local field potentials also in the primary auditory cortex of anaesthetized congenitally deaf cats (Rebillard et al., 1977, 1980). To resolve this discrepancy, we undertook a study using controlled visual stimulation (Kral et al., 2003). In this study on lightly halothane-anaesthetized cats eye refraction was assessed and
293 corrected with contact lenses to assure a sharp picture on the retina. Backprojection of the fovea on the screen was used to assure accurate stimulus presentation site. Visual flashes were presented at the fovea, followed by presentation of phase-reversal gratings (inducing the impression of a motion) of different spatial frequencies and orientations. These stimuli are effective in activating both the ventral and the dorsal streams of cortical visual processing, as they include both pattern and movement. To stimulate the peripheral parts of the visual field, light bars of different orientations, movement directions, and speed were manually presented in the periphery of the visual field.
Both local field potentials and unit activity were recorded in all cortical layers of field A1. From all cell clusters recorded in field A1, 98% did not show any modulation of spontaneous activity by the visual stimulus in deaf cats. In the remaining 2% of units, visual modulation could not be excluded with confidence (see Kral et al., 2003). The proportion of these ‘‘unclear responses’’ was the same in deaf and hearing animals. Additionally, subthreshold activations also were analyzed using the CSD method. Also here, no responses that could be classified as visually evoked were found. These findings at least indicate that visual reorganization does not include the primary auditory cortex in deaf animals. The primary auditory cortex has the capability to reorganize to process visual inputs if the visual input is redirected to the auditory thalamus, such as may occur when the inferior colliculus is removed by aspiration (Sur et al., 1990; Roe et al., 1992; Pallas and Sur, 1993; Pallas et al., 1999; von Melchner et al., 2000; Pallas, 2001). This situation is, however, not comparable to congenital auditory deprivation, when all anatomical pathways are morphologically intact. The lemniscal auditory pathway is most likely heavily structurally patterned by molecular markers (review in Sur and Rubenstein, 2005), which include repulsive factors. These repulsive factors possibly prevent the invasions of new fibers from the nonauditory nuclei of the thalamus or from nonauditory cortical areas into the primary auditory cortex. Extensive morphological distortions
are necessary to overcome these barriers. The consequence is a weaker capacity for cross-modal reorganization in the primary auditory cortex than in higher order auditory areas, which are normally bi- or polymodal. Reduced afferent input causes the primary cortex to become hypersensitive to cochlear-implant stimulation in congenitally and neonatally deaf animals (Kral et al., 2005; Raggio and Schreiner, 1999; cf. also Kotak et al., 2005; for further mechanisms cf. Turrigiano and Nelson, 2004). A slight but significant increase in spontaneous activity has been demonstrated even in anaesthetized animals (Kral et al., 2003). These results further support the assumption that the primary auditory cortex has less ability of cross-modal reorganization than higher order cortices. If cross-modal reorganization of the primary auditory cortex can be ruled out as a direct source of the deficits in congenitally deaf, two alternative explanations of the deficits that are typical for congenitally deaf animals are functional degeneration and functional misguided maturation. Developmental plasticity The development of the normal auditory system in hearing animals has been studied extensively (reviews in e.g., Payne, 1992; Cant, 1998; Sanes and Walsh, 1998), but information about the development of a naı¨ ve auditory system is sparse and the deficits resulting from the congenital deafness were unclear for a long time. Below, we will review the developmental sequence of a hearing auditory system. Cat’s middle ear is filled with a viscous, embryonic mesenchymal tissue till the end of the second week of life and the cochlea only slowly gains functionality during this time (Brugge et al., 1978). The first sound-evoked cortical responses at very high stimulus intensity can be elicited between day P3 and P8 in the cat; however, the thresholds decline under 100 dB sound pressure level (SPL) first after P10 (Konig et al., 1972; Brugge et al., 1988; Brugge, 1992). The development of hearing sensitivity (lowest unit thresholds) proceeds in different nuclei of the afferent auditory pathway with a similar time course
294 and reaches maturity around between P15 and P20 (auditory nerve: Kettner et al., 1985; Walsh and McGee, 1987; cochlear nucleus: Brugge et al., 1978; Brugge and O’Connor, 1984; central nucleus of the inferior colliculus: Moore and Irvine, 1979; Blatchley and Brugge, 1990; A1 field of the auditory cortex: Brugge et al., 1988; Eggermont, 1996). Interestingly, these data also correspond well with behavioral changes in hearing sensitivity of the cat (Ehret and Romand, 1981). Also the thresholds of auditory brainstem-evoked responses in the cat follow the same time course (Walsh et al., 1986). Consequently, it can be concluded that the development of sound sensitivity (in terms of neuronal thresholds) is determined by cochlear sensitivity in the cat. However, the sensitivity to low-frequency sounds develops before the sensitivity to high-frequency sounds in many vertebrates (review in Brugge, 1992). There is no evidence of units with sensitivity to stimuli of frequencies 410 kHz before P10 in the cat. Central mechanisms contribute little, if at all, to these developmental changes, and the development of functional properties in the central auditory system follows tightly the development of the cochlea. Spontaneous activity represents a property that is difficult to evaluate, as it is strongly influenced by anesthesia. Nonetheless in anesthetized cats, the maturation of spontaneous activity was noted to reach adult values at P70 in the auditory cortex (Eggermont, 1996). This property follows the increase in synaptic densities in the visual cortex of the cat during the first 30 days (see above, Cragg, 1975). Minimum latency for tone pips is decreasing steeply in the auditory cortex of the cat, from 40–60 ms between P9–12 to 18 ms at P40, when mature values are reached (Brugge et al., 1988; Eggermont, 1996; for electrical stimulation see Kral et al., 2005). This maturational sequence is most probably related to maturation of synaptic currents, and less to myelination of geniculocortical radiation, which continues beyond this age (visual system: Tsumoto and Suda, 1982) and possibly is counterbalanced by an increase in length of this projection (Eggermont, 1996).
Studies of spectral filtering by cortical cells have shown that the proportion of broadly tuned units increase with age, thus causing the mean width of tuning curves to increase (Brugge et al., 1988; Eggermont, 1996; Bonham et al., 2004). Broadly tuned units are mainly found in the ventral and dorsal parts of adult A1 (Schreiner and Mendelson, 1990; Heil et al., 1992; Schreiner and Sutter, 1992). These parts of the cortex are not responsive in young animals (Bonham et al., 2004). The finding of increasing bandwidth of units in field A1 contrasts the findings of decreasing bandwidth of tuning curves in the cat inferior colliculus during the first 30–35 days post natal (Moore and Irvine, 1979). Several factors are involved in shaping cortical tuning curves. Inhibition, thalamic divergence, and type of interaction (corticocortical vs. thalamocortical) are the most important ones. Also, the range of audible frequencies increases within the first weeks of life; a factor can have biased investigations of frequency tuning in the cortex (for rats, cf. Zhang et al., 2001). However, the spatial extension of excitation at the auditory cortex from peripheral stimulation is larger in young animals than in adults, both in cats and rats (between 30 and 90 days of age in the cat stimulated electrically through a cochlear implant, Kral et al., 2005). This indicates a higher thalamocortical divergence in young animals. The temporal properties of cortical units develop slowly postnatally. The best modulation frequency increases after birth to reach adult-like values at the age of 60 days, but the maximum best modulation frequencies were observed first at 150 days of age (Eggermont, 1991, 1996). It may be related to changes in inhibitory function after birth, which cause a suppression of the spontaneous activity after the onset response and result in a rebound response at 120–150 ms after the stimulus (Eggermont, 1992). Rebound response matures at approximately 150 days of age (see also Kral et al., 2005). This means that temporal properties are among the slowest to develop in the primary auditory field. Owing to the fact that the auditory system undergoes a massive reorganization during development (especially in altricial animals), the ability to adapt to external influences
295 during development is much higher than in mature (adult) animals. The basic underlying mechanism for plasticity, at least in its first step, is the modification of synaptic efficacy by repetitive stimulation of the synapse (long-term potentiation, LTP, Bliss and Lomo, 1973) and the opposite process, long-term depression (LTD, Ito et al., 1982). For LTP, the stimulation of the synapse has to be frequent, and the preand postsynaptic elements have to be activated successively in a short temporal window (10 ms, Markram et al., 1997; Zhang et al., 1998). For LTD, the stimulation has to be sparse and the coupling of presynaptic and postsynaptic activation has to be weak. In young animals, LTP and LTD can be elicited more easily than in adults (Crair and Malenka, 1995; Sermasi et al., 1999b). The time span during which LTD can be elicited more easily is longer than the corresponding period for LTP (Rittenhouse et al., 1999), indicating a period of life when the capability for synaptic depression is still larger but the LTP is already adult-like. Higher susceptibility to LTP/LTD in young animals is related to the above-mentioned changes in the composition of N-methyl-D-aspartate (NMDA) receptors and the exchange of NMDA receptors by
AMPA receptors in the early postnatal development, and many other mechanisms further participate in this process (for review see Kaczmarek et al., 1997; Syka, 2002).
Deafness and cortical development Functional development of the primary cortical areas (field A1) was significantly different in auditory deprived animals (congenitally deaf cats) compared with animals with normal hearing (Kral et al., 2005). The controls in these experiments were normal hearing animals whose hair cells were acutely (at the beginning of the experiment) destroyed by intracochlear application of neomycin. The auditory nerve was stimulated electrically using a cochlear implant. In these ‘‘hearing’’ controls, electrical stimulation of the auditory nerve led to smallamplitude (o100 mV) long-latency (450 ms) responses on postnatal day 8 (before hearing thresholds have fallen under 100 dB SPL). Later in life, amplitudes increased and latencies decreased in these animals. The same stimulation activated an increasingly larger cortical area up to the age of 2–3 months (Fig. 6). Afterwards, the activated cortical area shrunk to reach the size it has in adults at 4
Fig. 6. Representative example of activation maps computed from maximal amplitudes of local field potentials (Pa amplitudes) in 10 hearing and 10 deaf animals at different ages. Top: hearing controls show largest activations between 1 and 2 months post natal. Afterwards, maximal amplitudes and activated areas decrease. Bottom: in congenitally deaf cats, maximal activated areas were observed at 3 months post natal (data from Kral et al., 2005). See Plate 18.6 in Colour Plate Section.
296
months post natal. This pattern corresponds to the one obtained with acoustical stimulation in cats (Bonham et al., 2004). In very young deaf kitten (up to 3 months post natal), the activated areas were smaller than in their hearing counterparts. At approximately 3 months after birth, the activated region became significantly larger in congenitally deaf cats than in the hearing counterparts. From 4 months on, the activated area droped below the peak values at 3 months (Fig. 6). The above-mentioned experiments confirm that the developmental sequence of changes in cortical activated areas is heavily modified by deafness. Preliminary data with independent component analysis of the local field potentials recorded in different cortical layers indicate that a substrate of the above differences may be a desynchronized thalamocortical and corticocortical excitation in congenitally deaf cats (Hubka et al., 2004). Additionally, a less variable pattern of sink and sources between and within neighboring columns of the field A1 was observed in deaf animals, with a particular reduction in occurrence of current sources. These findings correspond well with the abovementioned disinhibition in the auditory cortex following auditory deprivation. Also, this pattern may indicate that an early developmental stage of cortical wiring has not been patterned by sensory stimuli and therefore could not mature properly (comp. Kalisman et al., 2005). When local field potentials were analyzed for their morphology, additional developmental deficits were revealed in congenitally deaf cats (Kral et al., 2005); the regular developmental sequence in morphology of local field potentials evoked by electrical stimulation via a cochlear implant was delayed and modified in deaf cats (Fig. 7). Especially, the development of a mature-like wave Nb in the middle latency evoked responses was incomplete and delayed by two months in deaf animals. Wave P1 in the long-latency evoked potentials appeared early in development in a similar way as in hearing animals. However, the amplitude of this component decreased with increasing age and it nearly disappeared in adult deaf cats. Two conclusions can be drawn from the results of this study:
1. The development of the auditory cortex is sensitive to auditory experience (or its absence). Only under the influence of auditory experience an appropriate functional development of primary auditory cortex can occur. 2. The developmental sequence is altered in two ways by congenital auditory deprivation: maturation of certain properties of the auditory cortex is delayed and others show degenerative processes during development. An active shaping influence of the auditory experience can be inferred from these results. To clarify the substrate of these changes within the auditory cortex, CSD analyses with stimulation through a cochlear implant were performed within the ROI (see above) during development (Fig. 8). The comparison revealed that in hearing controls, the gross synaptic currents increase significantly during development, reaching a peak within the second month of life. Afterwards, the maximal currents decrease. Additionally, individual currents become more and more structured in the temporal domain, indicating that the underlying individual synaptic currents shorten in duration and overlap less in time. At approximately 3 months of age, the sinks reach a fine structure that corresponds to the one in adult hearing cats stimulated through a cochlear implant. The peak in the synaptic currents occurs at the time when the synaptic densities in the visual cortex reach their maximum values (Cragg, 1975; Winfield, 1981, 1983). That does not necessarily mean that the maximum gross synaptic currents are an accurate measure of synaptic densities, but it indicates that in hearing cats the synaptic densities and the time functions of postsynaptic currents developed coherently during the second month of life and in such a way that they produce the largest gross synaptic currents. It may be assumed that at this age, the synaptic currents and their synchronization reach maturity, and that the synaptic density is the property that then determines the maximum gross synaptic current. The development of synaptic function was different in congenitally deaf cats (Fig. 8). In these animals, the gross synaptic currents were small at 2 months of age, reached very large amplitudes at 3 months, and then decayed rapidly to give rise to
297
Fig. 7. Morphology of amplitude-normalized mean local field potentials in the ROI at 10 dB above threshold during development. (A) Nb waves of congenitally deaf cats at 2 and 3 months post natal compare well to the immature Nb waves obtained from hearing cats at 1.25 months post natal. (B) Age-matched deaf and hearing cats at 2–3 months post natal. Hearing controls are characterized by a mature-like broad Nb wave; the deaf cats, however, have still an immature shape of Nb wave. (C) With increasing age, the Nb wave develops in deaf cats, and the P1 wave demonstrates a degeneration with a decreasing and smeared relative amplitude. (D) Sample-tosample comparison of grand mean averages computed from four adult deaf cats and six adult hearing controls. Deaf cats have significantly smaller Nb wave and a smaller P1 wave (WilcoxonMannWhitney test, a ¼ 5%) (data from Kral et al., 2005).
patterns described in adult cats (see above). We interpret the synaptic currents to still be immature at 3 months of age, corresponding to a developmental delay. At that age, the synaptic densities are possibly still increasing in the auditory cortex, as at the corresponding age in naı¨ ve visual cortex (in contrast to that of sighted animals) the synaptogenesis is still in progress and the eventually reached peak synaptic density is amplified (Winfield, 1981, 1983). Immature synaptic densities and the abnormally large synaptic current may combine at this age to give rise to the large peak in both gross synaptic currents and activated cortical areas. Further morphological studies are necessary to verify this hypothesis and to determine the time course of synaptic development in the auditory cortex.
Is the ‘‘deaf’’A1 functionally decoupled from higher order auditory cortical areas? The deficits found in a naı¨ ve cortical microcircuitry may have functional significance with two important implications: 1. A desynchronization of activity between cortical layers, particularly a delay in activation of supragranular layers, disables the proper function of the cortical intrinsic microcircuitry in field A1 of deaf cats. A synchronous activation of the pyramidal cells of layer V at different portions of their dendritic tree (at the level of supragranular layers) switches the cell into a different processing mode (e.g.,
298
Fig. 8. CSD signals obtained during development in hearing controls and deaf cats. Largest signals obtained in hearing controls between the first and the second month post natal; in deaf cats, this peak is delayed to the age of 3 months post natal. Subsequently, the deficits in CSD profiles as described in adult deaf cats develop (reproduced with permission from Kral et al., 2005). For details see text.
Larkum et al., 1999; Llinas et al., 2002), which is important for the proper relay of information from the cortex to its targets (cortical and subcortical). Therefore, the naı¨ ve auditory cortex might not properly activate the ‘‘output cells’’ in the infragranular layers of the primary auditory cortex. 2. The observed reduction of activity in infragranular layers further supports the mplications above and additionally indicates that corticofugal projections (e.g., corticothalamic feedback) do not function properly. It is therefore probable that the function of thalamocorticothalamic and corticocortical loops is compromised in deafness. Reductions in infragranular layer activation also points to a reduced activity in descending cortical projections, which target these layers and which are thought to convey a cognitive top-down modulation of activity (Raizada
and Grossberg, 2003), further indicating that the primary auditory cortex is decoupled from other cortical fields in congenital deafness.
Hearing after congenital deafness: chronic stimulation with cochlear implants To what extent are all these deficits caused by the absence of auditory input, and to what extent the findings represent an intrinsic difference in the molecular or structural constitution between hearing and congenitally deaf cats? The differences in genetic makeup of congenitally deaf cats (CDCs) might show up also at other structures, in addition to the cochlea. However, attempts to reveal such differences were unsuccessful so far: Studies of the cerebellum (West and Harrison, 1973) and the nucleus of the trigeminal nerve (Saada et al., 1996)
299
Fig. 9. A cochlear implant developed for chronic electrical stimulation in congenitally deaf cat (after Behrendt, 1999). Scale in centimeters; bar in the inset ¼ 1 mm.
have not demonstrated any difference between congenitally deaf cats and hearing controls. It would be a strong argument for the hypothesis that the abnormalities are caused by deprivation of input to the auditory system if the known signs of functional deficits in congenitally deaf cats were reversed by auditory experience. For that purpose, the capacity for plastic reorganization during development has been investigated using cochlear-implanted animals to activate the auditory system in congenitally deaf cats. The auditory nerve was stimulated electrically in congenitally deaf cats using cochlear implants inserted through the round window (Fig. 9, cf. Behrendt, 1999). The implant was fed transcutaneously in the interscapular line
and covered by a tissue jacket with a backpack. After a healing phase of 3–7 days post implantation, the animals were first tested with a sinusoidal stimulus applied through the implant. The ‘‘hearing thresholds’’ for electrical stimulation in these animals were determined by observing the pinna orientation reflexes visually during presentation of these electrical stimuli. Thresholds of pinna orientation reflexes correspond to hearing thresholds in hearing cats (Ehret, 1985). Congenitally deaf cats have preserved pinna orientation reflexes (Klinke et al., 1999; Kral et al., 2002). The pinna orientation thresholds were well reproducible in each animal and the threshold increased with 6 dB/octave with increasing stimulus frequency, comparably to thresholds of auditory nerve fibers (Hartmann
300 and Klinke, 1990). Using the assumption that the threshold of these reflexes corresponds to the perceptual thresholds also in these animals, the reflex thresholds were considered as valid measures of the hearing thresholds. After determining the threshold of cochlear electrical stimulation, the animals were equipped with a signal processor using a compressed analog coding strategy delivered in a monopolar configuration to the most apical electrode of the implant (Klinke et al., 1999). The signal processors were adjusted to the individual threshold curve to reach thresholds at an acoustical stimulation level of 65 dB SPL at each frequency from 125 Hz to 8 kHz. The maximum electrical current was limited to 10 dB above threshold. The stimulation was applied without interruption (except the short times of battery control and impedance tests of the electrodes) for 1.0–5.5 months. During the stimulation period, the animals were conditioned to simple acoustic stimuli through their cochlear implants and they learned to react to an acoustic stimulus within 10–18 training sessions (i.e., in a two alternative forced-choice paradigm, the success rate exceeded 60%). The animals lived in the standard animal-house environment during the chronic stimulation. They heard, via electrical stimulation through the portable signal processor, all sounds above 65 dB SPL within the range of 125 Hz–8 kHz. These sounds included environmental sounds during handling of the animals, their own vocalizations, vocalizations of other cats from the colony, and sounds produced during play. In this respect, the animals with cochlear implants lived in a more or less ‘‘normal’’ acoustic environment of a standard animal-house condition. After 1.05.5 months of auditory experience, auditory cortices of these animals were investigated in final experiments. With the strategy described above, the auditory cortex was mapped using surface-recorded local field potentials. In general, the lowest cortical thresholds (in naive cats significantly lower than in their normal hearing counterparts acutely deafened and stimulated electrically) were not significantly affected by chronic electrical stimulation (Kral et al., 2002).
The functional organization of the auditory cortex was significantly changed by hearing experience. The most prominent difference between the stimulated CDCs and the unstimulated CDCs was the larger activated cortical area: with increasing stimulation duration, the area responding to the stimulation (biphasic pulse 200 ms/phase, monopolar configuration) grew up to a factor of 5 (Fig. 10). This finding agrees with the findings of enlarged representation of the stimulus in hearing animals after conditioning or pairing the stimulus with electrical stimulation of basal nucleus (e.g., Kilgard and Merzenich, 1998). The expansion of the activated area seems to be a meaningful reorganization when the stimulation is done with a single-channel electrical stimulation because it provides more neural tissue to process the incoming activity. By that, more computational power for processing of the stimuli is guaranteed. The increase in the activated areas was a slow process, taking many weeks to months. Therefore, this process has to involve extensive morphological reorganizations. The process was not confined to the auditory cortex contralateral to the implanted ear, but the representation of the stimulus at the ipsilateral cortex also expanded, although to a lesser degree than that at the contralateral cortex (Kral et al., 2002). It is not known if these changes in cortical representation are caused by subcortical reorganizations, or if both cortical reorganizations are solely of cortical origin. The findings on corticofugal plasticity in hearing animals (e.g., Ma and Suga, 2003; Suga and Ma, 2003) suggest that the reorganization that occurs after activation of the auditory nervous system in naı¨ ve animals is primarily of cortical origin, with subcortical reorganization following afterwards. Extension of the cortical representations does not mean that processing of auditory stimuli has changed. Therefore, within the most activated cortical area, single- and multiunit activity was further analyzed. Here, a more complex pattern of responses showed up in trained animals: the cortical units showed a higher diversity in their response patterns. Poststimulus time histograms revealed several different types of unit responses (Klinke et al., 1999). The occurrence of
301
Fig. 10. Increase in the activated area with increasing duration of chronic electrical stimulation (four, two and two animals, respectively). Activated areas were normalized to body weight (details in Kral et al., 2002), although the same effect was observed also without normalization. Significant differences between naı¨ ve animals and animals stimulated 2 months, and also between animals stimulated 2 months and 5 months (Wilcoxon-Mann-Whitney, a ¼ 5%; reprinted with permission from Kral et al., 2002).
long-latency responses increased after chronic electrical stimulation, indicating that activity within the trained auditory cortex can persist for a longer time and is more similar to activity evoked by acoustic stimulation in hearing animals (compare Eggermont, 1992). Since this increase in occurrence of long-latency responses was not connected with a significant decrease in cortical thresholds, we can assume that the observed signs of cortical reorganization are not the consequence of a general increase in cortical sensitivity. Long-latency responses in hearing animals have been attributed to a rebound of inhibition (e.g., Eggermont, 1992), and the observed increase of cells with these responses indicate a more complex excitatoryinhibitory interaction after training. Different units not only began to respond differently to the same stimulus, but the responses also became more complex (Fig. 11, cf. Kral et al., 2001). These findings indicate that the naı¨ ve cortex develops feature-detection abilities after training, as different units respond differently to the same stimulus, and different stimuli are responded differently by the same unit.
In addition to these changes, the processing of the incoming information within the intrinsic cortical neuronal networks changed after chronic electrical stimulation. With increasing stimulation duration, the CSD signals increased in amplitude (Fig. 12). This was true for both mean sink amplitudes and mean sink latencies. That means that chronic electrical stimulation (auditory experience) significantly increased the synchronized synaptic activity in the primary auditory cortex. These changes reached a plateau after approximately 3 months of stimulation. It is interesting that the synchronized synaptic activity saturated at a higher level than in hearing controls. This not only demonstrates that the cerebral cortex in naive animals has a high capacity for plastic reorganization, but also shows that chronically stimulated animals have specialized for processing of electrical stimuli. The other interesting finding relates to the structure of the CSD profiles: after 3 months of stimulation they also showed a profile that corresponded well to the one from hearing controls (Fig. 12). The latencies of the earliest sinks approached the one described in hearing controls.
302
Fig. 11. Single trace of a multi-unit response to electrical stimulation (left) and the corresponding post-stimulus time histogram (right) in field A1 of a naı¨ ve and a chronically-stimulate CDC. Below: Rate-intensity functions of single- and multiunits from ROI in naı¨ ve and chronically stimulated congenitally deaf cats. Top: in naı¨ ve animals, the variability of the rate-intensity functions is rather small, dynamic range is o6 dB, and thresholds increases by approximately 6 dB with doubling of the frequency of the stimulus. Middle: after 2 months of stimulation, rate-intensity functions changed; the dynamic range increases (middle), and units with selective to certain characteristics of the stimulus appear; however, there are also units comparable with those from naı¨ ve animals (right). Bottom: after 5 months of stimulation, units appear that respond differentially to different characteristics of stimuli, and the dynamic range of responses increases (data from Klinke et al., 1999 and Kral et al., 2001). Conditioned stimuli: 437 Hz and 732 Hz.
303
Fig. 12. CSD signals normalize after chronic electrical stimulation through cochlear implants (sinks are filled). Representative comparison between age-matched naı¨ ve congenitally deaf cat (A) and chronically stimulated deaf cat after 5 months of electrical stimulation (B). More activity was found in deep cortical layers of chronically stimulated cats. (C) Quantitative analysis of the above data. Mean sink temporal integrals increase with increasing duration of chronic electrical stimulation (mean data from ROI, 1 months stimulation: two animals; 2–5 months of stimulation: one animal; all animals shown here were implanted at 3 months after birth; WilcoxonMannWhitney test at a ¼ 5%). (D) Shortest sink latencies decreased after chronic electrical stimulation and became more comparable to hearing controls. See Plate 18.12 in Colour Plate Section.
These findings are relevant to normal listening situations because synchronized activation of cortical layers are important for the normal activation of pyramidal cells in deep cortical layers (Larkum
et al., 1999; Llinas et al., 2002). This can assure that the output activity in the auditory cortex is transferred not only to subcortical nuclei, but also to the ipsilateral and contralateral auditory cortex.
304
In addition, only an appropriately activated neuron in the deep cortical layers can receive and handle the top-down modulation from higher order auditory cortex targeting these cortical layers (review in Raizada and Grossberg, 2003). The decoupling of primary from higher order cortical areas, assumed in naı¨ ve animals, was overcome by early hearing experience. Sensitive periods When the above-described effects were related to the age at implantation, it was shown that the earlier the implantation took place, the more extensive were the reorganizations in field A1 of the auditory cortex that could be achieved (Kral et al., 2001, 2002) and the expansions of the activated areas were largest in the earliest implanted animals. The age effect was even more consistent in the cortex ipsilateral to the ear that was chronically stimulated. In animals that were implanted as adults (within the sixth month after birth), smaller expansions were found at both the ipsilateral and contralateral cortex when compared to young implanted animals. The developmental plasticity of the auditory cortex thus showed a sensitive period from the second to the sixth month of life in cats. The largest reorganizations occurred at the implantation ages when the largest cortical representations occurred in naı¨ ve CDCs. This is taken as a sign of the importance of this phenomenon for the recovery after deprivation. Nonetheless, large changes could also be achieved after the cortical representations in naı¨ ve animals have shrunken (cf. Kral et al., 2002, 2005), which demonstrates that the sensitive period for recovery (effects of chronic electrostimulation) is longer than the sensitive period in development (age span when cortical representations are large). Similar findings have been presented for the visual system (recent review in Lewis and Maurer, 2005). The sensitive period for recovery in congenitally deaf cats can be further demonstrated with the morphology of the local field potentials (evoked potentials). The latencies of the Pa wave of the field potential normally decrease with increasing stimulation duration (Fig. 13, Kral et al., 2002),
but this decrease is statistically significant only after stimulation for more than 2 months in congenitally deaf cats. However, even after 5 months of stimulation, no decrease in latency of Pa occurred in animals that were implanted after the fifth month of age. Morphology of the local field potentials not only in the long-latency but also in the middle-latency range, differed between earlyand late-implanted animals. The bases for these differences are changes in the spatiotemporal relation of current sinks and sources within the auditory cortex. Growth factors play an important role in cortical plasticity. Studies of monocular deprivation have shown that infusion of brain growth factors into the cortex moves the critical period to earlier ages (Huang et al., 1999). Brain-derived neurotrophic factor (BDNF) increases the rate of the development of inhibition (review in Berardi et al., 2000). Neurotrophic factors affect the development of those synapses that are active, more than those that are inactive (Boulanger and Poo, 1999; Zhang and Poo, 2001; Nagappan and Lu, 2005). Activity can also regulate the amount of neurotrophic substances produced by neurons and the number of receptors for neurotrophins (Zafra et al., 1992; Meyer-Franke et al., 1998). These substances have a trophic effect on, e.g., growth of dendrites (McAllister et al., 1996; Horch, 2004; Dijkhuizen and Ghosh, 2005). This differential expression of neurotrophic factors during development could contribute to the creation of sensitive periods (Sermasi et al., 1999a; Lein et al., 2000), although some neurotrophins do not change their expression during development (e.g., Ichisaka et al., 2003). The end of the sensitive period in congenital auditory deprivation coincides approximately with the onset of puberty (6 months in cats). This of course does not mean that the auditory cortex in the adult brain lacks plasticity, but certain factors limit the extent of expression of neural plasticity in adults. Early in development, sensory experience can affect the development of the overall synaptic organization, which then leads to more efficient and durable learning than in the adult. While plasticity can be induced by ‘‘passive’’ listening during development (Zhang et al., 2001, 2002), a
305
Fig. 13. The capacity for plastic reorganization with cochlear implant stimulation decreases with increasing age. (A) Massive (but slow) expansions of the cortical activated areas, as demonstrated for early implanted animals in Fig. 10, become smaller with increasing implantation age (asterisk marks the dorsal end of the posterior ectosylvian sulcus; Kral et al., 2001, 2002). (B) The sensitive period can also be demonstrated in latencies of Pa waves, that become significantly smaller after 5 months of stimulation at early implantation (age 3.5 months post natal), but this decrease was smaller after implantation age 5 months post natal and was no longer discernible with implantation in adult age (6 months post natal Wilcoxon-Mann-Whitney test, a ¼ 0.4%; reprinted with permission from Kral et al., 2006). See Plate 18.13 in Colour Plate Section.
306
similar extent of plasticity in adults can only be achieved by pairing of stimuli with some instruction factor like electrical activation of nucleus basalis or active conditioning (Bakin and Weinberger, 1996; Kilgard and Merzenich, 2002; Bao et al., 2003). Finally, it must be taken into account that the naive adult auditory cortex might be differently reorganizing than the ‘‘hearing’’ adult cortex (in normal hearing animals) under comparable stimulation condition. After certain developmental stages have been reached in absence of auditory experience, the starting point of cortical organization will differ from the normalhearing counterparts. Studies of the somatosensory system have revealed an important cause of difference between early and adult plasticity. The overall number of synapses remains relatively constant in the adult barrel cortex, despite the extensive plasticity and the formation of new synaptic contacts (Trachtenberg et al., 2002). This indicates that the number of synapses, at least in the primary sensory cortex of adult animals, is relatively stable over time (for further evidence in the visual cortex, see O’Kusky and Colonnier, 1982a, b) and represents one important limiting factor for expression of adult plasticity. During development, synaptic densities change dramatically under the influence of experience (see above), so that possibly much more extensive wiring changes result within a short time. We still did not reach full understanding of the mechanism of these processes. All expressions of plasticity that have been observed at the level of synapses and receptors took place within minutes to hours, but the plasticity demonstrated in the above experiments took place in course of months.
Clinical relevance The topics discussed in this text are of cardinal relevance in clinical practice. Diagnosis of congenital deafness is hampered by the fact that the human cochlea becomes functional already during the intrauterine life (Granier-Deferre et al., 1985), but the diagnosis of hearing loss is possible only after birth. This is why clinicians differentiate only
between prelingual (before language has started being acquired) and postlingual deafness or hearing loss. The incidence of congenital hearing loss is about one per thousand. Hearing screenings have been introduced in several countries to detect hearing disabilities early (e.g., O’Donoghue, 1996, 1999; O’Donoghue et al. 1998). There are now strong indications that the reorganizations of the auditory nervous system discussed in this chapter takes place also in cochlearimplanted prelingually deaf children. It has been known for a long time that cochlear implantation in prelingually deaf adults does not lead to ‘‘openset’’ speech understanding (Busby et al., 1992, 1993; Dawson et al., 1992; Tyler and Lowder, 1992; Gantz et al., 1993, 1994;), and it is consequently recommend that cochlear implantation is performed before the age of 5 years (FryaufBertschy et al., 1997; Waltzman and Cohen, 1998; Schauwers et al., 2004). These recommendations were further supported by electrophysiological investigations. Children that were prelingually deaf and received a cochlear implant before their teen ages showed a delay in the normal development of the morphology of cortical evoked potentials (Ponton et al., 1996a, b; Eggermont et al., 1997; for further developmental data in cochlear-implanted children, also cf. Gordon et al., 2002, 2005). The development of normal latencies of cortical evoked potentials was delayed approximately by the same amount of time as the duration of deafness. After stimulation using a cochlear implant, the latencies of the evoked potentials matured, yet the original delay in development of latencies remained as if the children were developmentally delayed by the duration of their deafness. That led the authors to hypothesize that the latency of P1 is reduced during development to an extension that corresponds to the duration of hearing experience (time in sound). Nonetheless, these authors did not study children implanted under 4 years of age. Once children implanted before their fourth year of age were evaluated in the same way as the children implanted later, it was found that the early-implanted children caught up with their maturational delays in P1 latency within few months after implantation (Sharma et al., 2002a, b, c, 2005). Consequently,
307
there is a sensitive period in the development of auditory evoked potentials, which corresponds to the sensitive period in speech comprehension (Fryauf-Bertschy et al., 1997). Development of a mature N1 wave represents another sensitive period. Prelingually deaf children implanted in their teen ages do not develop a mature N1 wave, which follows the P1 wave in individuals with hearing (Ponton and Eggermont, 2001). Children implanted in their teens also do not achieve ‘‘open-set’’ speech comprehension without lipreading (Busby et al., 1992, 1993; Dawson et al., 1992; Tyler and Lowder, 1992; Gantz et al., 1993, 1994). It is known that higher order auditory and nonauditory areas are recruited during speech recognition in cochlearimplanted individuals and their activation correlates with achievement of speech comprehension (Giraud et al., 2000). The N1 wave is known to be generated in higher order auditory areas (LiegeoisChauvel et al., 1994), and thus it appears as if these areas were not properly activated in prelingually deaf children implanted late (in their teens). Studies of prelingually deaf children allow further conclusions affecting the neurophysiological question on plasticity in the naive auditory system. There are at least two phases of plastic reorganizations in the auditory cortex. The first one is a fast and extensive one, taking place in the first few weeks after cochlear implantation. The P1 latencies matured fast in the first short phase of plasticity in implanted children (Sharma et al., 2005). This phase was also found in children who were implanted late in life (the first, fast decay of P1 latency during the first few weeks after implant activation did not show a sensitive period). The second phase of decrease in P1 latency, which took place later (during the months after implantation), was much slower and less extensive, but showed a sensitive period. Late-implanted children did show this phase only in a very rudimentary way.
Conclusions The development of the auditory system depends critically on auditory experience. In absence of
hearing, the primary auditory cortex remains capable of responding to auditory stimuli, but the functionality of the auditory cortex is massively affected by the deprivation. In this respect, the naive auditory cortex represents a significantly different starting point for plastic cortical reorganizations compared to cortex in animals with hearing (acoustically-competent). Studies of neonatally deaf animals and studies of congenitally (or prelingually) deaf children indicate that the changes in the function of the auditory cortices that occur after restoration of hearing via cochlear implants can be differentiated into three phases: 1. The first phase spans the first days and weeks after implantation in humans. This very fast reorganization process (e.g., the fast decrease of P1 latency) does not have a specific sensitive period. The changes are most likely related to restoration of inhibitory function in the cortex and restoration of homeostatic regulation of neuronal excitability, increasing the synchronization of evoked activity. It may be connected with a fast reduction of the large (immature) gross synaptic currents and large activated areas found in young congenitally deaf animals. 2. The second phase is a slower reorganization, taking place within weeks to months after implantation. These changes constitute a sensitive period in humans and animals. The corresponding reorganization most probably includes an increase in cortical representation of the stimulated cortical region, restoration of the functionality of the cortical intrinsic microcircuitry, changes in the latencies of field potentials, and CSD signals and reappearance of the long-latency responses. It is in this phase that the primary areas most probably reorganize and sharpen their feature-detection abilities. 3. The third phase is the slowest, being a consequence of the restoration of the functionality of the primary cortical areas. It involves recruitment of higher order auditory cortices by the stimulus (e.g., Giraud et al., 2001a, b c; Giraud and Truy, 2002) and involves
308
formation of descending projections from higher order areas to the primary areas. It is probably connected with the appearance of N1 and later waves of evoked potentials. This phase initially overlaps with phase 2.
Acknowledgments The work has been supported by Deutsche Forschungsgemeinschaft and in part by DAAD. The authors wish to thank Dr. Anu Sharma and Dr. Peter Hubka for many fruitful discussions; Dr. Aage R. Moller and Dr. Andrew Sharott for critical comments on an earlier version of this manuscript; and N. Krimmel, H. Schalk, and D. Schiemann for excellent technical help. References Bakin, J.S. and Weinberger, N.M. (1996) Induction of a physiological memory in the cerebral cortex by stimulation of the nucleus basalis. Proc. Natl. Acad. Sci. USA, 93: 11219–11224. Bao, S., Chang, E.F., Davis, J.D., Gobeske, K.T. and Merzenich, M.M. (2003) Progressive degradation and subsequent refinement of acoustic representations in the adult auditory cortex. J. Neurosci., 23: 10765–10775. Behrendt, M., 1999. Entwicklung und Herstellung eines Cochlea-Implantates zur chronischen Stimulation von geho¨rlosen weiXen Katzen. J.W.Goethe University, Frankfurt am Main, Germany. Berardi, N., Pizzorusso, T. and Maffei, L. (2000) Critical periods during sensory development. Curr. Opin. Neurobiol., 10: 138–145. Blatchley, B.J. and Brugge, J.F. (1990) Sensitivity to binaural intensity and phase difference cues in kitten inferior colliculus. J. Neurophysiol., 64: 582–597. Bliss, T.V. and Lomo, T. (1973) Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J. Physiol., 232: 331–356. Bonham, B.H., Cheung, S.W., Godey, B. and Schreiner, C.E. (2004) Spatial organization of frequency response areas and rate/level functions in the developing AI. J. Neurophysiol., 91(2): 841–854. Boulanger, L. and Poo, M.M. (1999) Presynaptic depolarization facilitates neurotrophin-induced synaptic potentiation. Nat. Neurosci., 2: 346–351. Brugge, J.F. (1992) Development of the lower auditory brainstem of the cat. In: Romand, R. (Ed.), Development of Auditory and Vestibular Systems 2. Elsevier Science Publishers B.V., Amsterdam, pp. 173–296.
Brugge, J.F., Javel, E. and Kitzes, L.M. (1978) Signs of functional maturation of peripheral auditory system in discharge patterns of neurons in anteroventral cochlear nucleus of kitten. J. Neurophysiol., 41: 1557–1559. Brugge, J.F. and O’Connor, T.A. (1984) Postnatal functional development of the dorsal and posteroventral cochlear nuclei of the cat. J. Acoust. Soc. Am., 75: 1548–1562. Brugge, J.F., Reale, R.A. and Wilson, G.F. (1988) Sensitivity of auditory cortical neurons of kittens to monaural and binaural high frequency sound. Hear. Res., 34: 127–140. Busby, P.A., Tong, Y.C. and Clark, G.M. (1992) Psychophysical studies using a multiple-electrode cochlear implant in patients who were deafened early in life. Audiology, 31: 95–111. Busby, P.A., Tong, Y.C. and Clark, G.M. (1993) Electrode position, repetition rate, and speech perception by earlydeafened and late-deafened cochlear implant patients. J. Acoust. Soc. Am., 93: 1058–1067. Cant, N.B. (1998) Structural development of the mammalian auditory pathways. In: Rubel, E.W., Popper, A.N. and Fay, R.R. (Eds.), Development of the Auditory System. Springer Verlag, New York, pp. 315–413. Cragg, B.G. (1975) The development of synapses in the visual system of the cat. J. Comp. Neurol., 160: 147–166. Crair, M.C. and Malenka, R.C. (1995) A critical period for long-term potentiation at thalamocortical synapses. Nature, 375: 325–328. Cruikshank, S.J., Rose, H.J. and Metherate, R. (2002) Auditory thalamocortical synaptic transmission in vitro. J. Neurophysiol., 87: 361–384. Dauman, R., Tyler, R.S. and Aran, J.M. (1993) Intracochlear electrical tinnitus reduction. Acta Otolaryngol., 113: 291–295. Dawson, P.W., Blamey, P.J., Rowland, L.C., Dettman, S.J., Clark, G.M., Busby, P.A., Brown, A.M., Dowell, R.C. and Rickards, F.W. (1992) Cochlear implants in children, adolescents, and prelinguistically deafened adults: speech perception. J. Speech Hear. Res., 35: 401–417. Dijkhuizen, P.A. and Ghosh, A. (2005) BDNF regulates primary dendrite formation in cortical neurons via the PI3-kinase and MAP kinase signaling pathways. J. Neurobiol., 62: 278–288. Dodson, H.C. (1997a) Loss and survival of spiral ganglion neurons in the guinea pig after intracochlear perfusion with aminoglycosides. J. Neurocytol., 26: 541–556. Dodson, H.C. (1997b) Total cochlear hair cell destruction results in apoptosis in the spiral ganglion of the guinea pig. Brit. J. Audiol., 31: 79. Dodson, H.C. and Mohuiddin, A. (2000) Response of spiral ganglion neurones to cochlear hair cell destruction in the guinea pig. J. Neurocytol., 29(7): 525–537. Eggermont, J.J. (1991) Maturational aspects of periodicity coding in cat primary auditory cortex. Hear. Res., 57: 45–56. Eggermont, J.J. (1992) Stimulus induced and spontaneous rhythmic firing of single units in cat primary auditory cortex. Hear. Res., 61: 1–11.
309 Eggermont, J.J. (1996) Differential maturation rates for response parameters in cat primary auditory cortex. Auditory Neurosci., 2: 309–327. Eggermont, J.J., Ponton, C.W., Don, M., Waring, M.D. and Kwong, B. (1997) Maturational delays in cortical evoked potentials in cochlear implant users. Acta Otolaryngol., 117: 161–163. Ehret, G. (1985) Behavioural studies on auditory development in mammals in relation to higher nervous system functioning. Acta Otolaryngol. Suppl., 421: 31–40. Ehret, G. and Romand, R. (1981) Postnatale development of absolute auditory thresholds in kittens. J. Comp. Physiol. Psychol., 95: 304–311. Finney, E.M., Clementz, B.A., Hickok, G. and Dobkins, K.R. (2003) Visual stimuli activate auditory cortex in deaf subjects: evidence from MEG. Neuroreport, 14(11): 1425–1427. Finney, E.M., Fine, I. and Dobkins, K.R. (2001) Visual stimuli activate auditory cortex in the deaf. Nat. Neurosci., 4: 1171–1173. Friauf, E. and Shatz, C.J. (1991) Changing patterns of synaptic input to subplate and cortical plate during development of visual cortex. J. Neurophysiol., 66: 2059– 2071. Fryauf-Bertschy, H., Tyler, R.S., Kelsay, D.M., Gantz, B.J. and Woodworth, G.G. (1997) Cochlear implant use by prelingually deafened children: the influences of age at implant and length of device use. J. Speech Lang. Hear. Res., 40: 183–199. Gantz, B.J., Tyler, R.S., Woodworth, G.G., Tye-Murray, N. and Fryauf-Bertschy, H. (1994) Results of multichannel cochlear implants in congenital and acquired prelingual deafness in children: five-year follow-up. Am. J. Otol., 15(Suppl 2): 1–7. Gantz, B.J., Woodworth, G.G., Knutson, J.F., Abbas, P.J. and Tyler, R.S. (1993) Multivariate predictors of audiological success with multichannel cochlear implants. Ann. Otol. Rhinol. Laryngol., 102: 909–916. Giraud, A.L., Price, C.J., Graham, J.M. and Frackowiak, R.S.J. (2001a) Functional plasticity of language-related brain areas after cochlear implantation. Brain, 124: 1307–1316. Giraud, A.L., Price, C.J., Graham, J.M., Truy, E. and Frackowiak, R.S.J. (2001b) Cross-modal plasticity underpins language recovery after cochlear implantation. Neuron, 30(3): 657–663. Giraud, A.L. and Truy, E. (2002) The contribution of visual areas to speech comprehension: a PET study in cochlear implants patients and normal-hearing subjects. Neuropsychologia, 40(9): 1562–1569. Giraud, A.L., Truy, E. and Frackowiak, R. (2001c) Imaging plasticity in cochlear implant patients. Audiol. Neuro. Otol., 6(6): 381–393. Giraud, A.L., Truy, E., Frackowiak, R.S., Gregoire, M.C., Pujol, J.F. and Collet, L. (2000) Differential recruitment of the speech processing system in healthy subjects and rehabilitated cochlear implant patients. Brain, 123(Pt 7): 1391–1402.
Gordon, K.A., Papsin, B.C. and Harrison, R.V. (2002) Auditory brain stem and midbrain development after cochlear implantation in children. Ann. Otol. Rhinol. Laryngol. Suppl., 189: 32–37. Gordon, K.A., Papsin, B.C. and Harrison, R.V. (2005) Effects of cochlear implant use on the electrically evoked middle latency response in children. Hear. Res., 204: 78–89. Granier-Deferre, C., Lecanuet, J.P., Cohen, H. and Busnel, M.C. (1985) Feasibility of prenatal hearing test. Acta Otolaryngol. (Stockh.) Suppl., 421: 93–101. Hardie, N.A., Martsi-McClintock, A., Aitkin, L.M. and Shepherd, R.K. (1998) Neonatal sensorineural hearing loss affects synaptic density in the auditory midbrain. Neuroreport, 9: 2019–2022. Hartmann, R. and Klinke, R. (1990) Response characteristics of nerve fibers to patterned electrical stimulation. In: Miller, J.M. and Spelman, F.A. (Eds.), Cochlear Implants. Models of the Electrically Stimulated Ear. Springer Verlag, New York, pp. 135–160. Hartmann, R. and Kral, A. (2004) Central responses to electrical stimulation. In: Zeng, F.G., Popper, A.N. and Fay, R.R. (Eds.), Cochlear Implants: Auditory Prostheses and Electric Hearing. Springer, New York, pp. 213–285. Hartmann, R., Shepherd, R.K., Heid, S. and Klinke, R. (1997) Response of the primary auditory cortex to electrical stimulation of the auditory nerve in the congenitally deaf white cat. Hear. Res., 112: 115–133. Hartmann, R., Topp, G. and Klinke, R. (1984) Discharge patterns of cat primary auditory fibers with electrical stimulation of the cochlea. Hear. Res., 13: 47–62. Heid, S., Hartmann, R. and Klinke, R. (1998) A model for prelingual deafness, the congenitally deaf white cat — population statistics and degenerative changes. Hear. Res., 115: 101–112. Heid, S., Jahnsiebert, T.K., Klinke, R., Hartmann, R. and Langner, G. (1997) Afferent projection patterns in the auditory brainstem in normal and congenitally deaf white cats. Hear. Res., 110: 191–199. Heil, P., Rajan, R. and Irvine, D.R. (1992) Sensitivity of neurons in cat primary auditory cortex to tones and frequencymodulated stimuli. II: Organization of response properties along the ‘isofrequency’ dimension. Hear. Res., 63: 135–156. Horch, H.W. (2004) Local effects of BDNF on dendritic growth. Rev. Neurosci., 15: 117–129. Huang, Z.J., Kirkwood, A., Pizzorusso, T., Porciatti, V., Morales, B., Bear, M.F., Maffei, L. and Tonegawa, S. (1999) BDNF regulates the maturation of inhibition and the critical period of plasticity in mouse visual cortex. Cell, 98: 739–755. Hubka, P., Kral, A. and Klinke, R. (2004) Input desynchronization and impaired columnar activation in deprived auditory cortex revealed by independent component analysis. In: Syka, J. and Merzenich, M.M. (Eds.), Plasticity and Signal Representation in the Auditory System. Springer Verlag, Berlin, pp. 161–165. Hultcrantz, M., Snyder, R., Rebscher, S. and Leake, P. (1991) Effects of neonatal deafening and chronic intracochlear
310 electrical stimulation on the cochlear nucleus in cats. Hear. Res., 54: 272–280. Ichisaka, S., Katoh-Semba, R., Hata, Y., Ohshima, M., Kameyama, K. and Tsumoto, T. (2003) Activity-dependent change in the protein level of brain-derived neurotrophic factor but no change in other neurotrophins in the visual cortex of young and adult ferrets. Neuroscience, 117: 361–371. Ito, J. and Sakakihara, J. (1994) Tinnitus suppression by electrical stimulation of the cochlear wall and by cochlear implantation. Laryngoscope, 104: 752–754. Ito, M., Sakurai, M. and Tongroach, P. (1982) Climbing fibre induced depression of both mossy fibre responsiveness and glutamate sensitivity of cerebellar Purkinje cells. J. Physiol., 324: 113–134. Kaczmarek, L., Kossut, M. and Skangiel-Kramska, J. (1997) Glutamate receptors in cortical plasticity: molecular and cellular biology. Physiol. Rev., 77: 217–255. Kalisman, N., Silberberg, G. and Markram, H. (2005) The neocortical microcircuit as a tabula rasa. Proc. Natl. Acad. Sci. USA, 102: 880–885. Kettner, R.E., Feng, J.Z. and Brugge, J.F. (1985) Postnatal development of the phase-locked response to low frequency tones of auditory nerve fibers in the cat. J. Neurosci., 5: 275–283. Kiefer, J., Gstoettner, W., Baumgartner, W., Pok, S.M., Tillein, J., Ye, Q. and von Ilberg, C. (2004) Conservation of lowfrequency hearing in cochlear implantation. Acta Oto. Laryngol., 124(3): 272–280. Kiernan, A.E. and Steel, K.P. (2000) Mouse homologues for human deafness. Adv. Otorhinolaryngol., 56: 233–243. Kilgard, M.P. and Merzenich, M.M. (1998) Cortical map reorganization enabled by nucleus basalis activity. Science, 279: 1714–1718. Kilgard, M.P. and Merzenich, M.M. (2002) Order-sensitive plasticity in adult primary auditory cortex. Proc. Natl. Acad. Sci. USA, 99: 3205–3209. Klinke, R., Kral, A., Heid, S., Tillein, J. and Hartmann, R. (1999) Recruitment of the auditory cortex in congenitally deaf cats by long- term cochlear electrostimulation. Science, 285(5434): 1729–1733. Knudsen, E.I. (2004) Sensitive periods in the development of the brain and behavior. J. Cogn. Neurosci., 16: 1412– 1425. Konig, N., Pujol, R. and Marty, R. (1972) A laminar study of evoked potentials and unit responses in the auditory cortex of the postnatal cat. Brain Res., 36: 469–473. Kotak, V.C., Fujisawa, S., Lee, F.A., Karthikeyan, O., Aoki, C. and Sanes, D.H. (2005) Hearing loss raises excitability in the auditory cortex. J. Neurosci., 25: 3908–3918. Kral, A., Hartmann, R., Mortazavi, D. and Klinke, R. (1998) Spatial resolution of cochlear implants: the electrical field and excitation of auditory afferents. Hear. Res., 121: 11–28. Kral, A., Hartmann, R., Tillein, J., Heid, S. and Klinke, R. (2000) Congenital auditory deprivation reduces synaptic
activity within the auditory cortex in a layer-specific manner. Cereb. Cortex, 10: 714–726. Kral, A., Hartmann, R., Tillein, J., Heid, S. and Klinke, R. (2001) Delayed maturation and sensitive periods in the auditory cortex. Audiol. Neurootol., 6: 346–362. Kral, A., Hartmann, R., Tillein, J., Heid, S. and Klinke, R. (2002) Hearing after congenital deafness: central auditory plasticity and sensory deprivation. Cereb. Cortex, 12(8): 797–807. Kral, A., Schroder, J.H., Klinke, R. and Engel, A.K. (2003) Absence of cross-modal reorganization in the primary auditory cortex of congenitally deaf cats. Exp. Brain Res., 153(4): 605–613. Kral, A., Tillein, J., Heid, S., Hartmann, R. and Klinke, R. (2005) Postnatal cortical development in congenital auditory deprivation. Cereb. Cortex, 15: 552–562. Kral, A., Hartmann, R. and Klinke, R. (2006) Recruitment of the auditory cortex in congentially deaf cats. In: Lomber, S.G. and Eggermont, J.J. (Eds.), Reprogramming the Cerebral Cortex. Oxford University Press, pp. 191–210. Kuhl, P.K. (2004) Early language acquisition: cracking the speech code. Nat. Rev. Neurosci., 5: 831–843. Larkum, M.E., Zhu, J.J. and Sakmann, B. (1999) A new cellular mechanism for coupling inputs arriving at different cortical layers. Nature, 398: 338–341. Larsen, S.A. and Kirchhoff, T.M. (1992) Anatomical evidence of synaptic plasticity in the cochlear nuclei of white-deaf cats. Exp. Neurol., 115: 151–157. Leake, P.A. and Hradek, G.T. (1988) Cochlear pathology of long term neomycin induced deafness in cats. Hear. Res., 33: 11–34. Leake, P.A., Hradek, G.T. and Snyder, R.L. (1999) Chronic electrical stimulation by a cochlear implant promotes survival of spiral ganglion neurons after neonatal deafness. J. Comp. Neurol., 412(4): 543–562. Leake, P.A., Snyder, R.L. and Schreiner, C.E. (1987) Cochlear pathology of sensorineural deafness in cats: coadministration of kanamycin and aminooxyacetic acid. Ann. Otol. Rhinol. Laryngol., 96(Suppl 128): 48–50. Leake-Jones, P.A., Vivion, M.C., O’Reilly, B.F. and Merzenich, M.M. (1982) Deaf animal models for studies of a multichannel cochlear prosthesis. Hear. Res., 8: 225–246. Lein, E.S., Hohn, A. and Shatz, C.J. (2000) Dynamic regulation of BDNF and NT-3 expression during visual system development. J. Comp. Neurol., 420: 1–18. Lewis, T.L. and Maurer, D. (2005) Multiple sensitive periods in human visual development: evidence from visually deprived children. Dev. Psychobiol., 46: 163–183. Liegeois-Chauvel, C., Musolino, A., Badier, J.M., Marquis, P. and Chauvel, P. (1994) Evoked potentials recorded from the auditory cortex in man: evaluation and topography of the middle latency components. Electroencephalogr. Clin. Neurophysiol., 92: 204–214. Llinas, R.R., Leznik, E. and Urbano, F.J. (2002) Temporal binding via cortical coincidence detection of specific and nonspecific thalamocortical inputs: a voltage-dependent dye-
311 imaging study in mouse brain slices. Proc. Natl. Acad. Sci. USA, 99: 449–454. Lustig, L.R., Leake, P.A., Snyder, R.L. and Rebscher, S.J. (1994) Changes in the cat cochlear nucleus following neonatal deafening and chronic intracochlear electrical stimulation. Hear. Res., 74: 29–37. Ma, X.F. and Suga, N. (2003) Augmentation of plasticity of the central auditory system by the basal forebrain and/or somatosensory cortex. J. Neurophysiol., 89(1): 90–103. Maffei, A., Nelson, S.B. and Turrigiano, G.G. (2004) Selective reconfiguration of layer 4 visual cortical circuitry by visual deprivation. Nat. Neurosci., 7: 1353–1359. Markram, H., Lubke, J., Frotscher, M. and Sakmann, B. (1997) Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs. Science, 275: 213–215. McAllister, A.K., Katz, L.C. and Lo, D.C. (1996) Neurotrophin regulation of cortical dendritic growth requires activity. Neuron, 17: 1057–1064. McKerrow, W.S., Schreiner, C.E., Snyder, R.L., Merzenich, M.M. and Toner, J.G. (1991) Tinnitus suppression by cochlear implants. Ann. Otol. Rhinol. Laryngol., 100: 552–558. Meyer-Franke, A., Wilkinson, G.A., Kruttgen, A., Hu, M., Munro, E., Hanson Jr, M.G., Reichardt, L.F. and Barres, B.A. (1998) Depolarization and cAMP elevation rapidly recruit TrkB to the plasma membrane of CNS neurons. Neuron, 21: 681–693. Middlebrooks, J.C., Bierer, J.A. and Snyder, R.L. (2005) Cochlear implants: the view from the brain. Curr. Opin. Neurobiol., 15: 488–493. Mitzdorf, U. (1985) Current source-density method and application in cat cerebral cortex: investigation of evoked potentials and EEG phenomena. Physiol. Rev., 65: 37–100. Miyamoto, R.T., Kirk, K.I., Renshaw, J. and Hussain, D. (1999) Cochlear implantation in auditory neuropathy. Laryngoscope, 109: 181–185. Moore, D.R. and Irvine, D.R. (1979) The development of some peripheral and central auditory responses in the neonatal cat. Brain Res., 163: 49–59. Nagappan, G. and Lu, B. (2005) Activity-dependent modulation of the BDNF receptor TrkB: mechanisms and implications. Trends Neurosci., 28: 464–471. Nicholson, C. and Freeman, J.A. (1975) Theory of current source-density analysis and determination of conductivity tensor for anuran cerebellum. J. Neurophysiol., 38: 356–368. Nishimura, H., Hashikawa, K., Doi, K., Iwaki, T., Watanabe, Y., Kusuoka, H., Nishimura, T. and Kubo, T. (1999) Sign language ‘heard’ in the auditory cortex. Nature, 397(6715): 116. O’Donoghue, G.M. (1996) Cochlear implants in children: principles, practice and predictions. J. R. Soc. Med., 89: P345–P347. O’Donoghue, G.M. (1999) Hearing without ears: do cochlear implants work in children? BMJ, 318: 72–73.
O’Donoghue, G.M., Nikolopoulos, T., Archbold, S.M. and Tait, M. (1998) Congenitally deaf children following cochlear implantation. Acta Otorhinolaryngol. Belg., 52: 111–114. O’Kusky, J. and Colonnier, M. (1982a) A laminar analysis of the number of neurons, glia, and synapses in the adult cortex (area 17) of adult macaque monkeys. J. Comp. Neurol., 210: 278–290. O’Kusky, J. and Colonnier, M. (1982b) Postnatal changes in the number of astrocytes, oligodendrocytes, and microglia in the visual cortex (area 17) of the macaque monkey: a stereological analysis in normal and monocularly deprived animals. J. Comp. Neurol., 210: 307–315. Pallas, S.L. (2001) Intrinsic and extrinsic factors that shape neocortical specification. Trends Neurosci., 24: 417–423. Pallas, S.L., Littman, T. and Moore, D.R. (1999) Cross-modal reorganization of callosal connectivity without altering thalamocortical projections. Proc. Natl. Acad. Sci. USA, 96: 8751–8756. Pallas, S.L. and Sur, M. (1993) Visual projections induced into the auditory pathway of ferrets .2. Corticocortical connections of primary auditory cortex. J. Comp. Neurol., 337: 317–333. Payne, B.R. (1992) Development of the auditory cortex. In: Romand, R. (Ed.), Development of Auditory and Vestibular Systems 2. Elsevier Science Publishers B.V., Amsterdam, pp. 357–390. Petitto, L.A., Zatorre, R.J., Gauna, K., Nikelski, E.J., Dostie, D. and Evans, A.C. (2000) Speech-like cerebral activity in profoundly deaf people processing signed languages: implications for the neural basis of human language. Proc. Natl. Acad. Sci. USA, 97: 13961–13966. Pitts, W. (1952) Investigations on synaptic transmission. In: von Foerster H. (Ed.), Cybernetics, Trans. 9th Conf. Josiah Macy, New York, pp. 159–162. Ponton, C.W., Don, M., Eggermont, J.J., Waring, M.D., Kwong, B. and Masuda, A. (1996a) Auditory system plasticity in children after long periods of complete deafness. Neuroreport, 8: 61–65. Ponton, C.W., Don, M., Eggermont, J.J., Waring, M.D. and Masuda, A. (1996b) Maturation of human cortical auditory function: differences between normal-hearing children and children with cochlear implants. Ear. Hear., 17: 430–437. Ponton, C.W. and Eggermont, J.J. (2001) Of kittens and kids: altered cortical maturation following profound deafness and cochlear implant use. Audiol. Neurootol., 6: 363–380. Prieto, J.J., Peterson, B.A. and Winer, J.A. (1994a) Laminar distribution and neuronal targets of GABAergic axon terminals in cat primary auditory cortex (AI). J. Comp. Neurol., 344: 383–402. Prieto, J.J., Peterson, B.A. and Winer, J.A. (1994b) Morphology and spatial distribution of GABAergic neurons in cat primary auditory cortex (AI). J. Comp. Neurol., 344: 349–382. Raggio, M.W. and Schreiner, C.E. (1994) Neuronal responses in cat primary auditory cortex to electrical cochlear
312 stimulation .1. Intensity dependence of firing rate and response latency. J. Neurophysiol., 72: 2334–2359. Raggio, M.W. and Schreiner, C.E. (1999) Neuronal responses in cat primary auditory cortex to electrical cochlear stimulation. III. Activation patterns in short- and long-term deafness. J. Neurophysiol., 82: 3506–3526. Raizada, R.D.S. and Grossberg, S. (2003) Towards a theory of the laminar architecture of cerebral cortex: computational clues from the visual system. Cereb. Cortex, 13: 100–113. Rebillard, G., Carlier, E., Rebillard, M. and Pujol, R. (1977) Enhancement of visual responses on the primary auditory cortex of the cat after an early destruction of cochlear receptors. Brain Res., 129: 162–164. Rebillard, G., Rebillard, M. and Pujol, R. (1980) Factors affecting the recording of visual-evoked potentials from the deaf cat primary auditory cortex. Brain Res., 188: 252–254. Rittenhouse, C.D., Shouval, H.Z., Paradiso, M.A. and Bear, M.F. (1999) Monocular deprivation induces homosynaptic long-term depression in visual cortex. Nature, 397: 347–350. Roe, A.W., Pallas, S.L., Kwon, Y.H. and Sur, M. (1992) Visual projections routed to the auditory pathway in ferrets — receptive fields of visual neurons in primary auditory cortex. J. Neurosci., 12: 3651–3664. Rozas, C., Frank, H., Heynen, A.J., Morales, B., Bear, M.F. and Kirkwood, A. (2001) Developmental inhibitory gate controls the relay of activity to the superficial layers of the visual cortex. J. Neurosci., 21: 6791–6801. Ruben, R.J. (1997) A time frame of critical/sensitive periods of language development. Acta Oto. Laryngol., 117: 202–205. Rubinstein, J.T., Wilson, B.S., Finley, C.C. and Abbas, P.J. (1999) Pseudospontaneous activity: stochastic independence of auditory nerve fibers with electrical stimulation. Hear. Res., 127: 108–118. Ruckenstein, M.J., Hedgepeth, C., Rafter, K.O., Montes, M.L. and Bigelow, D.C. (2001) Tinnitus suppression in patients with cochlear implants. Otol. Neurotol., 22: 200–204. Ryugo, D.K., Pongstaporn, T., Huchton, D.M. and Niparko, J.K. (1997) Ultrastructural analysis of primary endings in deaf white cats: morphologic alterations in endbulbs of Held. J. Comp. Neurol., 385: 230–244. Saada, A.A., Niparko, J.K. and Ryugo, D.K. (1996) Morphological changes in the cochlear nucleus of congenitally deaf white cats. Brain Res., 736: 315–328. Sanes, D.H. and Walsh, E.J. (1998) The development of central auditory processing. In: Rubel, E.W., Popper, A.N. and Fay, R.R. (Eds.), Development of the Auditory System. Springer, New York, pp. 271–314. Schauwers, K., Gillis, S., Daemers, K., De Beukelaer, C. and Govaerts, P.J. (2004) Cochlear implantation between 5 and 20 months of age: the onset of babbling and the audiologic outcome. Otol. Neurotol., 25: 263–270. Schreiner, C.E. and Mendelson, J.R. (1990) Functional topography of cat primary auditory cortex: distribution of integrated excitation. J. Neurophysiol., 64: 1442–1459. Schreiner, C.E. and Sutter, M.L. (1992) Topography of excitatory bandwidth in cat primary auditory cortex: single-
neuron versus multiple-neuron recordings. J. Neurophysiol., 68: 1487–1502. Sermasi, E., Tropea, D. and Domenici, L. (1999a) A new form of synaptic plasticity is transiently expressed in the developing rat visual cortex: a modulatory role for visual experience and brain-derived neurotrophic factor. Neuroscience, 91: 163–173. Sermasi, E., Tropea, D. and Domenici, L. (1999b) Long term depression is expressed during postnatal development in rat visual cortex: a role for visual experience. Brain Res. Dev. Brain Res., 113: 61–65. Shallop, J.K., Peterson, A., Facer, G.W., Fabry, L.B. and Driscoll, C.L. (2001) Cochlear implants in five cases of auditory neuropathy: postoperative findings and progress. Laryngoscope, 111: 555–562. Sharma, A., Dorman, M.F. and Kral, A. (2005) The influence of a sensitive period on central auditory development in children with unilateral and bilateral cochlear implants. Hear. Res., 203: 134–143. Sharma, A., Dorman, M.F. and Spahr, A.J. (2002a) A sensitive period for the development of the central auditory system in children with cochlear implants: implications for age of implantation. Ear. Hear., 23: 532–539. Sharma, A., Dorman, M., Spahr, A. and Todd, N.W. (2002b) Early cochlear implantation in children allows normal development of central auditory pathways. Ann. Otol. Rhinol. Laryngol. Suppl., 189: 38–41. Sharma, A., Dorman, M.F. and Spahr, A.J. (2002c) Rapid development of cortical auditory evoked potentials after early cochlear implantation. Neuroreport, 13: 13651368. Shepherd, R.K., Baxi, J.H. and Hardie, N.A. (1999) Response of inferior colliculus neurons to electrical stimulation of the auditory nerve in neonatally deafened cats. J. Neurophysiol., 82(3): 1363–1380. Sininger, Y.S. and Trautwein, P. (2002) Electrical stimulation of the auditory nerve via cochlear implants in patients with auditory neuropathy. Ann. Otol. Rhinol. Laryngol., 111(5): 29–31. Skuse, D.H. (1993) Extreme deprivation in early childhood. In: Bishop, D. and Mogford, K. (Eds.), Language Development in Exceptional Circumstances. Erlbaum, Hillsdale, pp. 29–46. Snyder, R.L., Rebscher, S.J., Cao, K., Leake, P.A. and Kelly, K. (1990) Chronic intracochlear electrical stimulation in the neonatally deafened cat.1. Expansion of central representation. Hear. Res., 50: 7–34. Snyder, R.L., Rebscher, S.J., Leake, P.A., Kelly, K. and Cao, K. (1991) Chronic intracochlear electrical stimulation in the neonatally deafened cat.2. Temporal properties of neurons in the inferior colliculus. Hear. Res., 56: 246–264. Somogyvari, Z., Zalanyi, L., Ulbert, I. and Erdi, P. (2005) Model-based source localization of extracellular action potentials. J. Neurosci. Methods, 147: 126–137. Stewart, D.L. and Starr, A. (1970) Absence of visually influenced cells in auditory cortex of normal and congenitally deaf cats. Exp. Neurol., 28: 525–528. Suga, N. and Ma, X.F. (2003) Multiparametric corticofugal modulation and plasticity in the auditory system. Nat. Rev. Neurosci., 4(10): 783–794.
313 Sur, M., Pallas, S.L. and Roe, A.W. (1990) Cross-modal plasticity in cortical development: differentiation and specification of sensory neocortex. Trends Neurosci., 13: 227–233. Sur, M. and Rubenstein, J.L. (2005) Patterning and plasticity of the cerebral cortex. Science, 310: 805–810. Syka, J. (2002) Plastic changes in the central auditory system after hearing loss, restoration of function, and during learning. Physiol. Rev., 82(3): 601–636. Thedinger, B., House, W.F. and Edgerton, B.J. (1985) Cochlear implant for tinnitus. Ann. Otol. Rhinol. Laryngol., 94: 10–13. Trachtenberg, J.T., Chen, B.E., Knott, G.W., Feng, G., Sanes, J.R., Welker, E. and Svoboda, K. (2002) Long-term in vivo imaging of experience-dependent synaptic plasticity in adult cortex. Nature, 420: 788–794. Trautwein, P.G., Sininger, Y.S. and Nelson, R. (2000) Cochlear implantation of auditory neuropathy. J. Am. Acad. Audiol., 11: 309–315. Tsumoto, T. and Suda, K. (1982) Postnatal development of the corticofugal projection from striate cortex to lateral geniculate nucleus in kittens. Brain Res., 256: 323–332. Turrigiano, G.G. and Nelson, S.B. (2004) Homeostatic plasticity in the developing nervous system. Nat. Rev. Neurosci., 5: 97–107. Tyler, R.S. (1995) Tinnitus in the profoundly hearing-impaired and the effects of cochlear implants. Ann. Otol. Rhinol. Laryngol., 104: 25–30. Tyler, R.S. and Lowder, M.W. (1992) Audiological management and performance of adult cochlear-implant patients. Ear Nose Throat J., 71: 117–118. von Ilberg, C., Kiefer, J., Tillein, J., Pfennigdorff, T., Hartmann, R., Sturzebecher, E. and Klinke, R. (1999) Electric-acoustic stimulation of the auditory system. ORL, 61: 334–340. von Melchner, L., Pallas, S.L. and Sur, M. (2000) Visual behaviour mediated by retinal projections directed to the auditory pathway. Nature, 404(6780): 871–876. Walsh, E.J. and McGee, J. (1987) Postnatal development of auditory nerve and cochlear nucleus neuronal responses in kittens. Hear. Res., 28: 97–116.
Walsh, E.J., McGee, J. and Javel, E. (1986) Development of auditory-evoked potentials in the cat. I. Onset of response and development of sensitivity. J. Acoust. Soc. Am., 79: 712–724. Waltzman, S.B. and Cohen, N.L. (1998) Cochlear implantation in children younger than 2 years old. Am. J. Otol., 19: 158–162. West, C.D. and Harrison, J.M. (1973) Transneuronal cell atrophy in the congenitally deaf white cat. J. Comp. Neurol., 151: 377–398. Winfield, D.A. (1981) The postnatal development of synapses in the visual cortex of the cat and the effects of eyelid closure. Brain Res., 206: 166–171. Winfield, D.A. (1983) The postnatal development of synapses in the different laminae of the visual cortex in the normal kitten and in kittens with eyelid suture. Brain Res., 285: 155–169. Zafra, F., Lindholm, D., Castren, E., Hartikka, J. and Thoenen, H. (1992) Regulation of brain-derived neurotrophic factor and nerve growth factor messenger RNA in primary cultures of hippocampal neurons and astrocytes. J. Neurosci., 12: 4793–4799. Zhang, L.I., Bao, S. and Merzenich, M.M. (2001) Persistent and specific influences of early acoustic environments on primary auditory cortex. Nat. Neurosci., 4: 1123–1130. Zhang, L.I., Bao, S. and Merzenich, M.M. (2002) Disruption of primary auditory cortex by synchronous auditory inputs during a critical period. Proc. Natl. Acad. Sci. USA, 99: 2309–2314. Zhang, L.I. and Poo, M.M. (2001) Electrical activity and development of neural circuits. Nat. Neurosci., 4: 1207–1214. Zhang, L.I., Tao, H.W., Holt, C.E., Harris, W.A. and Poo, M. (1998) A critical window for cooperation and competition among developing retinotectal synapses. Nature, 395: 37–44.
Plate 18.6. Representative example of activation maps computed from maximal amplitudes of local field potentials (Pa amplitudes) in 10 hearing and 10 deaf animals at different ages. Top: hearing controls show largest activations between 1 and 2 months post natal. Afterwards, maximal amplitudes and activated areas decrease. Bottom: in congenitally deaf cats, maximal activated areas were observed at 3 months post natal (data from Kral et al., 2005).
Plate 18.12. CSD signals normalize after chronic electrical stimulation through cochlear implants. Representative comparison between age-matched naı¨ ve congenitally deaf cat (A) and chronically stimulated deaf cat after 5 months of electrical stimulation (B). More activity was found in deep cortical layers of chronically stimulated cats. (C) Quantitative analysis of the above data. Mean sink temporal integrals increase with increasing duration of chronic electrical stimulation (mean data from ROI, 1 months stimulation: two animals; 2–5 months of stimulation: one animal; all animals shown here were implanted at 3 months after birth; WilcoxonMannWhitney test at a ¼ 5%). (D) Shortest sink latencies decreased after chronic electrical stimulation and became more comparable to hearing controls.
Plate 18.13. The capacity for plastic reorganization with cochlear implant stimulation decreases with increasing age. (A) Massive (but slow) expansions of the cortical activated areas, as demonstrated for early implanted animals in Fig. 10, become smaller with increasing implantation age (Kral et al., 2001, 2002). (B) The sensitive period can also be demonstrated in latencies of Pa waves, that become significantly smaller after 5 months of stimulation at early implantation (age 3.5 months post natal), but this decrease was smaller after implantation age 5 months post natal and was no longer discernible with implantation in adult age (6 months post natal).
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 19
Disrupting the brain to guide plasticity and improve behavior Alvaro Pascual-Leone Department of Neurology, Center for Non-Invasive Brain Stimulation, Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, MA 02215, USA
Abstract: Neurones may be highly stable and nonplastic cellular structures, but they are engaged in dynamically changing, intrinsically plastic neural networks that provide a most energy efficient, spatially compact, and precise means to process input signals and generate adaptable responses to a changing environment. Neural plasticity is evolution’s invention to enable the nervous system to escape the restrictions of its own genome (and its highly specialized cellular specification) and thus adapt to environmental pressures, physiologic changes, and experiences. At neural system level two steps of plasticity can be identified: unmasking existing connections that may be followed by establishment of new ones possibly even with integration of new neural structures and neurons. In any case, plastic changes may not necessarily represent a behavioral gain for a given subject, as they represent the mechanism for development and learning, as much as a cause of pathology and disease. The challenge is to learn enough about the mechanisms of plasticity to be able to guide them, suppressing changes that may lead to undesirable behaviors while accelerating or enhancing those that result in a behavioral benefit for the subject or patient. Neurostimulation, including noninvasive brain stimulation techniques, provide an opportunity to modulate brain plasticity in a controlled and specific manner. Such interventions to guide behavior or treat pathological symptomatology might be more inmediate in their behavioral repercusion and thus more effective than approaches intent on addressing underlying genetic predispositions. Keywords: neural plasticity; genetics; stroke; aphasia; brain injury; brain–behavior relations any sensory or cognitive theory has to build into its framework the fact that the nervous system, and particularly the brain, undergoes continuous changes in response to modifications in its input afferents and output targets. The nervous system might then be viewed as a continuously changing structure of which plasticity is an integral property and the obligatory consequence of each sensory input, motor act, association, reward signal, action plan, or awareness. In this framework, notions such as psychological processes as distinct from organic-based functions or dysfunctions cease to be informative. Behavior will lead to changes in brain circuitry, just as changes in brain circuitry
An intrinsically plastic nervous system I shall argue that plasticity is not an occasional state of the nervous system. Instead, I conceive of plasticity as an intrinsic property of the nervous system retained throughout the lifespan, as the normal ongoing state of the nervous system. Therefore, it is not possible to understand normal psychological function or the manifestations or consequences of disease without invoking the concept of brain plasticity. A full, coherent account of Tel.: +1-617-667-0203; Fax: +1-617-975-5322; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57019-0
315
316
will lead to behavioral modifications. Ultimately, depending on the circumstances, neural plasticity can confer no perceptible change in the behavioral output of the brain, can lead to changes demonstrated only under special testing conditions, or can cause behavioral changes that constitute symptoms of disease. Plasticity might be studied at multiple different levels, from neural systems to molecules. I shall focus at the level of neural systems, and argue that the brain is organized in dynamically shifting neuronal networks as they provide a most energy efficient, spatially compact, and precise means to process input signals and generate responses (for review see Pascual-Leone et al., 2005). I conceptualize nodes in such networks as operators that contribute a given computation independent of the input (for discussion of the notion of a ‘‘metamodal brain’’ see Pascual-Leone and Hamilton, 2001). Inputs shift depending on the integration of a region in a distributed neural network, and the layered and reticular structure of the cortex with rich reafferent loops provides the substrate for rapid modulation of the engaged network nodes. In this conceptualization, plasticity represents a most efficient way to utilize the brain’s limited resources. It might in fact be argued that the high degree of plasticity at the neural network level is a most beneficial adaptation to the resistance to plastic change at the cellular level. Individual neurons are highly complex, and exquisitely optimized cellular elements, and their capacity of change and plastic modification is necessarily very limited. Integration of highly stable and nonplastic cellular elements into dynamically changing, intrinsically plastic neural networks assures functional stability while providing a subtrate for rapid adaptation to shifting demands. We may thus consider representation of function in the brain as best conceptualized by the notion of distributed neural networks, a series of assemblies of neurons that might be widely dispersed anatomically but are structurally interconnected and can be functionally integrated to serve a specific behavioral role. For example, spatial attention appears to be supported by the parietal lobes connected by callosal fibers and via the inferior colliculus, the prefrontal cortex (particularly
on the right) and cingulate gyrus, along with connections via the superior occipito-frontal fasciculus and the cingulum. Another common example is language, subserved by Broca’s and Wernicke’s areas in the dominant hemisphere and connections along the arcuate fasciculus and the extreme capsule. However, it is important to recognize that depending on behavioral demands, neuronal assemblies can be integrated into different functional networks by shifts in weighting of connections (functional and effective connectivity). Indeed, timing of interactions between elements of a network, beyond integrity of structural connections, might be a critical binding principle for the functional establishment of given network action and behavioral output. Such notions of dedicated, but multifocal, networks, which can dynamically shift depending of demands for a given behavioral output, provide a current resolution to the long-standing dispute between localizionists and equipotential theorists. Function comes to be identified with a certain pattern of activation of specific, spatially-distributed but interconnected neuronal assemblies in a specific time window and temporal order. In such distributed networks, specific nodes may be critical for a given behavioral outcome. Knowledge of such instances is clinically useful to explain findings in patients and localize their lesions, but provides an oversimplified conceptualization of brain–behavior relations. We may be better served realizing that behavior is never the result of the lesion, but rather the consequence of how the rest of the brain is capable of sustaining function following a given lesion. The challenge is to understand enough about the involved mechanisms to be able to promote adaptive changes and suppress maladaptive ones, ultimately optimizing outcome for a given individual.
Variable brain–behavior relations In this context, it is important to recognize that the relation between brain activity and structure on the one hand, and brain function on the other, is not one to one, but in itself variable. Neurones may be highly stable and nonplastic cellular structures, but they are engaged in dynamically
317
changing, intrinsically plastic neural networks. As long as an output pathway to manifest the behavior is preserved (even if following injury alternate pathways need to be unmasked or facilitated), changes in the activity across a distributed neural network may be able to establish new patterns of brain activation and sustain function. Such plastic processes may follow two distinct steps (Pascual-Leone et al., 2005): unmasking existing connections possibly followed by establishment of new ones including integration of new neural structures and neurons. The former is very rapid and represent the range of core aspects of normal physiology (e.g., the role of cross-modal interactions in visual perception). The latter develops slowly over months and years of sustained conditions, and can give rise to unexpected capacities (e.g., supranormal auditory abilities or verbal memory in the blind). The rapid modulation of established network interactions to sustain behavior can be illustrated with the following experiments. We asked normal subjects to open and close their fist deliberately at a self-paced rhythm of approximately one movement every second while lying in an fMRI scanner. As compared with rest, during movement there was a significant activation of the motor cortex (M1) contralateral to the moving hand and of the rostral supplementary motor cortex (SMA). If motor cortex activity is modified by repetitive transcranial magnetic stimulation (rTMS), the pattern of brain activation changes as behavioral integrity is maintained (Fig. 1). Application of slow rTMS to the contralateral M1 (presumed to suppress neuronal firing; Walsh and PascualLeone, 2003) results in increased activation of the rostral SMA and of M1 ipsilateral to the moving hand. Conversely, increasing excitability in the contralateral M1 (by application of fast rTMS) leads to a decrease in activation of rostral SMA. In a very elegant study, Lee et al. (2003) combining TMS and positron emission tomography (PET) have provided supporting evidence to these notions and critically extended them by revealing the shifts in cortico–cortical and cortico–subcortical connectivity underlying the changes in cortical activation patterns. Therefore, in the face of a change in motor cortex activity (in these cases
Fig. 1. Brain activation in functional magnetic resonance imaging while subjects performed the same rhythmic hand movement (under careful kinematic control) before and after repetitive transcranial magnetic stimulation (rTMS) of the contralateral motor cortex. Following sham rTMS (top row) there is no change in the significant activation of the motor cortex (M1) contralateral to the moving hand and of the supplementary motor cortex (SMA). After M1 activity is suppressed using 1 Hz rTMS (1600 stimuli, 90% of motor threshold intensity; middle row), there is an increased activation of the rostral SMA and of M1 ipsilateral to the moving hand. Increasing excitability in the contralateral M1 using highfrequency rTMS (20 Hz, 90% of motor threshold intensity, 1600 stimuli; bottom row) results in a decrease in activation of rostral SMA. Importantly, despite the modulation of brain activity, behavior remains unchanged. The shift in activity at the targeted brain region and across network might be considered an example of rapid plasticity to sustain behavioral integrity. See Plate 19.1 in Colour Plate Section.
transient disruption induced by rTMS; Walsh and Pascual-Leone, 2003) performance of a relatively simple movement task can be maintained by rapid operational remapping of motor representations, recruitment of additional motor areas, and taskrelated changes in cortico–cortical and cortico–muscular coherence (Strens et al., 2002; Chen et al., 2003; Lee et al., 2003; Oliviero et al., 2003). Under other circumstances, modulation of activity in a focal node of a distributed neural
318
network will lead to specific modulation of cortico–cortical and cortico–subcortical interactions and give rise to changes in behavior in a controlled and specific manner. For example, it has been hypothesized that the phenomenon of ‘‘extinction to double simultaneous stimulation’’ and ‘‘neglect’’ is related to an imbalance between the hemispheres resulting from the release of reciprocal inhibitory influences (Kinsbourne, 1977). Lesion of one hemisphere results in transhemispheric release of inhibition onto the healthy hemisphere that becomes ‘‘hyperactive,’’ creating ‘‘hyper-attention’’ to the ipsilesional side. In humans, support for this hypothesis first came from the report of a patient who suffered from severe spatial neglect (the failure to explore contralesional space) following a right parietal lesion (Vuilleumier et al., 1996). Following a second lesion to the left frontal cortex, the neglect symptoms completely and abruptly disappeared, lending creedence to the notion of a dynamic balance between the two hemispheres for the allocation of attentional resources. Animal studies by Sprague and later Payne and Lomber (Sprague, 1966; Payne et al., 1996) have provided critical insights into the underlying physiology and guided more recent applications in humans. If failure to orient to the contralesional side is the result of hyperactivity of the healthy hemisphere, then transient disruption of left cortical areas in right parietal-damaged patients may also temporarily alleviate extinction symptoms. In a group of 14 right brain-damaged patients, it was shown that application of single-pulse TMS to the left prefrontal cortex significantly reduced contralateral extinction when the TMS pulse was applied 40 ms after bilateral electrical stimulation of the fingers (Oliveri et al., 1999). These results were later replicated by the same group in a visuospatial task using high-frequency repetitive TMS (Oliveri et al., 2001). The performance of five right braindamaged patients in a line bisection task was significantly improved following parietal rTMS of the unaffected hemisphere. Again in right brain-damaged patients suffering from visuospatial neglect, Brighina et al. (2003) showed that a 2-week regimen of low-frequency repetitive TMS to the healthy hemisphere could reduce visuospatial
neglect beyond the period of stimulation. One hertz rTMS was applied to the left parietal cortex in three patients with a right parieto-temporal lesion every other day for 14 days. Visuospatial performance (clock drawing and line bisection tasks) was significantly improved immediately after treatment and for at least 15 days. Given the reciprocal interhemispheric inhibition and the proposed link to attentional performance, we hypothesized that suppression of one parietal cortex would lead to contralateral neglect, but at the same time, the disinhibition of structures involved in interhemispheric competition might lead to a functional release in the opposite hemisphere, which could result in a measurable ipsilateral behavioral enhancement (Hilgetag et al., 2001). To verify this hypothesis, normal subjects had to detect small rectangular stimuli briefly presented on a computer monitor either unilaterally in the left or right periphery, or bilaterally in both. Spatial detection performance was tested before and immediately after a 10 min, 1 Hz rTMS train to: (a) right parietal cortex; (b) left parietal cortex; (c) right primary motor cortex; and (d) sham stimulation. We observed a clear extinction phenomenon for stimuli presented contralaterally to the stimulated hemisphere (right or left parietal cortex). This deficit was accompanied by increased detection for unilateral stimuli presented on the side of the stimulated hemisphere compared to baseline (Fig. 2). None of the control stimulation sites had any effect on the detection performance. Detailed investigation revealed that although trends were mirror-symmetric for rTMS of left and right parietal cortex, the enhancement produced by right-hemispheric rTMS was significantly greater than that after left hemisphere (LH) and only right hemispheric stimulation produced a significant ipsilateral detection enhancement. These data suggest that in normal subjects, decreasing left parietal cortex excitability with rTMS disinhibits the contralateral cortex leading to improvements in performance, and provide further illustration to the notion of rapid plasticity leading to shifts in activity across a distributed neural network which can sustain behavior, improve it, or disrupt it, depending on the circumstances.
319
Fig. 2. Modified from Hilgetag et al. (2001) with permission. Changes in correct stimulus detection after parietal rTMS. The diagrams are based on changes in the number of correctly detected stimuli (relative to the total number of presented stimuli) averaged for both stimulus sizes and all subjects. (a) The pooled data show a significant increase in performance ipsilateral to the parietal rTMS location (increase in relative percentage points: 7.3% SEM: 2.6%), and a trend to decreased contralateral performance (reduction by 2.5%, SEM: 2.3%). In addition, detection of bilateral stimuli decreased significantly (11.7%, SEM: 2.0%). These trends are also apparent after separating data for (b) left parietal TMS and (c) right parietal rTMS. Significant trends (as determined by z-tests), are marked by asterisks.
Plasticity and genetics I believe that neural plasticity maybe thought of a ‘‘evolution’s invention’’ to enable the nervous system to escape the restrictions of its own genome (and its highly specialized cellular specification), and thus adapt to environmental pressures, physiologic changes, and experiences. Therefore, plasticity is not an occasional state of the nervous system; instead, it is the normal ongoing state of the nervous system throughout life. Very rapid, ongoing changes in neural systems in response to shifts in afferent input or efferent demand (e.g., by dynamic shifts in the strength of preexisting connections across distributed neural networks, changes in task-related cortico–cortical and cortico–subcortical coherence, or modifications of the mapping between behavior and neural activity) may be followed by the establishment of new connections through dendritic growth and arborization resulting in structural changes and
establishment of new pathways (Pascual-Leone et al., 2005). Thus, the scope of possible plastic changes is initially determined by existing connections, which are the result of genetically controlled neural development and thus different across individuals. Genetically controlled aspects of brain development define neuronal elements and initial patterns of connectivity. Subsequently, afferent input, efferent demand, contextual and environmental changes, and brain activity in general shape neural network stuctures and brain–behavior relations by mechanims of plasticity. Given the initial, genetically determined, individually different brain substrate, the same events will result in diverse consequences as plastic brain mechanisms act upon individually distinct neural substrates. Furthermore, genetic factor may regulate and define the range of plastic changes, their magnitude, stability, and chronometry. In a sense, the conceptualization of genetically determined starting point and subsequent
320
environmentally and behaviorally driven plastic changes is reminiscent of the debate over the relative roles of ‘‘nature versus nurture,’’ which remains unresolved in many fields of study, from childhood education to animal behavior or neurodegenerative disorders. Points of view are frequently polarized, rather then exploring the ways in which both sides play critical and complementary roles. Proponents of ‘‘nature’’ argue that human attributes are uniquely and primarily conditioned by genetics, opposing the view of those who believe that environmental influences and experience determine individual differences. Arguments similar to those brought to bear regarding acquisition of skills, language, or cognitive styles also apply to human pathology. Often, in the discussion of disease the terms used are ‘‘genetic’’ versus ‘‘environmental,’’ but the implication is the same as in the ‘‘nature’’ versus ‘‘nurture’’ debate. Disorders considered to be primarily genetic are ones in which the presence or absence of genetic mutations is the primary determinant of disease, independent of environmental circumstances. A disease considered to be primarily environmental is one in which people of virtually any genetic background can develop the disease provided that they are exposed to the necessary environmental factor or factors. However, for many disorders, one’s risk is strongly influenced by both genetic and environmental factors. For example, a susceptibility gene may strongly influence one’s risk of developing a disease only in response to a specific environmental exposure. If the environmental exposure occurs infrequently, the gene will be of low penetrance, and it may appear that the environmental exposure is the primary determinant of the disease, even though the gene is required for developing the disease. Therefore, even when environmental agents are suspected to be a major cause of a particular disease, this does not exclude the possibility that genetic factors also play a major role, particularly genetic mutations with low penetrance. Similarly, a critical mix of nature and nurture is likely to determine personal characteristics. Genetic factors may be thought of as laying the foundation on which environmental agents exert their influence. If so, then although certain environmental factors alone (regardless of genetic
factors) and certain genetic factors alone (regardless of environmental influences) may explain some behaviors and disease states, most of the time the interaction of both genetic and environment factors will be required. Deafness due to aminoglycoside toxicity is a particularly clear illustration of this interdependence and provides a simple, yet suitable illustration of the circumstances that may apply to complex behaviors and conditions. Prolonged exposure to aminoglycosides is toxic to cochlear cells. However, a mutation at nucleotide position 1555 in the mitochondrial 12S ribosomal RNA gene is associated with an extremely high susceptibility to aminoglycoside-induced deafness, even at exposure levels that would not be toxic to most individuals (Hutchin et al., 1993; Prezant et al., 1993; Inoue et al., 1996). In most cases, the mutation does not cause deafness without aminoglycoside exposure. The ‘‘penetrance’’ of this mutation is therefore dependent on the frequency with which individuals are exposed to an aminoglycoside antibiotic. If exposure to aminoglycoside antibiotics were rare, then the mutation would have low penetrance, and the critical role of this mutation in determining susceptibility to nonsyndromic deafness might be difficult to recognize. Similarly in complex behaviors, personality traits, and a vast majority of neuropsychiatric disorders, one or many genes may convey a susceptibility for certain environmental factors, experiences, afferent inputs, or efferent demands, to results in plastic changes that ultimately define personal characteristics or constitute manifestations of disease. Focal hand dystonia in musicians (Chamagne, 2003) may be a good example of pathological consequences of plasticity that can be promoted by suitable genetic predispositions, such as DYT-1 or other. In focal hand dystonia in musicians, ‘‘faulty’’ practice or excessive demand in the presence of certain predisposing factors may result in unwanted cortical rearrangement and lead to disease (Pascual-Leone, 2001). Autism may be another example of similar, plasticity-mediated pathology: genetic factors may lead to a predisposition such that developmentally-mediated plasticity (possibly in itself controlled by abnormal regulators) results in pathological complex
321
behaviors affecting social interactions, language acquisition, etc. Specific neuronal populations, for example mirror neurons, might be affected primarily or secondarily as manifestation of maladaptive changes. A consequence of such formulation is the notion that human bevarior and the manifestations of human disease, are ultimately heavily defined by brain plasticity, and that an intervention to guide behavior or treat pathological symptomatology might be more inmediate in its behavioral repercusion and thus more effective if aimed at modulating plasticity, than if intent on addressing underlying genetic predispositions. The challenge we face is to learn enough about the mechanisms of plasticity and the mapping relations between brain activity and behavior to be able to guide it, suppressing changes that may lead to undesirable behaviors while accelerating or enhancing those that result in a behavioral benefit for the subject or patient.
Plasticity in the setting of brain injury: the risk of change and the opportunity for intervention I believe that it is a misconception to consider plasticity is a capacity of the brain that can be activated in response to an insult to promote functional recovery or compensate for lost function. Rather, plasticity is always activated and it is unlikely primarily aimed at coping with injury. Presumably the nervous system is intrinsically plastic, as discussed above, to adapt to and cope with the challenges of a continuouly shifting environement. The environment is capable of change in ways unpredictable at the time of conception of the individual and thus, mechanisms to cope with such changes cannot be anticipated and provided for in genetic attributes. However, as an intrinsically plastic system, plasticity does need to be considered in the setting of coping with brain injury. Importantly, the plastic nature of the brain provides, following injury, a risk of maladaptive change and perpetuation of deficits, but also an opportunity for intervention and overcoming of symptoms. Following brain injury, behavior (regardless of whether normal or manifesting injury-related
deficits) remains the consequence of the functioning of the entire brain, and thus the consequence of a plastic nervous system. Ultimately, symptoms are not the manifestation of the injured brain region, but rather the expression of plastic changes in the rest of the brain. As long as efferent, output pathways exist to manifest the behavior, cortico–cortical and cortico–subcortico–cortical interactions will shift weights across the involved functional network, aiming to establish a suitable brain activation map for a desired behavioral result. Conceptually it might be worth thinking of processes occurring after brain injury and leading to restoration of function as fitting different mechanisms that may proceed partly in parallel but which have variable time frames. Initial plastic changes aim to minimize damage. Rapid functional improvement is likely to occur as dysfunctional, but not damaged, neuronal elements recover from the postinjury shock and penumbra processes resolve. Partially damaged neural elements may be able to be repaired relatively quickly after the insult as well, thus contributing to early functional improvement. Subsequent processes, once the final damage has been established, involve relearning (rather than recovery) and may follow a two steps process: initial unmasking and strengthening of existing neural pathways, and eventually the establishment of new structural changes. At all these stages of plastic adaptation, neurostimulation can guide the neural processes and promote adaptive, desirabel outcomes for a given individual. Influencing recovery of hand function The concepts discussed above can be illustrated by examining the role of the ipsilateral motor cortex in the recovery of hand motor function following stroke. After stroke, there is an increase in the excitability of the unaffected hemisphere, presumably owing to reduced transcallosal inhibition from the damaged hemisphere and increased use of the intact hemisphere. Several studies have demonstrated the increased cortical excitability in the unaffected hemisphere after a stroke. For example, in patients with acute cortical stroke, intracortical inhibition is decreased and intracortical
322
facilitation (ICF) increased in the unaffected hemisphere (Liepert et al., 2000). Furthermore, the interhemispheric inhibitory drive from the unaffected to the affected motor cortex in the process of voluntary movement generation is abnormal (Murase et al., 2004). Interestingly, the duration of stroke is inversely correlated to the imbalance of the excitability between the hemispheres. A disease duration of more than 4 months after stroke onset results in a tendency to normalization of the ICF of the unaffected hemisphere (Shimizu et al., 2002). Acutely after a stroke, increased inhibitory input from the undamaged to the damaged hemisphere makes conceptual sense if one considers it a manifestation of a neural attempt to control perilesional activity, reduce oxygen and glucose demands in the penumbra of the stroke, and thus limit the extension of the lesion. However, after an acute phase, and once the injury is stable, input to the perilesional area would seem to be best as excitatory in nature to maximize the capability of the preserved neurons in the injured tissue to drive behavioral output. If so, following the acute phase, we might expect a shift of interhemispheric (and many intrahemispheric) interactions, from inhibitory to excitatory. Should such a shift fail to take place, the resulting functional outcome may be undesirable, with limited behavioral restoration, in part owing to persistent inhibitory inputs from the intact to the damaged hemisphere. In fact, some neuroimaging studies demonstrate that long-term, persistent activation of the ipsilateral cortex during motor tasks is associated with poor motor outcomes, whereas a good motor recovery is associated with a decrease in activity in the unaffected and an increase in the affected primary sensorimotor cortex activity (Carey et al., 2002; Rossini and Dal Forno, 2004). Furthermore, the pattern of activation in well-recovered patients is similar to healthy subjects (Ward et al., 2003). More longitudinal studies of patients following a stroke and correlation of interhemispheric interactions with functional measures are needed to explore these issues further. If correct, neuromodulatory approaches targeting the intact hemisphere may be useful to limit injury and promote recovery after a stroke.
For instance, suppression of the ipsilateral motor cortex through slow rTMS (Pascual-Leone et al., 1998; Maeda et al., 2000) may enhance motor performance in patients stable following the acute phase of a stroke. In patients 1–2 months after a stroke, Mansur et al. (2005) applied 0.5 Hz rTMS for 10 min to the unaffected hemisphere to suppress cortical activity and thus release the damaged hemisphere from potentially excessive transcallosal inhibition. The results of this study support the notion that the overactivity of the unaffected hemisphere (ipsilateral hemisphere) may hinder hand function recovery, and neuromodulation can be an interventional tool to accelerate this recovery. The findings are consistent with results in normal subjects, where ipsilateral motor cortex activation on functional MRI during unilateral hand movements is indeed related primarily to interhemispheric interactions (Kobayashi et al., 2003), and disruption of the activity of one hemisphere reduces transcallosal inhibition to the contralateral hemisphere and can indeed improve ipsilateral motor function (Fig. 3; Kobayashi et al., 2004). However, Werhahn et al. (2003) conducted a similar study to evaluate the modulation effects of 1 Hz rTMS of the unaffected hemisphere on the paretic hand and found different results. In that study, 1 Hz rTMS of the unaffected hemisphere did not affect the finger tapping in the paretic hand in a small sample of five patients more than 1 year after a stroke. The time since the brain insult is likely to be a critical variable to consider. Studies with larger samples of patients are needed to investigate this question further. Behavioral motor therapy may also shift cortical excitability balance between hemispheres and thus influence outcome. For example, the beneficial effects of constraint-induced therapy on motor function (Grotta et al., 2004; Mark and Taub, 2004) are achieved through immobilization of the unaffected arm, which results in a reduction of the excitability of the contralateral (undamaged) motor cortex owing to the decreased efferent demand and afferent input (Liepert et al., 2001). The reduced activity of the undamaged motor cortex may decrease transcallosal inhibition of the damaged motor cortex and thus promote recovery,
323
entertained. Khedr et al. (2005) have reported extremely encouraging results along these lines, and results of a pilot study in primates support the feasibility of using a therapy approach, combining peri-infarct electrical stimulation with rehabilitative training to alleviate chronic motor deficits and promote recovery from cortical ischemic injury (Plautz et al., 2003). Very early experiments with invasive cortical stimulation in humans reveal similarly encouraging findings (Brown et al., 2003). In this setting, functional neuroimaging might be useful, among other things, to identify the perilesional areas to be targeted (Baron et al., 2004), and EEG or fMRI may allow investigators to define precisely and optimize the physiologic effects of TMS (Bestmann et al., 2004). Similar principles of neuromodulation can be applied to the recovery of nonmotor strokes and other focal brain lesions as illustrated by the studies on the effects of cortical stimulation on neglect discussed above (Hilgetag et al., 2001; Oliveri et al., 2001; Brighina et al., 2003) or the following experience with aphasia (Knecht et al., 2002; Kurland et al., 2004; Naeser et al., 2005a, b). Neuromodulation in nonfluent aphasia
Fig. 3. Modified from Kobayashi et al. (2003) with permission. (a) Ratio of execution times following rTMS at three different sites (ipsilateral M1, ipsilateral premotor cortex and Cz). Reaction times were significantly shorter after ipsilateral rTMS over primary motor cortex. (b) Changes in MEP sizes of the left first dorsal interosseus muscle with various interstimulus intervals.
ultimately by mechanisms similar to those recruited by suppressing cortical excitability through slow rTMS. Of course, the alternative neuromodulatory approach, directly aimed to enhance excitability of the damaged hemisphere perilesionally, can also be
Conclusive evidence for specific patterns of neuronal activation that predict recovery in aphasic patients remains elusive, but recent studies reveal that patients with better recovery often have higher activation in the Left Hemisphere (LH) (Miura et al., 1999; Warburton et al., 1999) and, in particular, the left superior temporal gyrus (Heiss et al., 1999). Some studies report that new LH activation is associated with language improvement following speech therapy (Small et al., 1998; Leger et al., 2002; Cornelissen et al., 2003). On the other hand, functional neuroimaging studies of patient with residual nonfluent aphasia have observed unusually high activation levels in right perisylvian language homologues (Belin et al., 1996; Rosen et al., 2000), which were not correlated with improved language performance (Rosen et al., 2000; Perani et al., 2003; Naeser et al., 2004). Indeed, this right hemisphere (RH) activation might be hindering recovery. It is with
324
Fig. 4. Representative example of T1-weighted structural MRI scan for the one of the six chronic aphasia patients studied. The figure shows axial slices and a left lateral view for the patient’s 3-D reconstructed MRI revealing the extend of the lesion. The lower right quadrant illustrates the anatomical locations of the different brain regions targeted with rTMS in patients with nonfluent aphasia including Motor Cortex for the mouth in green, BA 44 in pink, BA45 in blue and BA 22 in orange. In all patients studied the lesions are cortical in all patients affecting the distribution of the middle cerebral artery and encompassing subcortical lesions of the medical subcallosal fasciculus and middle 1/3 of the peri-ventricular white matter. See Plate 19.4 in Colour Plate Section.
this hypothesis of RH maladaptive plasticity that we conducted studies to decrease RH activity in specific RH language homologues in order to encourage functional rehabilitation in nonfluent aphasia. The first study aimed at investigating the effect of temporary suppression of four RH cortical regions on naming pictures using slow, 1 Hz rTMS. We studied six chronic, nonfluent aphasia patients with LH lesion (four men, two women; age 51–67 years; 5–30 years; post stroke onset). They had mild-to-severe nonfluent speech with a maximum phrase length of 1–4 words. All had cortical LH lesions that included a portion of left Broca’s area (pars triangularis and pars opercularis), and/or white matter deep to it (Fig. 4). In different rTMS sessions, we applied slow, 1Hz rTMS to transiently suppress activity in right pars triangularis; right pars opercularis; right motor cortex-mouth (M1, orbicularis oris); and right
posterior, superior temporal gyrus (R Brodman’s area (BA) 22) or anterior supramarginal gyrus (BA 40). The rTMS was applied for 10 min (600 pulses) at 90% motor threshold using a 7 cm diameter, figure 8-shaped coil (MagStim, NY). A frameless stereotaxic system (Brainsight, Rogue Industries, Montreal) was used to guide the position of the coil on the patient’s scalp and to document its accurate targeting of the specified brain regions throughout rTMS. Prior to the first rTMS session, the patient’s Baseline Naming ability was determined by asking them to name pictures from a standardized list (Snodgrass and Vanderwart pictures). Five sets of pictures were used (20 items per set). Each set had the same level of difficulty/complexity. The patient’s ability to name pictures was immediately tested after a 10-min rTMS treatment, with a different set of 20 pictures. The internal order and the set presentations were randomized.
325
All six patients correctly named the highest number of pictures following application of rTMS to right pars triangularis, right BA 45 (Fig. 5). There was a significant effect of site of
Fig. 5. Mean percent change from Baseline Naming score (and RT) for Snodgrass & Vanderwart lists pre-rTMS, and post-rTMS naming scores (and RT) after each of four right ROIs was stimulated at 1 Hz rTMS for 10 minutes. Note increase in number of pictures named correctly and decrease in RT for right BA 45; however, the reverse for right BA 44. Modified from Naeser et al., 2005b.
rTMS application on both number of pictures named correctly (repeated measures ANOVA F-value 8.3; p ¼ 0.001) and RT (repeated measures ANOVA F-value 3.6; po0.05). Patients named significantly more items after rTMS to right BA 45 than to right BA 44 (po0.0005), right M1 (po0.005), and right BA 22 (po0.001). On average this resulted in three more items named correctly after rTMS to right BA 45 than at Baseline or after any of the other areas. Following rTMS to right BA 44 subjects tended to name fewer items and were significantly slowed in their RTs (vs. right BA 22 po0.05, vs. right BA 45 and vs. right M1, Po0.01, Fig. 5). Further confirmation of this pattern of effect of TMS to RH language homologues a single patient study was conducted examining the effect of placement of the TMS coil on the scalp with locations differing from each other by only 1 cm.Picture naming for a 20-item S&V list was tested before and after 10 min of 1-Hz rTMS application to each area. The targeted area, which consistently produced the highest number of pictures named correctly by this patient post-rTMS, was right BA 45 (see Fig. 6). Once the target area of right BA 45 was determined to be most effective in causing a beneficial behavioral outcome in the nonfluent aphasic patients, we set out to determine if longer trains
Fig. 6. Mild-moderate nonfluent aphasia patient, P2. Performance level on naming after 10 minutes, 1 Hz rTMS applied to neighboring brain regions, which were only 1 or 2 cm apart. Note that stimulation of two areas separated by less than 1 cm (marked in yellow) within the right pars triangularis area, consistently led to the highest number of pictures named correctly (20 pictures per test). See Plate 19.6 in Colour Plate Section.
326
Fig. 7. Naming scores for four chronic aphasia patients, pre-rTMS, and at 2 weeks, 2 months and 8 months following ten, 1 Hz rTMS treatments to right BA 45 (R pars triangularis). Note the continued improvement clearly seen in 3 of the 4 patients longtime after the TMS despite the longstanding deficit and the lack of any additional treatment. Modified from Naeser et al., 2005b.
of rTMS specifically localized to right BA 45 could produce long-term benefits for the patients (Naeser et al., 2004). We report on four righthanded patients with moderately severe residual nonfluent aphasia at 5–11 years post left middle cerebral artery (MCA) stroke. Patients received ten, 20-min, 1-Hz rTMS treatments, 5 days a week for 2 weeks (1200 pulses per session) at 90% of motor threshold using the same TMS coil and Brainsight stereotaxic program described above. The primary outcome measures were standardized language tests including the first 20 items on the Boston Naming Test (BNT) and parts of the Boston Diagnostic Aphasia Exam (BDAE).
The patients were examined with these language tests by a blinded speech pathologist within 1–2 weeks prior to the 1st rTMS treatment and again at 2 weeks, 2 months, and 8 months postthe 10th treatment. The patients did not receive any individualized speech therapy during the study. A series of univariate one-way repeated measures Analyses of Variances were conducted, with time of testing (pre-rTMS; post-2 weeks; 2 months; 8 months) as the repeated measure. At 2 months post-rTMS, there was significant improvement on the BNT, and on the Animals and Tools/Implements naming subtests on the BDAE (Fig. 7). At 8 months, all three naming scores continued to
327
improve relative to pre-rTMS testing. Improvement was observed in number of words per longest phrase. For example, one of the patients increased from a 1- to a 3-word phrase length, when describing the cookie theft picture at 2 months postrTMS. This increase was sustained at 8 months post-rTMS. This is the first study to report lasting, improved naming at 2 and 8 months following application of rTMS treatments in chronic aphasia patients. Results suggest that the 2-month period following a series of rTMS treatments might be an optimum time to provide speech therapy, in order to promote greater potential for language improvement. Results should be considered preliminary, as this was an open-protocol study. However, all four patients were in the chronic, stable phase of aphasia, many years beyond the spontaneous recovery of 3–6 months poststroke (Demeurisse and Capon, 1987). An ongoing NIH-supported, parallel group, randomized, double-blind, and sham-stimulation controlled trial, is supporting the same conclusions. Furthermore, functional magnetic resonance imaging studies in these patients before and after the rTMS will provide insights onto the substrate of the behavioral effects, both acutely after the rTMS and those progressively developing over time. The continued behavioral benefit for months following the rTMS suggests that disruption of a specific node in the involved neural network, the right pars triangularis, may have led to a shift in weighted activity across the involved neural network and possibly promoted a shift in brain–behavioral mapping. Such a shift may be conceptualized as a change in strategy, which in this case gave rise to a more adaptive strategy that allowed for continued benefit and functional improvement.
Conclusions I have argued that the brain is highly plastic and that plasticity represents evolution’s invention to enable the nervous system to escape the restrictions of its own genome (and its highly specialized cellular specification) and adapt to rapidly shifting
and often unpredictable environmental and experiential changes. Plastic changes may not necessarily represent a behavioral gain for a given subject, and they represent a main mechanism for development and learning, as much as a cause of pathology and disease. Recovery of function after a focal brain injury, for example, a stroke, is essentially learning with a partially disrupted neural network and illustrates the dangers and opportunities of such a plastic brain. A main neural mechanism underlying relearning of skills and preservation of behavior involves shifts of distributed contributions across a specific neural network (fundamentally, the network engaged in learning the same skills in the healthy brain). Intra- and particularly interhemispheric interactions may shift from being initially inhibitory (to minimize damage) to later excitatory (to promote functional recovery). Changes in the time course of such connectivity shifts may result in the establishment of dead-end strategies and limit functional recovery. Ultimately, activation of brain areas that are not normally recruited in normal subjects may represent a nonadaptative strategy resulting in a poor prognosis. Neurostimulation provides a unique tool to disrupt specific brain regions and guide brain plasticity, promoting shifts in brain– behavior mapping and the establishement of new strategies that might be more adaptive for a given individual.
Acknowledgments This work was supported in part by grants from the National Institutes of Health (K24 RR018875, R01-MH069898, RO1-DC05672, RO1-NS 47754, RO1-NS 20068, RO1-EY12091, R01-EB 005047) and the Harvard-Thorndike General Clinical Research Center (NCRR MO1 RR01032). Critical aspects of the experimental work described herein was conducted in collaboration with Margaret Naeser, Paula Martin, Felipe Fregni, Hugo Threoret, Claus Hilgetag, Antoni Valero-Cabre, and Masahito Kobayashi. I thank Mark Thivierge for his invaluable administrative support. Parts of this chapter are contained in Pascual-Leone et al. Annual Review of Neuroscience, 2005.
328
References Baron, J.C., Cohen, L.G., Cramer, S.C., Dobkin, B.H., Johansen-Berg, H., Loubinoux, I., Marshall, R.S. and Ward, N.S. (2004) Neuroimaging in stroke recovery: a position paper from the First International Workshop on Neuroimaging and Stroke Recovery. Cerebrovasc. Dis., 18: 260–267. Belin, P., Van Eeckhout, P., Zilbovicious, M., Remy, P., Francois, C., Guillaume, S., Chain, F., Rancurrel, G. and Samson, Y. (1996) Recovery from nonfluent aphasia after melodic intonation therapy: a PET study. Neurology, 47: 1504–1511. Bestmann, S., Baudewig, J., Siebner, H.R., Rothwell, J.C. and Frahm, J. (2004) Functional MRI of the immediate impact of transcranial magnetic stimulation on cortical and subcortical motor circuits. Eur. J. Neurosci., 19: 1950–1962. Brighina, F., Bisiach, E., Oliveri, M., Piazza, A., La Bua, V., Daniele, O. and Fierro, B. (2003) 1 Hz repetitive transcranial magnetic stimulation of the unaffected hemisphere ameliorates contralesional visuospatial neglect in humans. Neurosci. Lett., 336: 131–133. Brown, J.A., Lutsep, H., Cramer, S.C. and Weinand, M. (2003) Motor cortex stimulation for enhancement of recovery after stroke: case report. Neurol. Res., 25: 815–818. Carey, J.R., Kimberley, T.J., Lewis, S.M., Auerbach, E.J. and Dorsey, L. (2002) Analysis of fMRI and finger tracking training in subjects with chronic stroke. Brain, 125: 773–788. Chamagne, P. (2003) Functional dystonia in musicians: rehabilitation. Hand Clin., 19: 309–316. Chen, W.H., Mima, T., Siebner, H.R., Oga, T., Hara, H., Satow, T., Begum, T., Nagamine, T. and Shibasaki, H. (2003) Low-frequency rTMS over lateral premotor cortex induces lasting changes in regional activation and functional coupling of cortical motor areas. Clin. Neurophysiol., 114: 1628–1637. Cornelissen, K., Laine, M., Tarkiainen, A., Jarvensivu, T., Marin, Salmelin R. (2003) Adult brain plasticity elicited by anomia treatment. J. Cogn. Neurosci.,15(3): 444–461. Demeurisse, G. and Capon, A. (1987) Language recovery in aphasic stroke patients: clinical, CT, and CBF studies. Aphasiology, 1(4): 301–315. Grotta, J.C., Noser, E.A., Ro, T., Boake, C., Levin, H., Aronowski, J. and Schallert, T. (2004) Constraint-induced movement therapy. Stroke, 35: 2699–2701. Heiss, W.D., Kessler, J. and Theil, A. (1999) Differential capacity of left and right hemispheric areas for compensation of poststroke aphasia. Ann.Neurol., 45: 430–438. Hilgetag, C.C., Theoret, H. and Pascual-Leone, A. (2001) Enhanced visual spatial attention ipsilateral to rTMS-induced ‘virtual lesions’ of human parietal cortex. Nat. Neurosci., 4: 953–957. Hutchin, T., Haworth, I., Higashi, K., Fischel-Ghodsian, N., Stoneking, M., Saha, N., Arnos, C. and Cortopassi, G. (1993) A molecular basis for human hypersensitivity to aminoglycoside antibiotics. Nucleic. Acids Res., 21: 4174–4179.
Inoue, K., Takai, D., Soejima, A., Isobe, K., Yamasoba, T., Oka, Y., Goto, Y. and Hayashi, J. (1996) Mutant mtDNA at 1555 A to G in 12S rRNA gene and hypersusceptibility of mitochondrial translation to streptomycin can be co-transferred to rho 0 HeLa cells. Biochem. Biophys. Res. Commun., 223: 496–501. Khedr, E.M., Ahmed, M.A., Fathy, N. and Rothwell, J.C. (2005) Therapeutic trial of repetitive transcranial magnetic stimulation after acute ischemic stroke. Neurology, 65: 466–468. Kinsbourne, M. (1977) Hemi-neglect and hemisphere rivalry. In: Weinstein, E.A. and Friedland, R.P. (Eds.) Hemi-inattention and hemisphere specialization. Advances in Neurology, Vol. 18. Raven Press, New York, pp. 41–49. Knecht, S., Floel, A., Drager, B., Breitenstein, C., Sommer, J., Henningsen, H., Ringelstein, E.B. and Pascual-Leone, A. (2002) Degree of language lateralization determines susceptibility to unilateral brain lesions. Nat. Neurosci., 5: 695–699. Kobayashi, M., Hutchinson, S., Schlaug, G. and PascualLeone, A. (2003) Ipsilateral motor cortex activation on functional magnetic resonance imaging during unilateral hand movements is related to interhemispheric interactions. Neuroimage, 20: 2259–2270. Kobayashi, M., Hutchinson, S., Theoret, H., Schlaug, G. and Pascual-Leone, A. (2004) Repetitive TMS of the motor cortex improves ipsilateral sequential simple finger movements. Neurology, 62: 91–98. Kurland, J., Naeser, M.A., Baker, E.H., Doron, K., Martin, P.I., Seekins, H.E., Bogdan, A., Renshaw, P. and YurgelunTodd, D. (2004) Transcranial magnetic stimulation as a complementary treatment for aphasia. Semin. Speech Lang., 25: 181–191. Lee, L., Siebner, H.R., Rowe, J.B., Rizzo, V., Rothwell, J.C., Frackowiak, R.S. and Friston, K.J. (2003) Acute remapping within the motor system induced by low-frequency repetitive transcranial magnetic stimulation. J. Neurosci., 23: 5308–5318. Leger, A., Demonet, J.-F., Ruff, S., Aithamon, B., Touyeras, B., Puel, M., Boulanouar, K. and Cardebat, D. (2002) Neural substrates of spoken language rehabilitation in an aphasic patient: an fMRI study. Neuroimage, 17: 174–183. Liepert, J., Storch, P., Fritsch, A. and Weiller, C. (2000) Motor cortex disinhibition in acute stroke. Clin. Neurophysiol., 111: 671–676. Liepert, J., Uhde, I., Graf, S., Leidner, O. and Weiller, C. (2001) Motor cortex plasticity during forced-use therapy in stroke patients: a preliminary study. J. Neurol., 248: 315–321. Maeda, F., Keenan, J.P., Tormos, J.M., Topka, H. and Pascual-Leone, A. (2000) Modulation of corticospinal excitability by repetitive transcranial magnetic stimulation. Clin. Neurophysiol., 111: 800–805. Mansur, C.G., Fregni, F., Boggio, P.S., Riberto, M., GallucciNeto, J., Santos, C.M., Wagner, T., Rigonatti, S.P., Marcolin, M.A. and Pascual-Leone, A. (2005) A sham-stimulation controlled trial of rTMS of the unaffected hemisphere
329 on hand motor function after stroke. Neurology, 64: 1802–1804. Mark, V.W. and Taub, E. (2004) Constraint-induced movement therapy for chronic stroke hemiparesis and other disabilities. Restor. Neurol. Neurosci., 22: 317–336. Miura, K., Nakamura, Y., Miura, F., Yamada, I., Takahashi, M., Yoshikawa, A. and Mizobata, T. (1999) Functional magnetic resonance imaging to word generation task in a patient with Broca’s aphasia. J. Neurol., 246: 939–942. Murase, N., Duque, J., Mazzocchio, R. and Cohen, L.G. (2004) Influence of interhemispheric interactions on motor function in chronic stroke. Ann. Neurol., 55: 400–409. Naeser, M.A., Martin, P.I., Baker, E.H., Hodge, S.M., Sczerzenie, S.E., Nicholas, M., Palumbo, C.L., Goodglass, H., Wingfield, A., Samaraweera, R., Harris, G., Baird, A., Renshaw, P. and Yurgelun-Todd, D. (2004) Overt propositional speech in chronic nonfluent aphasia studied with the dynamic susceptibility contrast fMRI method. Neuroimage, 22: 29–41. Naeser, M.S., Martin, P.I., Nicholas, M., Baker, E.H., Seekins, H., Helm-Estabrooks, N., Cayer-Meade, C., Kobayashi, M., Theoret, H., Fregni, F., Tormos, J.M., Kurland, J., Doron, K.W. and Pascual-Leone, A. (2005a) Improved naming after TMS treatments in a chronic, global aphasia patient—case report. Neurocase, 11(3): 182–193. Naeser, M.S., Martin, P.I., Nicholas, M., Baker, E.H., Seekins, H., Helm-Estabrooks, N., Cayer-Meade, C., Kobayashi, M., Theoret, H., Fregni, F., Tormos, J.M., Kurland, J., Doron, K.W. and Pascual-Leone, A. (2005b) Improved psicture naming in chronic aphasia after TMS to part of right Broca’s area, an open-protocol study. Brain Lang., 93: 95–105. Oliveri, M., Bisiach, E., Brighina, F., Piazza, A., La Bua, V., Daniele, O. and Fierro, B. (2001) rTMS of the unaffected hemisphere transiently reduces contralesional visuospatial hemineglect. Neurology, 57: 1338–1340. Oliveri, M., Rossini, P.M., Traversa, R., Cicinelli, P., Filippi, M.M., Pasqualetti, P., Tomaiuolo, F. and Caltagirone, C. (1999) Left frontal transcranial magnetic stimulation reduces contralesional extinction in patients with unilateral right brain damage. Brain, 122: 1731–1739. Oliviero, A., Strens, L.H., Di Lazzaro, V., Tonali, P.A. and Brown, P. (2003) Persistent effects of high frequency repetitive TMS on the coupling between motor areas in the human. Exp. Brain Res., 149: 107–113. Pascual-Leone, A. (2001) The brain that plays music and is changed by it. In: Zatorre, R. and Peretz, I. (Eds.), Music and the Brain. Academy of Sciences, New York. Pascual-Leone, A., Amedi, A., Fregni, F. and Merabet, L. (2005) The plastic human brain cortex. Ann. Rev. Neurosci., 28: 377–401. Pascual-Leone, A. and Hamilton, R.H. (2001) The metamodal organization of the brain. Prog. Brain Res., 134: 427–445. Pascual-Leone, A., Tormos, J.M., Keenan, J., Tarazona, F., Canete, C. and Catala, M.D. (1998) Study and modulation of human cortical excitability with transcranial magnetic stimulation. J. Clin. Neurophysiol., 15: 333–343.
Payne, B.R., Lomber, S.G., Geeraerts, S., van der Gucht, E. and Vandenbussche, E. (1996) Reversible visual hemineglect. Proc. Natl. Acad. Sci. USA, 93: 290–294. Perani, D., Cappa, S.F., Tettamanti, M., Rose, M., Scifo, P., Miozzo, A., Basso, A. and Fazio, F. (2003) An fMRI study of word retrieval in aphasia. Brain Lang., 85: 357–368. Plautz, E.J., Barbay, S., Frost, S.B., Friel, K.M., Dancause, N., Zoubina, E.V., Quaney, B.M. and Nudo, R.J. (2003) Postinfarct cortical plasticity and behavioral recovery using concurrent cortical stimulation and rehabilitative training: a feasibility study in primates. Neurol. Res., 25: 801–810. Prezant, T.R., Agapian, J.V., Bohlman, M.C., Bu, X., Oztas, S., Qiu, W.Q., Arnos, K.S., Cortopassi, G.A., Jaber, L. and Rotter, J.I. (1993) Mitochondrial ribosomal RNA mutation associated with both antibiotic-induced and non-syndromic deafness. Nat. Genet., 4: 289–294. Rosen, H.J., Petersen, S.E., Linenweber, M.R., Snyder, A.Z., White, D.A., Chapman, L., Dromerick, A.W., Fiez, J.A. and Corbetta, M.D. (2000) Neural correlates of recovery from aphasia after damage to left inferior frontal cortex. Neurology, 55: 1883–1894. Rossini, P.M. and Dal Forno, G. (2004) Neuronal post-stroke plasticity in the adult. Restor. Neurol. Neurosci., 22: 193–206. Shimizu, T., Hosaki, A., Hino, T., Sato, M., Komori, T., Hirai, S. and Rossini, P.M. (2002) Motor cortical disinhibition in the unaffected hemisphere after unilateral cortical stroke. Brain, 125: 1896–1907. Small, S.L., Flores, D.K. and Noll, D.C. (1998) Different neural circuits subserve reading before and after therapy for acquired dyslexia. Brain Lang., 62: 298–308. Sprague, J.M. (1966) Interaction of cortex and superior colliculus in mediation of visually guided behavior in the cat. Science, 153: 1544–1547. Strens, L.H., Oliviero, A., Bloem, B.R., Gerschlager, W., Rothwell, J.C. and Brown, P. (2002) The effects of subthreshold 1 Hz repetitive TMS on cortico-cortical and interhemispheric coherence. Clin. Neurophysiol., 113: 1279–1285. Vuilleumier, P., Hester, D., Assal, G. and Regli, F. (1996) Unilateral spatial neglect recovery after sequential strokes. Neurology, 46: 184–189. Walsh, V. and Pascual-Leone, A. (2003) Neurochronometrics of Mind: TMS in Cognitive Science. MIT Press, Cambridge, MA. Warburton, E., Price, C. and Swinburn, K. (1999) Mechanisms of recovery from aphasia: evidence from positron emission tomography studies. J. Neurol., Neurosurg. Psychiatry, 66: 155–161. Ward, N.S., Brown, M.M., Thompson, A.J. and Frackowiak, R.S. (2003) Neural correlates of motor recovery after stroke: a longitudinal fMRI study. Brain, 126: 2476–2496. Werhahn, K.J., Conforto, A.B., Kadom, N., Hallett, M. and Cohen, L.G. (2003) Contribution of the ipsilateral motor cortex to recovery after chronic stroke. Ann. Neurol., 54: 464–472.
Plate 19.1. Brain activation in functional magnetic resonance imaging while subjects performed the same rhythmic hand movement (under careful kinematic control) before and after repetitive transcranial magnetic stimulation (rTMS) of the contralateral motor cortex. Following sham rTMS (top row) there is no change in the significant activation of the motor cortex (M1) contralateral to the moving hand and of the supplementary motor cortex (SMA). After M1 activity is suppressed using 1 Hz rTMS (1600 stimuli, 90% of motor threshold intensity; middle row), there is an increased activation of the rostral SMA and of M1 ipsilateral to the movig hand. Increasing excitability in the contralateral M1 using high-frequency rTMS (20 Hz, 90% of motor threshold intensity, 1600 stimuli; bottom row) results in a decrease in activation of rostral SMA. Importantly, despite the modulation of brain activity, behavior remains unchanged. The shift in activity at the targeted brain region and across network might be considered an example of rapid plasticity to sustain behavioral integrity.
Plate 19.4. Representative example of T1-weighted structural MRI scan for the one of the six chronic aphasia patients studied. The figure shows axial slices and a left lateral view for the patient’s 3-D reconstructed MRI revealing the extend of the lesion. The lower right quadrant illustrates the anatomical locations of the different brain regions targeted with rTMS in patients with nonfluent aphasia including Motor Cortex for the mouth in green, BA 44 in pink, BA45 in blue and BA 22 in orange. In all patients studied the lesions are cortical in all patients affecting the distribution of the middle cerebral artery and encompassing subcortical lesions of the medical subcallosal fasciculus and middle 1/3 of the peri-ventricular white matter.
Plate 19.6. Mild-moderate nonfluent aphasia patient, P2. Performance level on naming after 10 minutes, 1 Hz rTMS applied to neighboring brain regions, which were only 1 or 2 cm apart. Note that stimulation of two areas separated by less than 1 cm (marked in yellow) within the right pars triangularis area, consistently led to the highest number of pictures named correctly (20 pictures per test). See Plate 19.6 in Colour Plate Section.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 20
Plasticity in brain processing and modulation of pain Donald D. Price1,, G. Nicholas Verne2 and Jeffrey M. Schwartz3 1
Oral and Maxillofacial Surgery, College of Dentistry, University of Florida, Gainesville, FL, USA 2 Gastroenterology, University of Florida, Gainesville, FL, USA 3 College of Medicine, and UCLA Neuropsychiatric Institute, University of Florida, Gainesville, FL, USA
Abstract: Brain processing of pain in humans is based on multiple ascending pathways and brain regions that are involved in several pain components, such as sensory, immediate affective, and secondary affective dimensions. These dimensions are processed both serially and in parallel. They include spinal ascending pathways that directly target limbic and brainstem structures involved in pain-related emotions as well as a pathway proceeding from the somatosensory cortices to limbic cortical areas. Superimposed on this neural organization is the capacity to process the dimensions of pain in multiple ways, as in patients who lack one cerebral hemisphere but can nevertheless locate and rate pain intensity and pain unpleasantness on both sides of the body. The dimensions of pain also can be psychologically modulated in multiple ways and these changes are accompanied by corresponding changes in relevant brain structures. Finally, understanding psychological modulation of pain and pain-related brain activity is optimized by a scientific framework that integrates principles of contemporary physics, neuroscience, and human experiential science. Keywords: parallel processing; serial processing; multiple representations; pain; quantum zeno effect; selfdirected neuroplasticity
experiments, human experiments involving microstimulation and microelectrode recordings within the brain, and animal and human neural-imaging studies. This chapter reviews brain mechanisms of pain, with emphasis on neuroplastic mechanisms. It focuses on divergent sources of knowledge about brain mechanisms of pain and synthesizes this information in a manner that helps explain how pain is highly modified by psychological factors and how different dimensions of pain can be represented in multiple ways. First, the dimensions of pain will be described and briefly explained. Second, the ascending pathways for pain and functional types of nociceptive neurons of their central-target brain structures will be discussed in relation to the dimensions of pain. Third, brain regions and networks that underlie the representation and processing of different dimensions and
Introduction Beyond the levels of primary afferents and spinal cord neurons, information concerning the processing of nociceptive information and representation of different dimensions of pain within the brain, until recently, has been scant and controversial. Prior to the advent of modern neural imaging of pain, most of the information about brain mechanisms of pain in humans was derived from studies that evaluated effects of destructive lesions or electrical stimulation. Within the last 15 years, however, considerable neurophysiological and neuroanatomical information about brain mechanisms of pain has accumulated from animal Corresponding author. Tel.: (352) (846-2718); Fax: (352) (8460588); E-mail:
[email protected]fl.edu DOI: 10.1016/S0079-6123(06)57020-7
333
334
Fig. 1. Schematic representation of major dimensions of pain and their interactions.
stages of pain will be discussed. Neuroplasticity of these networks will be emphasized in two ways. First, evidence will be presented that similar types of pain can be represented in multiple ways and in multiple circuits, for example, in different pathways of split-brain patients and persons with early cerebral damage. Second, evidence will be presented that neural activity in pain-related areas of the brain can be powerfully altered by psychological methods, including placebo, hypnosis, and changes in attention. Thus, evidence for considerable plasticity of brain mechanisms underlying pain will be reviewed. The schematics in Figs. 1 and 2 serve as guides to discussions of the dimensions of pain and their associated anatomical pathways and brain structures.
What are the dimensions of pain and how do they interact? Pain is an experience that is comprised of unique somatic or visceral sensory qualities, such as ‘‘burning,’’ ‘‘stinging,’’ or ‘‘aching’’, combined with a sense of intrusion or threat, or both, and an associated feeling of unpleasantness or other negative emotional feelings (Price, 1999). Some emotional feelings pertain to the immediate present and others may pertain to the long-term
future implications and past. Thus, pain includes a sensory dimension, an immediate affective dimension (e.g., unpleasantness), and sometimes a secondary affective dimension, termed pain-related suffering. The latter is based on rumination or reflection and pertains to the long-term implications of having pain, such as interruption of life activities and other concerns with future consequences. Psychophysical studies demonstrate that the first two pain dimensions have reliably different relationships to nociceptive stimulus intensity and can be differentially influenced by various psychological factors (Fig. 1; Price, 1999, 2000). Numerous studies support the view that the sensory and immediate affective dimensions of pain are separate and unique, even though they are often closely associated (See Price and Harkins, 1992; Price, 1999, 2000; for reviews). Two related experiments clearly illustrate this view and help establish the direction of causation between the two dimensions (Rainville et al., 1992). Both experiments were part of a hypnosis study in which pain was induced in study participants by immersing their left hands in a moderately painful waterbath heated to 47 1C. In the first experiment, hypnotic suggestions were alternately given to enhance or decrease pain unpleasantness without changing pain-sensation intensity. In the second, the hypnotic suggestions were targeted specifically toward
335
Fig. 2. Schematic of ascending pathways, subcortical structures, and cerebral cortical structures involved in processing pain. Symbols are those defined and referred to within text. See Plate 20.2 in Colour Plate Section.
enhancing or decreasing pain-sensation intensity and nothing was stated about pain unpleasantness. Pain unpleasantness but not pain-sensation intensity ratings were changed in the directions suggested in the first experiment, a result that was not surprising. However, both pain-sensation intensity and pain unpleasantness ratings changed in parallel in the second experiment despite the fact that the suggestions did not mention pain unpleasantness. This study helps to establish that the direction of causation-pain sensation is more of an immediate cause of pain unpleasantness than is the latter a cause of pain sensation. Thus, there is a serial relationship between the sensation of pain and its associated unpleasantness. Other psychophysical experiments and studies of pain patients also support a serial relationship between painsensation intensity and pain unpleasantness (Price et al., 1987; Price, 1999, 2000). At the same time, the first experiment showing selective effects on unpleasantness suggests parallel influences on pain unpleasantness (Fig. 1). If pain unpleasantness can
be selectively altered, pain sensation cannot be the only cause of unpleasantness. Parallel influences result from physical and psychosocial context and memory. Likewise, studies of pain patients show the distinction between immediate pain unpleasantness and pain-related suffering and their sequential interactions. Similar to the selective effects of some types of hypnotic suggestion on pain unpleasantness, personality traits and some demographic characteristics such as age can selectively influence pain-related suffering. For example, neuroticism enhances (Harkins et al., 1989) and age reduces (Riley et al., 2000) pain-related suffering without changing the sensory intensity or immediate unpleasantness of pain. Pain unpleasantness maintains and is more of a cause of pain-related suffering than vice versa. However, Rainville et al. (2005) have shown some recursive effects between pain-related suffering and immediate pain unpleasantness, because increases in pain-related secondary emotions such as depression, anxiety,
336
anger, and sadness were shown to enhance immediate pain unpleasantness. The effects on painsensation intensity were not significant. Thus, recursive effects of the model (Fig. 1) do not extend across all stages, whereas changes in magnitudes of nociceptive sensations would normally affect all subsequent stages. The sequential model of sensory intensity–unpleasantness–pain related suffering also is supported by multivariate (linear structural relations) analyses of ratings of these dimensions in large samples of pain patients (Wade et al., 1996; Riley et al., 2000, 2002; Lackner et al., 2005). The sequential model (Fig. 1) has been repeatedly confirmed and scores high on several indices of goodness-of-fit (Wade et al., 1996; Riley et al., 2000, 2002; Lackner et al., 2005). This model provides a psychological framework that will be discussed in relation to ascending pathways and brain circuits, as well as to mechanisms by which pain can be modified. Ideally, physiological activity in ascending pathways and brain circuits should help explain psychological models of pain, such as the serial/parallel processing model of Fig. 1.
Ascending nociceptive pathways to the brain The lateral spinothalamic tract and its thalamic projection to somatosensory cortices For the most part, the central nervous system origin of pain-related pathways is the dorsal horn of the spinal gray matter. A major pathway projects in the lateral spinothalamic tract to the ventral posterior lateral (VPL) nucleus of the thalamus. Dorsal-horn neurons of origin of this pathway are comprised of wide dynamic range (WDR) neurons that respond differentially to gentle and nociceptive levels of stimulation as well as nociceptive-specific (NS) neurons that respond predominantly to nociceptive stimuli (Price and Verne, 2002; Price et al., 2003). WDR neurons receive synaptic input from both primary-afferent nociceptive neurons and low-threshold mechanoreceptive neurons. As a consequence of this convergence, they respond with increasing impulse frequency over a very wide range of stimulation
intensity, such as very gentle touch or hair movement to tissue-damaging stimuli (Price et al., 1976, 1978, 1979; Willis, 1985). Hence, the term ‘‘wide dynamic range’’ aptly applies to this class of neurons. Based on a review by Willis et al. (2003), their anatomical work (Willis et al., 2001), and numerous studies of nociceptive neurons in the dorsal horn (See references in Price and Verne, 2002; Price et al., 2003), VPL (Casey and Morrow, 1983; Chung et al., 1986; Lenz et al., 2004), and primary somatosensory cortex (S1) (Kenshalo and Isensee, 1983; Kenshalo et al., 1988, 2000), there is considerable evidence that the large majority (470%) of neurons in the lateral spinothalamic tract and its thalamic extension to S1 are WDR neurons and most of the remaining neurons are NS. Spinal-cord WDR and NS neurons at the origin of this pathway are located in superficial and deep layers of the dorsal horn (Price et al., 1976, 1978, 1979; Willis, 1985). WDR and NS neurons within the lateral spinothalamic tract and its central targets are intermingled at all levels of this pathway (Kenshalo et al., 2000; Willis et al., 2003; Price et al., 2003). Similar to psychophysical responses of both monkeys and humans, monkey WDR neurons at dorsal horn and cortical levels can detect very small changes in nociceptive stimulus intensity. For example, they reliably increase their impulse frequencies in response to very small 0.2–0.4 1C temperature shifts during nociceptive skin temperatures such as 46 and 47 1C (Bushnell et al., 1984; Dubner et al., 1986; Kenshalo et al., 1988; Maixner et al., 1989). In contrast, NS neurons did not make such fine discriminations. Based on the capacity of WDR to precisely encode stimulus features of nociceptive stimuli, the parallels between WDR neuron and psychophysical responses to nociceptive stimuli, and the strong predominance of WDR neurons in the lateral spinothalamic tract, it has become abundantly clear that WDR neurons have important roles in the sensory processing of pain. The predominance of WDR neurons in this pathway has large implications for neuroplasticity in the processing of pain. The reason is that WDR neurons have a distinct receptive-field organization that can change over time as a result of several
337
physiological and psychological factors. The receptive fields of WDR neurons consist of a relatively small skin area (e.g., 15 cm2) within which nociceptive stimuli (e.g., 45–511C) evoke higher impulse frequencies than gentle mechanical stimuli (e.g., guard hair deflection or indentation with von Frey filaments with forces less than 50 mg), surrounded by a larger area (e.g., 50–70 cm2) wherein intense (e.g., 41 g von Frey filament) or nociceptive stimuli are needed to evoke impulse responses. However, the sizes of these receptive-field areas are not constant but critically depend on the history of stimulation and upon psychological factors such as attention. The factors that induce changes in WDR-receptive field sizes will be discussed in the section on pain modulation. The organization and sizes of WDR-receptive fields are integral to their ability to distinguish nociceptive from non-nociceptive stimulation and to encode the intensity of nociceptive stimulation (Price et al., 2003). Populations of WDR neurons are capable of encoding these distinctions as well as localizing the peripheral source of nociceptive stimulation (Price et al., 1978; Price 1988, 1999; Coghill et al., 1993). Unlike WDR neurons, NS neurons receive exclusive synaptic input from primary nociceptive afferent neurons and they comprise 20–30% of neurons within the spinothalamocortical pathway to S1 (Kenshalo et al., 2000). They also have an important role in pain, probably that of identifying some of the stimulus features of nociceptive stimuli. Because of their small receptive fields and specific responsiveness to one or more forms of nociceptive stimulation (e.g., heat, pinch), they may fine-tune the ability to recognize the body location and particular type of nociceptive stimulus (e.g., mechanical). NS neurons are likely to function in concert with WDR neurons. In general, the physiological characteristics of both types of neurons observed at the level of the spinal-cord dorsal horn appear to be preserved within brain areas that receive input from these spinal-cord neurons. Thus, both WDR and NS neurons are present at several brainstem and cortical levels and their physiological characteristics resemble those of dorsal horn nociceptive neurons in many ways.
Other ascending nociceptive pathways to the brain Several other ascending spinal pathways are important for pain (Fig. 2). An interesting difference between some of these pathways and the lateral spinothalamic tract is that unlike the latter, the former often reflect a preponderance of NS neurons. For example, Bernard and colleagues have characterized two novel pathways for pain, termed the spino-parabrachio-amygdaloid and the spinoparabrachio-hypothalamic pathway (Bernard et al., 1989; Bernard and Besson, 1990). The most remarkable and consistent feature of nociceptive neurons within these two pathways is that they mainly comprised of neurons predominantly responsive nociceptive stimulation, that is, NS neurons. This pattern is nearly the reverse of the lateral spinothalamic tract. Consistent with this unique feature is the finding that spinal-cord neurons of the origin of these pathways are exclusively within layer I of the dorsal horn, a region containing more NS than WDR neurons (Price and Dubner, 1977). Neurons within these pathways appear to encode nociceptive stimulus intensity with some degree of precision. These two pathways offer a striking comparison with the lateral spinothalamic tract because of the very different proportions of WDR and NS neurons and because their central targets appear functionally very different. Whereas the lateral spinothalamic tract and pathway to S1 is likely to be heavily involved in sensory-perceptual aspects of pain, the two pathways to the amygdala and hypothalamus are more likely to be heavily involved in autonomic processes and behaviors related to fear and defense. The central nucleus of the amygdala has been strongly implicated in fear, emotional memory and behavior, and autonomic and somatomotor responses to threatening stimuli (Bernard and Besson, 1990). Various hypothalamic nuclei have also been implicated in these functions. Based largely on their central targets, these pathways are very likely to strongly participate in the affective dimension of pain, particularly the initial affect associated with acute pain. There is also a spinohypothalamic pathway that contains neurons like those of the spinothalamic tract, and in fact many of the latter send axonal projections to both
338
thalamus and hypothalamus (Burstein et al., 1987; Dado et al., 1994; Giesler et al., 1995). Spinothalamic tract neurons projecting to more ventral and medial thalamic nuclei also appear to have much greater proportions of NS neurons (Willis, 1985; Giesler et al., 1981; Lenz et al., 1993).
Which central targets of ascending pathways participate in sensory and emotional aspects of pain? The foregoing analysis suggests the existence of several functionally different ascending pathways for pain. WDR neurons are critical for appreciating the intensity of painful sensations, which in turn contribute to pain-related emotions. Other pathways that contain mostly NS neurons are also important for several aspects of pain, including affect, motivation, autonomic and somatomotor activation, and possibly pain sensation. Although these different pathways are functionally diverse, we do not interpret them as supporting the classic view of two ascending spinothalamic systems for pain, a lateral system for sensory discrimination and a medial one for pain-affect (Melzack and Casey, 1968). We are in strong disagreement with this classic view because it is overly simplistic and does not account for numerous anatomical, physiological, and psychological observations. Most critically, it does not take into account the possibility of multiple central representations of the same pain dimension (e.g., pain-sensation intensity and pain unpleasantness) and the role of neuroplasticity. The following discussion provides an alternative account of how sensory and emotional dimensions of pain are represented within the brain and depend on interactions within a distributed network. As illustrated in the schematic of Fig. 2, some ascending spinal pathways target limbic structures such as the amygdala, insular cortex (IC), and anterior cingulate cortex. Some of these structures, such as the amygdala, are likely to be involved in rudimentary emotions such as fear. Others terminate in brainstem areas involved in arousal (reticular-formation nuclei), motoric orientation (superior colliculus), autonomic nervous-system responses (parabrachial nucleus, hypothalamus),
and neuroendocrine responses (hypothalamus). All of these structures are involved in monitoring the state and integrity of the body, and are part of the regulation of emotions and consciousness itself (Damasio, 1994; Price 1999, 2000). However, the somatosensory cortices and posterior parietal cortices are equally important inputs to limbic cortical structures, such as the insula and anterior cingulate cortex (Friedman et al., 1986; Neal et al., 1987, 1990). Therefore, there are two general ways that ascending pain-related information can access brainstem and limbic cortical structures involved in pain affect. One type of pathway targets these structures without directly activating somatosensory cortices, and the other proceeds through the somatosensory and posterior parietal cortices. As shown in Fig. 2, the lateral spinothalamocortical pathway is anatomically interconnected from S1 to a ventrally directed corticolimbic, somatosensory pathway that targets the very same limbic cortical areas that are more directly targeted by ascending spinal pathways. It has been proposed that this corticolimbic pathway integrates somatosensory input with other sensory modalities such as vision and audition and with learning and memory (Friedman et al., 1986). This pathway proceeds from S-1/S-2 to posterior parietal cortical areas and to IC and from IC to amygdala, perirhinal cortex, and hippocampus (Friedman et al., 1986; Neal et al., 1987, 1990; Price, 2000). One can conceptualize the schematic model of Fig. 2 as supporting the existence of both serial and parallel pathways for processing the different dimensions of pain. The serial pathway projects via a somatosensory–limbic connection and the parallel pathway through lower brainstem and thalamic nuclei to limbic structures. In our view, the serial pathway contributes to both sensory and affective dimensions of pain and other parallel pathway also contributes to multiple components of pain, including emotional feelings. This explanation is also consistent with the way pain dimensions are psychologically processed, as shown in Fig. 1. Neurological evidence supports this view as well. Thus, it shows involvement of the lateral spinothalamic tract to VPL and pathway to S1 in both sensory and affective dimensions of pain. Damage to S1 or to the lateral thalamus produces reduced capacity to appreciate both the sensory intensity and
339
unpleasantness of a nociceptive stimulus (Head and Homes, 1911; Echols and Clogclough, 1947; Kenshalo et al. 1989; Greenspan et al., 1997; Ploner et al., 1999). Lesions of the postcentral gyrus produce a temporary reduction in ongoing clinical pain intensity (Lewin and Phillips, 1952; White and Sweet, 1966). Both Ploner et al. (1999) and Head and Holmes (1911) found patients with lesions of either the somatosensory cortices or lateral somatosensory thalamus (one of these was histologically verified to be within VPL) to have deficits in pain appreciation, including both pain-sensation intensity and pain unpleasantness. This deficit was manifested as an inability to experience either sensations or unpleasantness in response to mild or moderate nociceptive stimulus intensities, such as 45–471C skin temperatures (Head and Holmes, 1911). These same patients, however, developed vague feelings of unpleasantness or ‘‘pain-like’’ feelings when the stimulus intensity was raised to still higher levels of stimulation, such as 50 1C. Thus, interruption of the pathway to VPL and S-1/S-2 cortex produces a deficit in pain sensation and pain-related unpleasantness throughout most of the nociceptive range. When the stimulus intensity reaches a high-enough level, other pathways contribute in a parallel manner to unpleasantness. Finally, recent observations from lesion and brain-imaging studies in humans suggest that S1, S2, and IC may be necessary for the recognition of negative emotions in others as well as self-generated negative emotions (Damasio, 1994; Adolphs et al., 2000). Electrical stimulation of some sites within thalamic VPL elicits pain, often with normal sensory qualities (Lenz et al., 2004). These sites were stimulated in patients who were undergoing brain surgery for movement disorders (n ¼ 50) or chronic pain (n ¼ 16). They were awake and rated evoked pain on visual analog scales. Two types of responses were observed. At some ‘‘pain sites’’ increasing frequencies of stimulation evoked graded sensations that extended from nonpainful to painful, similar to the graded responses of WDR neurons recorded in VPL of the same participants. Stimulation of other sites specifically evoked only pain, consistent with the selective nociceptive responses of NS neurons in VPL of these participants. Using the same technique of combining
single-neuron recording and electrical stimulation in awake patients, neurons have been found within the region posterior and inferior to VPL that respond exclusively to noxious stimuli (Lenz et al., 1988, 1993). Microstimulation of these sites at or near these NS neurons evoked only pain sensations (e.g., heat but not touch). These sites are likely to correspond to the posterior nuclear complex of other primates. Although the evidence is more scant and controversial, electrical stimulation within the postcentral gyrus sometimes evokes pain in humans. Though a systematic search was not undertaken to find cortical sites involved in pain, Penfield and Boldrey (1937) found 11 out of 426 stimulation sites wherein electrical stimulation of the exposed postcentral gyrus produced painful sensations in patients. These reports were such rare occurrences that the investigators concluded that appreciation of pain was not represented in the postcentral gyrus (S1). However, the fact that stimulation of at least some postcentral gyrus sites produced pain is evidence in favor of a role of the postcentral gyrus in sensory discriminative aspects of pain. The paucity of specific pain-related cortical sites could partly reflect the fact that Penfield and Boldrey were not specifically searching for pain sites. Another attempt to evoke pain by stimulation of the postcentral gyrus has shown that intense and unpleasant pains could be consistently elicited by stimulation of some sites within this region (Echols and Cogclough, 1947). The problem with this study is that it was carried out on patients suffering from phantom limb pain, a condition that may have potentiated pain or lowered the cortical threshold for pain. Thus, both lesion and electrical-stimulation studies generally support the involvement of thalamic VPL and somatosensory cortices in sensory and affective dimensions of pain, though the deficits from lesions are usually partial and often there is recovery of function. That these deficits are partial and temporary attests to the existence of multiple representations of pain and the potential for neuroplastic reorganization of brain circuitry, as will be discussed later. More selective deficits in emotional dimensions of pain are observed in patients with large lesions of the IC (Weinstein et al., 1955; Berthier et al.,
340
1988), anterior cingulate cortex (Foltz and White, 1962; White and Sweet, 1966), or prefrontal cortex (Hardy et al., 1952; White and Sweet, 1966). Damage to large parts of IC has been found among patients with pain asymbolia (Rubins and Friedman, 1948; Weinstein et al., 1955; Berthier et al., 1988). Patients with this condition do not display behavior indicative of threat or intrusion in response to painful stimuli. They no longer appreciate the destructive significance of pain and do not withdraw from pain stimuli or threatening gestures. This is so despite their capacity to detect the sensory features of nociceptive stimulation. Likewise, focal damage to the posterior parietal cortical area 7b in the monkey results in an absence of escape responses to normally painful temperatures (i.e., 51–521C) despite preservation of the ability to detect the offset of noxious thermal stimuli (Dong et al., 1996). Infraparietal area 7b receives input from S1 and S2 somatosensory cortex and projects to the IC. Therefore, this region of the parietal cortex and the IC may be a critical interface between sensory-discriminative and the immediate affective-motivational dimension of pain. However, electrical stimulation of several sites within the posterior insula has been shown to evoke unpleasant pains, often with distinct sensory qualities (Ostrowsky et al., 2002). These observations suggest that the posterior insula may function in to some extent in both sensory and affective components of pain. The involvement of the prefrontal cortical lobes in complex aspects of cognitive-evaluative and hence pain-related suffering is supported by detailed observations carried out on patients before and after prefrontal lobotomy. Hardy et al. (1952) found that prefrontal lobotomy produced no overall change in heat-induced experimental pain thresholds in eight patients tested. Nevertheless, there was some reduction in perceived intensity of clinical pain in four of five patients studied. However, the most striking changes were in patients’ attitudes, emotional reactions, and cognitive processing of pain. The lobotomized patients were emotionally indifferent to low-intensity pains, which, though perceived, evoked few affective reactions. A statement that epitomized this attitude was ‘‘Yes, I feel the pain but it doesn’t bother me’’.
Moderate to high intensity pains sometimes evoked overreactions manifested by a show of grimacing, fears, and agitation when direct questions forced them to focus on the pain. However, when patients were left alone, spontaneous suffering and pain-related behaviors were nearly absent. They showed little spontaneous concern about the negative implications of pain as regards damage to the body or threat to life. White and Sweet (1966) have corroborated these observations. Evidently, lobotomy somehow interferes with the spontaneous ongoing cognitive evaluations that are related to the long-term implications of having a persistent pain condition. It is possible that lobotomy selectively reduces pain-related suffering, a stage that is partly based on reflective processes related to memory of past consequences of having pain as well as future implications (Fig. 1). The lack of spontaneous pain-related suffering of patients with prefrontal lobotomy represents a deficit that differs from that which occurs in pain asymbolia (Rubins and Friedman, 1948; Weinstein et al., 1955). Lobotomized patients appear to perceive the immediate threat of pain once it is brought to their attention (Hardy et al., 1952), whereas patients with pain asymbolia appear incapable of perceiving the threatening nature of nociceptive stimuli under any circumstances. Immediate threat and sense of intrusion are cognitive mediators of what is unpleasant about pain and this component of pain affect is supported by insular and anterior cingulate cortical activity (Fig. 2). These areas, in turn, interact with posterior parietal, temporal, and prefrontal cortex, to integrate inputs from multiple sensory modalities related to present context with inputs related to past events. These interactions support the interface between immediate unpleasantness and pain-related suffering.
Electrophysiological studies of central neurons that integrate nociceptive with other inputs Most of the studies described in the preceding section are based on mainly on observations of human responses to nociceptive stimuli after central lesions and after electrical stimulation of central
341
pathways and brain regions. The emphasis of electrophysiological studies of pain-related pathways has been on sensory processing and early stages of pain. However, beyond the immediate processing associated with pain-sensation intensity, it seems reasonable that further processing of pain requires an evaluation of the pain sensation in relationship to the overall context in which the nociceptive stimulus occurs. This evaluation may represent part of the interface between sensory and affective dimensions of pain. Thus, one might expect that neurons involved in such a function would integrate sensory inputs from multiple sources, that is, they would be multisensory. For example, if the perceived degree of threat of a bee sting is enhanced by hearing and seeing the stinging bee, then such enhancement could occur among multisensory neurons and could be verified in neurophysiological experiments that use combinations of visual, auditory, and nociceptive stimuli.
Infraparietal cortex In direct support of this possibility, some neurons of the infraparietal cortex in area 7b in the monkey respond optimally to nociceptive stimuli yet also respond to visual stimuli (Dong et al., 1989, 1994). Dong et al. (1994) found that the responses of 7b neurons to mildly noxious heat stimuli (44–451C) were enhanced by antecedent or concurrent visual stimuli. However, this enhancement only occurred if the target location or direction of motion within the visual receptive field was spatially aligned with the cutaneous receptive field. The enhancement was much greater for mild nociceptive stimuli (44–451C) than for stronger stimuli (471C). Thus, it would appear that the neural organization of this region of the posterior partietal cortex is that of integrating nociceptive inputs with other sensory inputs in a manner that conveys information about the overall degree of threat presented to an organism (i.e., seeing a bee while feeling its sting). This integration is especially critical at the low end of the nociceptive stimulus range, wherein an organism must make a behaviorally relevant decision about the extent of threat presented by an object. This interpretation is based on principles of
multisensory integration that have been elaborated in great detail by Stein and colleagues (Stein and Meredith, 1993). Although insular cortical neurons are often multimodal (Hicks et al., 1988), possibly because they receive multimodal sensory input from parietal 7b cortex, they have yet to be tested for multisensory integration. Given the types of deficits in patients with pain asymbolia and the multimodal integration in parietal area 7b, it is certainly reasonable to hypothesize that it would occur in the IC.
Anterior cingulate cortex The physiology of the ACC and its role in pain is even more complex than that of the posterior parietal and IC. Electrophysiological studies of neurons in ACC suggest that they are involved in advanced stages of pain-related processing (Isomura and Takada, 2004). Similar to posterior parietal and insular cortices, ACC neurons appear to integrate multiple sensory inputs. They integrate sensory input with attention, memory, and alternative motor responses. Whereas S1 somatosensory neurons are involved in processing the discriminative features of nociceptive stimuli, such as intensity encoding and detection of differences, neuronal activity in the ACC is more closely associated with escape responses, attention, and response selection in monkeys (Iwata et al., 1980; Isomura and Takada, 2004; Nakamura et al., 2005). For example, ACC neuronal activity was found to be significantly higher when monkeys escaped from a nociceptive heat stimulus than when they detected a small change in nociceptive temperature (Iwata et al., 1980). Neurons of the rostral anterior cingulate motor area appear to be involved in the selection of appropriate motor responses as well as in the planning of movements (Isomura and Takada, 2004). In the context of a pain condition, all of these functions may be interrelated with the immediate affective dimension of pain. After all, the unpleasantness of pain is associated with domination of attention and motivation for escape or termination. Some ACC neurons respond during cues that signal impending pain and therefore during
342
anticipation of pain, while others respond to anticipation of a reward (Koyama et al., 2001), consistent with neuroimaging studies showing that the rostral ACC is activated during anticipation of pain. Thus, the anterior cingulate cortex may be a part of an attention-evaluation network. It may coordinate input from IC with output to the prefrontal cortex. As discussed earlier, the IC may combine nociceptive inputs with those of other sensory modalities to provide an integrated output related to a sense of intrusion or threat to the physical body (Dong et al., 1994). The ACC may coordinate input from the insula and medial thalamic nuclei with prefrontal cortical areas and limbic structures such as the amygdala to provide an appropriate motor response or plan of action (Fig. 2). The concomitant activation of the ACC and prefrontal cortex may be related to a part of a feedback network related to attention, cognitive evaluation, and self-awareness (Cohen, 1993). This possibility is consistent with the effects of prefrontal damage that include impairments in planning, behavioral control, affective attachment, and directed/sustained attention (White and Sweet, 1966; Cohen, 1993; Damasio, 1994).
Are there multiple ways that pain can be processed or represented in the brain? As shown in Fig. 2, the dimensions of pain are related to multiple ascending pathways and a distributed network of multiple brain regions. This network contains both serial and parallel processing of visceral and somatosensory information. This neurophysiological model has at least a general correspondence with the psychological manner in which pain is processed (Fig. 1). For example, there is psychophysical, neuroanatomical, neurological, and physiological evidence for both serial and parallel relationships between sensory and immediate unpleasantness of pain, as has been discussed so far. Although all regions of the pain matrix are involved in pain affect, some are more proximal to the affective feelings and expressions of pain. This issue is complicated still further by the possibility that there exist alternative ways that the sensory and affective dimensions
of pain can be represented in the brain. This possibility also is of interest to philosophers who question whether a given type of subjective experience, such as pain, can be realized in multiple ways. Two studies support this possibility. Psychophysical analysis of pain ratings of a man with total corpus callosum section provided an unusual opportunity to assess the contributions of alternative ascending pathways to pain and alternative neural representations of the dimensions of pain (Stein et al., 1989). As discussed earlier, a major ascending pathway for pain is that of the crossedspinothalamic tract to VPL and somatosensory cortices S1/S2 (Fig. 2). Normally, the two cerebral hemispheres can share nociceptive information arriving at S1/S2 by means of the corpus callosum. When the collosum is sectioned, the experience of nociceptive stimuli presented ipsilateral to the responding hemisphere can only result from impulses from ascending pathways that are not part of the crossed pathway to VPL and S1/S2 cortex, whereas the experience of contralateral nociceptive stimuli would reflect input from all nociceptive pathways. Two contact thermodes were placed on both feet in some trials or on both hands in others. Nociceptive temperatures of 43–511C could then be delivered to either the right or left foot (or right or left hand) and both the intensities and the stimulus locations were randomized. As might be expected, his ratings of stimuli presented contralateral to the responding hemisphere were normal and like those of most participants, that is, they increased monotonically from 43 to 511C. In contrast, when these same stimuli were presented ispsilateral to the responding hemisphere, he gave much lower unpleasantness and intensity ratings to low to moderate nociceptive temperatures (45–471C). This deficit reflects an absence of the contribution of the major contralateral pathways and S1/S2 to pain sensation and unpleasantness. Recall that similar deficits occur with lateral thalamic lesions (Head and Holmes, 1911; Greenspan et al., 1997). Surprisingly, when high-intensity nociceptive stimuli (49 to 511 skin temperatures) were presented ipsilateral to the responding cerebral hemisphere, he gave high pain sensation and pain unpleasantness ratings, similar to those
343
obtained from contralateral stimuli. This pattern of responses suggests that at high enough nociceptive levels, recruitment of pathways other than the classical crossed spinothalamic-cortical pathway may make up for the pain sensory deficit that occurs when one cerebral hemisphere is no longer able to share nociceptive information with the other via the corpus callosum. The other implication of this pattern of results is that both sensory and unpleasantness components of pain can be represented in multiple pathways. If that is the case, then effects of damage to one or the other pathway may be temporary if remaining pathways increase their responsiveness as a result of functional or anatomical reorganization. This possibility is supported by a study of pain and tactile sensitivity in four patients with complete unilateral hemispherectomy (Olausson et al., 2001). All four patients were easily able to perceive tactile and painful stimuli on both legs and there was a deficit in ability to localize pain in two of the patients. Ratings of pain intensity and unpleasantness ratings induced by contact heat and ratings of brushing were similar for both legs and similar to that of normal individuals. fMRI brain imaging of these patients revealed that brushing and painful heat-activated normal pain-related areas, S1, S2, insula, and anterior cingulate cortex, regardless of which leg was stimulated. These results show that ascending pathways and brain regions can develop the capacity for processing tactile and nociceptive information arriving as a result of stimuli presented ipsilateral to the responding cerebral hemisphere.
Modulating pain and neural representations of pain The widely distributed activations of multiple pathways and brain areas associated with pain can also be modulated on a moment-by-moment basis by numerous psychological factors, including, attention, expectations, anticipations of pain, and emotional states. Each of the locations of synaptic interaction shown in Fig. 2 represents a potential point at which processing of pain-related information can be modified by psychological factors. Presumably, this modification is accompanied by
changes in different aspects of pain experience. Demonstrations of parallel changes in pain experience and neural processing reflect both psychological and neural plasticity. The following section discusses these parallel changes in the context of how attention, expectations, placebo, and hypnotic suggestions modulate central neural activity and pain.
Brain-to-spinal-cord modulation of pain Descending modulation of spinal-cord processing of nociception has been a major topic in pain research over the last 35 to 40 years and has a voluminous literature (see Basbaum and Fields, 1978; Fields and Price, 1997 for reviews). Detailed explanations of descending modulation are beyond the scope of this chapter. Instead, a few examples will be given of studies that characterize the dynamic nature of brain-to-spinal-cord modulation of pain. Descending modulation of spinalcord nociceptive processing begins in the forebrain of mammals and includes such structures as the amygdala and rostral anterior cingulate cortex. These areas are also those which receive input from ascending nociceptive pathways. A major target of these structures includes the periaqueductal grey (PAG) of the midbrain. The PAG, in turn, projects to the rostroventral medulla (RVM) onto three types of neurons, on-cells, offcells, and neutral cells. All three types of RVM neurons project to the dorsal horn of the spinal cord, where on-cells amplify and off-cells inhibit nociceptive transmission. Amplification and inhibition presumably occur under different psychological circumstances, such as stress, fear, anticipation, and different types of attentional focus. Thus, this descending control system has bidirectional control of pain at the first synapse on neurons of origin of ascending spinal pathways. This system also is integrally related to endogenous opioid mechanisms, as demonstrated by a long history of research. Descending modulatory influences from attention on dorsal-horn sensory projection neurons have been shown in experiments in awake-trained monkeys (Hayes et al., 1981; Hoffman et al., 1981). These studies showed that responses of
344
trigeminothalamic WDR and NS neurons of the medullary dorsal horn could be modified by different attentional sets and conditions of stimulus relevance (Bushnell et al., 1984). Nociceptive temperatures (45–491C) applied to the monkeys’ face evoked impulse discharges in these neurons that were an increasing function of stimulus temperature. For a given temperature, a neuron would increase its firing rate and then reach a steady plateau frequency. At a randomly determined time during the plateau, a small temperature increase was presented. This small increase was accompanied by a corresponding increase in impulse frequency. WDR neurons were especially sensitive and could detect small temperature increases of 0.2–0.3 1C (e.g., from 46.0 to 46.3 1C). However, these responses to small temperature shifts increased during a task in which the monkey had to detect them in order to receive a reward (a relevant attentional set). They decreased when they performed a similarly demanding visual-discrimination task (an irrelevant attentional set). These opposing effects predict that attending to small differences in nociceptive temperatures would enhance pain and attending to small differences in light intensity during the same nociceptive stimulus would decrease pain. A psychophysical study by Miron et al. (1989) confirmed this prediction. The impulse frequencies elicited in WDR and NS trigeminothalamic neurons by nociceptive temperatures (45–491C) also can be modified by attentional factors related to anticipation. Impulse frequencies of WDR neurons were higher and their receptive fields were larger when a warning light preceded nociceptive stimuli as opposed to when these stimuli occurred unexpectedly. As discussed earlier, receptive-field sizes of WDR neurons are somewhat large and complex and this complexity is compounded by the fact that they are under moment-by-moment modulatory control. Thus, the magnitudes of the stimulus–response relationships are under dynamic modulatory control and may play as direct a role in determining pain perception as do the peripheral stimulus events themselves. The dorsal horn is part of a network involved in shaping pain perception, a network that includes the first synapse in ascending pathways for pain. This type of modulation had been anticipated by
existing theories of pain (e.g., the gate control theory), but these studies were the first to demonstrate this principle in a behavioral context. It is more difficult to determine modulatory effects at a spinal level in humans. However, an important consequence of inhibition (or facilitation) at the level of the dorsal horn is that all of the central targets of ascending pathways should be inhibited as a consequence of inhibition at the earliest central level. This potential consequence is also important for reducing the physiologically detrimental consequences of pain, such as reduced immune responses and deleterious autonomic and neuroendocrine responses. If a given type of analgesic mechanism involves brain-to-spinal cord inhibition, then one should observe reductions in many painrelated areas of the brain, including thalamus, S1/ S2, insula, and ACC. Thus, if attention modulates human pain in a manner suggested by studies of dorsal horn neurons in awake monkeys, inhibition should occur at all of these central regions.
Effects of attention and distraction at multiple painrelated brain regions This prediction is supported by imaging studies of the effects of attention and distraction in humans on pain-evoked activity in the thalamus and several cortical regions, such as the primary somatosensory cortex (S1), anterior cingulate cortex (ACC), and IC (Bushnell et al., 1999; Bantick et al., 2002; Brooks et al., 2002;Hoffman et al., 2004; Seminowicz et al., 2004; Valet et al., 2004). In one study, several pain-related brain regions were more activated during a pain-attention condition compared to a distraction condition (Peyron et al., 1999). In a similar study, Bushnell et al. (1999) showed the greatest attentional modulation of pain-evoked activity in S1 cortex, perhaps because participant’ attention was directed toward a sensory feature of the stimulus. Other regions, including the periaqueductal gray matter (PAG), ACC, and orbitofrontal cortex, may also be involved in the modulatory circuitry related to attention, as they have been shown to be activated during pain-distraction tasks (Petrovic et al., 2000; Tracey et al., 2002).
345
Expectation effects at multiple pain-related brain regions A study by Koyama et al. (2005) similarly demonstrates that expected pain levels powerfully shape pain-related neural activity in multiple brain regions. By combining psychophysical and fMRI techniques, brain activation associated with the intensity of expected pain and experienced actual pain was characterized. Expectation was manipulated by providing cues that impending skin-temperature stimuli would be higher or lower than those actually delivered. When cues were provided that signaled lower stimulation intensities, both ratings of expected pain and ratings of pain in response to actual stimuli were significantly reduced. The latter were reduced by 28%, approximately equivalent to a clearly analgesic dose of morphine (0.08 mg/kg of body weight) (Price et al., 1985). Nearly 85% of the variability of changes in experience of pain could be accounted for by changes in expected magnitudes of pain. Corresponding large reductions in neural activity occurred in multiple pain-related brain areas, including the thalamus, S1/S2 somatosensory cortices, insula, and dorsolateral prefrontal cortex. Consistent with findings of other studies of mental representation, brain regions involved in processing of expectations overlapped with those involved in the processing of afferent sensory information (Kosslyn et al., 2002). Koyama et al. propose that expectation-related information is integrated with afferent sensory information within the pain matrix of the brain to provide a complete cognitive experience of pain. However, widespread modulation within this matrix also could be a result of brain-to-spinal-cord descending modulation, as described above for attention. It is not clear yet whether these are two mutually exclusive explanations and how either possibility can be ruled out.
Placebo effects at multiple pain-related brain regions Similar to both brain-imaging studies of effects of attention and expectation on pain, two studies found that placebo manipulations produced
reductions in pain ratings and widespread decreases in neural activity of pain-related areas of the brain (Wager et al., 2004; Robinson et al., 2006). In the first study, decreases in pain-related brain activity were only statistically significant after termination of stimuli and at a later time when individuals rated pain (Wager et al., 2004). This type of effect could reflect modulation of cognitive processes related to the experience of representing pain and could be at least partly related to report bias. A second brain-imaging study examined whether placebo analgesia is accompanied by reductions in neural activity in pain-related areas of the brain during the time of stimulation (Robinson et al., 2006). Brain activity of irritable-bowel syndrome patients was measured in response to rectal distension by a balloon barostat. A large placebo effect produced by placebo suggestion was accompanied by large reductions in regional cerebral blood flow in thalamus, somatosensory cortices, insula, and anterior cingulate cortex, providing evidence for active afferent inhibition in placebo analgesia. These results constitute evidence against the hypothesis that placebo analgesic effects reflect nothing beyond report bias and provide further evidence for widespread inhibition of pain processing within the brain. The similarity between effects of expectation described above and results of placebo manipulation provide further support to the hypothesis that expectation is salient factor in placebo analgesia. Although it is a major factor in placebo analgesia, expectation is unlikely to operate alone. Desire for pain relief and the positive (or less negative) feelings associated with perception of the therapeutic intervention have also been proposed to contribute to placebo analgesia. Elsewhere, we have argued that a way of accounting for psychological mechanisms of placebo analgesia is by means of an emotion model (Price and Barrell 1984; Price et al., 1985; Vase et al., 2003). This model proposed that positive and negative emotional feelings are often co-determined by strength of desire and degree of expectation about an outcome, such as pain relief (Price and Barrell 1984; Price et al., 1985). The model accounts for the observations made in some studies that expectation ratings account for 44–60% of the variability in
346
pain ratings during the placebo condition (Price, 1999), for observations in other studies that desire and expectation ratings account for even more of this variance (Vase et al., 2003), and observations in some studies that changes in emotions account for placebo responses (Vase et al., 2003, 2005). Finally, it accounts for observations that changing emotional feelings by means other than placebo interventions can modulate pain (Rainville et al., 2005). Expectations, desires, and emotional feeling states are all examples of ‘qualia’ (Price, 1999). In phenomenological terms, there is something it is like to have desires, expectations, and emotional feeling states and their existence and magnitude can only be directly reported by those who experience them (Price and Barrell, 1980, 1984; Price et al., 1985). These first-person mentalistic concepts of course have potential neural correlates. A satisfying explanation of placebo mechanisms would involve integrating both experiential (first person) and neural (third person) cause–effect relationships. This type of integration is currently a highly discussed topic among psychologists, neuroscientists, and philosophers (Velmans, 2000).
Can the different dimensions of pain be selectively modulated? Some psychological factors and interventions can selectively change the unpleasantness of pain without altering the strength of the painful sensation, as was explained in the first section of this chapter (Price et al., 1987; Rainville et al., 1992; Price 1999, 2000). When this happens, it would be informative to know which brain structures are altered under these conditions. This question also relates the problem of determining which brain structures are most proximately related to the different dimensions of pain. Attempts to analyze the differential representations of sensory and affective dimensions of pain are provided by two positron emission tomography (PET) studies. In both studies, significant activation occurred in the somatosensory area I, anterior cingulate cortex (area 24), and anterior IC both before and during hypnosis. In the first study, hypnotic suggestions were given to selectively alter the affective dimension of pain
without changing the perceived intensity of the painful sensation (Rainville et al., 1997). Recall that this approach was successfully used in a psychophysical study by Rainville et al. (1992). The experimental conditions included immersion of the left hand in moderately painful water (47 1C) without hypnosis, with hypnosis but without suggestions, with hypnosis with suggestions for increased unpleasantness, and finally with hypnosis with suggestions for decreased unpleasantness. These manipulations were successful in providing much larger unpleasantness ratings during the high unpleasantness condition as compared to the low unpleasantness condition, but no differences in pain-sensation ratings. Consistent with these changes, no differences occurred across high and low unpleasantness conditions in somatosensory area I, a region considered to process sensory components of pain. In striking contrast and consistent with unpleasantness ratings, activity in anterior cingulate cortical area 24b was much greater in the high as compared to the low unpleasantness condition. A separate regression analysis, controlling for factors such as individual differences in global cerebral blood flow and pain-sensation intensity ratings, showed that pain unpleasantness ratings were significantly associated only with anterior cingulate activity in a specific region of area 24 (R ¼ 0.55, Po0.001). These results indicate that hypnotic modulation of the affective dimension of pain is at least partly reflected by changes in cortical limbic regional activity (i.e., anterior cingulate cortical area 24b). As discussed earlier, the anterior cingulate cortex may be a part of an attention–evaluation–affect network that attributes emotional valence and attention to pain. It also may represent a region that coordinates inputs from parietal areas involved in perception of bodily threat with frontal cortical areas involved in plans and response priorities for painrelated behavior. Thus, the role of the ACC in pain-related affect is likely to be both pivotal and complex. This view contrasts with that which simply attributes the function of pain affect to the ACC. The strategy of using psychological manipulations to selectively alter the immediate unpleasantness of pain without changing the intensity of painful sensation is a potentially powerful one
347
because it could be used to selectively alter other dimensions of pain affect, for example the secondary stage associated with suffering. This strategy also demonstrates that hypnotic modification of pain is both a real psychological and biological event. Mental events change the activity of the brain in a dynamic manner. A second similarly designed study used hypnotic suggestions that were selectively targeted toward the sensory intensive dimension of pain (Hofbauer et al., 2001). However, the suggestions were effective in modulating both sensory and affective dimensions of pain experience, as measured by the participants’ ratings of these dimensions. Unlike the first study, both somatosensory S1 neural activity as well as pain-sensation ratings were higher in the high as compared to the low sensory intensity condition. There was a similar but nonsignificant trend in the anterior cingulate cortex. The most important result of the second study is that it provides a confirmation of the role of S-I somatosensory cortex in the sensory-discriminative dimension of pain.
view presented here takes into account neuroplastic mechanisms and the likelihood that there exist multiple ways that the dimensions of pain can be represented, such as in split-brain individuals and persons with total absence of one cerebral hemisphere. Finally, our account suggests that the pain matrix is involved not only appreciation the qualities and intensity of actual pain, as suggested by classical theories, but additionally in several other pain-related experiences. These include the anticipation of pain and expectation of specific pain intensities (Koyama et al., 2005), as discussed in this chapter, and the act of representing pain through rating it on a scale. The latter function is elegantly illustrated by a study by Moulton et al. (2005). They found that pain-related areas, such as S1 somatosensory cortex, were activated not only when a painful stimulus was presented, but also a second time when participants rated the pain. Apparently, the act of representing a pain through a rating requires some of the same brain structures that are activated during pain itself. Since the pain is rated after it is experienced, the somatosensory cortex must somehow rerepresent the pain. This would entail integration with memory.
What are the relationships between activity in the pain matrix and pain-related experiences? Anticipation, expectation, and magnitude judgments of pain The classic view of neural representations of pain suggests that the different dimensions of pain are represented in distinctly separate parallel pathways and brain centers. The alternative view presented in this chapter is different from this classic view in several ways. First, it proposes that there exists both serial and parallel processing of sensory and affective dimensions of pain, both at a psychological and neurobiological level. If that is the case, then somatosensory processing regions of the brain are also integral components of affective processing. This view is consistent with neurological evidence that the somatosensory cortices are involved in emotions. For example, the ability to recognize emotional states of others is reduced after damage to the somatosensory cortices (Damasio, 1994; Adolphs et al., 2000). Second, the alternative
Two types of observations in explaining pain and its neural representations Explanations of interrelationships between neural activity within the pain matrix and different aspects of pain and pain-related experiences require two types of observations. One type of observation is neurophysiological and consists of such measurements as the recording of impulse activities in neurons, global neural activity, and measurements of lesion size and location. The other type of observation is the reports of participants, who provide accounts and ratings of their experiences of pain intensity, pain unpleasantness, and other qualitative aspects of pain. This second type of observation also applies to some of the independent variables of studies, such as expected levels of pain, magnitude of desire for relief, and degree of attention to one type of stimulus or another (e.g., light versus sound). The reports of participants are about their phenomenal experience and are
348
described in mentalistic terms. The relationships between the two types of observation are often correlational, as in fMRI experiments, but can also be causal, as when unpleasant pains are evoked by electrical stimulation of VPL of the thalamus. The combinations of experiential and neurophysiological observations in experiments where causal and correlational relationships are determined offer a strategy whereby we can explain how pain is represented and dynamically modified.
How do we optimally conceptualize mind–brain relationships in efforts to understand psychological modulation of pain? This review of brain mechanisms of pain has emphasized neuroplasticity in the neural processing associated with different pain dimensions. As we have discussed, this plasticity is manifested by multiple ways that the brain can process pain and by multiple ways that pain can be modulated by attention, emotions, and expectations. As outlined by Velmans for consciousness itself (Velmans, 2000), causal relationships occur between experiential dimensions of pain (mind–mind relationships), between different areas within the brain (brain–brain relationships), between brain activity and experiential dimensions (brain–mind relationships), and finally between experiential dimensions and brain activity (mind–brain relationships). Of these four types of causal relationships, mind–brain relationships take on particular interest given recent theoretical advances in our understanding of possible effects of focused attention on nervous system function (Schwartz and Begley, 2002; Schwartz et al., 2005). Specifically, the application of well-established principles of contemporary physics to neuroscience allows us to better understand how phenomenal experience (e.g., an ‘‘image’’ of pain relief) could modify brain responses, and over time systematically change them. The reason why principles of contemporary quantum physics are fully applicable to brain physiology stems from the fact that ion channels in the brain are, at their narrowest point, less than 1 nm in diameter, or not much more than a single ion wide. Due to this profound limitation in the
uncertainty in the position of all the trillions of ions in the brain as they pass through their relevant ion channels, the brain itself must be treated as a quantum physical environment. One important and well-verified law in quantum physics called the quantum Zeno effect is a key to understanding how focused attention can systematically alter the brain’s response to environmental inputs (see Schwartz and Begley, 2002; Schwartz et al., 2005 for discussion). Quantum Zeno effect was first described nearly 30 years ago and has been extensively studied many times since then. One classic example of it is the fact that rapid repeated observation of a molecule will hold the molecule in a stable state. It does this by markedly slowing the rate of fluctuation demonstrated in molecules when they are not measured (i.e., observed) in a repetitive fashion. This is a basic principle of quantum physics — the rate of observation has marked measurable effects on the phenomenon being observed. The quantum Zeno effect for neuroscience application states that the mental act of focusing attention can hold in place brain circuits associated with what is focused on (e.g., pain versus pain relief). Focusing attention on mental experience maintains the brain state arising in association with that experience. What this means is that if one focuses attention on an experience, the set of relevant brain circuitry with which that experience is associated will be held in a dynamically stable state. For example, an expectation of pain relief can elicit a focusing of attention on actual experiences of pain relief that are associated with patterns of activity in a given brain circuitry (Koyama et al., 2005). When sufficient attention is focused on the experience of pain relief, the associated brain circuitry becomes dynamically stable. This acute effect of focused attention can then enable the well-validated principle of Hebb (1955), namely that repeated patterns of neural activity can cause neuroplastic changes and new connectivities to form in well-established neural circuits (‘‘cells that fire together wire together’’). This type of attention-based mechanism of neuroplastic change has been termed self-directed neuroplasticity to emphasize that alterations in CNS function can be readily driven by and dynamically modified
349
by willfully directed mental events (Schwartz and Begley, 2002; Schwartz et al., 2005). As was stated above, mental events change the activity of the brain in a dynamic manner. Basic principles of contemporary physics now enable us to place this empirically well-validated fact within theoretically coherent, scientifically grounded, and technically described context. The term attention density has recently been coined to help clarify this mechanism. Attention density is defined as the number of observations per unit time, and thus the higher the intensity of focus of attention the higher the attention density. The term is important because it is this increased number of observations per unit time, or increased focus that brings the quantum Zeno effect into play. This increased focus causes the relevant neural circuitry to become stabilized. It is this fact that allows us to explain how the brain’s response to pain can be systematically modified by the quality of attention that is focused (or not focused) on it. These principles of self-directed neuroplasticity are fully testable in experimental contexts. Many of the experimental findings described in this chapter can be more clearly understood in light of these theoretical advances, and many new empirically based investigations of pain-related events in the CNS can be organized within this new theoretical frame. For example, the rapidly growing data base on the effects of expectation on S1, ACC, and insular cortices can be better understood within this physics-based reasoning. Specifically, if participants’ expectation of pain relief during placebo analgesia causes them to focus attention on experiential aspects of pain relief, then attention density associated with pain relief will be increased. This can activate the quantum Zeno effect that will dynamically stabilize the patterns of neural activity arising in circuitry associated with pain relief. This can then call Hebbian mechanisms into play, which can lead to neuroplastic changes in the brain’s response to nociceptive stimuli, or even the mental events associated with those stimuli. Findings such as those of Price (2000) Rainville et al. (1992, 2005) and Moulton et al. (2005) could be understood and explained to a significant extent by this mechanism. It is our hope that investigators in a variety of related fields will find
this kind of reasoning conducive to the creation and pursuit of a variety of new and clinically useful experimental paradigms.
References Adolphs, R., Damasio, H., Tranel, D., Cooper, G. and Damasio, A.R. (2000) A role for somatosensory cortices in the visual recognition of emotion as revealed by three-dimensional lesion mapping. J. Neurosci., 20: 2683–2690. Bantick, S.J., Wise, R.G., Ploghaus, A., Clare, S., Smith, S.M. and Tracey, I. (2002) Imaging how attention modulates pain in humans using functional MRI. Brain, 125: 310–319. Basbaum, A.I. and Fields, H.L. (1978) Endogenous pain control mechanisms: review and hypothesis. Ann. Neurol., 4: 451–462. Bernard, J.F. and Besson, J.M. (1990) The spino(trigemino)pontoamygdaloid pathway: electrophysiological evidence for an involvement in pain processes. J. Neurophysiol., 63: 473–490. Bernard, J.F., Peschanski, M. and Besson, J.M. (1989) A possible spino (trigemino)-ponto–amygdaloid pathway for pain. Neurosci. Lett., 100: 83–88. Berthier, M., Starkstein, S. and Leiquarda, R. (1988) Asymbolia for pain: a sensory-limbic disconnection syndrome. Ann. Neurol., 24: 41–49. Brooks, J.C., Nurmikko, T.J., Bimson, W.E., Singh, K.D. and Roberts, N. (2002) fMRI of thermal pain: effects of stimulus laterality and attention. Neuroimage, 15: 293–301. Burstein, R., Cliffer, K.D. and Giesler, G.J. (1987) Direct somatosensory projections from the spinal cord to the hypothalamus and telencephalon. J. Neurosci., 7: 4159–4164. Bushnell, M.C., Duncan, G.H., Dubner, R. and He, L.F. (1984) Activity of trigeminothalamic neurons in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. J. Neurophysiol., 52: 170–187. Bushnell, M.C., Duncan, G.H., Hofbauer, R.K., Ha, B., Chen, J. and Carrier, B. (1999) Pain perception: is there a role for primary somatosensory cortex? Proc. Natl. Acad. Sci. USA,, 96: 7705–7709. Casey, K.L. and Morrow, T.J. (1983) Ventral posterior thalamic neurons differentially responsive to noxious stimulation in the awake monkey. Science, 221: 675–677. Coghill, R.C., Mayer, D.J. and Price, D.D. (1993) Wide dynamic range but not nociceptive-specific neurons encode multidimensional features of prolonged repetitive heat pain. J. Neurophysiol., 69: 703–716. Cohen, R.A. (1993) The Neuropsychology of Attention. Plenum Press, New York. Chung, J.M., Lee, K.H. and Surmeier, D.J. (1986) Response characteristics of neurons in the ventral posterior lateral nucleus of the monkey thalamus. J. Neurophysiol., 56: 370–390. Dado, R.J., Katter, J.T. and Giesler Jr., G.J. (1994) Spinothalamic and spinohypothalamic tract neurons in the cervical enlargement in rats. II. Responses to innocuous and
350 noxious mechanical thermal stimuli. J. Neurophysiol., 71: 981–1002. Damasio, A. (1994) Descartes Error. Avon Books, New York. Dong, W.K., Chudler, E.H. and Sugiyama, K. (1994) Somatosensory, multisensory, and task-related neurons in cortical area 7b (PF) of anesthetized monkeys. J. Neurophysiol., 72: 542–564. Dong, W.K., Roberts, V.J. and Hayashi, T. (1996) Behavioral outcome of injury in area 7B cortex of monkeys. Pain, 64: 579–587. Dong, W.K., Salonen, L.D., Kawakami, Y., et al. (1989) Nociceptive responses of trigeminal neurons in SII-7b cortex of awake monkeys. Brain Res., 484: 314–324. Dubner, R., Bushnell, M.C., Duncan, G.H. (1986) Sensorydiscriminative capacities of nociceptive pathways and their modulation by behavior, In: Yaksh T.L. (Ed.), Spinal Afferent Processing. Plenum, New York, pp. 331–321. Echols, D.H. and Cogclough, J.A. (1947) Abolition of painful phantom limb by resection of the sensory cortex. JAMA, 134: 1476–1477. Fields, H.L. and Price, D.D. (1997) Toward a neurobiology of placebo analgesia. In: Harrington, A. (Ed.), The Placebo Effect. Harvard University Press, Boston, pp. 93–116. Foltz, E.L. and White, L.E. (1962) Pain ‘‘relief’’ by frontal cingulumotomy. J. Neurosurg., 19: 89–100. Friedman, D.P., Murray, E.A. and O’Neill, T. (1986) Cortical connections of the somatosensory fields of the lateral sulcus of macaques: evidence for a corticolimbic pathway for touch. J. Comp. Neurol., 252: 323–347. Giesler Jr., G.J., Yezierski, R.P., Gerhart, K.D. and Willis, W.D. (1981) Spinothalamic tract neurons that project to medial and/or lateral thalamic nuclei: evidence for a physiologically novel population of spinal cord neurons. J. Neurophysiol., 46: 1285–1308. Greenspan, J.D., Joy, S.E. and McGillis, S.L. (1997) A longitudinal study of somesthetic perceptual discorders in an individual with a unilateral thalamic lesion. Pain, 72: 13–25. Hardy, J.D., Wolff, H.G. and Goodell, H. (1952) Pain Sensations and Reactions. Williams and Wilkins, Baltimore, pp. 322–333. Harkins, S.W., Price, D.D. and Braith, J. (1989) Effects of extraversion and neuroticism on experimental pain, clinical pain, and illness behavior. Pain, 36: 209–218. Hayes, R.L., Dubner, R. and Hoffman, D.S. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. II. Behavioral modulation of responses to thermal and mechanical stimuli. J. Neurophysiol., 46: 428–443. Head, H. and Holmes, G. (1911) Sensory disturbances from cerebral lesions. Brain, 34: 102–154. Hebb, D.O. (1955) Drives and the C. N.S. (conceptual nervous system). Psych. Rev., 62: 243–254. Hicks, T.P., Benedek, G. and Thurlow, G.A. (1988) Modality specificity of neuronal responses within the cat’s insula. J. Neurophysiol., 60: 422–437. Hofbauer, R.K., Rainville, P., Duncan, G.H. and Bushnell, M.C. (2001) Cognitive modulation of pain sensation alters
activity in human cerebral cortex. J. Neurophysiol., 86: 402–411. Hoffman, D.S., Dubner, R., Hayes, R.L. and Medlin, T.P. (1981) Neuronal activity in medullary dorsal horn of awake monkeys trained in a thermal discrimination task. I. Responses to innocuous and noxious thermal stimuli. J. Neurophysiol., 46: 409–427. Hoffman, H.G., Richards, T.L., Coda, B., Bills, A.R., Blough, D., Richards, A.L. and Sharar, S.R. (2004) Modulation of thermal pain-related brain activity with virtual reality: evidence from fMRI. Neuroreport, 15: 1245–1248. Isomura, Y. and Takada, M. (2004) Neural mechanisms of versatile functions in primate cingulate cortex. Rev. Neurosci., 15: 279–291. Iwata, K., Kamo, H., Ogawa, A., Tsuboi, Y., Noma, N., Mitsuhashi, Y., Taira, M., Kenshalo Jr., D.R., Giesler, G.J., Leonard, R.B. and Willis, W.D. (1980) Responses of neurons in primate ventroposteriolateral nucleus to noxious stimuli. J. Neurophysiol., 43: 1594–1614. Kenshalo Jr., D.R., Chudler, E.H., Anton, F. and Dubner, R. (1988) SI nociceptive neurons participate in the encoding process by which monkeys perceive the intensity of noxious thermal stimulation. Brain Res., 454: 378–382. Kenshalo Jr., D.R. and Isensee, O. (1983) Responses of SI cortical neurons to noxious stimuli. J. Neurophysiol., 50: 1479–1496. Kenshalo, D.R., Thomas, D.A. and Dubner, R. (1989) Somatosensory cortical lesions change the monkeys’ reaction to noxious stimulation. J. Dent. Res., 68: 897. Kosslyn, S.M., Cacioppo, R.J., Davidson, R.J., Hugdahl, K., Lovallo, W.R., Spiegel, D. and Rose, R. (2002) Bridging psychology and biology. The analysis of individuals in groups. Am. Psychol., 57: 341–351. Koyama, T., Kato, K., Tanaka, Y.Z. and Mikami, A. (2001) Anterior cingulate activity during pain-avoidance and reward tasks. Neurosci. Res., 39: 421–430. Koyama, T., McHaffie, J.G., Laurienti, P.J. and Coghill, R.C. (2005) The subjective experience of pain: where expectations become reality. Proc. Natl. Acad. Sci., 102(36): 12950–12955. Lackner, J.M., Jaccard, J. and Blanchard, E.B. (2005) Testing the sequential model of pain processing in irritable bowel syndrome: a structural equation modeling analysis. Eur. J. Pain, 9: 207–218. Lenz, F.A., Dostrovsky, J.O. and Tasker, R.R. (1988) Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses. J. Neurophysiol., 59: 299–316. Lenz, F.A., Ohara, S., Gracely, R.H. and Dougherty, P.M. (2004) Pain encoding in the human forebrain: binary and analogue exteroceptive channels. J. Neurosci., 24: 6540–6544. Lenz, F.A., Seike, M., Lin, Y.C., Baker, F.H., Rowland, L.H. and Gracely, R.H. (1993) Neurons in the area of human thalamic nucleus ventralis caudalis respond to painful heat stimuli. J. Neurophysiol., 70: 200–212. Lewin, W. and Phillips, C.G. (1952) Observations on partial removal of the postcentral gyrus for pain. J. Neurol. Neurosurg. Psychiatr., 15: 143–147.
351 Maixner, W., Dubner, R., Kenshalo Jr., D.R. and Oliveras, J.L. (1989) J. Neurophysiol., 62: 437–449. Melzack, R. and Casey, K.L. (1968) Sensory, motivational, and central control determinants of pain. In: Kenshalo, D.R. (Ed.), The Skin Senses. C. C. Thomas, Springfield, pp. 423–439. Miron, D., Duncan, G.H. and Bushnell, M.C. (1989) Effects of attention on the intensity and unpleasantness of thermal pain. Pain, 39: 345–352. Moulton, E.A., Keaser, M.L., Gullapalli, R.P. and Greenspan, J.D. (2005) Regional intensive and temporal patterns of functional MRI activation distinguishing noxious and innocuous contact heat. J. Neurophysiol., 93: 2183–2193. Nakamura, K., Roesch, M.R. and Olson, C.R. (2005) Neuronal activity in macaque SEF and ACC during performance tasks involving conflict. J. Neurophysiol., 93: 884–908. Neal, J.W., Pearson, R.C. and Powell, T.P. (1987) The cortico–cortico connections of area 7b, PF, in the parietal lobe of the monkey. Brain Res., 419: 341–346. Neal, J.W., Pearson, R.C. and Powell, T.P. (1990) The ipsilateral cortico–cortico connections of area 7 with the frontal lobe in the monkey. Brain Res., 509: 31–40. Olausson, H., Ha, B., Duncan, G.H., Morin, C., Ptito, A., Ptito, M., Marchand, S. and Bushnell, M.C. (2001) Cortical activation by tactile and painful stimuli in hemispherectomized patients. Brain, 124: 916–927. Ostrowsky, K., Magnin, M., Ryvlin, L., Isnard, J., Guenot, M. and Mauguiere, F. (2002) Representation of pain and somatic sensation in the human insula: a study of responses to direct electrical cortical stimulation. Cerebral Cortex, 12: 376–385. Penfield, W. and Boldrey, E. (1937) Somatic motor and sensory representation in the cerebral cortex of man. Brain, 60: 116–126. Petrovic, P., Petersson, K.M., Ghatan, P.H., Stone-Elander, S. and Ingvar, M. (2000) Pain-related cerebral activation is altered by a distracting cognitive task. Pain, 85: 19–30. Peyron, R., Garcia-Larrea, L., Gregoire, M.C., Costes, N., Convers, P., Lavenne, F., Mauguire, F., Michel, D. and Laurent, B. (1999) Haemodynamic brain responses to acute pain in humans: sensory and attentional networks. Brain, 122: 1765–1780. Ploner, M., Freund, H.J. and Schnitzler, A. (1999) Pain affect without pain sensation in a patient with a postcentral lesion. Pain, 81: 211–214. Price, D.D. (1999) Psychological Mechanisms of Pain and Analgesia. I.A.S.P. Press, Seattle. Price, D.D. (2000) Psychological and neural mechanisms of the affective dimension of pain. Science, 288: 1769–1772. Price, D.D. and Barrell, J.J. (1984) Some general laws of human emotion: interrelationships between intensities of desire, expectations, and emotional feeling. J. Person., 52: 389–409. Price, D.D., Barrell, J.E. and Barrell, J.J. (1985) A quantitative—experiential analysis of human emotions. Motiv. Emotion,, 9: 19–38. Price, D.D. and Barrell, J.J. (1980) An experiential approach with quantitative methods: a research paradigm. J. Humanistic Psych., 20: 75–95.
Price, D.D. and Dubner, R. (1977) Neurons that subserve the sensory discriminative aspects of pain. Pain, 3: 307–388. Price, D.D., Dubner, R. and Hu, J. (1976) Trigeminothalamic neurons in nucleus caudalis responsive to tactile, thermal, and nociceptive stimulation of the monkey’s face. J. Neurophysiol., 39: 936–953. Price, D.D., Greenspan, J. and Dubner, R. (2003) Neurons involved in exteroreceptive functions of pain. Pain, 106: 215–219. Price, D.D. and Harkins, S.W. (1992) The affective-motivational dimension of pain: a two stage model. APS J., 1: 229–239. Price, D.D., Harkins, S.W. and Baker, C. (1987) Sensory–affective relationships among different types of clinical and experimental pain. Pain, 28(3): 291–299. Price, D.D., Hayes, R.L., Ruda, M.A. and Dubner, R. (1978) Spatial and temporal transformations of input to spinothalamic tract neurons and their relation to somatic sensation. J. Neurophysiol., 41: 933–947. Price, D.D., Hayashi, H., Dubner, R. and Ruda, M.A. (1979) Functional relationships between neurons of laminae I, II, and III of the primate dorsal horn. J. Neurophysiol., 42: 1590–1608. Price, D.D. and Verne, G.N. (2002) Does the spinothalamic tract to VPL and somatosensory cortex have roles in both pain sensation and pain-related emotions? J. Pain,, 3: 105–108. Price, D.D., von der Gruen, A., Miller, J., Rafii, A. and Price, C. (1985) A psychophysical analysis of morphine analgesia. Pain, 22: 320–330. Rainville, P., Bao, Q.V. and Chetrien, P. (2005) Pain-related emotions modulate experimental pain perception and autonomic responses. Pain, 118: 306–318. Rainville, P., Duncan, G.H., Price, D.D., Carrier, B. and Bushnell, M.C. (1997) Pain affect encoded in human anterior cingulate but not somatosensory cortex. Science, 277: 968–971. Rainville, P., Feine, J.S., Bushnell, M.C. and Duncan, G.H. (1992) A psychophysical comparison of sensory and affective responses to four modalities of experimental pain. Somatosens. Motor Res., 9: 265–277. Riley, J.L., Wade, J.B., Myers, C.D., Sheffield, D., Papas, R. and Price, D.D. (2002) Ethnic differences in the experience of chronic pain. Pain, 100: 291–298. Riley III., J.R., Wade, J.B., Robinson, M.E. and Price, D.D. (2000) The stages of pain processing across the adult lifespan. J. Pain, 1: 162–170. Robinson, M.E., Craggs, J., Verne, G.N., Perlstein, W.R. and Price, D.D. (2006) Placebo analgesia is accompanied by active afferent inhibition. J. Pain, 7(4): 887. Rubins, J.L. and Friedman, E.D. (1948) Asymbolia for pain. Neurol. Psychiatr., 60: 554–560. Schwartz, J.M. and Begley, S. (2002) The Mind and the Brain: Neuroplasticity and the Power of Mental Force. Harper Collins, New York. Schwartz, J.M., Stapp, H.P. and Beauregard, M. (2005) Quantum physics in neuroscience and psychology: a neurophysical
352 model of mind-brain interaction. Phil. Trans. R. Soc., 360: 1309–1327. Seminowicz, D.A., Mikulis, D.J. and Davis, K.D. (2004) Cognitive modulation of pain-related brain responses depends on behavioral strategy. Pain, 112: 48–58. Stein, B.E. and Meredith, M.A. (1993) The Merging of the Senses. MIT Press, Cambridge, MA. Stein, B.E., Price, D.D. and Gazzaniga, M. (1989) Pain perception in a man with total corpus callosum transection. Pain, 38: 51–55. Tracey, I., Ploghaus, A., Gati, J.S., Clare, S., Smith, S., Menon, R.S. and Matthews, P.M. (2002) Imaging attentional modulation of pain in the periaqueductal gray in humans. J. Neurosci., 22: 2748–2752. Valet, M., Sprenger, T., Boecker, H., Wiloch, F., Rummeny, E., Conrad, B., Erhard, P. and Tolle, T.R. (2004) Distraction modulates connectivity of the cingulo-frontal cortex and the midbrain during pain—an fMRI analysis. Pain, 109: 399–408. Vase, L., Robinson, M.E., Verne, G.N. and Price, D.D. (2003) The contribution of suggestion, desire, and expectation to placebo effects in irritable bowel syndrome patients. Pain, 105: 17–25. Vase, L., Robinson, M.E., Verne, G.N. and Price, D.D. (2005) Increased placebo analgesia over time in irritable bowel
syndrome (IBS) patients is associated with desire and expectation but not endogenous opioid mechanisms. Pain, 115: 338–347. Velmans, M. (2000) Understanding Consciousness. Routledge, London. Wade, J.B., Dougherty, L.M., Archer, C.R. and Price, D.D. (1996) Assessing the stages of pain processing: a multivariate approach. Pain, 68: 157–168. Wager, T.D., Rilling, J.K., Smith, E.E., Sokolik, A., Casey, K.L., Davidson, R.J., Kosslyn, S.M., Rose, R.M. and Cohen, J.D. (2004) Placebo-induced changes in fMRI in the anticipation and experience of pain. Science, 303: 1162–1167. Weinstein, E.A., Kahn, R.L. and Slate, W.H. (1955) Withdrawal, inattention, and pain asymbolia. Arch. Neurol. Psychiatr., 74: 235–240. White, J.C. and Sweet, W.H. (1966) Pain and the Neurosurgeon. Thomas, Springfield, Illinois. Willis Jr., W.D. (1985) The Pain System. Karger, New York. Willis, W.D., Zhang, X., Honda, C.N. and Giesler, G.J. (2001) Projections from the marginal zone and deep dorsal horn to the ventrobasal nuclei of the primate thalamus. Pain, 92: 267–276. Willis, W.D., Zhang, X., Honda, C.N. and Giesler, G.J. (2003) A critical review of the role of the proposed VMpo nucleus in pain. J. Pain,, 3: 79–94.
Plate 20.2. Schematic of ascending pathways, subcortical structures, and cerebral cortical structures involved in processing pain. Symbols are those defined and referred to within text.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 21
Plasticity of pain-related neuronal activity in the human thalamus W.S. Anderson1, S. O’Hara1, H.C. Lawson1, R.-D. Treede2 and F.A. Lenz1, 1 Department of Neurosurgery, Johns Hopkins University, Baltimore, MD, USA Institut fu¨r Physiologie und Pathophysiologie, Johannes Gutenberg-Universita¨t, Mainz, Germany
2
Abstract: Strokes and other forms of injury to the central nervous system cause changes in function because of the injuries themselves and indirectly because injuries cause expression of neural plasticity. Studies in humans undergoing neurosurgical procedures for implantation of electrodes for deep brain stimulation and for making lesions in the brain have contributed understanding of both normal and abnormal functions of the somatic sensory system. This chapter will specifically discuss the reorganization of the ventral caudal (Vc) sensory nucleus of the thalamus that occurs in connection with pain conditions after strokes and spinal cord injuries. It is shown that pain is associated with expression of neural plasticity that alters maps of noxious and innocuous stimulation in the thalamus and affect processing of sensory information. Results from studies of neural activity in the thalamus in humans will be compared with results from animal studies. Keywords: amputation; dystonia; neurophysiology; thalamus; sensory reorganization; single neuron recordings. PF maps indicate the image of the body contained in thalamo-cortical assemblies. Many studies of cortical plasticity secondary to injuries of the nervous system have been published (Kaas, 1991; Recanzone et al., 1992; Nudo et al., 1996). Although the reorganization of cortical maps (Jenkins et al., 1990) may reflect changes in thalamic organization (Pons et al., 1991), Only a few studies have been published of thalamic plasticity in response to injuries to the nervous system (Rasmusson, 1996a, b; Willis et al., 2001). In this chapter we review literature regarding human thalamic plasticity as studied during stereotactic procedures for the treatment of chronic pain or movement disorders. Chronic pain is a pain that occurs daily over a 6-month period. Intractable chronic pain and movement disorders can be treated by implantation of stimulating electrodes in or adjacent to the principal sensory
Introduction Studies of plasticity of the somatosensory system in monkeys have focused on maps of cortical function determined by examining neuronal receptive fields – the part of the body where peripheral stimulation evokes a neuronal response (receptive field, RF) (Kaas, 1991). Studies in humans can explore both maps determined from RFs and from sensations evoked by microstimulation using electrical impulses (current strength at mA current levels) measured in terms of the location and quality of sensations evoked by stimulation of the brain (projected field – PF) (Lenz et al., 1994c, 1998a). While RF maps are a reflection of the organization of inputs to the central nervous system, Corresponding author. Tel.: +1 410 955 2257; Fax: +1 410 614 9877; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57021-9
353
354
nucleus of the thalamus (ventral caudal – Vc) (deep brain stimulation) (Hosobuchi, 1986; Young et al., 1985) or by thalamotomy (Cardoso et al., 1995; Tasker et al., 1988). The borders of Vc were explored in both patient groups. In patients with chronic pain, Vc was the target for implantation of stimulating electrodes. In patients with movement disorders Vc was explored as a predictor of the location of the cerebellar (Ventralis intermedius – Vim) and pallidal receiving nuclei of the thalamus (Ventral oral – Vo composed of anterior and posterior subnuclei Voa and Vop). These nuclei were then lesioned or stimulated for treatment of movement disorders (Vitek and Lenz, 1998; Zirh et al., 1999). Below we will discuss the characteristics of cells in the Vc region in the thalamus and the importance of this region of the thalamus for pain and movement disorders.
The Vc region of the thalamus The region of Vc is divided into a cutaneous core region and a posterior–inferior region. In the core the majority of cells responded to innocuous, cutaneous, mechanical stimuli. The core region probably corresponds to Vc proper (Lenz et al., 1993b, 1994a; Willis et al., 2001), which is the human nucleus analogous to the ventral posterior (VP) nucleus in the monkey (Hirai and Jones, 1989). The region posterior and inferior to the core may correspond to the subnuclei of Vc, the ventral caudal portae (Vcpor posterior to Vc) and ventral caudal parvocellular nucleus (Vcpc inferior to Vc). In addition, the posterior–inferior Vc region may include the posterior nucleus, magnocellular medial geniculate (Lenz et al., 1993b; Mehler 1966) and ventral medial posterior (Vmpo) nucleus (Craig et al., 1994; Graziano and Jones, 2004; Willis et al., 2001)). The thalamic activity was characterized by a baseline firing, the neuronal RF, and the projected field for microstimulation at different thalamic sites. Below we review evidence that activity in the subnuclei in the Vc region and adjacent nuclei is different in patients with pain following an injury to the nervous system, and in patients with movement disorders.
Evidence that the region of Vc is involved in pain processing Involvement of a forebrain structure in the mechanism of pain can be demonstrated by its connections with pain-signaling pathways, by neurons in the structure responding to painful stimuli, by decreases in pain following lesions of the structure, and by microstimulation-evoked pain. Lesion studies have demonstrated that the spinothalamic tract (STT) has a dense termination in the Vc region of the thalamus (Bowsher, 1957; Mehler, 1962, 1966; Mehler et al., 1960), and in regions that are posterior to the Vc in the magnocellular medial geniculate nucleus (Mehler, 1962, 1969), the limitans and Vc portae nuclei, (Mehler, 1966) and in the Vcpc (Mehler, 1966). The human Vmpo (Blomqvist et al., 2000) may or may not be a terminus for the part of the STT originating from neurons in lamina I of the spinal cord (Craig et al., 1994; Graziano and Jones, 2004; Willis, et al., 2001).In humans, cells in Vc respond differentially or selectively to painful stimuli (Lee et al., 1999; Lenz et al., 1994b) as well as to innocuous (cool and mechanical) stimuli (Lenz and Dougherty, 1998a). Fig. 1 shows an example of a cell in Vc responding to noxious stimuli of the thermal (D) and mechanical submodalities (B and C). Cells in the posterior–inferior region can have selective response to noxious heat (Lenz et al., 1993a) or cold stimuli (Davis et al., 1999). These observations are consistent with those of monkey studies that have demonstrated that cells within VP (Casey and Morrow, 1983) (Gautron and Guilbaud, 1982; Bushnell et al., 1993; Apkarian and Shi, 1994) (Bushnell and Duncan, 1987) or posterior–inferior to VP respond to noxious stimuli (Apkarian et al., 1991; Apkarian and Shi, 1994; Casey, 1966). Lesioning and stimulation studies suggest that the cells in the Vc region of the thalamus signal sensory aspects of pain. Stimulation of cells within or posterior–inferior to Vc can evoke sensations of pain, warm, or cold (Dostrovsky et al., 1991) (Lenz et al., 1993b; Davis et al., 1999; Ohara and Lenz, 2003). Injection of local anesthetic into monkey VP (corresponding to human Vc (Hirai and Jones, 1989)) interferes with the animal’s ability to discriminate both innocuous and noxious temperature stimuli (Duncan et al., 1993).
355
Fig. 1. Activity of a cell (061093) in Vc responding to painful mechanical and thermal stimuli. (A) Location of the cell (arrow) relative to the positions of trajectories, nuclear boundaries, and other recorded cells. The anterior commissure–posterior commissure (ACPC) line is indicated by the horizontal line, and the trajectories are shown by the oblique lines (left-anterior, up-dorsal). Nuclear location was approximated from the position of the ACPC line. Lateral location of the cell (in millimeters) is indicated above each map. Trajectories have been shifted along the ACPC line until the most posterior cell with a cutaneous RF is aligned with the posterior border of Vc. Since cells responding to innocuous sensory stimuli may be located posterior to Vc (Apkarian and Shi, 1994), this map represents a first approximation of nuclear location and dimensions. The locations of cells are indicated by ticks to the right of each trajectory. Cells with cutaneous RFs are indicated by long ticks, those without definable RFs by short ticks. Filled circles attached to the long ticks indicate that somatic sensory testing was carried out. The scale is as indicated. The shape of action potentials recorded at the beginning of the recording on this cell during application of the brush (upper) and at the end of the recording, during a 12 1C stimulus (lower). Data were collected from the up going stroke of the action potential by using a voltage threshold of 0.15 mV. The RF and PF for the natural, surface or deep, nonpainful, or tingling sensation evoked by micro-stimulation at the recording site (threshold – 15 mA) are also shown. B, response to the brush, LC (large arterial clip, nonpainful), MC (medium clip, sometimes painful), and SC (small clip, always painful). (C) the response of the neuron to progressive increases in pressure applied with a nonpenetrating towel clip, indicated by the number of steps. (D) responses to heat stimuli at 421, 451, and 48 1C. E, responses to cold stimuli at 121, 181, and 24 1C. The upper trace in each panel is a footswitch signal indicating the onset and duration of the stimulus in panels (B) and (C) and the thermode signal in panels (D) and (E). The scales for the axes for all histograms (bin width 100 ms) are indicated in each panel (adapted with permission from John Wiley and Sons, Inc. Lee et al. (1999) Responses of neurons in the region of human thalamic principal somatic sensory nucleus to mechanical and thermal stimuli graded into the painful range. J. Comp. Neurol., 410: 541–555).
356
Thus, the region of Vc receives input from pain signaling pathways, and contains cells that respond to noxious stimuli. Injections of local anesthetic disable the discrimination of pain and temperature, and stimulation of this region evokes the sensations of temperature and pain. Therefore, this region is an important forebrain structure mediating acute, and perhaps chronic pain sensations.
Reorganization of Vc after nervous system injury Spontaneous activity, RF and PF maps
Fig. 2. Map of receptive and projected fields for trajectories in the region of the Vc in a patient with spinal cord transection at T8. (A) A trajectory in the 15-mm lateral parasagittal plane (labeled Lat 15 mm) through the ‘region of Vc’, which represents the arm. (B) A trajectory 2 mm lateral to the first (Lat 17 mm). The upper panels in A and B show the position of the trajectories (labeled S3 in A, and S4 in B) relative to nuclear boundaries as predicted radiologically. The AC-PC line is indicated by the horizontal line in the panel while the trajectories are shown by the oblique lines. The locations of cells are indicated by ticks small lines to the right of, and at right angles to the trajectories. Cells with RFs are indicated by long ticks; those without are indicated by short ticks. Stimulation sites are shown to the left of the trajectory. Long ticks indicate a somatosensory response to stimulation, while short ticks indicate no response to microstimulation. Scales are as indicated. The lower panel in A and B shows paired figurines for sites as numbered in the middle panel. The figurine to the right indicates the RF; NR indicates that the cell had no RF. The figurine
Recordings from the thalamus in patients with pain secondary to injury of the central nervous system (central pain) suggest mechanisms of these syndromes. Studies of patients with central pain secondary to spinal cord transection have demonstrated plasticity of spontaneous activity, RFs and PFs (Lenz et al., 1994c). Many cells that represented the part of the body did not have RFs (Lenz et al., 1994c) so that the presence of RFs could not be used to define the location of Vc. Therefore, the region of Vc was defined as the cellular thalamic region where sensations were evoked at 25 mA or less. The ‘region of Vc’ in spinal patients can be divided into different areas according to RF and PF locations. Areas that were distant from the representation of the anesthetic part of the body were termed ‘spinal control’ areas while those that were adjacent to or included in the representation of the area of absolute sensory loss were termed ‘borderzone/anesthetic’ areas. The ‘region of Vc’ in patients with a movement disorder was termed the ‘control’ area. The terms ‘control,’ ‘spinal control,’ and ‘borderzone/anesthetic’ in parenthesis were used to refer to these parts of the thalamus throughout this chapter. to the left indicates the PF for microstimulation at that site while the number below the figurine indicates the threshold in mA. At all sites along both trajectories where sensations were evoked that sensation was described as tingling (adapted with permission from The American Physiological Society. Lenz et al. (1994c) Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J. Neurophys., 72: 1570–1587).
357
Fig. 2 shows the thalamic map in a patient with functional spinal cord transection at the T7 level leading to central pain (Weng et al., 2000, 2003). In the ‘control’ and ‘spinal control’ areas the locations of RFs and PFs were usually well matched. However, there was frequently a mismatch between the location of RFs and PFs (RF/PF mismatch) in ‘borderzone/anesthetic’ areas (Fig. 2B). For example, in panel C, there is a large representation of the chest and abdominal wall as judged by RFs while PFs are located in the lower extremity, below the spinal transection. This area is therefore regarded as a borderzone between the normal and the anesthetic region with a representation of the torso that is much larger than usual (Lenz et al., 1988a). ‘Borderzone/anesthetic’ areas of the thalamus often exhibit increased RF representations of the border of the anesthetic part of the body while PF representations show the anesthetic part of the body specifically. This leads to a mismatch between the RFs and PFs. Two sites were said to have a consistent RF if the RF of both sites included the same part of the body. The length of a trajectory with consistent RFs is the distance along the trajectory where each RF continues to include the same part of the body. Human thalamic somatotopy in Vc is arranged with tongue and intra-oral structures medially and leg laterally. Somatotopy is constant within sheets of cells in the parasagittal plane parallel to rows of trajectories during our exploration (see Fig. 2 and (Lenz et al., 1988b)). The width of these sheets increases with the size of the representation of a part of the body. Therefore, the maximal distance along a trajectory over which the RF or the PF stays consistent is longer for body parts with larger representations (Lenz et al., 1994c). The lengths of trajectories with a consistent RF in a particular part of the body were significantly longer in ‘borderzone/anesthetic’ than in ‘control’ or ‘spinal control’ areas. In ‘control’ and ‘spinal control’ areas the locations of RFs and PFs were usually well matched (Fig. 2, S3). In ‘borderzone/anesthetic’ areas of the thalamus, there was frequently a mismatch between the location of RFs and PFs (RF/PF mismatch). In ‘borderzone/anesthetic’ areas (Fig. 2, S4), RFs were often located on the border of the
anesthetic part of the body while PFs were referred to anesthetic parts of the body. The RF and PF are plotted against each other in Fig. 3, which shows the mismatch in the ‘borderzone/anesthetic’ regions but not in the ‘control’ regions. There were numerous PFs representing the anesthetic part of the body, suggesting that the perceptual image of that part of the body is unchanged. In contrast, the representation of peripheral inputs (RFs) changed dramatically to represent the borderzone of the anesthetic part of the body. Similar changes are seen in patients with amputations (Lenz et al., 1998a). Neurons with RFs adjacent to the area of sensory loss in amputation patients occupied a larger part of the thalamic homunculus (Lenz et al., 1998a) than found for the same part of the body in patients without sensory abnormality, i.e., patients with movement disorders (Lenz et al., 1988a, 1994c). This result is consistent with somatotopic reorganization of afferent inputs from the limb. Similarly, the large area over which PFs include the stump (cf (Lenz, et al. 1993b, 1994c)) suggest that there has been reorganization of the image of the limb embedded in the central nervous system (Jensen and Rasmussen, 1994). It has also been reported that phantom sensations can be evoked by stimulation of the region where stump RFs are located in the region of Vc (Davis et al., 1998). Thus, amputations and spinal cord injury show similar patterns of thalamic reorganization. Specifically, the reorganization/plasticity in the image of the body embedded in the thalamus is less than that for the representation of sensory inputs following loss of sensory input. Increased thalamic bursting activity has been observed in patients with pain secondary to spinal transection (Steriade et al., 1990). The highest rate of bursting occurs in cells that do not have peripheral receptive fields and that are located in the ‘borderzone/anesthetic’ region. These cells also have the lowest firing rates in the interval between bursts (primary event rate) (Lenz et al., 1994c). The low firing rates suggest that these cells have decreased tonic excitatory drive and are hyperpolarized, perhaps due to loss of excitatory input from the STT ( Eaton and Salt, 1990; Ericson et al., 1993; Blomqvist et al., 1996; Dougherty et al.,
358
1996). Therefore, the available evidence suggests that affected thalamic cells in patients with spinal transection are dominated by bursting consistent with membrane hyperpolarization (Steriade and Deschenes, 1984; Steriade and Llinas, 1988;
Steriade et al., 1990; Davis et al., 1998; Lenz et al. 1998b). Bursting activity is maximal in cells located in the region posterior and inferior to the core nucleus of Vc (Table 4 in ref (Lenz et al., 1994c)). This region is strongly associated with thermal and pain sensations (see above). Thus, increased thalamic bursting activity may be correlated with some aspects of the abnormal sensations that these patients experience (e.g., dysesthesia or allodynia). Since the area to which pain is referred and the area of sensory loss overlap (Lenz et al., 1994c), thalamic bursting might be related to sensory loss, rather than to pain. This is consistent with a report of increased bursts occurring in old world monkeys with thoracic anterolateral cordotomies, which interrupt the STT (Weng et al., 2000). Some of these animals showed increased responsiveness to electrocutaneous stimuli and perhaps indicating hyperalgesia associated with central pain (Vierck, 1991). The most pronounced changes in firing pattern were found in VP cells, which respond to both cutaneous brushing and painful compressive stimuli with activity that is not graded into the noxious range (multireceptive cells). In comparison with normal controls, multireceptive cells in the monkeys with cordotomies showed significant increases in the number of bursts that occurred during spontaneous and evoked activities. These findings have been questioned in a recent study of thalamic bursting in patients with chronic pain of multiple different diagnoses (Radhakrishnan et al., 1999). In this report the number of bursting cells per trajectory was similar in patients with movement disorders (controls) and in patients with
Fig. 3. Mismatch of RF and PF to microstimulation in patients with spinal cord injuries and in controls with movement disorders. Both RFs and PFs are numbered by their representation within Vc from tongue medial to leg lateral, as in the inset to panel (A). The RFs are plotted against the location of the part of the PF closest to the RF. Results are shown separately for ‘control’ (A), control spinal (B), and ‘borderzone/anesthetic’ (C), as labeled (adapted with permission from The American Physiological Society. Lenz et al. (1994c) Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J. Neurophys., 72: 1570–1587).
359
chronic pain. However, the earlier report and this recent report are much different (Lenz et al., 1994c; Radhakrishnan et al., 1999) in terms of patient population (spinal cord injury vs mixed chronic pain), location of cells studied (Vc vs anterior and posterior to Vc), and burst indices (incidence of bursting cells/trajectory vs bursting parameters). Finally the criterion for identifying a burst was questionable since a preburst silent period was not required. Such criterion has not previously been adopted by studies of thalamic low-threshold spike bursts (Steriade et al., 1990; Lenz et al., 1994c; Jeanmonod et al., 1996; Sherman and Guillery, 2001). Clearly, the earlier study and the studies in the monkey both demonstrate that increased thalamic low threshold spike bursting occurs in the somatic sensory nucleus of the thalamus in patients with, and primate models of, central pain secondary to spinal cord injury (Lenz et al., 1994c).
Modalities of sensations evoked by thalamic stimulation in relation to symptoms of central pain Stimulation of a region of Vc is more likely to evoke pain in patients with chronic neuropathic pain than in patients without somatic sensory abnormalities, such as patients with movement disorders (Lenz et al., 1998b). The region of the VC was divided on the basis of projected fields into a region representing the part of the body where the patients experienced chronic pain (‘pain affected’) and a region representing the body parts that were not affected with the pain (‘pain unaffected’). The region of Vc was also divided into the core and posterior–inferior regions (see above) on anatomical grounds. The ‘control’ region was defined as Vc region in patients with movement disorders and no experience of chronic pain In both the core and posterior–inferior regions, the proportion of sites where threshold microstimulation-evoked pain was larger in pain-affected and unaffected areas than in ‘control’ areas (Fig. 4). The number of sites where thermal (warm or cold) sensations were evoked was correspondingly smaller, so that the total of pain plus thermal sites was not significantly different across all areas. Therefore, sites where stimulation-evoked pain in patients with
neuropathic pain may correspond to sites where thermal sensations were evoked by stimulation, in patients without somatic sensory abnormality. In the posterior–inferior region the number of sites where cold was evoked by stimulation decreased significantly while the number of sites where pain was evoked increased significantly. These findings are consistent with another study of the microstimulation-evoked pain in patients with central pain (Davis et al., 1996; Lenz et al., 1998b) Sensory abnormalities including increased sensitivity to painful stimuli (hyperalgesia) or pain evoked by stimuli, which are normally nonpainful (allodynia), were studied in patients by microstimulation of the Vc region of the thalamus. Microstimulation of the region of Vc in patients with hyperalgesia-evoked pain more frequently than similar stimulation in patients without hyperalgesia (Lenz et al., 1998b). Stimulation in the region of Vc that represented the part of the body where the patient experienced hyperalgesia-evoked pain more frequently than did stimulation in the region of the Vc that represented other parts of the body (Lenz et al., 1998b). Therefore, microstimulation-evoked pain and hyperalgesia may both be the result of stimulation of an intact sensory pathway, which does not normally mediate pain. A stimulus, which normally activates such a pathway, might produce pain rather than the nonpainful sensation normally evoked by the stimulus, which evokes activity in that pathway. This is consistent with our studies of sensory function in patients with poststroke central pain, which demonstrate that allodynia occurs in those sensory modalities that have intact sensory pathways (Greenspan et al., 2004). In this series of patients quantitative tactile sensory testing showed hypoesthesia to tactile stimuli in 60% (6/10) and normal tactile sensibility in 40% (4/10). Tactile allodynia to brushing occurred in 54% (7/13) of the cases and to von Frey hairs in one patient. The incidence of tactile allodynia was higher in patients with normal tactile thresholds (4/4) than in those with tactile hypoesthesia (1/5, po0.05, Fisher), demonstrating that tactile allodynia occurred in cases with spared tactile pathways more than those pathways which were presumably normal because of their normal tactile thresholds.
360
associated with the absence, not the presence, of cold allodynia (Po0.05, Binomial). The two patients with cold hypoesthesia and cold allodynia had modest cold sensitivity deficits (cold threshold approximately 10 1C below the adapting temperature) when compared with the other patients with hypoesthesia, six of whom had thresholds 20–30 1C below the adapting temperature. These two patients had unequivocal evidence of a stroke but because the cold hypoesthesia was symmetrical it was most likely not due to the stroke. The most extreme example of cold allodynia occurred in a patient who first detected the cold stimulus as painful cold at 30.2 1C, and who was unable to tolerate her hand in a water bath at 30 1C. In this case cold allodynia was evoked at temperatures cold enough to activate receptors in the cold pathway but not those of the pathway signaling thermal and mechanical noxious plus cold stimuli (Bartoschuk, 1993; Craig and Bushnell, 1994; Lenz and Dougherty, 1998b). These three cases suggest that normal or limited compromise of cold sensibility is associated with cold allodynia, consistent with the results of the thalamic microstimulation results.
Fig. 4. Percentages of pain, cold, and warm sensations evoked by stimulation in the core (panel C) and posterior–inferior areas separately (B) and combined (A). These percentages are shown for movement disorder patients (control) and for areas of thalamus representing the part of the body where the patient does (pain affected) or does not experience chronic pain (pain unaffected) (adapted with permission from John Wiley and Sons, Inc. Lenz et al. (1998b) Reorganization of sensory modalities evoked by microstimulation in region of the thalamic principal sensory nucleus in patients with pain due to nervous system injury. J. Comp. Neurol., 399: 125–138).
A similar pattern was found in quantitative sensory testing for cold. Most patients (11/13) exhibited hypoesthesia for the perception of cool stimuli, but a few of these (3/11) showed cold allodynia. Even with the assumption of liberal probability of occurrence of allodynia in the presence of cold hypoesthesia (P ¼ 0.5 or 0.75), our results demonstrated that cold hypoesthesia was significantly
Microstimulation-evoked sensations explain sensory abnormalities in patients with poststroke central pain In our patients with poststroke central pain, brush allodynia occurred in the presence of normal tactile thresholds, suggesting that the dorsal columnthalamic-cortical pathway was intact (Nathan et al., 1986) Therefore, the results in poststroke central pain suggest that brush-evoked allodynia involves input to the forebrain through a pathway including dorsal column-thalamic Vc – post central gyrus and parietal operculum (Jones, 1985; Lenz et al., 1988a; Van Buren and Borke, 1972). Activation of afferents known to project through the dorsal columns was associated with unpleasant dysesthesias only in stroke patients with poststroke dysesthesia, a variant of CPSP (Triggs and Beric, 1994). This suggests that poststroke dysesthesias result from transmission of activity through the dorsal columns.
361
Tactile sensibility was measured by von Frey hairs and by a camel hair brush, both of which are largely mediated through the dorsal column pathway (Lee et al., 1999; Lenz et al., 1988a; Mountcastle, 1984), although some STT cells respond to both of these stimuli (Lee et al., 1999; Lenz et al., 1994b; Willis et al., 1974; Willis and Coggeshall, 1991). Sensory testing in patients with lesions of the dorsal columns reveals mild deficits in tactile sensation, while lesions of the STT (sparing the dorsal columns) were associated with no deficit in tactile sensation (Nathan et al., 1986). Therefore, the reduced tactile thresholds in these patients are likely due to decreased transmission of stimuli through the dorsal column-thalamic-cortical pathway.The tactile allodynia observed here could be due to disinhibition of the thalamic nucleus Vc, secondary to loss of corticothalamic inputs to Vc from the insula (Sherman and Guillery, 2001). Layer 6 pyramidal cells in insular cortex send axons expressing excitatory amino acids to the monkey and cat VP (Clasca et al., 1997; Mesulam and Mufson, 1985; Mufson and Mesulam, 1984), corresponding to human Vc (Hirai and Jones, 1989). Insular corticothalamic fibers terminate on thalamo-cortical relay cells and on inhibitory interneurons either intrinsic to VP or in the reticular nucleus of thalamus (Sherman and Guillery, 2001). Loss of corticothalamic input from primary and secondary somatic sensory cortex can lead to disinhibition, as measured by an increased response to tactile stimuli or the increased RF size (Burchfiel and Duffy, 1974; Ergenzinger et al., 1998; Sherman and Guillery, 2001) cf (Ghosh et al., 1992; Yuan et al., 1986)). Therefore, loss of corticothalamic input from nonprimary somatic sensory cortex can disinhibit VP, corresponding to human Vc (Hirai and Jones, 1989). The presence of allodynia in patients with relatively intact cold pathways is consistent with studies of the STT and related centers in the brain. Patients with central pain uniformly show deficits of STT associated pain and temperature sensibility (Beric et al., 1988; Boivie et al., 1989; Greenspan et al., 2004; Vestergaard et al., 1995). Stimulation of the STT evokes pain in patients with central pain (Tasker, 1982), but evokes nonpainful thermal sensations in patients without neuropathic
pain (Tasker, 1988). Anterolateral cordotomy relieves pain in a much greater proportion of patients with somatic pain than those without neuropathic pain (Sweet et al., 1994; Tasker et al., 1980). The failure of cordotomy to relieve neuropathic pain might be anticipated from the occurrence of central pain in patients with impaired function of the STT (Cassinari and Pagni, 1969)(Andersen et al., 1995; Boivie et al., 1989). In sum, these results suggest that the generator for pain and cold allodynia in patients with central pain is the terminus of the STT. These results imply a model in which sparing of a submodality by lesions causing central pain is associated with the occurrence of allodynia in that submodality. This model is built on a broad base of evidence including sensory testing studies in patients with poststroke central pain, lesion studies, stimulation, and neuronal recordings in the human central nervous system. The studies demonstrate that expression of neural plasticity from injury to the nervous system is related to different sensory pathways in patients with chronic pain. In patients with poststroke central pain the loss of some STT sensory pathways leads to chronic pain while the preservation of an ascending sensory pathway leads to allodynia or hyperalgesia in the modality subserved by that pathway. Thus forebrain sensory reorganization produced by nervous system injury directly contributes to the resulting chronic pain and sensory abnormalities.
Acknowledgments Supported by grants to FAL from the Eli Lilly Corporation and the National Institutes of Health (NS28598, K08-NS1384, P01 NS32386-Project 1, NS38493, NS40059).
References Andersen, P., Vestergaard, K., Ingeman-Nielsen, M. and Jensen, T.S. (1995) Incidence of poststroke central pain. Pain, 61: 187–193. Apkarian, A.V. and Shi, T. (1994) Squirrel monkey lateral thalamus. I. somatic nociresponsive neurons and their relation to spinothalamic terminals. J. Neurosci., 14: 6779–6795.
362 Apkarian, A.V., Shi, T., Stevens, R.T., Kniffki, K.-D. and Hodge, C.J. (1991) Properties of nociceptive neurons in the lateral thalamus of the squirrel monkey. Soc. Neurosci. Abstr., 17: 838. Bartoschuk, L.M. (1993) The biological basis of food perception and acceptance. Food Qual. Pref., 4: 21–32. Beric, A., Dimitrijevic, M.R. and Lindblom, U. (1988) Central dysesthesia syndrome in spinal cord injury patients. Pain, 34: 109–116. Blomqvist, A., Ericson, A.C., Craig, A.D. and Broman, J. (1996) Evidence for glutamate as a neurotransmitter in spinothalamic tract terminals in the posterior region of owl monkeys. Exp. Brain Res., 108: 33–44. Blomqvist, A., Zhang, E.T. and Craig, A.D. (2000) Cytoarchitectonic and immunohistochemical characterization of a specific pain and temperature relay, the posterior portion of the ventral medial nucleus, in the human thalamus. Brain, 123: 601–619. Boivie, J., Leijon, G. and Johansson, I. (1989) Central poststroke pain–a study of the mechanisms through analyses of the sensory abnormalities. Pain, 37: 173–185. Bowsher, D. (1957) Termination of the central pain pathway in man: the conscious appreciation of pain. Brain, 80: 606–620. Burchfiel, J.E. and Duffy, F.H. (1974) Corticofugal influence upon cat thalamic ventrobasal complex. Brain Res., 70: 395–411. Bushnell, M.C. and Duncan, G.H. (1987) Mechanical response properties of ventroposterior medial thalamic neurons in the alert monkey. Exp. Brain Res., 67: 603–614. Bushnell, M.C., Duncan, G.H. and Tremblay, N. (1993) Thalamic VPM nucleus in the behaving monkey. I. Multimodal and discriminative properties of thermosensitive neurons. J. Neurophysiol., 69(3): 739–752. Cardoso, F., Jankovic, J., Grossman, R.G. and Hamilton, W. (1995) Outcome after stereotactic thalamotomy for dystonia and hemiballismus. Neurosurg., 36: 501–508. Casey, K.L. (1966) Unit analysis of nociceptive mechanisms in the thalamus of the awake squirrel monkey. J. Neurophysiol., 29: 727–750. Casey, K.L. and Morrow, T.J. (1983) Ventral posterior thalamic neurons differentially responsive to noxious stimulation of the awake monkey. Science, 221: 675–677. Cassinari, V. and Pagni, C.A. (1969) Central Pain. A neurosurgical survey Harvard University Press, Cambridge, MA. Clasca, F., Llamas, A. and Reinoso-Suarez, F. (1997) Insular cortex and neighboring fields in the cat: a redefinition based on cortical microarchitecture and connections with the thalamus. J. Comp Neurol., 384: 456–482. Craig, A.D. and Bushnell, M.C. (1994) The thermal grill illusion: unmasking the burn of cold pain. Science, 265: 252–255. Craig, A.D., Bushnell, M.C., Zhang, E.T. and Blomqvist, A. (1994) A thalamic nucleus specific for pain and temperature sensation. Nature, 372: 770–773. Davis, K.D., Kiss, Z.H., Luo, L., Tasker, R.R., Lozano, A.M. and Dostrovsky, J.O. (1998) Phantom sensations generated by thalamic microstimulation. Nature, 391: 385–387.
Davis, K.D., Kiss, Z.H.T., Tasker, R.R. and Dostrovsky, J.O. (1996) Thalamic stimulation-evoked sensations in chronic pain patients and nonpain (movement disorder) patients. J. Neurophysiol., 75: 1026–1037. Davis, K.D., Lozano, R.M., Manduch, M., Tasker, R.R., Kiss, Z.H. and Dostrovsky, J.O. (1999) Thalamic relay site for cold perception in humans. J. Neurophysiol., 81: 1970–1973. Dostrovsky, J.O., Wells, F.E.B. and Tasker, R.R. (1991) Pain evoked by stimulation in human thalamus. In: Sjigenaga, Y. (Ed.), International symposium on processing nociceptive information. Elsevier, Amsterdam, pp. 115–120. Dougherty, P.M., Li, Y.J., Lenz, F.A., Rowland, L. and Mittman, S. (1996) Evidence that excitatory amino acids mediate afferent input to the primate somatosensory thalamus. Brain Res., 278: 267–273. Duncan, G.H., Bushnell, M.C., Oliveras, J.L., Bastrash, N. and Tremblay, N. (1993) Thalamic VPM nucleus in the behaving monkey. III. effects of reversible inactivation by lidocaine on thermal and mechanical discrimination. J. Neurophysiol., 70: 2086–2096. Eaton, S.A. and Salt, T.E. (1990) Thalamic NMDA receptors and nociceptive sensory synaptic transmission. Neurosci. Lett., 110: 297–302. Ergenzinger, E.R., Glasier, M.M., Hahm, J.O. and Pons, T.P. (1998) Cortically induced thalamic plasticity in the primate somatosensory system. Nat. Neurosci., 1: 226–229. Ericson, A.C., Blomqvist, A., Craig, A.D., Ottersen, O.P. and Broman, J. (1993) Enrichment of Glutamate-like immunoreactivity in spinothalamic tract terminals in the nucleus submedius of cat. Soc. Neurosci. Abstr., 18: 832. Gautron, M. and Guilbaud, G. (1982) Somatic responses of ventrobasal thalamic neurones in polyarthritic rats. Brain Res., 237: 459–471. Ghosh, S., Turman, A.B., Vickery, R.M. and Rowe, M.J. (1992) Responses of cat ventroposterolateral thalamic neurons to vibrotactile stimulation of forelimb footpads. Exp. Brain Res., 92: 286–298. Graziano, A. and Jones, E.G. (2004) Widespread thalamic terminations of fibers arising in the superficial medullary dorsal horn of monkeys and their relation to calbindin immunoreactivity. The J. Neurosci., 24: 248–256. Greenspan, J.D., Ohara, S., Sarlani, E. and Lenz, F.A. (2004) Allodynia in patients with poststroke central pain (CPSP) studied by statistical quantitative sensory testing within individuals. Pain, 109: 357–366. Hirai, T. and Jones, E.G. (1989) A new parcellation of the human thalamus on the basis of histochemical staining. Brain Res. Rev., 14: 1–34. Hosobuchi, Y. (1986) Subcortical electrical stimulation for control of intractable pain in humans. Report of 122 cases (1970–1984). J. Neurosurg., 64: 543–553. Jeanmonod, D., Magnin, M. and Morel, A. (1996) Low-threshold calcium spike bursts in the human thalamus. Common physiopathology for sensory, motor and limbic positive symptoms. Brain, 119: 363–375. Jenkins, W.M., Merzenich, M.M., Ochs, M.T., Allard, T. and Guic-Robles, E. (1990) Functional reproganization of primary
363 somatosensory cortex in adult owl monkeys after behaviorally controlled tactile stimulation. J. Neurophysiol., 63: 82–104. Jensen, T.S. and Rasmussen, P. (1994) Phantom pain and related phenomena after amputation. In: Wall, P.D. and Melzack, R. (Eds.), Textbook of pain (3rd Edn.). Churchill Livingstone, New York, pp. 651–665. Jones, E.G. (1985) The Thalamus. Plenum, New York. Kaas, J.H. (1991) Plasticity of sensory and motor maps in adult mammals. Annu. Rev. Neurosci., 14: 137–167. Lee, J., Dougherty, P.M., Antezana, D. and Lenz, F.A. (1999) Responses of neurons in the region of human thalamic principal somatic sensory nucleus to mechanical and thermal stimuli graded into the painful range. J. Comp. Neurol., 410: 541–555. Lenz, F.A., Dostrovsky, J.O., Tasker, R.R., Yamashiro, K., Kwan, H.C. and Murphy, J.T. (1988a) Single-unit analysis of the human ventral thalamic nuclear group: somatosensory responses. J. Neurophysiol., 59: 299–316. Lenz, F.A. and Dougherty, P.M. (1998a) Cells in the human principal thalamic sensory nucleus (ventralis caudalis––Vc) respond to innocuous mechanical and cool stimuli. J. Neurophysiol., 79: 2227–2230. Lenz, F.A. and Dougherty, P.M. (1998b) Neurons in the human thalamic somatosensory nucleus (ventralis caudalis) respond to innocuous cool and mechanical stimuli. J. Neurophysiol., 79: 2227–2230. Lenz, F.A., Garonzik, I.M., Zirh, T.A. and Dougherty, P.M. (1998a) Neuronal activity in the region of the thalamic principal sensory nucleus (ventralis caudalis) in patients with pain following amputations. Neurosci., 86(4): 1065–1081. Lenz, F.A., Gracely, R.H., Baker, F.H., Richardson, R.T. and Dougherty, P.M. (1998b) Reorganization of sensory modalities evoked by microstimulation in region of the thalamic principal sensory nucleus in patients with pain due to nervous system injury. J. Comp Neurol., 399: 125–138. Lenz, F.A., Gracely, R.H., Hope, E.J., Baker, F.H., Rowland, L.H., Dougherty, P.M. and Richardson, R.T. (1994a) The sensation of angina can be evoked by stimulation of the human thalamus. Pain, 59(1): 119–125. Lenz, F.A., Gracely, R.H., Rowland, L.H. and Dougherty, P.M. (1994b) A population of cells in the human thalamic principal sensory nucleus respond to painful mechanical stimuli. Neurosci. Lett., 180: 46–50. Lenz, F.A., Kwan, H.C., Martin, R., Tasker, R., Richardson, R.T. and Dostrovsky, J.O. (1994c) Characteristics of somatotopic organization and spontaneous neuronal activity in the region of the thalamic principal sensory nucleus in patients with spinal cord transection. J. Neurophysiol., 72: 1570–1587. Lenz, F.A., Seike, M., Lin, Y.C., Baker, F.H., Rowland, L.H., Gracely, R.H. and Richardson, R.T. (1993a) Neurons in the area of human thalamic nucleus ventralis caudalis respond to painful heat stimuli. Brain Res., 623: 235–240. Lenz, F.A., Seike, M., Richardson, R.T., Lin, Y.C., Baker, F.H., Khoja, I., Jaeger, C.J. and Gracely, R.H. (1993b) Thermal and pain sensations evoked by microstimulation in the area of human ventrocaudal nucleus. J. Neurophysiol., 70: 200–212.
Lenz, F.A., Tasker, R.R., Kwan, H.C., Schnider, S., Kwong, R., Murayama, Y., Dostrovsky, J.O. and Murphy, J.T. (1988b) Single unit analysis of the human ventral thalamic nuclear group: correlation of thalamic ‘‘tremor cells’’ with the 3–6 Hz component of parkinsonian tremor. J. Neurosci., 8: 754–764. Mehler, W.R. (1962) The anatomy of the so-called ‘‘pain tract’’ in man: an analysis of the course and distribution of the ascending fibers of the fasciculus anterolateralis. In: French, J.D. and Porter, R.W. (Eds.), Basic Research in Paraplegia. Thomas, Springfield, pp. 26–55. Mehler, W.R. (1966) The posterior thalamic region in man. Confinia Neurol., 27: 18–29. Mehler, W.R. (1969) Some neurological species differences- A posteriori. Ann. NY Acad. Sci., 167: 424–468. Mehler, W.R., Feferman, M.E. and Nauta, W.H.J. (1960) Ascending axon degeneration following anterolateral cordotomy. An experimental study in the monkey. Brain, 83: 718–750. Mesulam, M. and Mufson, E.J. (1985) Insula of Riel in man and monkeys. Cereb. Cortex, 4: 179–225. Mountcastle, V.B. (1984) Central nervous mechanisms in mechanoreceptive sensibility. In: Brookhart, J.M., et al. (Eds.), Handbook of Physiology. Sensory Processes. American Physiological Society, Bethesda, MD, pp. 744–789. Mufson, E.J. and Mesulam, M.M. (1984) Thalamic connections of the insula in the rhesus monkey and comments on the paralimbic connectivity of the medial pulvinar nucleus. J. Comp. Neurol., 227: 109–120. Nathan, P.W., Smith, M.C. and Cook, A.W. (1986) Sensory effects in man of lesions of the posterior columns and of some other afferent pathways. Brain, 109(Pt 5): 1003–1041. Nudo, R.J., Milliken, G.W., Jenkins, W.M. and Merzenich, M.M. (1996) Use-dependent alterations of movement representations in primary motor cortex of adult squirrel monkeys. J. Neurosci., 16: 785–807. Ohara, S. and Lenz, F.A. (2003) Medial lateral extent of thermal and pain sensations evoked by microstimulation in somatic sensory nuclei of human thalamus. J. Neurophysiol., 90: 2367–2377. Pons, T.P., Garraghty, P.E., Ommaya, A.K., Kaas, J.H., Taub, E. and Mishkin, M. (1991) Massive cortical reorganization after sensory deafferentation in adult macaques. Science, 252: 1857–1860. Radhakrishnan, V., Tsoukatos, J., Davis, K.D., Tasker, R.R., Lozano, A.M. and Dostrovsky, J.O. (1999) A comparison of the burst activity of lateral thalamic neurons in chronic pain and nonpain patients. Pain, 80: 567–575. Rasmusson, D. (1996a) Changes in the organization of ventroposterior lateral thalamic nucleus after digit removal in the adult raccoon. J. Comp. Neurol., 364: 92–103. Rasmusson, D.D. (1996b) Changes in the response properties of neurons in the ventroposterior lateral thalamic nucleus of the raccoon after peripheral deafferentation. J. Neurophysiol., 75: 2441–2450. Recanzone, G.H., Merzenich, M.M., Jenkins, W.M., Grajski, K.A. and Dinse, H.R. (1992) Topographic reorganization of
364 the hand representation in cortical area 3b of owl monkeys trained in a frequency-discrimination task. J. Neurophysiol., 67: 1031–1056. Sherman, S.M. and Guillery, R.W. (2001) Exploring the thalamus. Academic Press, San Diego, CA. Steriade, M. and Deschenes, M. (1984) The thalamus as a neuronal oscillator. Brain Res. Rev., 8: 1–63. Steriade, M., Jones, E.G. and Llinas, R.R. (1990) Thalamic oscillations and signaling. Wiley, John & Sons, New York. Steriade, M. and Llinas, R.R. (1988) The functional states of the thalamus and the associated neuronal interplay. Physiol. Rev., 68: 649–742. Sweet, W.H., Poletti, C.E. and Gybels, G.M. (1994) Operations in the brainstem and spinal canal with an appendix on the relationship of open and percutaneous cordotomy. In: Wall, P.D. and Melzack, R. (Eds.), Textbook of Pain (3 Edn). Churchill and Livingston, New York, pp. 1113–1136. Tasker, R.R. (1982) Identification of pain processing systems by electrical stimulation of the brain. Hum. Neurobiol., 1: 261–272. Tasker, R.R. (1988) Percutaneous Cordotomy: the lateral high cervical technique. In: Schmidek, H.H. and Sweet, W.H. (Eds.), Operative Neurosurgical Techniques Indications, Methods, and Results. Saunders, W.B., Philadelphia, PA, pp. 1191–1205. Tasker, R.R., Doorly, T. and Yamashiro, K. (1988) Thalamotomy in generalized dystonia. Adv. Neurol., 50: 615–631. Tasker, R.R., Organ, L.W. and Hawrylyshyn, P. (1980) Deafferentation and causalgia. In: Bonica, J.J. (Ed.), Pain. Raven Press, New York, pp. 305–329. Triggs, W.J. and Beric, A. (1994) Dysaesthesiae induced by physiological and electrical activation of posterior column afferents after stroke. J. Neurol. Neurosurg. Psych., 57: 1077–1080. Van Buren, J.M. and Borke, R.C. (1972) Variations and Connections of the Human Thalamus. Springer Verlag, Berlin. Vestergaard, K., Nielsen, J., Andersen, G., Ingeman-Nielsen, M., Arendt-Nielsen, L. and Jensen, T.S. (1995) Sensory abnormalities in consecutive unselected patients with central poststroke pain. Pain, 61: 177–186.
Vierck, C.J. (1991) Can mechanisms of central pain syndromes be investigated in animal models? In: Casey, K.L. (Ed.), Pain and central nervous system disease: the central pain syndromes. Raven Press, New York, pp. 129–141. Vitek, J.L. and Lenz, F.A. (1998) Contemporary pallidotomy for treatment of dystonia and other movement disorders. In: Grossman, R.G. and Krauss, G. (Eds.), Pallidotomy for treatment of Parkinson’s disease and other movement disorders (1 Edn). Lippincott-Raven, Philadelphia, PA, pp. 267–274. Weng, H.R., Lee, J.-I., Lenz, F.A., Vierck, C.J., Rowland, L.H. and Dougherty, P.M. (2000) Functional plasticity in primate somatosensory thalamus following chronic lesion of the ventral lateral spinal cord. Neurosci., 101: 393–401. Weng, H.R., Lenz, F.A., Vierck, C. and Dougherty, P.M. (2003) Physiological changes in primate somatosensory thalamus induced by deafferentation are dependent on the spinal funiculi that are sectioned and time following injury. Neurosci., 116: 1149–1160. Willis, W.D. and Coggeshall, R.E. (1991) Sensory Mechanisms of the Spinal Cord. Plenum Press, New York. Willis, W.D., Trevino, D.L., Coulter, J.D. and Maunz, R.A. (1974) Responses of primate spinothalamic tract neurons to natural stimulation of hindlimb. J. Neurophysiol., 37(2): 358–372. Willis Jr., W.D., Zhang, X., Honda, C.N. and Giesler Jr., G.J. (2001) Projections from the marginal zone and deep dorsal horn to the ventrobasal nuclei of the primate thalamus. Pain, 92: 267–276. Young, R.F., Kroening, R., Fulton, W., Feldman, R.A. and Chambi, I. (1985) Electrical stimulation of the brain in treatment of chronic pain. J. Neurosurg., 62: 389–396. Yuan, B., Morrow, T.J. and Casey, K.L. (1986) Corticofugal influences of S1 cortex on vetrobasal thalamic neurons in the awake rat. J. Neurosci., 6: 3611–3617. Zirh, A., Reich, S.G., Dougherty, P.M. and Lenz, F.A. (1999) Stereotactic thalamotomy in the treatment of essential tremor of the upper extremity: reassessment including a blinded measure of outcome. J. Neurol. Neurosurg. Psych., 66: 772–775.
A.R. Møller (Ed.) Progress in Brain Research, Vol. 157 ISSN 0079-6123 Copyright r 2006 Elsevier B.V. All rights reserved
CHAPTER 22
Neural plasticity in tinnitus Aage R. Møller University of Texas at Dallas, School of Behavioral and Brain Science, Richardson, TX, USA
Abstract: Two distinctly different kinds of tinnitus occur: objective and subjective tinnitus. Objective tinnitus is caused by sounds generated in the body while subjective tinnitus is caused by abnormal neural activity that is not evoked by sound. This chapter discusses subjective tinnitus. Subjective tinnitus has many forms. In most forms of tinnitus the anatomical location of the physiological abnormality is in the central nervous system, although the sensation is often referred to one ear or both ears. The cause of most forms of subjective tinnitus is the changes that have occurred as a result of expression of neural plasticity, thus a form of reprogramming of the brain that is not to the benefit of the individual person. Tinnitus often occurs together with hearing loss, indicating that the expression of neural plasticity has been evoked by deprivation of input. Tinnitus is often accompanied by hyperacusis, and sometimes phonophobia and depression, indicating altered processing of auditory information or rerouting of information. Several studies have brought evidence that some forms of tinnitus are associated with an abnormal involvement of the nonclassical (extralemniscal, diffuse, or polysensory) auditory pathways that bypass the primary auditory cerebral cortex and provide subcortical connections to limbic structures among others. There is no general treatment for tinnitus, but there are several treatments that can alleviate or reduce the tinnitus in some patients. Keywords: tinnitus; neural plasticity; hyperactive disorders; nonclassical auditory pathways intolerable sounds that interfere with sleep and intellectual work. It can be continuous or intermittent. While tinnitus can have the character of high-frequency tones or broadband noise, it is often difficult to describe and often tinnitus does not resemble any known physical sound. Subjective tinnitus is a phantom sensation that may be accompanied by abnormal sensations of sounds and sounds may be perceived unpleasantly loud (hyperacusis). Phantom sensations have been reported in vision (phosphene), olfaction (phantosmia or olfactory hallucinations), and taste (metallic taste). Tinnitus may also have similarities with vertigo and other vestibular disorders. Tinnitus has many similarities with the symptoms of neurologic disorders such as paresthesia, and in particular, with central neuropathic pain (Møller, 2006b). Sounds from outside may evoke fear
Introduction Tinnitus is the general name for many different forms of pathologies that have the following character in common: sounds are perceived in the absence of sounds from outside the body. There are two main kinds of tinnitus, known as objective and subjective tinnitus. Objective tinnitus is generated by sounds inside the body, such as from turbulent blood flow or muscle contractions (see Schleuning, 1991). An observer using a stethoscope can hear that kind of tinnitus. Objective tinnitus is rare and will not be discussed in this chapter. Subjective tinnitus has many expressions. It can be weak sounds only heard in quiet environments; it can be Corresponding author. Tel.: +1 972 883 2313; E-mail:
[email protected] DOI: 10.1016/S0079-6123(06)57022-0
365
366
(phonophobia) in some individuals with severe tinnitus. Severe tinnitus may cause depression and suicide (Møller, 2006a). Auditory hallucinations, which are perception of meaningful sounds such as music or speech (Braun et al., 2003), occur in connection with psychiatric disorders such as schizophrenia and may also be caused by drugs of various kinds. There is considerable evidence that the anatomical location of the abnormality that causes many forms of subjective tinnitus is the central nervous system (CNS), and that the changes in function are caused by expression of neural plasticity. This chapter will discuss the role of neural plasticity in causing tinnitus and in the possibilities of treating tinnitus.
Etiology and nature of central tinnitus Subjective tinnitus is a hyperactive sensory disorder that is similar to paresthesia. Subjective tinnitus has many forms: it can be intermittent or unrelenting; it can be weak or strong; it can cause varying degrees of discomfort from slight annoyance to preventing sleep and intellectual activities; and it can cause suicide (Møller, 2006a). Severe subjective tinnitus is often accompanied by hyperacusis (lowered threshold for discomfort from sound) or phonophobia (fear of sound). Subjective tinnitus is often accompanied by phonophobia and depression occurs in some patients. Tinnitus can be referred to one side (one ear) or it can be bilateral. It can have the character of a tone, noise, or cricket-like sounds. It is typical that the intensity of tinnitus is difficult to determine. When assessed by comparing the tinnitus to a sound presented to the ear, unrealistically low estimates of the tinnitus are obtained (Vernon, 1976; Goodwin and Johnson, 1980). The visual analog scale (VAS) similar to what is often used to assess the intensity of pain may provide a better measure of the severity of tinnitus than techniques that use matching of the tinnitus to a sound. The VAS can be applied to bilateral tinnitus as well as unilateral tinnitus. Subjective tinnitus is a phantom sensation (Jastreboff, 1990) with many similarities to other
phantom sensations such as the phantom limb and central pain. There is considerable evidence that the symptoms and signs of these disorders are caused by functional changes and reorganization of the CNS through expression of neural plasticity (Møller, 2006b). Subjective tinnitus is generated by abnormal neural activity and the anatomical location where this abnormal activity, which is perceived as a sound, is generated. The location is the CNS for most forms of severe subjective tinnitus. There are strong indications that these changes in function have occurred because of expression of neural plasticity. Tinnitus is therefore associated with a form of reprogramming of the CNS that causes symptoms and signs of disease. Some studies have shown that in some patients these functional changes can be reversed by sound exposure (Jastreboff and Jastreboff, 2000). Tinnitus can occur in connection with deprivation of sound or reduced input from the ear, such as may be caused by hearing loss of any etiology, including hearing loss caused by middle ear disorders such as otosclerosis (conductive types of hearing loss). Subjective tinnitus often occurs after exposure to loud sounds that causes hearing loss, but hearing loss from other causes such as agerelated changes (presbycusis) may also be accompanied by tinnitus (Møller, 2006a) and even deaf people may have tinnitus. Administration of drugs such as aspirin, some diuretics, aminoglycoside antibiotics, and cisplatin (Seligmann et al., 1996; Simpson and Davies, 1999) may cause tinnitus. Many forms of tinnitus that occur without any cause can be found (idiopathic tinnitus). Subjective tinnitus occurs almost always in patients with vestibular schwannoma (House and Brackmann, 1981; Møller, 2006a). Surgical manipulations of the intracranial portion of the auditory nerve in other operations can likewise cause tinnitus (Møller and Møller 1989). Vascular contact with the auditory nerve root may be involved in tinnitus (Møller, 1987, 1993), but such contacts occur commonly without any symptoms (Sunderland, 1948) indicating that factors other than vascular contact with the auditory nerve are necessary for the development of tinnitus (Møller, 1993). Hereditary disorders such as Williams Beuren syndrome (WBS) (infantile hypocalcaemia) have a
367
high incidence of hyperacusis ( Klein et al., 1990; Borsel van et al., 1997; Møller, 2006b). Tinnitus also occurs together with some developmental disorders such as various forms of autism.
Functional abnormalities associated with subjective tinnitus It is now recognized that most forms of severe subjective tinnitus are caused by abnormal neural activity that results from expression of neural plasticity. Subjective tinnitus has many similarities with central pain conditions (Møller, 1997, 2006b). The involvement of neural plasticity in many forms of pain is well known (Wall and Melzack, 1999; Woolf and Mannion, 1999; Møller, 2006b) (see Chapter 20).
Expression of neural plasticity in central tinnitus Most forms of subjective tinnitus are hyperactive disorders that are caused by expression of neural plasticity (Møller, 2006b), but it is rarely established with any certainty what caused the expression of neural plasticity. Tinnitus often occurs after overstimulation, and it may occur in connection with deprivation of auditory input that may be caused by disorders of the cochlea, but can also be caused by disorders that affect the middle ear, such as otosclerosis (Glasgold and Altman, 1966). Expression of neural plasticity may cause hyperactivity; changes in the ways of sound processing and rerouting of information. Animal experiments have shown that changes occur in evoked responses from the inferior colliculus after exposure to loud sounds, and these changes have been interpreted as signs of hyperactivity (Szczepaniak and Møller, 1996). Studies of place cells in the hippocampus of the rat have shown changes in function after similar sound exposure (Goble et al., 2004). Other animal experiments have shown that deprivation of input to the auditory nervous system causes hypersensitivity to electrical stimulation of the cochlear nucleus and the inferior colliculus (Gerken et al., 1984). Signs of changes in processing (altered temporal integration) have
also been demonstrated after suppression of input (Gerken et al., 1991). It is well known that abnormal sensory input (or deprivation of input) may cause functional changes such as changes in synaptic efficacy, or morphological changes such as elimination or creation of synapses, axons, or dendrites. Studies in the auditory system have shown that severance of auditory nerve can cause extensive morphological changes in the cochlear nucleus (Morest et al., 1979; Morest and Bohne, 1983). Hebb’s principle states that morphological connections may become established between neurons that are activated together (Hebb, 1949). This form of plastic changes has often been referred to as ‘‘neurons that fire together, wire together.’’ Functional and morphological changes may also occur when the correlation between neural activities in neurons that are connected with each other is decreased. This would correspond to the reverse of Hebb’s principle and implies that connections that are not used are eliminated (this may just be a version of the general rule of ‘‘use it or lose it’’). Age-related changes such as presbycusis could perhaps be explained in that way. Tinnitus is more prevalent in old individuals and that may be explained by such changes. Changes in the GABA system that occur normally with age (Caspary et al., 1990) may also contribute to hyperactive sensory functions such as tinnitus (see Møller, 2006a).
Hyperactivity Expression of neural plasticity can be triggered by reduced input to the auditory nervous system, as occurs in hearing loss, and that may increase the sensitivity of the ear and certain parts of the auditory nervous systems. If the sensitivity is increased beyond a certain degree, it may result in hyperactivity and the resulting change in neural activity may resemble that evoked by sound and be perceived as a sound, thus causing tinnitus. Many forms of subjective tinnitus have similarities with other hyperactive disorders. Tinnitus and the other signs that may accompany severe tinnitus (hyperacusis, phonophobia, etc.) may be caused by functional or morphological changes in the nuclei in the ascending auditory pathways (Gerken et al.,
368
1984; Kaltenbach, 2000). Animal experiments have shown signs of hyperactivity in auditory brainstem nuclei and changes in the tonotopic organization (Kaltenbach et al., 1996) after similar overstimulation that normally causes tinnitus in humans (Kaltenbach and Afman, 2000). Studies in humans and animals have shown indications that tinnitus may be associated with reorganization of auditory nuclei and the auditory cerebral cortex (Mu¨hlnickel et al., 1998). Change in processing Animal experiments have shown that deprivation of input can change the temporal integration in brainstem auditory nuclei (the cochlear nucleus and the inferior colliculus) (Gerken, 1979; Gerken et al., 1984) as does overstimulation of a kind that normally results in tinnitus in humans (Szczepaniak and Møller, 1996). Change in temporal integration is a sign of change in neural processing. Similar change in temporal integration of somatosensory input has been shown to occur in individuals with central pain (Møller and Pinkerton, 1997)), thus a sign of similarities between severe tinnitus and central pain (Møller, 1997, 2006b). Rerouting of neural activity Redirection of information to regions of the CNS may also occur in severe tinnitus, and structures that normally do not receive auditory input may process auditory input. Rerouting of information may occur by unmasking of dormant synapses that can open routes normally blocked because of synapses that do not conduct. Studies have indicated that abnormal activation of the nonclassical auditory pathways (also known as the extralemniscal, diffuse, or nonspecific pathways, Aitkin, 1986) may occur in some patients with tinnitus (Møller et al., 1992; Cacace et al., 1994). Some patients with tinnitus have signs of cross-modal interaction between the auditory and the somatosensory system. This may explain the presence of symptoms and signs of not only tinnitus, but also hyperacusis, phonophobia, and depression (see Møller, 2006b). Involvement of the neoclassical auditory pathway may cause an abnormal crossmodal interaction between the auditory system
and other sensory systems such as the somatosensory and visual systems. Phonophobia may result from an abnormal activation of the limbic system through the nonclassical auditory pathways, which are not normally activated by sound stimulation (Møller, 2003). In addition to being involved in some forms of tinnitus, there are also indications that abnormal activation of the nonclassical auditory pathways (Graybiel, 1972; Aitkin, 1986; Møller, 2003) occurs in some individuals with autism (Møller et al., 2005) and the nonclassical pathway is normally active in young children (Møller and Rollins, 2002) ([47]Møller et al., 1992). The fact that the nonclassical auditory system receives input from the somatosensory system has been used in these studies of the involvement of the nonclassical auditory system in perception of loudness. The nonclassical pathways ascend in parallel to the much better known classical pathways (Graybiel, 1972; Aitkin, 1986; Møller, 2003). The nonclassical pathways receive their main auditory input from the central nucleus of the inferior colliculus, which belongs to the classical ascending pathways. The thalamic nuclei of the nonclassical auditory pathways are the dorsal and medial parts of the medial geniculate bodies (dMGB and mMGB) while the classical auditory pathways use the ventral thalamic nucleus (vMGB) (see Møller, 2003). While the thalamic nuclei of the classical pathways project to the primary auditory cortex, the nonclassical pathways bypass the primary sensory cortices and project directly to secondary and association cortices. Less specific analysis of sounds is performed in the nonclassical pathways compared to that of the classical pathways ( Syka et al., 2000; Møller, 2003). While the neurons in classical sensory pathways only respond to one sensory modality, neurons in the nuclei of the nonclassical auditory pathways respond to several sensory modalities. The nonclassical auditory pathways have subcortical connections to limbic structures from the dorsal and medial MGB (‘‘low route’’, LeDoux, 1992), while the classical auditory pathways connect to limbic structures through a long chain of neurons where the
369 information is processed and modulated by intrinsic and extrinsic neural activity. Processing in the nonclassical pathways is different from that of the classical pathways. The nuclei of the nonclassical pathways perform less-specific analysis of sounds than those of the classical pathways (Syka et al., 2000; Møller, 2003). This is why these pathways also have been known as the diffuse pathways.
The finding that temporal mandibular joint (TMJ) disorders are often associated with tinnitus (Morgan, 1992) and that the tinnitus may be alleviated when the TMJ is treated (Morgan, 1973) may be explained by interaction between the somatosensory system (the sensory trigeminal nerve) and the auditory system (Shore et al., 2000). Interaction between neck muscles and tinnitus is also apparent (Levine, 1999; Levine et al., 2003). Such cross-modal interaction is assumed to involve the nonclassical auditory pathways (Møller et al., 1992) because this system receives input from other sensory systems, contrary to the classical ascending sensory systems, which only receive input of a single modality (Møller, 2003). Subjective tinnitus has similarities with other phantom sensations such as some forms of central pain in which rerouting of information in the CNS occurs. Allodynia is an example of rerouting of innocuous somatosensory information to pain circuits (see Chapter by Woolf). It seems likely that abnormal cross-modal interaction between the auditory and the somatosensory system (Møller et al., 1992; Szczepaniak and Møller, 1993; Cacace et al., 1994) is caused by expression of neural plasticity, which has caused changes in synaptic efficacy or created new synapses or new morphologic connections through sprouting of axons or dendrites. These studies thus indicate that the neural activity that causes tinnitus may not be generated in the same neural structures as those that normally process sound-evoked neural activity. Other studies have also indicated that some forms of tinnitus may be caused by activation of neural circuits that are not normally involved in processing of sound-evoked neural activity (Lockwood et al., 1998), and that may explain why many patients find it difficult to match an external sound
to their tinnitus. Such attempts result in estimates of the intensity of the sounds that the patients perceive to be very low compared with the patients’ complaints (10 30 dB above threshold, Vernon, 1976). The use of the same VAS that is often used in evaluation of pain seems to be a better way for assessing tinnitus than loudness matching. Me´nie`re’s disease is a progressive disorder that is defined by a triad of symptoms, namely vertigo with nausea, fluctuating hearing loss, and tinnitus (Minor et al., 2004). Expression of neural plasticity may be involved in creating the symptoms of Me´nie`re’s disease (Møller, 2006b) as indicated by the finding that application of air puffs to the middle ear cavity reduces the symptoms and even normalizes electrophysiologic signs (abnormal summating potential) (Densert et al., 1995). The applied air pressure affects the fluid pressure in the inner ear stimulating receptors in the labyrinth. Traditionally, the symptoms of Me´nie`re’s disease have been related to abnormal pressure (or rather volume) of the fluid in the inner ear. Cause of individual variations The relationship between known factors that can cause tinnitus and the actual occurrence of tinnitus varies among individuals. For example, while tinnitus is associated with hearing loss, not all individuals with the same hearing loss acquire tinnitus. Individuals also react differently to tinnitus. Some may have severe reactions in the form of hyperacusis, depression, and other affective symptoms, while other people can cope with tinnitus without such reactions. While most people have experienced tinnitus intermittently, some people have never experienced tinnitus. Treatment of subjective tinnitus Many kinds of treatment for subjective tinnitus and hyperacusis have been tried but the results have in general been poor. Treatment of tinnitus is hampered by lack of understanding of the pathophysiology of the disorder and the absence of detectable morphological abnormalities. The efficacy of treatments varies between individuals supporting the
370
hypothesis that subjective tinnitus has different etiology, although the symptoms may be similar. Lidocaine was one of the first medications that was found to alleviate tinnitus in some individuals, but it never found practical use because it had to be administrated intravenously. Since then many different drugs have been tried with little beneficial effect. Administration of benzodiazepines has had some success in alleviating tinnitus (Alprazalam, Clonazepam) (Vernon and Meikle, 2003). Microvascular decompression (MVD) of the auditory nerve root (Jannetta, 1977; Kondo et al., 1980; Møller et al., 1993) can alleviate tinnitus or reduce its intensity in some patients. A study of the outcome of MVD operations in selected group of patients showed that an average of 40.3% of 12 men and 17 women had marked improvement of their tinnitus. The outcome was different in men and women with only 29.3% of men having relief, while 54.8% of the women in this group of patients had relief of the tinnitus from MVD operations (Møller et al., 1993). This study also showed that the outcome was less favorable in patients who had had their tinnitus for a long period (40.3% for patients with duration of tinnitus of less than 4 years and 59.7% for patients who had had tinnitus for more than 4 years). The average success rate was less than what was achieved by MVD in patients with hemifacial spasm and trigeminal neuralgia, for which MVD has a success rate of approximately 85% ( Møller, 1991; Barker et al., 1995; Barker et al., 1996). MVD operations were much less effective in relieving tinnitus in patients with bilateral tinnitus (Vasama et al., 1998) supporting the hypothesis that pathophysiology of unilateral and bilateral tinnitus are fundamentally different. Some forms of severe tinnitus can be ameliorated by exposure to specific sounds together with psychological treatment (Jastreboff and Jastreboff, 2000). This treatment known as tinnitus retraining treatment (TRT) (Jastreboff and Jastreboff, 2000) is based on the hypothesis that proper stimulation can reverse plastic changes in the nervous system that were caused by deprivation of input. This kind of treatment has similarities with the use of transderm electric nerve stimulation (TENS) for treatment of central pain. (Willer, 1988; Møller, 2006b) (see Chapter by Price).
More recently, electrical stimulation of the cerebral auditory cortex has been shown to ameliorate tinnitus in selected cases of tinnitus (De Ridder et al., 2004; De Ridder et al., 2005; Kleinjung et al., 2005). Transcranial magnetic stimulation was used as a test in selection of patients for implantation of the electrodes for stimulation of the cerebral cortex. It is possible that electrical stimulation of the cerebral cortex has its effect by affecting the function of the thalamus. There are indications that the primary cerebral cortex is not involved in generating tinnitus and the effect of cortical electrical stimulation could be explained by its activation of the abundant descending fiber tracts from the cortex to the MGB (corticothalamic pathways). Lack of controlled studies (Dobie, 1999) and firm differential diagnoses of tinnitus have made the choice of drugs and other treatments more of an art than a science. As the expression of tinnitus varies widely among individuals also, the efficacy of the same treatment for individuals with similar kind and severity of tinnitus varies widely. Treatments that are effective in one individual may lack beneficial effect in another individual with similar tinnitus. It has been difficult to evaluate any form of treatment for tinnitus because of the diversity of the symptoms and probably also of the pathophysiology, which may include different anatomical locations of the physiological abnormalities that cause the generation of the abnormal neural activity, which causes the sensation of sounds without any sound reaching the ear.
References Aitkin, L.M. (1986) The Auditory Midbrain, Structure and Function in the Central Auditory Pathway. Humana Press, Clifton, NJ. Barker, F.G., Jannetta, P.J., Bissonette, D.J., Larkins, M.V. and Jho, H.D. (1996) The long-term outcome of microvascular decompression for trigeminal neuralgia. N. Eng. J. Med., 334: 1077–1083. Barker, F.G., Jannetta, P.J., Bissonette, D.J., Shields, P.T. and Larkins, M.V. (1995) Microvascular decompression for hemifacial spasm. J. Neurosurg., 82: 201 210. Borsel van, J., Curfs, L.M.G. and Fryns, J.P. (1997) Hyperacusis in Williams syndrome: a sample survey study. Genet. Couns., 8: 121–126.
371 Braun, C.M., Dumont, M., Duval, J., Hamel-Hebert, I. and Godbout, L. (2003) Brain modules of hallucination: an analysis of multiple patients with brain lesions. J. Psychiatry Neurosci., 28: 432–449. Cacace, A.T., Lovely, T.J., McFarland, D.J., Parnes, S.M. and Winter, D.F. (1994) Anomalous cross-modal plasticity following posterior fossa surgery: some speculations on gazeevoked tinnitus. Hear. Res., 81: 22–32. Caspary, D.M., Raza, A., Lawhorn, Armour, B.A., Pippin, J. and Arneric, S.P. (1990) Immunocytochemical and neurochemical evidence for age-related loss of GABA in the inferior colliculus: implications for neural presbycusis. J. Neurosci., 10: 2363–2372. De Ridder, D., De Mulder, G., Walsh, V., Muggleton, N., Sunaert, S., Verlooy, J., VandeHeyning, P. and Møller, A.R. (2005) Transcranial magnetic stimulation for tinnitus: a clinical and pathophysiological approach: influence of tinnitus duration on stimulation parameter choice and maximal tinnitus suppression. Otol. Neurotol., 147: 495–501. De Ridder, D., De Mulder, G., Walsh, V., Muggleton, N., Sunaert, S. and Møller, A. (2004) Magnetic and electrical stimulation of the auditory cortex for intractable tinnitus. J. Neurosurg., 100: 560–564. Densert, B., Sass, K. and Arlinger, S. (1995) Short term effects of induced middle ear pressure changes on the electrocochleogram in Meniere’s disease. Acta Otolaryngol., 115: 732–737. Dobie, R.A. (1999) A review of randomized clinical trials in tinnitus. Laryngoscope, 109: 1202–1211. Gerken, G.M. (1979) Temporal summation of pulsate brain stimulation in normal and deafened cats. J. Acoust. Soc. Am., 66: 728–734. Gerken, G.M., Saunders, S.S. and Paul, R.E. (1984) Hypersensitivity to electrical stimulation of auditory nuclei follows hearing loss in cats. Hear. Res., 13: 249–260. Gerken, G.M., Solecki, J.M. and Boettcher, F.A. (1991) Temporal integration of electrical stimulation of auditory nuclei in normal hearing and hearing-impaired cat. Hear. Res., 53: 101–112. Glasgold, A. and Altman, F. (1966) The effect of stapes surgery on tinnitus in otosclerosis. Laryngoscope, 76: 1524–1532. Goble, T.J., Farmer, G.E., Frank, J., Møller, A.R. and Thompson, L.T. (2004) Acute noise exposure alters hippocampal placefields: evidence for extralemniscal sensory pathway plasticity. Program No. 209.8.2004. Goodwin, P.E. and Johnson, P.M. (1980) The loudness of tinnitus. Acta Otolaryngol. (Stockh.), 90: 353–359. Graybiel, A.M. (1972) Some fiber pathways related to the posterior thalamic region in the cat. Brain Behav. Evol., 6: 363–393. Hebb, D.O. (1949) The Organization of Behavior. New York, Wiley. House, J.W. and Brackmann, D.E. (1981) Tinnitus: surgical treatment. In: Tinnitus, Ciba Foundation Symposium 85. Pitman Books Ltd., London. Jannetta, P.J. (1977) Observations on the etiology of trigeminal neuralgia, hemifacial spasm, acoustic nerve dysfunction and
glossopharyngeal neuralgia. Definitive microsurgical treatment and results in 117 patients. Neurochirurgia (Stuttg.), 20: 145–154. Jastreboff, P.J. (1990) Phantom auditory perception (tinnitus): mechanisms of generation and perception. Neurosci. Res., 8: 221–254. Jastreboff, P.J. and Jastreboff, M.M. (2000) Tinnitus retraining therapy (TRT) as a method for treatment of tinnitus and hyperacusis patients. J. Am. Acad. Audiol., 11: 162 177. Kaltenbach, J.A. (2000) Neurophysiologic mechanisms of tinnitus. J. Am. Acad. Audiol., 11: 125–137. Kaltenbach, J.A. and Afman, C.E. (2000) Hyperactivity in the dorsal cochlear nucleus after intense sound exposure and its resemblance to tone-evoked activity: a physiological model for tinnitus. Hear. Res., 140: 165–172. Kaltenbach, J.A., Meleca, R.J. and Falzarano, P.R. (1996) Alterations in the tonotopic map of the cochlear nucleus following cochlear damage, in auditory system plasticity and regeneration. In: Salvi, R.J., Henderson, D., Fiorino, F. and Colletti, V. (Eds.), Auditory System Plasticity and Regeneration. Thieme Medical Publishers, New York, pp. 317–332. Klein, A.J., Armstrong, B.L., Greer, M.K. and Brown, F.R. (1990) Hyperacusis and otitis media in individuals with Williams syndrome. J. Speech Hear. Dis., 55: 339–344. Kleinjung, T., Eichhammer, P., Langguth, B., Jacob, P., Marienhagen, J., Hajak, G., Wolf, S.R. and Strutz, J. (2005) Longterm effects of repetitive transcranial magnetic stimulation (rTMS) in patients with chronic tinnitus. Otolaryngol. Head Neck Surg., 132: 566–569. Kondo, A., Ishikawa, J., Yamasaki, T. and Konishi, T. (1980) Microvascular decompression of cranial nerves, particularly of the seventh cranial nerve. Neurol. Med. Chir. (Tokyo), 20: 739–751. LeDoux, J.E. (1992) Brain mechanisms of emotion and emotional learning. Curr. Opin. Neurobiol., 2: 191–197. Levine, R.A. (1999) Somatic (craniocervical) tinnitus and the dorsal cochlear nucleus hypothesis. Am. J. Otolaryngol., 20: 351–362. Levine, R.A., Abel, M. and Cheng, H. (2003) CNS somatosensory-auditory interactions elicit or modulate tinnitus. Exp. Brain Res., 153: 643–648. Lockwood, A., Salvi, R., Coad, M., Towsley, M., Wack, D. and Murphy, B. (1998) The functional neuroanatomy of tinnitus. Evidence for limbic system links and neural plasticity. Neurology, 50: 114–120. Minor, L.B., Schessel, D.A. and Carey, J.P. (2004) Meniere’s disease. Curr. Opin. Neurol., 17: 9 16. Morest, D.K., Ard, M.D. and Yurgelun-Todd, D. (1979) Degeneration in the central auditory pathways after acoustic deprivation or over-stimulation in the cat. Anat. Rec., 193: 750. Morest, D.K. and Bohne, B.A. (1983) Noise-induced degeneration in the brain and representation of inner and outer hair cells. Hear. Res., 9: 145–152. Morgan, D.H. (1973) Temporomandibular joint surgery. correction of pain, tinnitus, and vertigo. Dent. Radiogr. Photogr., 46: 27–46.
372 Morgan, D.H. (1992) Tinnitus of TMJ origin. J. Craniomandibular Pract., 10: 124–129. Møller, A.R. (1991) The cranial nerve vascular compression syndrome: I. A review of treatment. Acta Neurochir. (Wien.),, 113: 18–23. Møller, A.R. (1993) Cranial nerve dysfunction syndromes: pathophysiology of microvascular compression. In: Barrow, D.L. (Ed.), Neurosurgical Topics Book 13, ‘Surgery of Cranial Nerves of the Posterior Fossa,’ Chapter 2. American Association of Neurological Surgeons, Park Ridge, IL, pp. 105–129. Møller, A.R. (1997) Similarities between chronic pain and tinnitus. Am. J. Otol., 18: 577–585. Møller, A.R. (2003) Sensory Systems: Anatomy and Physiology. Academic Press, Amsterdam. Møller, A.R. (2006a) Hearing: Anatomy, Physiology, and Disorders of the Auditory System (2nd Edn). Academic Press, Amsterdam. Møller, A.R. (2006b) Neural Plasticity and Disorders of the Nervous System. Cambridge University Press, Cambridge. Møller, A.R., Kern, J.K. and Grannemann, B. (2005) Are the non-classical auditory pathways involved in autism and PDD? Neurol. Res., 27: 625–629. Møller, A.R. and Møller, M.B. (1989). Does intraoperative monitoring of auditory evoked potentials reduce incidence of hearing loss as a complication of microvascular decompression of cranial nerves? Neurosurgery, 24: 257–263. Møller, A.R., Møller, M.B. and Yokota, M. (1992) Some forms of tinnitus may involve the extralemniscal auditory pathway. Laryngoscope, 102: 1165–1171. Møller, A.R. and Pinkerton, T. (1997) Temporal integration of pain from electrical stimulation of the skin. Neurol. Res., 19: 481–488. Møller, A.R. and Rollins, P. (2002) The non-classical auditory system is active in children but not in adults. Neurosci. Lett., 319: 41–44. Møller, M.B. (1987) Vascular compression of the eighth nerve as a cause of tinnitus. In: Harsch H.F. (Ed.), Proceedings of the III International Tinnitus Seminar, Munster, Western Germany, June 11–13, 1987, Verlag, Karlsruhe, West Germany, pp. 340 347. Møller, M.B., Møller, A.R., Jannetta, P.J. and Jho, H.D. (1993) Vascular decompression surgery for severe tinnitus: selection criteria and results. Laryngoscope, 103: 421–427.
Mu¨hlnickel, W., Elbert, T., Taub, E. and Flor, H. (1998) Reorganization of auditory cortex in tinnitus. Proc. Nat. Acad. Sci. USA 95: 10340–10343. Schleuning, A.J. (1991) Management of the patient with tinnitus. Med. Clin. N. Am., 75: 1225–1237. Seligmann, H., Podoshin, L., Ben-David, J., Fradis, M. and Goldsher, M. (1996) Drug-induced tinnitus and other hearing disorders. Drug Saf., 14: 198–212. Shore, S.E., Vass, Z., Wys, N.L. and Altschuler, R.A. (2000) Trigeminal ganglion innervates the auditory brainstem. J Comp. Neurol., 419: 271–285. Simpson, J.J. and Davies, E. (1999) Recent advances in the pharmacological treatment of tinnitus. Trends Pharmacol. Sci., 20: 12–18. Sunderland, S. (1948) Microvascular relations and anomalies at the base of the brain. J. Neurol. Neurosurg. Psychiatry,, 11: 243–257. Syka, J., Popelar, J. and Kvasnak, E. (2000) Response properties of neurons in the central nucleus and external and dorsal cortices of the inferior colliculus in guinea pig. Exp. Brain Res., 133: 254–266. Szczepaniak, W.S. and Møller, A.R. (1993) Interaction between auditory and somatosensory systems: a study of evoked potentials in the inferior colliculus. Electroencephologr. Clin. Neurophysiol., 88: 508–515. Szczepaniak, W.S. and Møller, A.R. (1996) Evidence of neuronal plasticity within the inferior colliculus after noise exposure: a study of evoked potentials in the rat. Electroenceph. Clin. Neurophysiol., 100: 158–164. Vasama, J.P., Møller, M.B. and Møller, A.R. (1998) Microvascular decompression of the cochlear nerve in patients with severe tinnitus. Preoperative findings and operative outcome in 22 patients. Neurol. Res., 20: 242–248. Vernon, J. (1976) The loudness of tinnitus. Hear. Speech Action, 44: 17–19. Vernon, J.A. and Meikle, M.B. (2003) Masking devices and alprazolam treatment for tinnitus. Otolaryngol. Clin. N. Am., 36: 307–320. Wall, P.D. and Melzack, R. (1999) Textbook of Pain (4th Edn.). Churchill-Livingstone, Edinburgh. Willer, J.C. (1988) Relieving effect of TENS on painful muscle contraction produced by an impairment of reciprocal innervation: an electrophysiological analysis. Pain, 32: 271–274. Woolf, C.J. and Mannion, R.J. (1999) Neuropathic pain: aetiology, symptoms, mechanisms, and managements. The Lancet, 353: 1959–1964.
Subject Index Amphetamines 117 Amygdala 4, 337 Amyotrophic lateral sclerosis (ALS) 208, 213 therapies of 213 Analgesia 224 Anatomical, physiological and behavioral consequences 5 Angelman syndrome (AS) 28 Neurofibromatosis 1 (NF1) Anterior and posterior cingulate 124 Anterior cingulate cortex 341 Anterior commissure–posterior commissure (ACPC) 355 Anterior–posterior networks 136 Anterolateral cordotomy 361 Anthocyanins 216 Anti-inflammatory 216 Anticipation 347 Antioxidant 216 Aphasia 143 Aphasic patient 152 Ascending nociceptive pathways 336 Associated 210 Attention 344 Attention-deficit disorder 124 Attentional modulation 111 Auditory deafferentation 5 hallucinations 366 memory 100 nerve 285 sensory deficits 89 thalamus 5 Autism 320, 367–368
a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) 288 50 -bromo-20 -deoxyuridine (BrdU) 190 a-motoneurons 232 g-motoneurons 232 Acetylcholine 152 Acetylcholine esterase (AChE) 96 Acetylcholine esterase inhibitors 152 Acquired hand movement disorders 93 movement disorders 93 Activation 187 ACTIVE (advanced cognitive training in vital elderly) 97 Activity-dependent plasticity 3, 261, 269 spinal cord 263, 268 significance of 272 Adaptive neuroplasticity 167 Adhesion molecule 188 Adult brain 188 stem cells 209 Age-related changes 61, 65, 71 in the eye 90 treatability of 66 cortical changes 74 decline 92 disorders 57 peripheral changes 63 Aging 46, 48, 50, 59, 82, 130 brain 43, 57 normal 124, 205, 208 on neural efficiency 131 theories 58 treatability of 66 succesfully 58 Allodynia 358, 369 Alzheimer’s disease (AD) 53, 83, 91, 96, 204 Parkinson’s disease (PD) 208 AMPA 25 receptors 26, 295
B-cell differentiation 209 Basal ganglion 91 BDNF transcription 16 Behavioral change 269 Berlin study 58 Bilateral tinnitus 370 Blood oxygen level-dependent (BOLD) 46, 148 Blueberries 216 Boston Diagnostic Aphasia Exam (BDAE) 326 373
374 Boston Naming Test (BNT) 326 Brachium of the IC (BIC) 5 Brain injury 143, 151, 187, 199 paediatric 173 treatment of 214 lesions 204 perfusion 175 plasticity 81, 84, 85, 100 with negative consequences 85 with positive consequences 85 reorganization 173 repair 207 Brain-derived growth factor (BDGF) 194 Brain-derived neurotrophic factor (BDNF) 304 Brain-to-spinal-cord modulation of pain 343 Broca’s aphasia 144 area 147, (147, 324) Brodmann’s Area (BA) 22, 40, 144, 324 Bromocriptine 152 Calcium-calmodulin kinase II (CaMKII) 25 Calmodulin (CaM) 26 CaMKII 26 CAMP 17 CBP 34, 35 Cell therapy 207 Central auditory system 88 neuropathic pain 365 pain 359, 370 post stroke 360 syndromes 224 reorganizations 285 tinnitus 366–367 visual system 90 Central pattern generator (CPG) 231 Cerebellum 203 Cerebral palsy 264 reprogramming 167 Changes in memory performance 50 CNS modification 261 persistent 261 Coactivation 72 Cochlear implants 283, 285 Coffin–Lowry syndrome (CLS) 33
Cognitive decline 82, 88 normal 83 pathological 83 neural plasticity 199 rehabilitation 123 reserve 91 training 134 Colliculus superior 4, 338 inferior 4 Competitive processes 85 Conduction velocity in fiber tracts 63 Congenital auditory deprivation 284 deaf mice 9 Cooling loops 159 Cortex prefrontal (PFC) 46, 123–124, 201 Cortical 225 ischemic infarction 225 map plasticity 60 maps 60 networks 3 plasticity 111, 283, 353 restorative 225 receptive fields (RFs) 61 remodeling 61 reorganization 57, 60–61, 70, 113 Cortico–cortical connectivity 317 Cortico–subcortical connectivity 317 CREB-binding protein 34 Critical period 84 Cross-modal interaction 368–369 plasticity 8 reorganization 292 Cryoloop placements 161 Current source density (CSD) 287 analysis 287 Cutaneous feedback 249 pathways 242 Deaf auditory cortex 287 Deafferentation 224 Deafness and cortical development 295 Deficits word comprehension 150 Degraded neuromodulatory control 87 Dementia 83 Dentate gyrus 59, 189
375 Depression 152, 366 rate sensitive 235 Deprivation of input 367 Developmental plasticity 5, 263, 293 Dextro-amphetamine 152 Diaschisis 145 Diffusion tensor imaging (DTI) 52 Diffusion-weighted imaging (DWI) 144 Disorders of memory 25 Distraction 344 Disuse 60 DNA 36 Dopamine (DOPA) 152, 244 Dopaminergic (DA) 94 Dorsal root ganglion (DRG) 18 Dorsolateral PFC (DLPFC) 48 Doublecortin (DCX) 188 DWI/PWI lesion 149 Dysesthesia 358 Dysfunction of frontal systems 130 DYT-1 320 Electrical stimulation 285 Electromyographic activity (EMG) 269 Embryonic development 3 Enhance memory 97 neuromodulatory function 98 Enhanced performance of training 225 Enriched environments 66, 68, 69, 72, 75, 91, 118 Epidermal growth factor (EGF) 188 Epigenetic chromatin remodeling 25 factors 208 Episodic encoding 48 retrieval 50 Etiology and nature 366 Event-related potentials, ERP 95, 127 Evoked responses middle latency 296 Excitability of the motor pool 246 Excitatory synaptogenesis 17, 19 Expectation 347 Expression of neural plasticity 45, 367 Facilitate reorganization 151 Fast speech 89 Fibroblast growth factor (FGF) 188, 194 Fear conditioning 114, 188, 194
Focal hand dystonia 320 Frontal lobe dysfunction 123 syndrome 124 Frontal systems dysfunctions, consequences of 131 Functional magnetic resonance imaging (fMRI) 49, 88, 95, 125, 146, 174, 203, 226, 322, 343 Fusiform face area or FFA 127 FM sweeps 116 G-protein coupled receptors 27 GABA 204 GABAergic 190 granular 188 neurons 7 Gamma activity 95 Gene expression 7 Genomic imprinting 25 Gerontology 58 Glial-derived neurotrophic factor (GDNF) 212 Gliosis 68 Globus pallidus 124 Glutamate 204 Goal-management training program (GMT) 135 Golgi tendon organs 232 Grammatical impairment 150 Group II pathway 241 Growth factors 304 Glycosaminoglycan (GAG) 215 H-reflexes 232, 235, 266 Hand movement disorders 93 Healthy aging 83 humans 70 seniors 45 Hearing after congenital deafness 298 loss 284, 366 screenings 306 Hearing-impaired 89 Hebb’s principle 367 Hebbian 116 Hebbian mechanisms 349 Hematopoietic progenitors/stem 209, 215 tissue 209 Hemifacial spasm 370 Hemiplegic spastic poststroke patients 242 HERA model 51 High-threshold afferents 244
376 Higher order auditory cortex 292 Hippocampal recruitment 50 Hippocampus 25, 52, 59, 187, 202 Histone acetyltransferases (HATs) 35 Histone deacetylases (HDACs) 35 Homosynaptic depression, HD 235 Human AC133 (CD133) antigen 210 disability 8 thalamic somatotopy 357 thalamus 353 umbilical cord blood (hUCB) 209 Huntington’s disease 208 Hyperactivity 367 Hyperacusis 367 Hypoesthesia 360 Hypoperfusion 144 Hypothalamus 337 Ibotenic acid 159 Immediate affective 334 Implantation of stimulating electrodes 353 Inferotemporal cortex (ITC) 128 Inferior colliculus (IC) 4 Infraparietal cortex 341 Influence of background stimuli 114 Injury 225 Inhibition reciprocal 238 recurrent 232, 238 Insular cortex 338, 361 Insulin-like growth factor (IGF-1) 194 Intentional encoding 48 Interlimb reflexes 244 Intervention 81 Intracortical connections 203 facilitation (ICF) 322 Ischemic 225 injury 224 penumbra 144 stroke 144 Kindling 204 L-type voltage-dependent Ca2+ channels 69 Language 173 function lost and recovered 150 rehabilitation 152
Language-impaired children 95 Latency, evoked responses 296 Lateral PFC 125 spinothalamic tract 336, 338 Lateral column (LC) 273 Lateral geniculate nucleus (LGN) 4 Learning 151, 199 and reading impairments 95 disabilities 31 Left-hemisphere-only process 148 LH 323 Life, normal 263 Limbic system 91 Lipofuscin 68 Locomotor recovery 247 Locus for cortical reprogramming 165 Long-term depression (LTD) 59, 247, 295 implications of 203 potentiation (LTP) 26, 114, 151, 247 Low route 368 Me´nie`re’s disease 369 M-wave 235 Magnetic resonance imaging (MRI) 144 MAPK 19 MAP2 210 MCA (stroke) 326 MCI 96 Medial geniculate nucleus (MGN) 4–5 Medullary reticular formation (MRF) 248 MEG 149 Memory 13 disorders of 25 enhancement of 101 working 46 Metabolic diseases 215 Methyl transferases (DNMTs) 36 Microtubule 210 Microvascular decompression (MVD) 370 Middle cerebral artery occlusion (MCAO) 211 Middle ectosylvian gyrus (MEg) 158 Middle suprasylvian gyrus (MSg) 158 Middle suprasylvian sulcus (pMSs) 158 Mild cognitive impairment (MCI) 83 Mind–brain relationships 348 Mini-mental state examination (MMSE) 99 Mitogen-activated protein kinase (MAPK) 25 Molecular mechanisms of rewiring 7
377 Mononuclear hUCB (MNC hUCB) 210 Motor map reorganization 114 Motor cortex (M1) 92, 225, 317 stimulation 224, 226 map 203 motor systems 88 MRI 146 structural 45 Mucopolysaccharidosis (MPS) 215 Multimodal cells 8 Mutations of genes 25
Normal aging 124, 205, 208 cognitive decline 83 life 263
NB neurons 115 Naı¨ ve auditory system 286 Negative learning 87, 91 plasticity 92–93 in children 95 Neonatal spinal cord transection 264 Nerve growth factor (NGF) 18, 210 Nestin 189 Neural plasticity (see also plasticity) 54, 365 progenitor cells 189 stem and progenitor cells 187 after injury 190 stem cells 208 Neurodegenerative disorders 208 Neuroendocrine responses 338 Neurofibromatosis type 1 (NF1) 31 Neurogenesis 207 Neuromodulation 323 Neuromodulatory control systems 91 Neurotrophic factor, brain derived (BDNF) 15 Neuroprotection 216 Neurotrophins 304 Nigrostriatal system 94 Nimodipine 69 Nitric oxide synthase (NOS) 69 NMDA receptors 18, 25, 26, 59, 295 Nociceptive responses 224 stimulus 334 Nociceptive-specific (NS) 336 Noisy processing 87 Nonclassical auditory pathways 368 Nonfluent aphasia 323 Noradrenergic (NA) 241 agonist 152 Norepinephrine 152
Pain 227 control 224 unpleasantness, of 334 judgement of 347 Pain-related neuronal activity 353 Parabrachial nucleus, hypothalamus 338 Parahippocampal cortex 124 Paresthesia 365 Parietal regions surrounding lesions in the left hemisphere 147 Parkinsonian symptoms in rat 94 Passive unattended intervention 74 Pathological cognitive decline 83 Pediatric brain injury 173 Perfusion-weighted imaging (PWI) 144 Periaqueductal grey (PAG) 343 Persistent inward currents (PICs) 233 PET 147, 149, 203 Phantom limb 8, 366 sensations 365, 366–369 Pharmacological intervention 69 Phonophobia 366, 368 Phytochemicals 216 Piracetam 152 PKC 26 Plastic-adaptive changes 63 Plasticity 114, 203 and genetics 319 following lesions of pMSs cortex 162 ocular dominance 5 use-dependent 60 Plasticity based training program 90, 92–93, 95 Positron emission tomography (PET) 48, 146, 174, 317, 346 Postcentral gyrus (S1) 339 Posterior parietal cortex 157 suprasylvian gyrus (dPSg) 158, 161
Objective tinnitus 365 Ocular dominance plasticity 5 Oculomotor delayed response (ODR) 126 Operant conditioning of the SSR 268 Over-recruitment 53–54 Overtrained humans 93
378 Postnatal development 200, 264, 286 Poststroke central pain 360 Prefrontal cortex (PFC) 46, 123–124, 201 medial 224 Presbycusis 366–367 Presynaptic inhibition (PSI) 235 Primary auditory cortex (A1) 5, 287, 295 motor cortex (M1) 202 sensory cortex 111 visual cortex (V1) 4–5 Projected field – PF 353 Protein 2, 210 kinase A (PKA) 25 kinase C (PKC) 25 Pruning of excessive synapses 200 Quantum Zeno effect 348 Rat aging 69 elderly 92 Rate-sensitive depression 235 RBANS (repeatable battery for the assessment of neuropsychological status) 99 Reading impairments 95 Reaction time 129 Receptive field, RF 353 Reciprocal inhibition 238 Recklinghausen’s disease 31 Recovery enhancement stroke 227 from TBI 202 in complex discourse skills 173 of hand function 321 of language 144 of locomotion 247 of motor function 223 Recurrent inhibition 232, 238 Reflex pathways 232, 249 Region of interest, ROI 289 Regional cerebral blood flow (rCBF) 177 Rehabilitation 111 of PFC function 134 Relearning 151 Relocation of cerebral functions 169 Renshaw cells 232, 238
Reorganization in aphasia 145 of aged brain 61 Repetitive tasks 135 Rerouting of information 367 neural activity 368 retinal afferents 5 Restorative cortical plasticity 225 Retardation syndromes 28 Retinal inputs into 5 Retinoic acid (RA) 210 Rett syndrome 35 Reversing behavioral and neurochemical losses 94 language learning 95 Reward pathway 152 Rewired A1 7 Rewiring model 8 molecular mechanisms of 7 retinal afferents 5, 9 RF and PF maps 356 RF/PF mismatch 357 Right hemisphere (RH) 323 Rostroventral medulla (RVM) 343 RSK-2 33 Repetitive transcranial magnetic stimulation (rTMS) 322 RTT 35 Rubinstein–Taybi syndrome (RTS) 34 Schizophrenia 124 Secondary somatosensory cortex 111 Sensitive periods 304 Sensorimotor cortex 202 organization 61 Sensory pathways 5 Serotonin 152 Shearing of white matter 181 SI cortical maps 71 Single photon emission computed tomography (SPECT) 173 Skill learning 114 Somatosensory cortex 85, 92, 202 secondary 111 cortices S1/S2 336, 342 system 306 Spasticity 224, 233 SPECT 173 Speech perception 89 Spinach 216
379 Spinal cord 261 injuries treatment of 214 lesions 231, 247 proprioceptive reflexes 264 Spinal locomotion 249 locomotor pattern generator (LPG) 262 reflexes 232 stretch reflex (SSR) 266 Spinal-cord injury (SCI) 214, 232–233 Spino-parabrachio-amygdaloid 337 Spino-parabrachio-hypothalamic pathway 337 Spinothalamic tract (STT) 224, 227, 354 Spirulina 216 Stepping 247 Stimulation (rTMS) 273 Stimulation therapy 151 Stretch hyperreflexia 234 reflexes 233 Striatum 124 Stroke recovery enhancement 227 Strokes 124, 144, 202, 211, 223, 226 Structural MRI 45 Subgranular layer (SGL) 189 Subjective tinnitus 365 Substance addiction 124 Substantia nigra 124 Subventricular zone (SVZ) 188 Successful aging 58 Suffering 336 Sulfate, a glycosaminoglycan (GAG) 215 Superior colliculus (SC) 4, 338 Supplementary motor cortex (SMA) 317 Synaptic morphogenesis 18 Synaptogenesis 15–16 Syndromes 227 Task-relevant representations 125 Temporal integration 368 modulation 65, 115 processing 65 resolution abilities 89 Temporal mandibular joint (TMJ) 369 Thalamic hyperactivity 227 organization 353 pain 224 plasticity 353 stimulation 359
Thalamotomy 354 Therapies for ALS 213 Tinnitus 365 objective 365 retraining treatment (TRT) 370 subjective 365 Tissue plasminogen activator or TPA 211 Tizanidine 242 Transcranial magnetic stimulation (rTMS) 146, 202, 317, 370 Transection of the 273 Transgenic mice 189 Traumatic brain injury (TBI) 124, 173, 199, 214 Treatability of age-related changes 66 Treatment of injured brain and spinal cord 214 Trigeminal neuralgia 370 Tyrosine kinase B (TrkB) 15 Umbilical cord blood cells 214–215 Up-regulating dopamine cell function 94 Use-dependent neural plasticity 60 Ventralis intermedius – Vim 354 Ventrolateral PFC [VLPFC] 48 Verbal memory 89 Vertigo 365 Vestibular schwannoma 366 Visual activity 3 cognition 90 functions 157 memory 90 orienting 159 rewiring 9 Visual analog scale (VAS) 366 Visuospatial skills 90 Wada testing 146 Wear and tear 84 Wernicke’s area 143, 146, 147 Whisker 113 White matter shearing of 181 Wide dynamic range (WDR) 336 Williams (Beuren syndrome (WBS) 366 Windup 245 Withdrawal from painful stimuli 264 Word comprehension deficits 150 Working memory 46 Zipcode-binding protein-1 (ZBP1) 18
