The lung: Development, Aging and the Environment

Contents Preface, vii List of Contributors, ix Introduction, xiii Part I: Critical Events in Normal Lung Development and Aging 1 Lung Morphogenesis, Role of Growth Factors and Transcription Factors 3-12 Wellington V. Cardoso 2 Development of Airway Epithelium 13-32, Charles G. Plopper, Michelle V. Fanucchi 3 Development of the Airway Innervation 33-53, Malcolm P. Sparrow, Markus Weichselbaum, Jenny Tollet, Peter K. McFawn, John T. Fisher 4 Development of Alveoli 55-73, Stephen E. McGowan, Jeanne M. Snyder 5 Development of the Pulmonary Basement Membrane Zone 75-79, Michael J. Evans, Philip L. Sannes

v

10 Development of the Pulmonary Surfactant System 149-167, Sandra Orgeig, Christopher B. Daniels, Lucy C. Sullivan 11 Development of the Pulmonary Immune System 169-176, Lisa A. Miller 12 Development of Antioxidant and Xenobiotic Metabolizing Enzyme Systems 177-185, Michelle V. Fanucchi 13 Compensatory Lung Growth: Relationship to Postnatal Lung Growth and Adaptation in Destructive Lung Disease 187-199, Connie C.W. Hsia, Robert L. Johnson Jr, Ewald R. Weibel 14 Pulmonary Transition at Birth 201-211, Stuart B. Hooper, Richard Harding 15 Normal Aging of the Lung 213-233, Kent E. Pinkerton, Francis H.Y. Green

6 Development of the Pulmonary Vasculature 81-103, Part II: Environmental Influences on Lung Rosemary Jones, Lynne M. Reid Development and Aging 7 Developmental Physiology of the Pulmonary Circulation 105-117, Steven H. Abman

16 Pulmonary Consequences of Preterm Birth 237-251, Kurt H. Albertine, Theodore J. Psysher

8 Development of Fluid Transport Pulmonary Epithelia 119-129, Jonathan H. Widdicombe

17 Role of Nutrition in Lung Development Before and After Birth 253-266, Richard Harding, Megan L. Cock, Cheryl A. Albuquerque

9 Physical, Endocrine and Growth Factors in Lung Development 131-148, Stuart B. Hooper, Megan J. Wallace

18 Influence of High Altitude on Lung Development and Function 267-275, David P. Johns, David W. Reid

vi

Contents

19 Genetic Factors Involved in Susceptibility to Lung Disease 277-289, Steven R. Kleeberger

25 Environmental Toxicants and Lung Development in Experimental Models 345-351, Michelle V. Fanucchi, Charles G. Plopper

20 Effects of Environmental Tobacco Smoke on Lung Development 291-299, Jesse Joad

26 Repair of Environmental Lung Injury During Development 353-362, Suzette Smiley-Jewell, Laura S. Van Winkle

21 Nicotine Exposure During Early Development: Effects on the Lung 301-309, Gert S. Maritz

27 Effects of Aging, Disease and the Environment on the Pulmonary Surfactant System 363-375, Sandra Orgeig, Christopher B. Daniels

22 Exposure to Allergens During Development 311-319, Laurel J. Gershwin

28 Environmental Determinants of Lung Aging 377-395, Francis H.Y. Green, Kent E. Pinkerton

23 Development of Atopy in Children 321-331, David B. Peden

Index, Pages 397-403

24 Effects of Air Pollution on Lung Function Development and Asthma Occurrence 333-343, Frank D. Gilliland, Rob McConnell

A color plate section follows the index

From the moment of birth, until the time of death, the lung is an essential organ that provides our bodies with oxygen and eliminates the carbon dioxide we produce. Our ability to reach our physical and mental potential throughout our entire life span is strongly influenced by the efficient functioning of our lungs. Over the last few decades it has become increasingly apparent that both environmental and genetic factors operating during early life can induce persistent alterations in the function and health of the respiratory system. The principal objectives of this book are, firstly, to concisely present current concepts of normal processes involved in the growth, maturation and aging of the lung, and secondly, to integrate the growing body of evidence relating to the influence o f environmental and genetic factors on the structure and function of the lung and on respiratory health in later life. A third objective is to identify future directions for research into factors influencing respiratory health. We consider that these are important objectives as respiratory illness is a major contributor to morbidity and death at all stages of life, from birth to senescence. For example, asthma affects millions of children and adults worldwide, with both the incidence of asthma and the number of asthma-related deaths increasing. In the USA alone, the number of asthma cases has been estimated at 12 million, with the estimated cost of asthma-related care rising from 6.2 billion dollars in 1990 to over 10 billion dollars in 1995. The dramatic increase in the incidence and cost of this chronic inflammatory disorder has resulted in enormous research funding being directed towards its prevention and treatment. With the increasing use of molecular and cellular technologies, our knowledge of biological processes involved in the development of the respiratory organs has expanded tremendously; as a result, new concepts regarding the control of lung development have rapidly evolved. In parallel with our greater understanding of normal development is the realization that a wide range of environ-

mental factors can impact upon the genetic program of lung development. Many such factors can result in persistent alterations in lung structure and function that can, in turn, lead to an increased susceptibility to respiratory illness throughout postnatal life. For example, an increasing body of epidemiological data suggests that early childhood events such as premature birth, early respiratory infections or exposure to allergens can predispose the individual to airway dysfunction and common respiratory disorders such as asthma and chronic obstructive airway disease (COPD), increasing the risk of death from respiratory causes. It is also evident that genetic polymorphisms can affect an individual's susceptibility to a range of environmental factots such as allergens, cigarette smoke, nutrient restriction and infection, and these are only now becoming better understood. With the increasing interest in early developmental origins of ill-health and the role of the environment in human biology, we believe it is timely to review the scientific literature relating to these important health issues. Our purpose is to integrate current knowledge of the impact of environmental factors that can influence lung development, susceptibility to respiratory illness and the rate of aging of the lung. Each of these aspects of lung biology is of direct relevance to an understanding of respiratory health, a matter that is likely to become increasingly important in an aging population. In preparing this book, we have aimed at making it accessible to not only those working in lung biology, but also to non-experts with a broad interest in human health. Our hope is that this book will be of value to all concerned with respiratory health, including thoracic physicians, respiratory scientists, members of the pharmaceutical industry, toxicological and environmental regulators, pediatricians, perinatologists and gerontologists.

Richard Harding, Kent E. Pinkerton, and Charles G. Plopper

List of Contributors

Steven H. Abman Department of Pediatrics University of Colorado School of Medicine The Children's Hospital Denver, CO USA Kurt H. Albertine Department of Pediatrics, Medicine, Neurobiology, and Anatomy University of Utah School of Medicine Salt Lake City, UT USA Cheryl A. Albuquerque Department of Obstetrics and Gynecology Santa Clara Valley Medical Center San Jose, CA 95128 USA

Christopher B. Daniels Department of Environmental Biology University of Adelaide Adelaide, SA Australia Michael I. Evans Department of Anatomy, Physiology & Cell Biology School of Veterinary Medicine University of California, Davis Davis, CA USA Michelle V. Fanucchi Department of Anatomy, Physiology & Cell Biology School of Veterinary Medicine University of California, Davis Davis, CA USA

Wellington V. Cardoso Pulmonary Center Boston University School of Medicine Boston, MA USA

John T. Fisher Department of Physiology Queens University Kingston Ontario Canada

Megan L. Cock Department of Physiology Monash University Clayton, VIC Australia

Laurel I. Gershwin School of Veterinary Medicine University of California, Davis Davis, CA USA

Frank D. Gilliland

Rosemary Jones

Department of Preventive Medicine USC Keck School of Medicine Los Angeles, CA USA

Harvard Medical School and Department of Anesthesia and Critical Care Massachusetts General Hospital Boston MA USA

Francis H.Y. Green Department of Pathology & Laboratory Medicine University of Calgary Calgary Alberta Canada Richard Harding Department of Physiology Monash University Clayton, VIC Australia Stuart B. Hooper Department of Physiology Monash University Clayton, VIC Australia Connie C.W. Hsia Department of Internal Medicine University of Texas Southwestern Medical Center Dallas, TX USA

Jesse Joad

Steven R. Kleeberger Laboratory of Pulmonary Pathobiology National Institute of Environmental Health Sciences Research Triangle Park NC USA Gert S. Maritz Department of Physiological Sciences University of the Western Cape Bellville South Africa Rob McConnell Department of Preventive Medicine USC Keck School of Medicine Los Angeles, CA USA Peter K. McFawn Department of Physiology University of Western Australia Nedlands, WA Australia

Department of Pediatrics School of Medicine University of California, Davis Davis CA USA

Stephen E. McGowan Department of Internal Medicine Veterans Affairs Research Service University of Iowa Iowa City, IA USA

David P. Johns Discipline of Medicine University of Tasmania Clinical School Hobart Tasmania Australia

Lisa A. Miller Department of Anatomy, Physiology & Cell Biology School of Veterinary Medicine University of California, Davis Davis, CA USA

Robert L. Johnson, Jr

Sandra Orgeig

Department of Internal Medicine University of Texas Southwestern Medical Center Dallas, TX USA

Department of Environment Biology University of Adelaide Adelaide, SA Australia

David B. Peden Division of Allergy, Immunology & Environmental Medicine The School of Medicine University of North Carolina Chapel Hill, NC USA Kent E. Pinkerton Center for Health and the Environment University of California, Davis Davis, CA USA Charles G. Plopper School of Veterinary Medicine University of California Davis, CA USA Theodore J. Pysher Department of Pathology University of Utah School of Medicine Primary Children's Medical Center Salt Lake City UT USA David W. Reid Discipline of Medicine University of Tasmania Clinical School Hobart Tasmania Australia tynne M. Reid Department of Pathology Harvard Medical School Children's Hospital Boston, MA USA Philip t. Sannes Department of Molecular Biomedical Sciences College of Veterinary Medicine North Carolina State University Raleigh, NC USA

Suzette Smiley-lewell Department of Anatomy, Physiology and Cell Biology School of Veterinary Medicine University of California, Davis Davis, CA USA Jeanne M. Snyder Department of Anatomy and Cell Biology University of Iowa College of Medicine Iowa City, IA USA Malcolm P. Sparrow Department of Medicine University of Western Australia Nedlands, WA Australia Lucy C. Sullivan Department of Microbiology and Immunology University of Melbourne Melbourne, VIC Australia Jenny Toilet Department of Physiology University of Western Australia Nedlands, WA Australia Laura S. Van Winkle Department of Anatomy, Physiology and Cell Biology School of Veterinary Medicine University of California, Davis Davis, CA USA Megan I. Wallace Department of Physiology Monash University Clayton, VIC Australia Ewald R. Weibel Institute of Anatomy University of Bern Bern Switzerland

Markus Weichselbaum Department of Medicine University of Western Australia Nedlands, WA Australia

Jonathan H. Widdicombe Department of Human Physiology University of California, Davis Davis, CA USA

The respiratory system is one of the most complex organ systems in the mammalian body. In the lower respiratory tract alone, well over forty distinct cell phenotypes have been identified. These cell populations are distributed throughout a very complex architectural framework which includes a series of branching contiguous tubular structures which terminate in an even more complex series of septated surfaces, the alveoli. The cell phenotypes are distributed in a highly heterogeneous fashion. Their phenotypic expression and functional interactions with other phenotypes and the associated matrix are unique to the microenvironment in which they reside. This book addresses two general questions. During the development of the respiratory system, what are the critical events that lead to the complex mature organ system? Secondly, what is the impact of environmental and genetic factors on these developmental events? Early in embryonic development, the respiratory system begins as an outpocketing of the primitive embryonic gut into the adjacent mesenchyme. This initial budding consists of two cell phenotypes, one epithelial in nature from the endoderm and the other mesenchymal from the surrounding splanchnopleure. The transformation from this small aggregation of cells, about the size of a period on this page, to the highly complex series of structures we know as the larynx, trachea, extrapulmonary bronchi, intrapulmonary bronchi, bronchioles and the alveolar gas exchange area involves tremendous change and growth. This can be best appreciated by realizing that for the average human adult, the surface area of the gas exchange area is similar to that of a tennis court. The key events that produce these changes are (a) overall growth, (b) branching morphogenesis, (c) cellular proliferation, (d) cellular differentiation and (e) matrix formation. In addition to enormous overall growth, the formation of this complex architecture is the result of branching morphogenesis of the budding epithelium into the surrounding matrix. The event of branching itself, and the fact that the cells forming this complex architecture remain the same size as the airways grow, requires a very high rate of cellular proliferation of both epithelial and mesenchymal components. In fact,

the process of branching morphogenesis itself relies heavily on differential proliferation of adjacent epithelial cell populations combined with focal programmed cell death. As the number and size of branches continues to grow and the number of cells continues to increase, many of the cellular populations in specific focal areas begin to differentiate for specific functional roles. The complex architecture that makes up the branching tubular tracheobronchial airway tree and septated alveolar gas exchange area relies on the differential formation and reorganization of the vast complex of matrix and fibrous connective tissue structures which are continually undergoing synthesis, reorganization, and, in some cases, removal. Epithelial tissue undergoes four general categories of branching morphogenesis during lung development: epithelium branches into mesenchyme to form airways, into mesenchyme to form airway secretory glands, into mesenchyme to form blood vessels, and, together with extracellular matrix, epithelium forms interalveolar septa. The growth of the tracheobronchial tree and associated blood vessels are the initial branching events; these two processes lead to the formation of the conducting airways and large vessels, pulmonary arteries, pulmonary veins, and bronchial arteries. These events occur early in gestation and are generally complete prior to the end of gestation. While the tracheobronchial airways grow and mature, branching morphogenesis forms the submucosal glands within the airway walls. The process of alveolarization, which produces interalveolar septa, is a branching event that involves not only epithelium, but vascular endothelium as well. Both cell types are in close interaction with fibroblasts and connective tissue elements for the reorganization of primary septa into secondary septa which leads to the formation of definitive alveoli. The very large increase in the number of cells required for the growth in size and complexity of the developing lung does not occur uniformly in the developmental process. It varies according to (a) tissue and cell compartment, (b) gestational age, (c) status of cellular differentiation and (d) the local microenvironment. In contrast to adults, in

which cell proliferation is very low, cell proliferation in the developing lungs of fetuses and infants is very high. The pattern, rate, and actual phenotypes undergoing active proliferation vary greatly within the developing lung, depending upon which tissue compartment and which cell population within that compartment are involved at any particular time. Active proliferation continues for a significant portion of in utero lung development and is highly dependent on the status of cellular differentiation occurring in the same site. The local microenvironment, determined by position within the airway or vascular trees, and the mix of cell types at each site, dictate the focal proliferation rate. These proliferative and morphogenetic events result in vast increases in almost all components of the respiratory system. This growth includes not only an increase in the volume of the lungs and trachea as a whole, but also increases in alveolar air space volume, capillary blood volume, and the volume of large airways and vessels. There are major increases in surface area including that on the epithelial side and that on the capillary endothelial side of the alveolar blood-air barrier. In addition, there are tremendous increases in the numbers of all structures, including alveoli, capillaries, and, of course, all of the cell components that make up the entire system. The processes of cellular differentiation and cellular proliferation appear to compete with each other during development. Cellular differentiation within the respiratory system begins primarily in late gestation and after birth, and involves a tremendous increase in the diversity of cell types. This includes expression of differentiated function for at least fifteen different epithelial phenotypes, including elements derived from the neuroectoderm, along with the endodermal derivatives from the gut. A wide variety of mesenchymal cells, including immune and inflammatory cells recruited from the circulation, can differentiate locally. As with the differentiation of most cell populations within the respiratory system, the differential expression of enzyme systems critical for interaction between the organism and its environment also show a primarily postnatal pattern of expression. This includes enzymes involved in bioactivation of a wide range of organic compounds and enzyme systems which manage antioxidant pools and Phase II detoxification systems which transform reactive metabolites into non-reactive excretable chemicals. These enzymes

include cytochrome P450 monooxygenases, epoxide hydrolase, glutathione pool regulation enzymes, glutathione s-transferases, glucuronyl transferases, SOD and GPx, catalase. In Part 2 of this book, it becomes abundantly clear that all of the processes and events involved in the development of the respiratory system are susceptible to perturbation by environmental influences. In considering the impact of different environmental influences on lung and airway development, we need to bear in mind that all of these developmental processes represent a continuum beginning early in embryonic life and continuing through old age to the death of the organism. It appears that a very restricted number of these events are critical for successful life outside the uterus. Birth disarranges some of these events, but does not substantially impede their progress. There appears to be a small number of these processes that during fetal life must reach a certain stage to permit viability of the newborn. The gas exchange area must have grown to sufficient size in relation to metabolic body size to promote adequate oxygenation. The blood-air barrier must be of sufficient area and be sufficiently thin to allow adequate exchange of gases between the alveolar air surface and the alveolar capillary blood. Pulmonary surfactant and associated proteins must be produced in sufficient quantity and with a wide enough distribution to ensure lowering of surface tension below that required to allow cyclic alveolar expansion and collapse with minimal expenditure of energy. An additional factor that needs careful consideration when evaluating the impact of environmental factors on the developing lung is that all of these processes, including timing of events and the pattern in which they unfold, is highly variable depending on the species. This becomes especially problematic when attempting to establish models for evaluating the level of risk which exposure to environmental factors may pose for the developing human lung and for identifying the mechanisms by which environmental factors modify developmental processes in the lung. Clearly further research and appropriate animal models are required before we can fully understand the important role that environmental and genetic factors play in the development of the human lung. It is hoped that such research will ultimately lead to a better quality of lifetime respiratory health and greater longevity for the human population.

Critical Events in Normal Lung Development and Aging

ISBN 0 12 324751 9

Part 1

Copyright © 2004 Elsevier

Chapter

Lung Morphogenesis, Role of Growth Factors and Transcription Factors W e l l i n g t o n V, ...C . ardoso

.

.

.

.

.

.

.

.

.

.

.

1

.

Pulmonary Center, Boston University School of Medicine, Boston, MA, USA

INTRODUCTION

The term 'morphogenesis' refers to a coordinated series of molecular and cellular events that during development shape the structure of tissues and organs. In the lung this represents the process by which epithelial tubules and blood vessels are formed and patterned to ultimately generate the airways and alveoli. In most species this encompasses the pre- and postnatal period of life (Fig. 1.1). This chapter focuses on the mechanisms that regulate the initial events of lung morphogenesis and discusses how signaling molecules present in the early lung influence this process. Because most of the information available in the literature has been generated in mouse models, this species will be used as a reference throughout the text. Subjects such as regulation of lung cell differentiation and blood vessel formation are discussed in detail in other chapters and therefore will be only superficially reviewed here. Lung, thyroid, liver, and pancreas are examples of organs derived from the primitive foregut. 1 At an early stage of embryonic development, when gastrulation is completed, a single sheet of endodermal cells located outside the embryo invaginates to form the primitive gut tube. A number of transcription factors start to be expressed in the endoderm in overlapping but distinct domains along the anterior-posterior (A-P) axis of the gut tube. This roughly subdivides the gut endoderm into organ-specific domains in which specific cell fates are assigned. Subsequently, the endoderm in these areas undergoes bud morphogenesis to form organ primordia. Endodermal development is influenced not only by locally expressed transcription factors, but also by soluble factors that diffuse from adjacent cell layers to the endoderm. Soluble factors are critical in mediating paracrine interactions and establishing feedback loops that control The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

morphogenesis. It has been recently reported that a close contact between vascular endothelial cells and the gut endoderm is essential for induction of liver and pancreatic primordial buds. 2'3 Moreover, soluble factors secreted by developing structures in the neighborhood of the gut tube play a major role in the development of foregut derivatives. For example, fibroblast growth factor (Fgf) and bone morphogenetic protein (Bmp) signals from the heart and transverse septum (future diaphragm) are critical for the initial steps of hepatogenesis; 4'5 the notochord provides Fgf and activin signals necessary for initiation of a pancreatic program of gene expression by the endoderm. 6'7 The signals and mechanisms involved in specification of the endoderm to become lung are currently undetermined.

ONSET

OF

LUNG

DEVELOPMENT

Primary lung buds are identified in humans around the fourth week of embryonic life; however, in species such as the mouse or rat, buds emerge much later, at midgestation (embryonic days E9.5 and E11.5, respectively). 8'9 Endoderreal buds form from each ventro-lateral side of the foregut and invade the adjacent mesoderm; these buds then grow caudally and ventrally, joining each other at the midline to form the primordial lung. At the site where the primary buds connect (future carina), the trachea develops. 1~ The endoderm generates a great variety of specialized cells that constitute the respiratory epithelium, such as alveolar type I and II cells, and airway secretory mucous, serous, Clara and ciliated cells, among others. The lung mesenchyme originates from cells of the lateral plate mesoderm (splanchnic mesoderm), which at an early developmental stage migrate to the primitive foregut. Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

The:Lung: DeVelopme,~t;:ABir~gan:d ihe Environnient .

.

.

.

:i

.

Mesodermal-derived mesenchyme of the distal lung generates blood vessels by vasculogenesis. 11 Vascular structures also form by angiogenesis from vessels that migrate from the aortic arches to the area of the nascent lung. Veins originate from the left atrium. As the lung develops, vascular and airway components become closer at the distal end of the respiratory tree to form the future alveolar-capillary barrier. Other mesenchymal-derived structures are cartilage that surrounds large airways, airway and vascular smooth muscle and the connective tissue of the lung. The pleura, like other mesothelial membranes such as the pericardium, originates from the somatic mesoderm.

Molecular regulation of lung bud initiation Little is known about the mechanisms that control the initial events of lung development. One of the earliest signs of lung development is local expression of thyroid transcription factor I (Ttfl) by the endoderm. 12 This homeoboxcontaining transcription factor, also known as Nkx2.1, or thyroid-specific enhancer binding protein-1 (T/ebp-1), is found in thyroid primordia, forebrain, pituitary gland, and the epithelium of the lung. Ttfl marks the sites where thyroid and lung will form (Fig. 1.2). Analysis of Ttfl knockout mice shows that this gene, although important for normal branching and differentiation, is not essential for initiation of bud morphogenesis of these organs. 13'14 Information from genetically altered mice also implicates signaling by retinoids, hepatocyte nuclear factor (Hnf3~), the zinc finger transcription factor Gli, and Fgf as critical regulators of this process. Retinoids are vitamin A derivatives that play an important role in development and homeostasis of a variety of organs. Retinoic acid (RA) is the active form of retinoids and exists as several isoforms, including all trans and 9-cis. R A results from a multi-step oxidation process from retinol (vitamin A). 15 Retinaldehyde dehydrogenase-2 (Raldh2) appears to be the critical enzyme in generating RA during organogenesis. 16-1s RA effects are mediated by two families of nuclear receptors, RARs and RXRs. These receptors, each comprising three isotypes (t~, ~ and y), heterodimerize to form the functional unit that transduces RA signaling. 15'19RARs and RXRs are expressed throughout lung development beginning at the earliest stages. 2~ High levels of Raldh2 and ubiquitous activation of a retinoic acid responsive element R A R E - l a c Z transgene in a reporter mouse at the onset of lung development (Fig. 1.2) suggest that RA signaling is highly active in the lung primordia. 22 Several studies implicate RA as an essential signal for lung bud initiation. Retinoid deprivation in vitamin A deficient rats or administration of retinoid antagonists in cultured embryos result in lung agenesis. 23'24 Disruption of RA signaling in double RAR t~ and [3 knockout mice results in unilateral lung agenesis and lung hypoplasia. 25 The mechanisms involved in these abnormalities have not been established. Hnf3[3, recently renamed Foxa2, is a member of the Hnf-3/winged helix/forkhead family of transcription factors. Hnf3~l plays an essential role in early gut morphogenesis.

In Hnf3~ null mutant mice the endoderm fails to invaginate and the gut tube does not form. Consequently development of all gut derivatives is impaired and embryos die by day E9.5. 26 Interestingly, in normal mice Hnf3~ expression is maintained in the lung epithelium throughout embryonic and adult life, where it controls expression of differentiation genes. 27 Glis are zinc finger transcription factors expressed in the foregut mesoderm and in the developing lung mesenchyme. 28'29 Glis are believed to transduce signaling by Sonic hedgehog (Shh, discussed below). Individually or in combination, the three members of this family (Glil-3) appear to play different roles in lung development. Remarkably, when Gli2 and Gli3 are simultaneously inactivated in knockout mice, no lungs or trachea are formed and other foregut derivatives such as stomach and pancreas are hypoplastic. 3~This phenotype is intriguing for two reasons: first, it is more severe than that found in Shh null mice, in which lungs are formed, 31 suggesting that Gli2 and 3 may be signals shared by other pathways besides Shh; secondly, Glis have not been shown to be locally expressed at prospective sites of lung formation and it is not known whether they induce local expression of soluble factors to act on the endoderm. Thus it is unclear how the lack of Glis so dramatically affects lung bud initiation. Presumably Glis are critical for maintaining Hnf3~ expression and consequently overall endodermal survival. Supporting this hypothesis, low levels of Hnf3 have been described in Gli2 and Gli3 null mice. 3~ Primary lung bud formation requires activation of Fgf signaling by foregut endodermal cells. Mice lacking Fgfl0 or its receptor Fgfr2b do not have lungs. 32'33 As discussed later, Fgfl0 induces budding by binding to and activating Fgfr2 signaling in the endoderm. Fgfr2 is expressed throughout the foregut endoderm, while Fgfl0 is found in the mesoderm at sites of prospective lung bud formation. 34'35 Fgfl0 is a chemoattractant and a proliferation factor for endodermal and endodermal-derived epithelial cells. 36'37 The mechanism elicited by Fgfl0-Fgfr2 appears to be a rather general strategy to form buds; disruption of Fgfr2b signaling dramatically affects development of other foregut derivatives including thyroid and pancreas, and structures such as the limbs. 38

Formation of the trachea There is morphological and genetic evidence suggesting that the trachea and lungs originate by independent processes. A classical study from Spooner and Wessels 1~ shows that in mice, formation of lung buds precedes tracheal formation. A striking observation from Fgfl0 knockout mice is the absence of lungs in the presence of a well-formed and apparently normal trachea. 32 Thus, the mechanism elicited by Fgfl0-Fgfr2 to generate buds seems to be dispensable to form the tracheal tube. Interestingly, Fgfr2 is highly expressed in the endoderm of the prospective trachea but, besides Fgfl0, no other Fgf with similar high-affinity binding for this receptor has been reported in this region.

Lung Morphogenesis

Several mechanisms have been proposed to explain how the tracheal tube forms and separates from the developing foregut. It is currently accepted that once lung buds form and fuse in the midline, a septum growing from caudal to cranial regions separates tracheal and esophageal compartments. Alternatively it is thought that separation occurs by fusion of endodermal ridges growing from each side of the foregut; as they meet in the midline, two tubes form. 39'4~ In addition, a mechanism involving local activation of programmed cell death in the endoderm has been proposed. 41 Tracheo-esophageal fistula, a relatively common abnormality of human tracheal development, results from partial to complete lack of separation of the respiratory tract from the esophagus. 42 This abnormality has been reported in a number of knockout mice, including Shh-/-31 (see below), Ttfl _/_13 and Gli2-/-; Gli3 +/_.30 Retinoids are also essential for normal tracheal development because in vitamin A deficient rat embryos and RARtx/I]2 double mutant null mice, tracheo-esophageal fistula is observed. 23'25

BRANCHING

MORPHOGENESIS

To generate the bronchial tree, primary buds undergo branching morphogenesis. This patterning event, also found in other tree-like developing structures, involves bud outgrowth, bud elongation and subdivision of the terminal units by reiterated budding or by formation of clefts between buds. In mice it starts at around day El0.5, as secondary buds arise, and extends up to day -~E17, when the distal lung expands to form saccules (Fig. 1.1). 9,43 The role of epithelial-mesenchymal interactions in regulating branching morphogenesis and differentiation has been well demonstrated by classical and recent studies using embryonic lung cultures. These studies have demonstrated that the respiratory epithelium is able to change its pattern of growth and differentiation when recombined

with mesenchyme of different origins. This is best exemplified by the ectopic induction of distal lung buds from the tracheal epithelium when trachea is cultured in the presence of distal lung mesenchyme. 1~ Diffusible factors in concert with transcription factors form local networks that mediate these epithelial-mesenchymal interactions.

FGF signaling as a driving mechanism for branching The fibroblast growth factor family comprises more than 20 ligands which signal via four tyrosine kinase receptors (Fgfrl-4). Interactions of ligand and receptor with heparan sulfate proteoglycans are fundamental to form a stable Fgf-Fgfr complex and to properly transduce Fgf signaling. 45 Fgfs are found in a wide variety of species and their role in epithelial branching has been remarkably conserved during evolution. In Drosophila, expression of the Fgf ligand branchless (Bnl) is detected in cells near tracheal epithelial tubules at prospective sites of budding. Bnl acts as a chemotactic factor for the epithelium. Bnl diffuses to bind and activate an Fgfr (breathless, btl) expressed by the epithelium; epithelial cells then elongate and migrate toward Bnlexpressing cells, resulting in bud formation. 46 An analogous mechanism involving Fgfl0 and Fgfr2 appears to control branching of the developing mouse lung. These regulators are already present at the onset of lung development and are essential for bud initiation. 3z During branching Fgfl0 is expressed in a dynamic fashion in the lung mesenchyme; transcripts have been localized to sites where distal epithelial buds will form (Figs 1.2 and 1.3). In turn Fgfr2, which binds to Fgfl0 with high affinity, is evenly expressed along the respiratory tract epithelium and is locally activated by Fgfl0. 35'47 Studies in organ culture show that a heparin bead soaked in recombinant Fgfl0 placed near a mesenchyme-free lung epithelial explant induces the epithelium to migrate and proliferate toward the bead. Moreover, if the dynamic pattern of Fgfl0 is disrupted by engrafting

Pseudoglandular ~'%'.'~ Saccular t i .....] Canalicular Illllmlll Alveolar

......

I T

.............ii ................

...........

CONCEPTION

"

.......

"

111111111111111111111111111111111111111111

.....

I

BIRTH

.....

: ........ ' - -

Human . . . .

-I

ADULT

Fig. 1.1. Stages of lung development in humans and mice. During the pseudoglandular stage most branching morphogenesis occurs, and the lung has a gland-like appearance with epithelial tubules separated by thick mesenchyme; during the canalicular stage, airway branching is completed, the mesenchyme becomes thinner leading to an approximation between the epithelial tubules and blood vessels. During the saccular stage the distal lung expands to form saccules, and type I and type II cells differentiate. During the alveolar stage, septation of saccules gives rise to mature alveoli, d: day; PN: postnatal; w: week; y: years. (Reproduced with permission from Malpel S, Cardoso WV. Lung development. In: Encyclopedia of Life Sciences, Volume 11, London: Nature Publishing Group, 2002; p. 164.)

Fig. 1.2. Molecular regulators of lung morphogenesis. Retinoic acid synthesis and utilization in the foregut at the onset of lung (lu) development: high levels of expression of Raldh2 (A) and ubiquitous activation of RARE-lacZ transgene (B). (C) Ttfl expression in lung and thyroid (th) primordia. During branching morphogenesis (D) Fgfr2b is expressed throughout the respiratory tract epithelium, while (E) Fgfl0 and (F) Bmp4 are localized to the distal mesenchyme and distal epithelium, respectively. (G) Distal epithelial expression of Shh in lung culture. (H) Fgfl0 is a chemoattractant factor for distal epithelial buds; an Fgfl0 bead engrafted onto lung organ culture is engulfed by distal buds (black arrowhead). (A, B) E9.5 days; (C) El0 days; (D-F) E11.5 days; (G, H) E11.5 + 24 h lung culture. A, C, D-G: whole mount in situ hybridization; B: X-gal staining. Arrowheads point to signal in each panel. (See Color plate 1.)

an Fgfl0-containing bead onto cultured embryonic lungs, the local 'static expression' of Fgfl0 deviates airway growth toward the bead and alters the branching pattern. 36'37 The unique pattern of expression of Fgfl0 in the early lung suggests that Fgfl0 is involved in the spatial control of lung bud formation in vertebrates. Fgfl 0 and the Drosophila branchless may be evolutionarily related and may regulate epithelial morphogenesis via a similar mechanism. However,

in contrast to mice, tracheal budding in flies does not involve cell proliferation and occurs solely by cell migration. 46 During branching, the mesenchyme seems to modulate the responses of the epithelium to Fgfl0. There are in vitro data showing that in the absence of mesenchyme, Fgfl0 induces budding in both proximal and distal epithelial rudiments. However, when the epithelium and mesenchyme are cultured together, Fgfl0 elicits budding only in

Lung Morphogenesis

distal lung. 36'37 This is intriguing since Fgfr2b is expressed in both proximal and distal epithelium at seemingly equal levels. 47 Presumably the proximal mesenchyme contains factors that inhibit Fgfr2 activation by Fgfl0 or prevent Fgfl 0-induced bud morphogenesis from occurring.

Control of branching morphogenesis

Fig. 1.3. Signaling molecules in early lung development. (A) Expression of Fgfl0 in the distal mesenchyme activates Fgfr2b signaling in the epithelium and budding (arrow) is initiated. (B) As the bud grows, the tip bud epithelium interacts with Fgfl0-expressing cells; Fgfl0 induces Bmp4 and epithelial cell proliferation is inhibited; Shh inhibits Fgfl0 expression to extinguish the chemoattractant source. (C) Fgfl 0 appears at different sites to induce a new generation of buds (arrow). A cleft is formed in between buds. Tgf[3 1 at the subepithelial mesenchyme prevents local budding by inhibiting Fgfl0 expression and inducing synthesis of extracellular matrix components, which stabilize the cleft. Other factors such as epidermal growth factors (EGO,hepatocyte growth factors (Hgf), Pdgf and Fgf7 are expressed in the distal mesenchyme in a more diffuse fashion and contribute to the overall growth of the branching tubules besides having other functions. Model based on Bellusci et al. 35 and Lebeche et al. 48 (See Color plate 2.)

As branching morphogenesis proceeds, the exchange of signals between epithelial and mesenchymal cell layers of nascent buds establishes feedback loops that control airway size, branching patterns and cell fate. Signaling molecules differentially expressed at the tips of the branching tubules form a distal signaling centre that is critical in controlling these events (Figs 1.2 and 1.3). Correct branching requires a precise control of Fgfl0 levels in time and space. There is evidence that at least some of the factors that regulate Fgfl0 expression are present in the epithelium of the developing bud. When embryonic lung mesenchymal cells are cultured in the absence of the epithelium, Fgfl0 mRNA levels markedly increase, 48 suggesting that the epithelium secretes diffusible factors that are inhibitory for Fgfl 0 gene expression. One of these factors seems to be Sonic hedgehog (Shh). Shh is an important signaling molecule expressed in a proximal-distal gradient with the highest levels at tips of the bud epithelium (Fig. 1.2). Its major receptor, Patched-1 (Ptc 1), and downstream mediators of Shh signaling, the transcription factors Glil-3, are expressed in the mesenchyme. 29'49 Shh induces mesenchymal cell proliferation and regulates the expression of a number of mesenchymal genes. 31 Shh, either as a recombinant protein in organ cultures or overexpressed in lungs of transgenic mice, inhibits Fgfl0 expression. 35'48 Lungs from Shh knockout mice show disrupted airway branching and resemble rudimentary sacs, but proximaldistal differentiation is preserved. Interestingly, in these mice Fgfl0 expression is no longer focal as in wild type, but becomes rather diffuse. 31 Thus, distal epithelial expression of Shh in the growing bud may function to locally inhibit Fgfl0 expression in the mesenchyme and prevent widespread distribution of Fgfl0 signals. Expression of different Glis in the lung occurs in somewhat overlapping domains; however, sites such as the subepithelial mesenchyme show the highest levels of Glil. 29 The role of Gli genes in lung bud initiation has been previously discussed. Disruption of Gli3 gene expression leads to defects in specific lobes of the lung. 29 Fgf signaling and airway branching are also controlled by a family of cysteine-rich proteins collectively called Sprouty (Spry). First described in Drosophila, it gained this name because Spry mutant flies show an increased number of tracheal branches. 5~ Spry is induced by the FGF ligand branchless at the bud tips and results in inhibition of lateral budding. In mice, four family members have been identified. 51 Spry2 and Spry4 are expressed in the developing distal lung in the epithelium and mesenchyme, respectively. 52-54 Disruption of Spry2 in lung cultures using anti-sense oligonucleotides shows a stimulatory effect on distal branching

: The~: E:Ung: Development; :Aging andthe Envi ~onment~

and differentiation. 52 By contrast, overexpression of Spry2 in the distal lung epithelium of transgenic mice inhibits branching and epithelial cell proliferation. 54 Bone morphogenetic protein-4 (Bmp4) is a member of the T g ~ superfamily of growth factors and is expressed during branching morphogenesis at high levels and in a dynamic fashion at the bud tip epithelium (Fig. 1.2). 55,56 Bmp4 signaling is transduced by type I and type II serinethreonine kinase receptors and Smad transcription factors. 57 In the developing lung Bmp4 appears to restrict cell proliferation and assign a distal cell fate to the bud epithelium. Targeted disruption of Bmp4 signaling in the lung epithelium of transgenic mice results in proximalization. In these mice, the peripheral lung is populated by proximal cell phenotypes, such as ciliated cells or secretory Clara cells. 56 In turn, overexpression of Bmp4 in the distal epithelium results in small lungs containing distal flat cells that resemble alveolar type I cells. 55 Fgfl0 controls Bmp4 levels. In organ cultures, Bmp4 expression is induced in the distal epithelium that surrounds an engrafted Fgfl0 bead. 37'48 Bmp4 antagonizes the Fgfl0 effects by inhibiting distal epithelial cell proliferation and preventing bud formation (Fig. 1.3). Bmp signaling is also controlled by antagonists such as Noggin, Chordin and the Cerberus-related factor Cerl, 56'57 all expressed in the developing lung. Noggin is a secreted molecule that binds to Bmp4 with high affinity and prevents it from binding to Bmp receptors. During branching Noggin is expressed at low levels in the distal lung mesenchyme until around day 13.5; Noggin is also expressed in the dorsal mesenchyme of the trachea in a pattern complementary to Bmp4. 56 Noggin null mutants, however, do not seem to have abnormal lungs. 58 Several other Tgf~ superfamily members are present in branching airways. Tgf~ 1 transcripts are uniformly expressed throughout the subepithelial mesenchyme, although the protein accumulates at sites of cleft formation and along proximal airways. 48's9 Tgffl 1 inhibits Fgfl0 expression and is a potent negative regulator of epithelial cell proliferation, differentiation and branching in lung organ cultures. 6~ Moreover, Tgf]] 1 induces synthesis of extracellular matrix which, when deposited in the epithelial-mesenchymal interface, is thought to prevent local branching (Fig. 1.3). 59 Recent data suggest that establishment of a distal signaling centre and branching are antagonized by RA signaling. In contrast to its role in primary lung bud initiation (see above), RA does not seem to be necessary for branching morphogenesis. As shown by an RARE-lacZ reporter mouse, the appearance of secondary buds and subsequent branching is marked by a sharp downregulation of RA signaling in the lung in both epithelium and mesenchyme. 22 Studies in RA-treated lung organ cultures show that preventing this downregulation from occurring with exogenous RA disrupts distal lung formation and maintains the lung in a proximallike immature stage. 61'62 In these cultures, levels and distribution of genes involved in distal lung formation are markedly altered; Fgfl0 expression is inhibited and Bmp4 signals at bud tips are low and diffuse. 22

:i:~

:::~::::::::~:~~::~ ~

Another factor that is essential for airway branching and differentiation is Ttfl (Nkx2.1). Ttfl is expressed in the lung epithelium from its earliest stages; as branching proceeds, a proximal-distal gradient is identified with the highest levels at the tips. 12 Ttfl knockout mice have highly abnormal dilated hypoplastic lungs, airway branching is markedly disrupted and overall development appears to be arrested at early pseudoglandular stage. 13 The role of Hox proteins in lung pattern formation remains controversial. In Drosophila, mutation of these transcription factors results in dramatic changes in anterior-posterior specification of the body plan. 63 Many Hox genes are expressed in the developing lung, 64 however, only Hoxa5 shows lung abnormalities when inactivated in knockout mice. Lungs from Hoxa5-/- mice are immature and tracheal development is disrupted. 65 Left-right a s y m m e t r y One of the least understood and most intriguing patterning events in organogenesis is the establishment of left-right (L-R) symmetry. Left and right lungs have highly stereotypical but different branching patterns and number of lobes, which vary according to species. L-R patterning of the lung is linked to the general body plan and is actually initiated well before lungs are formed. Lefty-1 and-2, nodal and Pitx-2 have been identified as major regulators of L-R asymmetry in viscera. 66 When expression of these transcription factors is disrupted in mice, laterality defects known as pulmonary isomerisms are found. 67 These defects are characterized by abnormally symmetric lungs. In wild-type mice, the left and right lungs consist of one and four lobes, respectively; however, in Lefty knockout mice single-lobed lungs are found on each side. 67 Paradoxically, while branching is influenced by Lefty-1, this regulator is not expressed by the developing lung. Like Lefty-2 and nodal, Lefty-1 is expressed only during a short window of time around days E8-8.5, on the left side of the prospective floor plate and lateral plate mesoderm. This suggests that some patterning decisions have already occurred when organ primordia arise. Other signaling molecules such as Shh, RA, Gli and activin receptor IIb have been implicated in L-R asymmetry in the lung. 3~

REGULATION OF PROXIMAL-DISTAL DIFFERENTIATION Although defined cell phenotypes are mostly seen around the time of birth, morphological and molecular features of differentiation can be detected much earlier, when the respiratory tree is being formed. The first molecular marker of differentiation expressed by lung epithelial cells is the surfactant protein gene SP-C. SP-C has been detected in the mouse lung in the distal epithelium at around day E10.5, when secondary buds form. 7~Other surfactant protein genes, SP-A, SP-B and SP-D, are consistently detected 3-4 days later. 71-73 By this time, morphological differeflces between

Lung Morphogenesis distal (cuboidal) and proximal (tall columnar) cells are obvious. Differentiation of the distal epithelium into type I and type II cells occurs around the period of sacculation (in mice day - E l 7 ) . 9 Type II cells express surfactant protein genes and contain lamellar bodies, cytoplasmic inclusions that store surfactant material. Type I cells are characteristically flat and express the markers Aquaporin 5 and TI~. 74'75 Regulation of the type I cell phenotype is still little understood (see Chapter 9). It has been suggested that factors such as Bmp4 may favor formation of type I over a type II cell phenotype. 55 By contrast, Fgf7 inhibits type I cell markers and induces surfactant protein gene expression. 75,76 Signaling molecules involved in early patterning events have been shown to also act as regulators of epithelial differentiation at later developmental stages. Ttfl, Hnf3[3 and Gata6 induce surfactant protein gene expression in lung epithelial cell lines. 77 T g ~ l , besides inhibiting branching in vitro, prevents distal lung maturation when ectopically expressed in the distal lung epithelium of transgenic mice. TM The Hnf3/forkhead homologue transcription factor Hfh4 (Foxjl) is essential for the development of ciliated cells and also serves as a proximal marker of lung differentiation. Hfh4 is expressed in the mouse lung from day El4 to El5 onwards. 79 Loss or gain of function of Hfh4 in genetically altered mice leads, respectively, to absence or ectopic formation of ciliated cells in target organs. Hfh4 knockout mice, besides lacking ciliated cells, also display laterality defects in internal organs, s~ In Spc-Hfh4 transgenic mice, misexpression of Hfh4 in the distal lung leads to the appearance of ciliated cells in peripheral areas and suppression of surfactant protein expression, sl Clara cell 10kDa protein (CC10) is expressed by non-ciliated proximal epithelial cells of secretory nature; transcripts are first identified around day El6 in murine lungs. Ttfl, Hnf3 and C/EBP transcription factors have been reported to be regulators of CC10 expression. 73'82

CONCLUSIONS The recent use of molecular and genetic tools to approach classical questions in developmental biology has advanced the current understanding of how signaling molecules control organogenesis. While much information has been generated regarding early lung morphogenesis, less is known about the molecular regulation of alveolarization. This formidable patterning event that takes place in the distal lung subdivides the pre-existing saccules into smaller units, the mature alveoli, and greatly increases the surface area for gas exchange. Septation of the primary saccules appears to be dependent on interstitial myofibroblasts and seems to occur under tight control of elastin levels. There is evidence that signaling by platelet-derived growth factor (Pdgf), Fgf and retinoic acid is involved in this process. Pdgf signaling is necessary for the development of lung myofibroblasts; RA is thought to promote elastin synthesis while Fgfr3 and -4 control elastin gene expression; glucocorticoid hormones

9

oppose the RA effects. 83-86 During neonatal and adult life, expression of signaling molecules such as Tgf~s, Fgfs and Shh has been reported in the lung, presumably to maintain homeostasis. 87-89 There is evidence that at least some of these signals are recruited to mediate cellular activities during injury-repair. 87 Studies aimed at understanding the molecular basis of lung development may provide useful insights into lung regeneration that can be potentially applied to areas such as stem cell therapy and tissue engineering for lung diseases.

REFERENCES 1. Wells JM, Melton DA. Vertebrate endoderm development. Ann. Rev. Cell Dev. Biol. 1993; 15:393-410. 2. Matsumoto K, Yoshitomi H, Rossant Jet al. Liver organogenesis promoted by endothelial cells prior to vascular function. Science 2001; 294:559-63. 3. Lammert E, Cleaver O, Melton D. Induction of pancreatic differentiation by signals from blood vessels. Science 2001; 294:564-7. 4. Jung J, Zheng M, Goldfarb M e t al. Initiation of mammalian liver development from endoderm by fibroblast growth factors. Science 1993; 284:1998-2003. 5. Rossi JM, Dunn NR, Hogan BL etal. Distinct mesodermal signals, including BMPs from the septum transversum mesenchyme, are required in combination for hepatogenesis from the endoderm. Genes Dev. 2001; 15:1998-2009. 6. Hebrok M, Kim SK, Melton DA. Notochord repression of endodermal Sonic hedgehog permits pancreas development. Genes Dev. 1998; 12:725-31. 7. Kim SK, Hebrok M, Melton DA. Notochord to endoderm signalling is required for pancreas development. Development 1997; 124:4243-52. 8. Pringle KC. Human fetal lung development and related animal models. Clin. Obst. Gynecol. 1986; 29:502-13. 9. Ten Have-Opbroek. The development of lung in mammals: an analysis of concepts and findings. Am. J. Anat. 1981; 162:201-19. 10. Spooner BS, Wessels NK. Mammalian lung development: interactions in primordium formation and bronchial morphogenesis. J. Exp. Zool. 1970; 175:445-54. 11. DeMello D, Sawyer D, Galvin Net al. Early fetal development of lung vasculature. Am. J. Resp. Cell Mol. Biol. 1997; 16:568-81. 12. Lazzaro D, Price M, de Felice M et al. The transcription factor TTF-1 is expressed at the onset of thyroid and lung morphogenesis and in restricted regions of the fetal brain. Development 1991; 113:1093-104. 13. Minoo P, Su G, Drum H et al. Defects in tracheoesophageal and lung morphogenesis in Nkx2.1(-/-) mouse embryos. Dev. Biol. 1999; 209:60-71. 14. Kimura S, Ward JM, Minoo P. Thyroid-specific enhancerbinding protein/thyroid transcription factor 1 is not required for the initial specification of the thyroid and lung primordia. Biochimie 1999; 81:321-7. 15. Chambon P. A decade of molecular biology of retinoic acid receptors. FASEB J. 1996; 10:940-54. 16. Niederreither K, McCaffery P, Drager U etal. Restricted expression and retinoic-induced downregulation of the retinaldehyde dehydrogenase type 2 (RALDH-2) gene during mouse development. Mech. Dev. 1997; 62:67-78. 17. Niederreither K, Subbarayan V, Dolle P etal. Embryonic retinoic acid synthesis is essential for early mouse postimplantation development. Nat. Genet. 1999; 21:444-8.

The Lung: Developmenti :Aging and~ihe Environment ~:~ ~ :~

18. Zhao D, McCaffery P, Ivins KJ et al. Molecular identification of a major retinoic acid-synthesizing enzyme: a retinaldehyde-specific dehydrogenase. Eur. J. Biochem. 1996b; 240:15-22. 19. Kastner P, Mark M, Ghyselinck N et al. Genetic evidence that the retinoid signal is transduced by heterodimeric RXR/RAR functional units during mouse development. Development 1997; 124:313-26. 20. Dolle P, Ruberte E, Leroy P et al. Retinoic acid receptors and cellular retinoid binding proteins. I. A systematic study of their differential pattern of transcription during mouse organogenesis. Development 1990; 110:1133-51. 21. Mollard R, Viville S, Ward SJ et al. Tissue-specific expression of retinoic acid receptor isoform transcripts in the mouse embryo. Mech. Dev. 2000; 94:223-32. 22. Malpel SM, Mendelsohn C, Cardoso WV. Regulation of retinoic acid signalling during lung morphogenesis. Development 2000; 127:3057-67. 23. Dickman ED, Thaller C, Smith SM. Temporally regulated retinoic acid depletion produces specific neural crest, ocular and nervous defects. Development 1997; 124:3111-21. 24. Mollard R, Ghyselinck NB, Wendling O et al. Stage-dependent responses of the developing lung to retinoic acid signalling. Int. J. Dev. Biol. 2000; 44:457-62. 25. Mendelsohn C, Lohnes D, Decimo D et al. Function of retinoic acid receptors (RARs) during development. II. Multiple abnormalities at various stages of organogenesis in RAR double mutants. Development 1994; 120:2749-71. 26. Ang S-L, Rossant J. HNF-3 is essential for node and notochord formation in mouse development. Cell 1994; 78:561-74. 27. Stahlman MT, Gray ME, Whitsett JA. Temporal-spatial distribution of hepatocyte nuclear factor-3beta in developing human lung and other foregut derivatives. J. Histochem. Cytochem. 1998; 46:955-62. 28. Hui C-C, Slusarski D, Platt KA et al. Expression of three mouse homologs of the Drosophila segment polarity gene cubitus interruptus, Gli, Gli-2, and Gli-3, in ectoderm- and mesoderm-derived tissues suggests multiple roles during postimplantation development. Dev. Biol. 1994; 162:402-13. 29. Grindley JC, Bellusci S, Perkins D etal. Evidence for the involvement of the Gli gene family in embryonic mouse lung development. Dev. Biol. 1997; 188:337-48. 30. Motoyama J, Liu J, Mo R et al. Essential function of Gli2 and Gli3 in the formation of lung, trachea and oesophagus. Nature Genet. 1998; 20:54-7. 31. Pepicelli CV, Lewis P, McMahon A. Sonic hedgehog regulates branching morphogenesis in the mammalian lung. Cur. Biol. 1998; 8:1083-6. 32. Sekine K, Ohuchi H, Fujiwara M et al. Fgf-10 is essential for limb and lung formation. Nature Genet. 1999; 21:138-4 1. 33. De Moerlooze L, Spencer-Dene B, Revest J-M etal. An important role for the IIIb isoform of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling during mouse organogenesis. Development 2000; 127:483-92. 34. Orr-Urtreger A, Bedford MT, Burakova T et al. Developmental localization of the splicing alternatives of fibroblast growth factor receptor-2 (FGFR2). Dev. Biol. 1993; 158:475-86. 35. Bellusci S, Grindley J, Emoto H et al. Fibroblast growth factor 10 (FGF10) and branching morphogenesis in the embryonic mouse lung.Development 1997; 124:4867-78. 36. Park WY, Miranda B, Lebeche D et al. FGF-10 is a chemotactic factor for distal epithelial buds during lung development. Dev. Biol. 1998; 201:125-34. 37. Weaver M, Dunn NR, Hogan BL. Bmp-4 and Fgfl0 play opposing roles during lung bud morphogenesis. Development 2000; 127:2695-704.

~

:i~

~

38. Ohuchi H, Hori Y, Yamasaki M e t al. FGF10 acts as a major ligand for FGF receptor 2 IIIb in mouse multi-organ development. Biochem. Biophys. Res. Commun. 2000; 277:643-9. 39. Sutliff KS, Hutchins GM. Septation of the respiratory and digestive tracts in human embryos: crucial role of the tracheoesophageal sulcus. Anat. Rec. 1994; 238:237-47. 40. Zaw-Tun HA. The tracheoesophageal septum - Fact or fantasy? ActaAnat. 1982; 114:1-21. 41. Zhou, Hutson JM, Farmer PJ, Hasthorpe S etal. Apoptosis in tracheoesophageal embryogenesis in rat embryos with or without adriamycin treatment, ft. Pediatr. Surg. 1999; 34:872-5. 42. Landing BH, Dixon LG. Congenital malformations and genetic disorders of the respiratory tract (larynx, trachea, bronchi and lungs).Am. Rev. Respir. Dis. 1979; 120:151-85. 43. Hogan BLM. Morphogenesis. Cell 1999; 96:225-33. 44. Shannon JM. Induction of alveolar type II cell differentiation in fetal tracheal epithelium by grafted distal lung mesenchyme. Dev. Biol. 1994; 166:600-14. 45. Szebenyi G, Fallon JF. Fibroblast growth factors as multifunctional signalling factors. Int. Rev. Cytol. 1999; 185:45-106. 46. Sutherland D, Samakovlis C, Krasnow MA. Breathless encodes a Drosophila FGF homolog that controls tracheal cell migration and the pattern of branching. Cell 1996; 87:1091-101. 47. Cardoso WV, Ito A, Nogawa H et al. FGF-1 and FGF-7 induce distinct patterns of growth and differentiation in embryonic lung epithelium. Dev. Dyn. 1997; 208:398-405. 48. Lebeche D, Malpel S, Cardoso WV. Fibroblast growth factor interactions in the developing lung. Mech. Dev. 1999; 86:125-36. 49. Bellusci S, Furuta Y, Rush MG et al. Involvement of Sonic hedgehog (Shh) in mouse embryonic lung growth and morphogenesis. Development 1997; 124:53-63. 50. Hacohen N, Kramer S, Sutherland D et al. Sprouty encodes a novel antagonist of FGF signalling that patterns apical branching of the Drosophila airways. Cell 1998; 92:253-63. 51. Minowada G, Jarvis LA, Candace L e t al. Vertebrate sprouty genes are induced by FGF signalling and can cause' chondrodysplasia when overexpressed. Development 1999; 126:4465-75. 52. Tefft DT, Lee M, Smith S etal. Conserved function ofmSpry-2, a murine homolog of Drosophila sprouty, which negatively modulates respiratory organogenesis. Current Biol. 1999; 9:219-22. 53. DeMaximy AA, Nakatake Y, Moncada S et al. Cloning and expression pattern of a mouse homologue of Drosophila sprouty in the mouse embryo. Mech. Dev. 1999; 81:213-16. 54. Mailleux AA, Tefft D, Ndiaye D etal. Evidence that SPROUTY2 functions as an inhibitor of mouse embryonic lung growth and morphogenesis. Mech Dev. 2001; 102:81-94. 55. Bellusci S, Henderson R, Winnier Get al. Evidence from normal expression and targeted misexpression that bone morphogenetic protein-4 (Bmp-4) plays a role in mouse embryonic lung morphogenesis.Development 1996; 122:1693-702. 56. Weaver M, Yingling JM, Dunn NR etal. Bmp signalling regulates proximal-distal differentiation of endoderm in mouse lung development. Development 1999; 126:4005-15. 57. Massague J, Chen YG. Controlling TGF signalling. Genes Dev. 2000; 14:627-44. 58. Brunet LJ, McMahon JA, McMahon AP et al. Noggin, cartilage morphogenesis, and joint formation in the mammalian skeleton. Science 1998; 280:1455-7. 59. Heine UI, Munoz EF, Flanders KC etal. Colocalization of TGF-[31 and collagen I and III, fibronectin, and glycosaminoglycans during lung branching morphogenesis. Development 1990; 109:29-36. 60. Serra R, Moses HL. pRb is necessary for inhibition of N-myc expression by TGF-1 in embryonic lung organ cultures. Development 1995; 121:3057-66.

Lung Morphogenesis 61. Cardoso WV, Williams MC, Mitsialis SA et al. Retinoic acid induces changes in the pattern of airway branching and alters epithelial cell differentiation in the developing lung in vitro.Am. J. Respir. Cell Mol. Biol. 1995; 12:464-76. 62. Cardoso WV, Mitsialis SA, Brody JS et al. Retinoic acid alters the expression of pattern-related genes in the developing rat lung. Dev. Dyn. 1996; 207:47-59. 63. Lewis EB. A gene complex controlling segmentation in Drosophila. Nature 1978; 276:565-70. 64. Kappen C. Hox genes in the lung. Am. J. Respir. Cell Mol. Biol. 1996; 15:156-62. 65. Aubin J, Lemieux M, Tremblay M etal. Early postnatal lethality in Hoxa-5 mutant mice is attributable to respiratory tract defects. Dev. Biol. 1997; 192:432-45. 66. Mercola M, Levin M. Left-right asymmetry determination in vertebrates. Annu. Rev. Cell Dev. Biol. 2002; 17: 779-805. 67. Meno C, Shimono A, Saijoh Y etal. Lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 1998; 94:287-97. 68. Tsukui T, Capdevila J, Tamura K etal. Multiple left-right asymmetry defects in Shh(-/-) mutant mice unveil a convergence of the Shh and retinoic acid pathways in the control of Lefty-1. Proc. Natl. Acad. Sci. USA 1999; 96:11376-81. 69. Suk OP, Li E. The signalling pathway mediated by the type IIB activin receptor controls axial patterning and lateral asymmetry in the mouse. Genes Dev. 1997; 11:1812-26. 70. Wert S, Glasser SW, Korfhagen TR etal. Transcriptional elements from the human SP-C gene direct expression in the primordial respiratory epithelium of transgenic mice. Dev. Biol. 1993; 156:426-43. 71. Wuenschell CW, Sunday ME, Singh G etal. Embryonic mouse lung epithelial progenitor cells co-express immunohistochemical markers of diverse mature cell lineages. J. Histochem. Cytochem. 1996; 44:113-23. 72. Botas C, Poulain F, Akiyama J e t al. Altered surfactant homeostasis and alveolar type II cell morphology in mice lacking surfactant protein D. Proc. Natl. Acad. Sci. USA 1998; 9:11869-74. 73. Zhou L, Lim L, Costa RH et al. Thyroid transcription factor1, hepatocyte nuclear factor-3beta, surfactant protein B, C, and Clara cell secretory protein in developing mouse lung. J. Histochem. Cytochem. 1996; 44:1183-93. 74. Williams MC, Cao Y, Hinds A et al. T1 alpha protein is developmentally regulated and expressed by alveolar type I cells, choroid plexus, and ciliary epithelia of adult rats. Am. J. Respir. Cell. Mol. Biol. 1996; 14:577-85. 75. Borok Z, Lubman RL, Danto SI et al. Keratinocyte growth factor modulates alveolar epithelial cell phenotype in vitro:

76. 77. 78. 79.

80.

81.

82. 83. 84. 85. 86. 87. 88.

89.

expression of aquaporin 5. Am. J. Respir. Cell Mol. Biol. 1998; 18:554-61. Borok Z, Danto SI, Lubman RL etal. Modulation of tlalpha expression with alveolar epithelial cell phenotype in vitro. Am. J. Physiol. 1998; 275:L155-64. Perl A-KT, Whitsett JA. Molecular mechanisms controlling lung morphogenesis. Clin. Genet. 1999; 56:14-27. Zhou LJ, Dey C, Wert SE et al. Arrested lung morphogenesis in transgenic mice bearing an SP-C-TGF-1 chimeric gene. Dev. Biol. 1996; 175:227-38. Hackett BB, Brody SL, Liang M e t a l . Primary structure of hepatocyte nuclear factor/forkhead homologue 4 and characterization of gene expression in the developing respiratory system and reproductive epithelium. Proc. Natl. Acad. Sci. 1995; 92:4249-53. Chen J, Knowles HJ, Hebert JL et al. Mutation of the mouse hepatocyte nuclear factor/forkhead homologue 4 gene results in an absence of cilia and random left-right asymmetry. J. Clin. Invest. 1998; 102:1077-82. Tichelaar JW, Lira L, Costa RH etal. HNF3/forkhead homologue-4 influences lung morphogenesis and respiratory epithelial cell differentiation in vivo. Dev. Biol. 1999; 213:405-17. Nord M, Cassel TN, Braun H e t al. Regulation of the Clara cell secretory protein/uteroglobin promoter in lung. Ann. N Y Acad. Sci. 2000; 923:154-65. Bostrom H, Willetts K, Pekny M e t al. PDGF-A signalling is a critical event in lung alveolar myofibroblast development and alveogenesis. Cell 1996; 85:863-73. Weinstein M, Xu X, Ohyama K et al. FGFR-3 and FGFR-4 function cooperatively to direct alveogenesis in the murine lung. Development 1998; 125:3615-23. Tschanz SA, Damke BM, Burri PH. Influence of postnatally administered glucocorticoids on rat lung growth. Biol. Neonate 1995; 68:229-45. Massaro GD, Massaro D. Retinoic acid treatment partially rescues failed septation in rats and in mice. Am. J. Physiol. 2000; 278:L955-60. Holgate ST, Davies DE, Lackie PM etal. Epithelialmesenchymal interactions in the pathogenesis of asthma. J. AUerg. Clin. Immunol. 2000; 105:193-204. Echelard Y, Epstein DJ, St-Jacques B etal. Sonic hedgehog, a member of a family of putative signalling molecules, is implicated in the regulation of CNS polarity. Cell 1993; 75:1417-30. Gebert JF, Moghal N, Frangioni JV et al. High frequency of retinoic acid receptor beta abnormalities in human lung cancer. Oncogene 1991; 6:1859-68.

Development of Airway Epithelium Charles G. Plopper* and Michelle V. Fanucchi SchoolUepartm ent ot An atomy, Ph ys,ology a n d Cell B,o, ogy, of Veterinary Med. icine, Uni:versitg:of:California, uaws, c,~, ua,~ r ' ~

_

_

.....

__,_

INTRODUCTION

A number of developmental processes are involved in the establishment of the tracheobronchial airway tree. The pattern of branching of the airways including the angle of branching and the proportions of daughter branches in relation to parent airway appear to be established relatively early by the process of branching morphogenesis. As summarized in detail in Chapter 1, this process is initiated with the earliest formation of respiratory tract structures in the thorax in the embryonic period and continues for a substantial period of time during early gestation. It is heavily dependent on epithelial-mesenchymal contact and continual interaction to regulate the rate and pattern of formation. The composition of the wall of the airways in adults varies substantially between different segments, with most of the differences being highly polarized from more proximal airways to more distal airways. The major components of the wall include: (1) the surface lining epithelium with its associated derivative, the submucosal gland, (2) the basement membrane zone, including basal lamina and an extended population of fibroblasts, and (3) bundles of smooth muscle and cartilage. The distribution of all these components varies substantially within the airway tree in adults. The entire wall is invested with a large number of nerves that appear to be in two separate distributional patterns, one associated with the epithelial surface and another associated with the glands and smooth muscle in the submucosa and adventitia. As detailed in Chapter 3, the formation of *To whom correspondence should be addressed. The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

the nerves occurs early in development, once the pattern of the airway tree has been laid down. The presence of nerves in the wall, however, does not establish that they have processes that extend into the epithelial compartment or directly to the smooth muscle. It is not clear when this occurs, but it apparently occurs during the differentiation process. Chapter 3 also addresses airway smooth muscle and establishes that it is differentiated early in development once the basic pattern of the wall airway has been laid down. Once the basic geometric pattern of the airways has been established, they undergo substantial enlargement through longitudinal and circumferential growth. The great increase in cell and tissue mass necessary to accomplish growth relies on active proliferation of resident cell populations and the ability of the same cell populations to synthesize and secrete matrix components. How these processes are established and regulated and how they are balanced with forces promoting differentiation of the same cell populations is not understood. Further, these complex processes continue for a substantial period of time after birth. The temporal pattern for the differentiation processes varies significantly by species, but always moves in a proximal to distal direction with time. What this means is that during pre- and postnatal development of the airways, different airway generations will be in different stages of development. At any given time point, more proximal airway generations will be more differentiated than the more distal generations. Because the other aspects of airway development have been defined in Chapters 1 and 3, this chapter will emphasize the epithelium and its pattern of differentiation and what is known about regulation of the differentiation process. Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

i~:;~ ~i~i:~ii~:~::i~~ ii....~ ~

DIFFERENCES EXPRESSION

IN IN

~

~ ~ ~ ~i i,ii~i~ :, i~ ~i~ i/~ ~

PHENOTYPIC ADULTS

This chapter is organized on the premise that understanding the development of cellularly and architecturally complex organ systems such as the respiratory system, especially in the case of tracheobronchial airways, requires definition of changes based on specific airway sites and clear distinction of the timing of events in these sites. One of the major considerations in evaluating the potential toxicity of environmental contaminants for the developmental process is understanding which of the compartments is in which stage of development and differentiation at the time of exposure. Further understanding of airway development and the mechanisms that regulate it needs to be based on a clear understanding of the architectural organization and microenvironment-related characteristics that are expressed in differentiated systems in adults and on how these microenvironments respond to toxic stressors in adults. Previous studies have clearly established that two of the major classes of respiratory toxicants, oxidant air pollutants and bioactivated polyaromatic hydrocarbons, produce patterns of acute cytotoxicity that are highly site- and cell-selective. To define the complexity of the respiratory toxic response, we have compared multiple sites with profoundly different responses to oxidant air pollutants and bioactivated cytotoxicants" respiratory mucosa and olfactory mucosa in the nasal cavity, the trachea, and proximal, mid-level bronchi and distal bronchioles in the lungs. 1-14 It is now well-recognized that the respiratory system of adult mammals contains over 40 different cell phenotypes

~i~:~ i~i~ii/i(~i~ii~~,i::~i ~i~i~:~:i~~I~I i ~i~~):~ii!i,~:i~i~:~i~:i~~!~i~/i~( ~/i~~ ~ii~ ~ ~I~,~I~~:I:~)~:~i:~)~:~i!~i:i:~!i~!/~'~i~!i)~i~,!/~

distributed within a large number of distinct microenvironments. Using microdissection, we have defined the complexity and microenvironment-dependent nature of phenotypic expression for potential target cell populations within different airway sites. 2'15-3z Virtually every aspect of the composition of the wall of the airways, including epithe' lium and glands, smooth muscle, and cartilage vary by species. This is especially true for the airway epithelium, which is highly varied in any one species depending on precisely where in the airway tree the cell populations are examined for these characteristics: the composition of the cell populations lining the luminal surface (Table 2.1), the composition of the secretory product found within these epithelial populations (Tables 2.1-2.3) where the same phenotypes are present in many different airways, their relative abundance and proportion of the luminal surface occupied by the specific phenotypes may vary widely. What this means is that the organization of tracheal epithelium is very different from that of terminal and respiratory bronchioles in the same animal. When different species are compared on an airway by airway basis it is clear that the same types of variability exist. In fact in many species, individuals that are free of respiratory disease have very different cell populations in the same microenvironment compared to healthy individuals of other species. This is also true for the potential for the metabolism of xenobiotics either by an activation system (cytochrome P-450 monooxygenases) or a variety of detoxification and antioxidant systems (see Chapter 12). This also applies to the distribution of submucosal glands, with many species having glands extensively down the airway tree as far as

Table 2.1. Carbohydrate content of tracheal epithelium. Carbohydrate content PAS

AB

HID

+

+

-

+

+

+

-

+

+

+

-

Mucous

+

+

+

+

Clara

I::

Species

Cell type

Abundance

Hamster

Clara

:::

Mucous

+

Rat

Serous

:::

Mucous Mouse

Mucous

Rabbit Dog

Mucous

++

+

+

+

Cat

Mucous

++

+

+

+ and-

Serous

+

ND

ND

ND

Mucous

-H-

+

+

+ and-

+ +

+ and+

Pig Sheep

Mucous

++

i:~ M o n k e y

Mucous

++

+ +

Mucous

III

+

ii i~i:~Huma:n I . . . .

: ....

+:

:i : + a n d _

Development

of A i r w a y

Epithelium

Table 2.2. Carbohydrate content of tracheal submucosal glands. i 84~ii :i/:i

:~iii~/!:ii~I i

Carbohydrate content Species

Abundance

Hamster Rat

+ +

Mouse

+

Rabbit Dog

+ ++

Cat Pig

', ~ ~ ',

~ ~~

Secretory cell

PAS

AB

HID

MUCOUS

+

+

m

+

+ and

Serous

+

Mucous

+

Serous

+

Mucous Mucous Serous Mucous Serous Mucous Serous Mucous

+

+

+ a n d _ : : ~:~i

+

+

4--

small bronchioles whereas in other species they are restricted to the most proximal portions of the trachea. Cartilage is not a prominent feature of the conducting airways distal to the trachea in most species the size of rabbits or smaller, but is found extensively throughout the intrapulmonary airways in larger species, including humans. The distribution and organization of smooth muscle appear to be relatively site-specific. The complexity of the cellular organization within even a restricted portion (bronchial airways) of a complex organ such as the lungs emphasizes the need for highly precise sampling methodology. This need is further emphasized by the wide variability in local exposure dose created by the architectural complexity of the tracheobronchial airway tree itself. 5'9'33 As would be expected from a highly complex cellular organization, the metabolic potential of cell populations in different microenvironments within the respiratory system varies widely. The principal enzyme system for xenobiotic bioactivation, the cytochrome P-450 monooxygenases, has broad variability in isozyme expression, substrate specificity and level of activity. 2'34-41 This is also true for the enzyme systems involved in detoxification, especially the glutathione S-transferases and epoxide hydrolases. 35'4~ The cells in each of these different microenvironments also manage their glutathione pools very differently. 9'41'44'45 The pattern of heterogeneity of metabolic function appears to be relatively unique for each species of mammal. Inflammatory responses

.... iii:~

_

+ +

....

+ .

: -]-

+

.

.

.

+ and' : : + :

generated by acute exposure to oxidant air pollutants also vary greatly by site within the tracheobronchial airway tree. 3'46 The biological uniqueness of the cell populations in local airway microenvironments is further emphasized by the fact that when epithelial populations are cultured with the surrounding matrix intact, they maintain the same phenotypic expression and response to toxicants that would be expected if they were still resident within the intact animal. 1~ This complexity emphasizes the need for precise sampling to establish meaningful cellular and metabolic profiles and to validate them for patterns of cytotoxicity. Cultures have been used for definition of local cytotoxicity, 12,17'30 metabolism, 49 maintenance of biological function in vitro 10'12 and definition of local exposure dose. 5'9'33 They have even been validated for obtaining nucleic acids for definition of gene expression at the level of the local microenvironment.50

OVERALL

DEVELOPMENT

OF

AIRWAYS

Early branching morphogenesis As outlined in Chapter 1, the early formation of the airways and the subsequent development of submucosal glands are produced by the process of branching morphogenesis. In essence, this involves the differential growth of an epithelial tube into an associated mesenchymal derivative

Table 2.3.

Lectin reactivity of airway epithelium and submucosai glands.

?~)::~,~:~/:::~i:i!/!ii::::(i i:~!~.... :! : .... : :: : ?i: ::i::~::::~::~)~:: :

84184184184 ~ ~ ~ ,

:i

LCA

WGA

Man GIc GalNAc

NANA GalNAc

-

: :::

~:i~:~: :~:~ ~i~

BSAI::

~:

SBA :i ~:~:::::~

~

::

DBA

Gai-

Gal

GalNAc

Gal GalNAc

Gal GalNAc

Gal :::::: : GalNAc

:I :

::::::: P N A : :

UEAi)I:::I:::::

GalNAc

Species/cell type

Sugar specificity

:::

::i:::

In a i r w a y surface e p i t h e l i u m Human bronchi

Mucous

+

+

~,::

:: ::

Sheep trachea

M u c o u s M1

-

::',

++

:I:

Mucous M2

-

',::

++

::'

Mucous M3

:':

::;

-

-

-

+

+

-(+++)

+

-

-(-H-I-) a

--(-H--l-)

bronchiole Rat trachea

Mucous

-

Rhesus trachea

Mucous

~,:::

,,,

i

::II

l'l|

O f a i r w a y s u b m u c o s a l glands Human bronchi

Mucous

Sheep

Serous Mucous M4

Mouse trachea

Serous Mucous

+ ; ' :

Mucous

++

Serous Rhesus trachea

+

:::

: ', : ',

:::

-H-

+

++

++ ++ +

_ -(+++)

-

-

:::: ::: fill

1111

IIII

:III

-

++

++

-(+-H-)

-

(::::)

-

-

Serous Rat trachea

+

: : :

.

.

.

.

'...................:::~:::~::: ::: = :i ~ : :~::~:::::::: :

.el :'~,,, m

!

|

Mucous

::::

::::

:::

-

-(::::)

-(::::)

:::I

Serous

++

++

+

-

-(++)

-(++)

+

|

.....:: ::

.....

Abbreviations: LCA, Lotus tetragonolobus; WGA, wheat germ agglutinin; SBA, Glycine max; BSA 1, Bandeirea simplicifolia i; DBAI ii Dolichos biflorus; PNA, Arachis hypogea; RCA, Ricinus communis; UEA1, Ulex europeus; Man, mannose; GIc, glucose; GalNAcI: ::~::

N-acetylgalactosamine; NANA, N-acetylneuraminic acid (sialic acid); Gal, galactose; Fuc, fucose. aReaction in parenthesis is after neuraminidase treatment.

containing both cells and matrix. The composition of the matrix appears to dictate where the growing tube will divide. The bifurcation process itself is produced by focal differences in proliferation and programmed cell death to produce rapid growth in areas adjacent to sites of no growth. The no growth sites appear to be associated with bands of newly formed collagen and elastin. Each branching of this growing tube is regulated by a variety of cytokines and growth factors. Subsequent development of other components that form the wall in adults occurs at later times in specific airways. This development appears to move in a proximal to distal pattern following the branching of the epithelial tube. For the formation of the airway tree, this process is thought to be complete prior to birth and varies from species to species as to the percentage of gestation during which the process is complete. As outlined in Chapter 1, subsequent branching produces alveolar septation in alveolar spaces. Once the general pattern of the tree has been established, subsequent developmental processes are essentially growth in two directions: either longitudinally

.

.

.

.

:::::

to extend the length of the tube or circumferentially to increase its diameter. What regulates these processes and how they are associated with differentiation and growth of the wall constituents are not yet clear. This would be of particular signifcance given the substantial impact that the size and angles of the airways have on the flow of air during the respiratory cycle. In most mammalian species, the majority of the growth of the airways is a postnatal event. This suggests that for an extended period after birth, these growth events are susceptible to perturbations by environmental contaminants.

Respiratory bronchioles The most distal airways, located at the junction between the gas exchange area and tracheobronchial airway tree, form an extensive transitional zone in the human lung. This zone, exceeding three generations of branching in humans, is characterized by intermixing of alveolar epithelium, simple cuboidal epithelium mixed with the pseudostratified cuboidal epithelium (with basal, mucus and ciliated cells)

. . . .

found in more proximal airways (see 51 for a review). The respiratory bronchioles are extensive (exceeding three generations) in humans, macaques, dogs, cats and ferrets. In rhesus monkeys, and possibly in other primates, including humans, the two epithelial populations, bronchiolar and alveolar, are distributed on opposite sides of the airway in relation to the position of the pulmonary arteriole. 52 A pseudostratified population with ciliated cells lines numerous generations of respiratory bronchioles on the side adjacent to the pulmonary arteriole. The alveolarized areas are surrounded by a simple cuboidal bronchiolar epithelial population on the side opposite the arteriole. In the majority of mammalian species, the bronchiolar epithelium occupies the proximal portion of the transitional bronchiole, and alveolar gas exchange epithelium lines the distal portion. This is the case for mice, hamsters, rats, guinea pigs, rabbits, pigs, sheep, cattle and horses, s3 The composition of the peribronchiolar region associated with Clara cells includes the presence of smooth muscle adjacent to the basal lamina, extensive collagen interspersed with elastin and few capillaries. Those capillaries that are present are not closely associated with the epithelial basal lamina. The principal vessel in the area is the pulmonary arteriole. In contrast, the alveolar portions of this transitional zone generally include a substantial capillary bed closely applied to the basal lamina of the alveolar epithelial populations. Although the matrix composition of the alveolar gas exchange portions of the lung have been studied in some detail, the same is not true for the matrix associated with the bronchioles. 54 In fetal animals, where the majority of epithelial cells are poorly differentiated or undifferentiated, the boundary between the epithelium lining presumptive distal conducting airway and that lining future gas exchange regions is defined relatively easily in some species. 54-57 The distinguishing features include differences in epithelial configuration and modifications in the surrounding mesenchymaUy derived components. Most of these components, including smooth muscle and fibroblast-like cells, appear to mature somewhat more quickly than do the associated epithelium. 52'58 The morphogenesis of the respiratory bronchiole during fetal lung development has been studied in detail in only one species: rhesus monkeys. 58 The respiratory bronchiole begins as a tube lined by glycogen-filled cuboidal cells intermixed with an occasional ciliated cell. Alveolarization begins in the most proximal aspect of the respiratory bronchiole, at approximately 60% gestation in rhesus monkeys and in humans. The alveolarization appears as a formation of outpocketings into surrounding extracellular matrix. The outpocketings, which are lined by cuboidal epithelium, occur only on the side of the potential respiratory bronchiole opposite the pulmonary arteriole. They begin at the same time that secondary septa are forming in the distal acinus. Outpocketing or alveolarization occurs over a very short period (5 days) in rhesus monkeys. As alveolarization progresses from proximal to distal in the potential respiratory bronchiole, the epithelial cells also differentiate. By

~

i i i 84 !~

:

~ii~

i~i ~ 84

ii

67% gestation, ciliated cells are confined to the epithelium adjacent to the pulmonary arteriole, and the cytodifferentiation of the epithelial cells characteristic of alveoli is beginning in the outpocketings. Contacts between epithelium and underlying fibroblastic cells are observed for a very brief period in regions of respiratory bronchiole development. Epithelium of proximal generations of respiratory bronchiole differentiates earlier than more distal generations, but much later than in the trachea.

Submucosal glands The developmental events involved in the formation of submucosal glands have been well described for a number of species, including rats, 59 opossums, 6~ ferrets, 6x rhesus monkeys 62 and humans. 63'64 The sequence of events in humans has been characterized sub-grossly 65'66 and histologically. 63'64'67'68 The ultrastructure and histochemistry of gland development have been characterized in more detail in rhesus monkeys. 62 In rhesus monkeys, most of the process occurs in the fetus between the end of the pseudoglandular stage and the beginning of the terminal sac stage of development. Gland development implies four phases summarized in Fig. 2.1: (a) the formation of buds by projections of undifferentiated cells from the maturing surface epithelium, (b) the outgrowth and branching of these buds into cylinders of undifferentiated cells, (c) the differentiation of mucus cells in proximal tubules associated with proliferation of tubules and acini, and with undifferentiated cells distally, and (d) differentiation of serous cells in peripheral tubules and acini, with continued proliferation in most distal areas. The cells forming gland buds are not basal cells, as first thought, but rather an undifferentiated cell similar to the surface epithelium (Fig. 2.1). 62 Connective tissue appears to play a role in this process, as evidenced primarily through the presence of cartilage plates in the areas of initial bud formation. Glands appear first at the junction of cartilage plate and smooth muscle, followed by areas over cartilage plates and then in the area over smooth muscle. The secretory cell population differentiates in a centrifugal pattern, with nearly mature cells lining proximal tubules and immature cells in more distal portions. Mucus cells in the proximal portion of the gland develop before serous cells. Glandular mucus cells and serous cells differentiate at different times during development and through a different sequence of events. 62

EPITHELIAL D I F F E R E N T I A T I O N Overview Of the over 40 different cellular phenotypes that have been identified in the lungs of adult mammals, the differentiation of the epithelial cells lining the air passages appears to be the most critical in the successful function of the lung. At least eight of these cell phenotypes line the tracheobronchial

The Lung: Development,

Aging and the Environment

Fig. 2.1. Morphogenesis of submucosal glands in the trachea of rhesus monkeys. In very young fetuses (72 days gestational age (DGA)), the initial phase is projection of buds (B) from the luminal (L) surface epithelium into the surrounding matrix accompanied by an invagination (arrow) of the surface epithelium. As fetuses age (80 and 87 DGA) the buds extend further into submucosal connective tissue, with an apparent lumen (arrowhead) and continue until the formation of a cylindrical projection. In mid-gestation (105 DGA), the tube (T) branches extensively into the matrix with a patent lumen (arrow) apparent throughout. In late gestation (125 DGA) and early postnatal age (12 DPN), differentiation between the duct (D) and more peripheral secretory structures including proximal tubes (P) and large numbers of secretory acini (A) are evident. Continuing growth includes expansion of a center area's proximal tubular structures and marked enlargement of the ducts in adults.

conducting airways, including ciliated cells, basal cells, mucus goblet cells, serous cells, Clara cells, small mucus granule cells, brush cells, neuroendocrine cells, and a number of undifferentiated or partially differentiated phenotypes that have not been well characterized. The abundance and distribution of these cell types within the conducting airway tree vary by position within the tree and by species. The pattern of differentiation of the tracheal epithelial lining has been characterized for a large number of species. 69 The general pattern appears to be the same for most species in terms of which phenotypes are identified earliest during development and which differentiate later (summarized in Fig. 2.2). The critical difference between differentiation of these epithelial cells during development

is the percentage of intrauterine life in which the differentiation occurs. The epithelium of the trachea is the earliest of all the epithelial populations to differentiate. In some species it is relatively differentiated prior to birth, and in other species the majority of the differentiation occurs postnatally. In most species, with the possible exception of ferrets, ciliated cells differentiate first. Nonciliated cells with secretory granules appear next. Basal and small mucus granule cells appear last. There is a polarity in the differentiation of ciliated cells in the trachea, with the epithelium over the smooth muscle undergoing ciliogenesis earlier than that on the cartilaginous side. The reverse appears to be true for the nonciliated secretory cells, with secretory granules appearing on the cartilaginous side of the trachea first.

Development of Airway Epithelium

DIFFERENTIATION

OF T R A C H E A L E P I T H E L I U M

Neuroendocrine (small granule) . . . . . . . . . . . . . . . Differentiating . . . . . . . . . . . . . .9 ciliate

.

\

M a t u r e ciliated " 99 : 9

Undifferentiated Columnar

~

9 9

Differentiating 9 secretory

9 1 4 9 1 4 9

/

\

9 1 4 9 1 4 9 1 4 9 1 4 9 1 4 9 1 4 9 1 4 9 1 4 9

M a t u r e secretory--

9 9

e w w,,w

.

w wwv,,w.l,w

w~ uw ww.,.,,wwv.w..,,w

o

I I e

"Differentiating . . . . . . . . . . . . . . . . .

/

kM a t u r e basal

basal Mouse

I - ............ 2 5 % . . . . . . . . . . ( 5 days)

Hamster

I-. ............ 1 2 . 5 % ....... ( 2 days)

Rat

i - ............. 2 0 % . . . . . . . . . . (4.5 days)

Rabbit

I-- ............ 4 0 % ......... (13 days)

Rhesus M o n k e y

j . ............................. 7 0 % ............... ~ : ~ (118 days)

Human

~. ..............................

75% . . . . . . . . . . . . . . . . . .

(26 wks)

Species

Duration of Gestation ( % ) (Days)

Birth

Fig. 2.2. Diagrammatic comparison of the pattern of differentiation and maturation of the principal cell phenotypes observed in the tracheobronchial airways, with an emphasis on epithelium in trachea. This compares the timing of events in relation to parturition (open arrow) for each species. The dotted line indicates the duration of time during gestation from the initial observation of the formation of cilia to parturition. It is represented both in terms of the percentage of gestation over which differentiation occurs in utero and the actual number of days it takes. For most species, a significant portion of differentiation is postnatal. The proportion of time during gestation when these events occur is very species-specific.

Pattern in trachea and bronchi The ultrastructural features of overall tracheal epithelial differentiation in developing fetuses have been described in rabbits, 7~ mice, 7] hamsters 72'73 and rats. 68 In view of the diversity in the airways in different species, small laboratory mammals may not be adequate models for the study of human tracheobronchial epithelium. 6s The most extensive study of the development of the mucus cell was performed on the trachea of rhesus monkeys TM and is reviewed here. Gestation for rhesus monkeys averages 168 days, with the stages of lung development as follows: embryonic period, 21-55 days gestational age (DGA); pseudoglandular, 56-80 DGA; canalicular, 80-130 DGA; and terminal sac, 131 DGA to term. 75 In the youngest fetuses, all cells appear as illustrated in Fig. 2.3A. The cells are columnar and the apices of most of

the cells reach the luminal surface. Nuclei have little heterochromatin, and the cytoplasm is filled from base to apex with glycogen. The few organelles present are located in the apex of the cell and include short narrow strands of granular endoplasmic reticulum (GER), small spherical mitochondria and a small Golgi apparatus located adjacent to the lateral surface of the cell (Fig. 2.3A). These cells were present in the epithelial lining in the youngest animals in the embryonic stage to the middle of the canalicular phase. Near the end of the embryonic period, many cells similar to those at younger ages, but containing larger numbers of apical organelles, are observed. The organelles include spherical mitochondria and increased amounts of GER with dilated cisternae. The cisternae of the Golgi apparatus are dilated and surrounded by enlarged membrane-bound vacuoles, and glycogen is concentrated near the nucleus and

it

Fig. 2.3. Differentiation of mucus goblet cells in trachea of fetal rhesus monkeys: (A) early in gestation (46 DGA) columnar cells contain large pools of glycogen (Gly) and a central nucleus (N); (B) with continued age, columnar cells taper at the base and begin to accumulate membranebound secretory granules (arrowheads) in apical cytoplasm near the luminal surface (62 DGA); (C) by approximately 50% of gestation (90 DGA) columnar cells have markedly tapered bases, numerous apical secretory granules (arrowhead) and little cytoplasmic glycogen; (D) in the perinatal period (141 DGA) many of the cells have a prominent Golgi apparatus (Go) and an abundance of secretory granules in their apical cytoplasm (arrowhead).

intermixed with the organelles. These cells are observed in fetuses up to early in the canalicular stage. Through most of the pseudoglandular stage, most nonciliated cells had increased numbers of apical membrane-bound secretory granules containing a flocculent matrix, with a small electron-dense spherical core (Fig. 2.3B). Most of the remaining cytoplasm is still filled with glycogen. The cytoplasm surrounding the glycogen is more electron-dense than in younger ages and occupies more of the apical portion of the cell. The nuclei exhibit prominent nucleoli and small patches of heterochromatin. The mitochondria exhibit noncircular profiles and appear to be tubular. The amount of GER appears to be the same as in younger ages, but the cisternae were no longer dilated. The Golgi apparatus is surrounded by vacuoles of various sizes. The luminal surface of these cells is covered by long, regular microvilli. In somewhat older animals (late pseudoglandular), the apices of a large proportion of the secretory cells are filled with spherical granules (Fig. 2.3C). Most of these cells have abundant cytoplasmic glycogen, most of which is basal to the nucleus. Apical to the nucleus, glycogen is interspersed among organelles and granules. The cells appear more fusiform than at younger ages, being wide at the luminal side and narrow at the base (Fig. 2.3C).

In fetuses from midcanalicular stage and older, secretory cells containing cytoplasmic glycogen were rare and, when observed, the glycogen content was minimal. From this time to parturition, only two forms of secretory cells are observed. Both cells have little cytoplasmic glycogen, and the cytoplasm was condensed. There is a distinct variation in the abundance of apical secretory granules in these cells, ranging from very few, in cells with a narrow cytoplasm and few organelles, to an abundance of these granules in other cells (Fig. 2.3D). The cytoplasm of these cells contains small mitochondria and varying amounts of GER. The Golgi apparatus is located on the apical side of the nucleus and show variable degrees of activity. In cells with more granules (Fig. 2.3D), the Golgi apparatus is larger, has more cisternal stacks, and has larger and more numerous adjacent vesicles. Long, regular microvilli are a characteristic feature of the surface of the secretory cells. There is considerable variability in the abundance of these cellular forms between 105 days and parturition. At earlier ages, they are of approximately equal abundance. Near parturition, most of the secretory cells resemble that in Fig. 2.3D. Some of the cells have an even larger percentage of their cytoplasm occupied by granules than illustrated in Fig. 2.3D.

Development

In the postnatal period, most of the secretory cells have an abundance of electron-lucent granules filling their apical cytoplasm. The majority of these granules have small electron-dense cores. A few have large electron-dense biphasic cores, as observed in the adult. In general, the nucleus and its surrounding cytoplasm are restricted to the basal portion of the cell and the Golgi apparatus, and other organelles occupy a small percentage of the cytoplasm. Up to 134 days after birth, there are, however, a few secretory cells the cytoplasm of which contained abundant organelles and a variable number of secretory granules, as is observed in the late fetal period (Fig. 2.3D). By 134 days of postnatal age, nearly all the secretory cells have a configuration similar to that observed in adults. The cytoplasm is filled with electron-lucent secretory granules that appeared to distend the cell's cytoplasm. The nucleus is compressed at the basal portion of the cell, and organelles are minimal. In most cases, the cytoplasmic granules contain a biphasic core. The central part of the core is the most electron-dense portion of the granules. Differentiation of mucus glycoprotein biosynthesis and secretion develops slightly behind the other aspects of

of A i r w a y

El~itheliun7

2i

cellular differentiation. Prior to the presence of secretory granules, the principal material reacting with periodic acid-Schiff (PAS) is the large store of glycogen surrounding the nucleus (Fig. 2.4). Once granules appear, based on ultrastructure, the contents of the granules resemble that in adults (Fig. 2.4). These granules are not only PAS positive, but also positive for alcian blue (AB), indicating acidic groups, and for high iron diamine (HID) indicating that they are sulfated. The range of distribution of these patterns of stain reactivity varies by airway depending on the mix in the adult. In core granules, the sulfated material is generally identified in the center, whereas in uncored granules it tends to be on the periphery. As the cells fill with secretory product the distribution of staining reaction tends to follow closely that of the granules. The pattern by which sugars are expressed during differentiation is quite variable, depending on the specific end group (Table 2.4). The majority of sugars are expressed during the early phases of biosynthesis and granule formation. Others are expressed much later and all are generally, if they will be expressed in the adult, present in cells with a differentiated secretory apparatus by birth. With the exception of N-acetylgalactosamine (bound by PNA

Fig. 2.4. Comparison of the distribution of reaction products for alcian blue/periodic acid-Schiff (AB/PAS) and high iron diamine (HID) in the tracheal epithelium of fetal rhesus monkeys during the period when granule formation is first observed (62 DGA). The granules found in the apex of the cell are primarily AB/PAS positive (A) (arrow). Highly PAS positive material is found on the apical and basal portions of the nucleus (arrowhead). The majority (but not all) of the AB positive material is also sulfated (B) (arrow).

i ~ii~h ~i:!!~i~~i ~ g iii:ID e~/o!p

~ !~ ~ i !!!~i~g~ ~ing !a!!~ i~ !!:i hi~ I~!IE!!~:V iroi~i m ~ ~!~ii~i~iil~!~ii84 i!~!ii:84~i!iii ill i~~iiif:i~i!ii ~i~:i!ilii!iiiI!!~i:ii~i~ili~~i ii ~i~iii~i~!i!!i i il~!:~i~IIi84 iI~84184

Table 2.4. Lectin reactivity in developing rhesus monkey trachea. i ~i~i~ili!i:i~/!i!~~iil~:ii:i~ii:iii!~~?i :~iiiii:i~~~i~ :~ i~84184 iii!~ii::i! i84 ~ii/i ii iI Days:igestational age

Lectin

50

LCA UEA 1 SWGA BSA 1 PNA DBA

-

80

- : + + + + + . . . . . . + + + . . .

90

135

155

18dPN

+ + +

+ + +

+ + + + -

+ + + + -

+ .

~ i ii: i:

.

.

Abbreviations: see Table 2.3.

lectin), all others are more prominent in postnatal animals than in neonatal. This particular terminal sugar appears to be expressed in the early phases of mucus cell differentiation and its expression is suppressed, at least in rhesus monkeys, in late gestation and early postnatal life. Use of monoclonal antibodies established against mucus antigens indicates that very early in differentiation the composition of the core proteins for the secretory product may not be incorporated into the granules which initially form (Fig. 2.5). Their incorporation appears to be somewhat later during the differentiation process. Tracheobronchial epithelium continually renews itself. To identify the progenitor cell types that are involved in the self-renewal in vivo, the traditional approach is to carry out mitotic index and nuclear labeling studies. For the nuclear labeling study, the incorporation of [3H] thymidine or bromodeoxyuridine is used. Using these approaches, most of the data suggest that less than 1% of the epithelial

A

cell population is involved in cell proliferation. 76-8~ Both basal and secretory cell types are capable of incorporating these nucleotide precursors and mitosis, whereas ciliated cells are considered to be terminally differentiated and incapable of division. 81 In fact, only under exceptional circumstances are the ciliated cells of isolated hamster trachea capable of synthesizing DNA, as evidenced by the incorporation of [3H] thymidine. 8z Differentiation and proliferation normally are inversely related. Based on this view, a number of investigators 83'84 suggest that it is the basal cell type that serves as the stem cells, or the progenitor cell type that is involved in normal maintenance as well as in the regeneration and redifferentiation of bronchial epithelium after injury. However, this view is inconsistent with data obtained from the developmental studies and studies of injury/repair. In the developing tracheas of a number of animal species, including humans and nonhuman primates, basal cells are derived from an undifferentiated columnar epithelium. 85 Furthermore, the appearance of the basal cell type in the tracheal surface lining layer occurs after the appearance of ciliated and nonciliated secretory cell types. 86 Furthermore, in the growing intrapulmonary airways, 87'88 the basal cell type is not found in the smallest airway. 85 In the injury models, such as the mechanical and toxic gases exposure models, hyperproliferation is seen in the secretory cell type, but not in the basal cell type. 80'89-91 This indicates that it is less likely for the basal cell type to serve as a progenitor cell type that initiates the growth of airway epithelium and the repair of epithelial damage. 85 Studies of the repopulation of epithelial ceils on denuded tracheal grafts have been used to assess the "progenitor" nature of various bronchial epithelial cell types. Denuded tracheal grafts are usually produced by removing the lining epithelial layer by repeated freezing and thawing of tracheal grafts. 92 Using this technique, combined with the cell

B

Fig. 2.5 Comparison of the distribution of immunoreactive mucin with an antibody that reacts with all mucus cells in adult rhesus monkeys and the distribution of AB/PAS positive material from serial sections of trachea of fetal rhesus monkey when secretory granules are just beginning to form (50 DGA). (A) The immunoreactive secretory product is in highly focal areas of a small number of positive cells (arrows). (B) On section serial to (A), it is clear that these sites are also positive for PAS (arrows). Cartilage (C) is negative. (E) epithelium.

:

:

:

(

: Devil

separation technique, the mucociliary epithelium can be repopulated in the denuded tracheal graft by enriched basal cell population from rabbits and rats. 93'94 These experiments clearly demonstrate the pluripotent nature of the basal cell type. However, there are several deficiencies in these experiments. First of all, the definition of basal cell type is based on the ultrastructural picture and the immunohistochemical stain. It is well known, though, that secretory cells lose their differentiated features upon isolation and culturing in vitro. The degranulated secretory cells may resemble the basal cell type, and the morphologic tools used in these studies cannot distinguish satisfactorily the basal one from the degranulated secretory cell type in dissociated and isolated cell preparations. Furthermore, for the preparations in these studies, the purity of basal cell type population is only 90%. Using flow cytometry to isolate basal cells, it has been found that basal cells from rat trachea had a colonyforming efficiency of 0.6%, whereas secretory cells and unsorted cells had efficiencies of 3.4 and 2.6%, respectively.95 From these results, it may be concluded that basal cells have less proliferative activity than secretory cells. It is therefore difficult to conclude from these tracheal graft repopulation studies that basal cell type is the progenitor cell type responsible for the initiation of airway epithelial cell growth and the repair in response to injury.

Pattern in bronchioles The process of cytodifferentiation of the nonciliated cells of distal bronchioles entails substantial rearrangement, loss and biogenesis of cellular organelles. Up to late fetal age, terminal bronchioles are lined by simple cuboidal to columnar epithelium composed of glycogen-filled nonciliated cells with few organelles. The shifts in cellular components with time for species in which the predominant cellular constituent in adults is agranular (smooth) endoplasmic reticulum (AER), such as in mice, hamsters, rats and rabbits, are summarized in Fig. 2.6. The pattern is essentially similar for these species. What varies from species to species is the timing of these events. The first event is a dramatic loss in cytoplasmic glycogen. In rabbits, this drop is from ---70% of cytoplasmic volume to less than 10% in adults. A similar substantial loss occurs in rats, hamsters and mice. In rabbits, this loss begins immediately prior to birth and continues for up to 4 weeks of postnatal age. 96 A similar change occurs in mice. 97 In rats, the loss of cytoplasmic glycogen begins at birth and drops to adult levels within the first week of postnatal life. 98 In hamsters, cytoplasmic glycogen is not detectable immediately after birth. 99 Associated with the drop in cellular glycogen is a substantial biogenesis of membranous organelles, especially AER. Smooth endoplasmic reticulum is not detected in nonciliated cells until immediately prior to birth in rabbits (Fig. 2.6). 1~176 At birth, fewer than 20% of the cells contain more than 10% AER. By 2 weeks, in almost 70% of the nonciliated cells, AER occupies more than 10% (up to 50%) of the cell volume. The adult configuration is reached at approximately 28 days postnatally in rabbits. In mice, the adult configuration of

Epiih ii m84 AER i s reached at approximately 3 weeks postnatally. 97 Granular endoplasmic reticulum in prenatal animals is approximately twice as abundant in rabbit Clara cells as it is in rats (Fig. 2.6). 1~ The decrease in cellular abundance of GER occurs gradually in rabbits and is still double the adult configuration (2% of cell volume) at 4 weeks postnataUy, but in rats the level decreases by 50% immediately postpartum and is at or near the adult configuration (less than 1%) by 10 days postnatally. The situation for rats and mice appears similar to that for rabbits, but for hamsters GER is near the adult configuration immediately postpartum. 99'1~ Secretory granule appearance also varies by species. The earliest at which secretory granules are detected in the Clara cells of rabbits and mice is within the first week of postnatal life, whereas in rats and hamsters granules are abundant prenatally. In rabbits as well as mice, granule abundance resembling adult levels occurs by 21 days postnatally. In rats, granule abundance reaches adult abundance by 7 days postnatally and is at adult configuration immediately postpartum in the hamster. The only species in which Clara cell differentiation has been characterized fully where the adult Clara cell population does not have an abundance of AER is rhesus monkeys. 52 In that species, the loss of cytoplasmic glycogen and an increase, rather than a decrease, in GER occurs over a substantial period both prenatally and postnatally. Studies in humans suggest that developmental events for Clara cells are similar to those in rhesus monkeys, but may extend longer than the 6 months to a year (postnatally) required for differentiation of all the nonciliated cells in terminal respiratory bronchioles of monkeys. As summarized in Chapter 12, the expression of cytochrome P-450 monooxygenases (CYP) in Clara cells during their differentiation has been evaluated in a number of species. 1~176 Protein for the NADPH P-450 reductase and CYP2B is detected earliest, with the reductase somewhat later than CYP2B in rabbits. CYP4B is detected 2-3 days of age later (see Chapter 12). The initial distribution is in the most apical border of a small percentage of the nonciliated cell population. During the period in which the amount of detectable protein increases, the distribution changes in two ways. First, an immunologically detectable protein is found in an increasing proportion of the nonciliated cells as animals become older. Second, the distribution of detectable protein within an individual cell increases from the apex to the base with increasing age. The youngest age at which intracellular protein can be detected immunohistochemically varies substantially between these species (rat, rabbit, monkey). Protein becomes detectable in hamsters approximately 3-4 days prior to birth and reaches the distribution and intensity observed in the adult by 3 days postnatally. CYP4B is not detectable before 1 day postnatally, but is at adult levels shortly thereafter. In rabbits and rats, the timing is somewhat different. NADPH reductase is found initially just prior to birth in rabbits, and CYP2B and 4B are not observed until after birth. All these proteins have an adult distribution and intensity by 28 days postnatally. In rats, CYP2B, CYP4B and NADPH reductase are detected in the first 2-3

T h e Lung: D e v e l o p m e n t ,

A g i n g and the E n v i r o n m e n t

27 DGA 100 I

80 60

40 20

0-.099

.1-.199

.2-.299

.3-.399

.4-.499

>.5

Late Gestational Age 10o-

1-2 D Postnatal

80" 60" 4020"

~

i1,, 0-.099

.1-.199

.2-.299

9 .3-.399

...... ,4-.499

>.5

Perinatal Period 2 weeks 10080 6040

0..099

.1-.199

.2-.299

,3-.399

1-2 Weeks Postnatal

.4-.499

,

,, ,

>.5

,

Adult

100 9

~

M

(12-17 w e e k s )

80 60 40 20

0-.099

.1-.199

.2-.299

.3-.399

.4-.499

>.5

Volume Fraction (vol/vol) of Agranular Endoplasmic Reticulum in Clara Cell Cytoplasm 4 Weeks and Adult Fig. 2.6. Diagrammatic comparison of Clara cellular organization during pre- and postnatal differentiation with correlation to morphometric analysis of cytoplasmic content of smooth endoplasmic reticulum from morphometric measurements in lungs of rabbit. The morphometric data (histograms) compare the percent of bronchiolar Clara cells with different amounts of smooth endoplasmic reticulum in developing and adult rabbits. AER, agranular (smooth) endoplasmic reticulum; BL, basal lamina; G, Golgi apparatus; GER, granular (rough) endoplasmic reticulum; Gly, glycogen; M, mitochondria; Gr, secretory granule; Nu, nucleus.

days of postnatal life and are apparently at adult densities and distributions by 21 days postnatally. CYP1AI is not detectable prenatally in rats but can be detected in increasing, but small, amounts until it reaches adult levels at approximately

21 days postnatally. Intracellular expression of protein precedes the appearance and increase in the abundance of AER by 2-4 days in each of these species. Activity for these proteins is first detected approximately 2-3 days after the

Development

of A i r w a y

Epithelium

Table 2.5. D e v e l o p m e n t of agranular endoplasmic reticulum (AER) P-450 reductase and monooxygenase enzymes in the rabbit lung.

Amount of enzyme activity 27-28 DGA

1-2 DPN

7 DPN

Assay

28 DPN

Adult

64.5

100

% of adult value

AERa'b P-450 reductase Immunohistochemistry b,c Western blot P-450 isozyme 2B

Immunohistochemistryb Western blot r~-Tll P-450 ;,sv._yme 4B immunohistochemistry b ~A~--..L--_"

14 DPN

L~--L

0.2

8.2

8.2

30.1

+ +

0:i~::~ ~ ~ ~ ~i~=

+

0

.

.

protein is immunologically detectable within Clara cells. The activity studies have been done with whole lung homogenates and reflect potential activity from other cell populations as well as from Clara cells. While both the AER abundance and antigenic protein intensity reach the adult configuration in ---3-4 weeks after birth in rats and rabbits, the activity for these isozymes is still considerably below that for adults. This suggests that the functionality of these proteins continues to increase after the protein density and organelle composition have reached adult levels of expression. Table 2.5 summarizes the relationship between changes in AER abundance, expression of immunoreactive protein, and microsomal P-450 activity for rabbits. The timing for rats is somewhat shifted to the left for postnatal time points and to the right for perinatal ones, compared with rabbits. The pattern of expression of Clara cell secretory protein is similar to that of the cytochrome P-450 monooxygenase system in relation to the appearance of cellular organelles. While there is substantial interspecies variability in the timing of expression, the general pattern is similar, at least for the four species studied in most detail: rats, rabbits, hamsters and mice. 97'1~176176176 The protein appears earliest in the central or apical portion of a few cells per bronchiole, and the number of cells in which antigen can be detected

+-H'+

:I::

"-t--H-+

+

+ .

I:I:

++

++

+

0 0

-H"

. ~ i:i .

.

.

.

.

.

:, : ~

, ::: ~

.

,~.

increases with increasing age. In hamsters, the secretory protein antigen can be detected in a number of cells by the beginning of the last trimester of pregnancy and reaches the adult configuration in terms of density and number of cells labeled at about 3-4 days postnatally. In rats, a small proportion of the cells are labeled prenatally, and the distribution observed in the adult is present at about 7 days postnatally. This adult configuration occurs between 3 and 4 weeks in rabbits, and the earliest detectable signal in the Clara cells is immediately prior to birth. The timing in mice and rats is similar to that in rabbits. Immunoreactive protein has also been detected in late gestation in human fetuses, but at what age the distribution resembles adults has not been determined. Intracellular expression of the protein follows the changes in GER and is closely related to the first appearance and increase in the abundance of secretory granules. Western blotting of this protein indicates that it is present earlier in lung homogenate than its appearance in bronchiolar Clara cells suggests. This is because secretory cells in proximal airways express the protein much earlier and, in general, are more differentiated in the perinatal period than are secretory cells of bronchioles. In hamsters, the situation is inverse, with the bronchioles differentiating in this respect prior to the bronchi.

REGULATION

OF

DIFFERENTIATION

Trachea and bronchi Defining the mechanisms regulating maturation of fetal and neonatal airway epithelium is an area of active research with studies focused primarily on (a) in vitro studies, using isolated cell populations, and (b) genetically manipulated mice. The factors controlling differentiation and the mechanisms for these controls are poorly understood, especially in vivo. Three areas of regulatory control have been explored in some detail: the interleukins, vitamin A and growth factors, and transcription factors. In vivo studies have been minimal for direct treatment. However, the effects of epidermal growth factor (EGF) on lung development have been examined in rhesus monkeys. EGF treatment in utero markedly stimulates the maturation of the tracheal secretory apparatus, including both the tracheal surface and submucosal glands. 1~ The secretory apparatus is more differentiated in that there are more mucus cells, increased secretory product stored in the epithelium and glands, and increased quantities of secretory product in lavage and amniotic fluid. By contrast, treatment with triamcinalone, a glucocorticoid, induces maturation of the gas exchange area, 11~but does not affect the maturation of the secretory apparatus (Table 2.6). Based on studies primarily in mice, it appears that most of the differentiation of proximal airway epithelium is under the regulation of two transcriptional factors, the homeodomain transcriptional factor NKX2.1 (otherwise known as TTF-1) and winged-helix family transcription factor HNF-3/forkhead homolog-4 (HFH-4). TM The former appears to promote the initial branching of the tracheal bud but also promotes other aspects of tracheal differentiation, possibly via the FGF pathway. The key regulatory mechanism by which these transcription factors modulate differentiation of epithelium in proximal airways is not clearly understood. There appears to be a few key differentiation processes that may be regulated by specific transcription factors. Some studies have suggested that NKX2.1 promotes differentiaTable 2.6. Effect of EGF and Triamcinalone Acetonide (TAC) on total glycoconjugate detectable in the trachea of fetal rhesus monkeys.

Days gestational age 128 150 128 150

150 p < 0.05

Total secretory product (mm 3x 103/mm 2) (~+ 1 SD)

Treatment None None EGF TAC (11 mg/day) T A C (I O:mg/day)

0.48 + 0.37 1.36 a+ 0.33 1.77a+0.28 1.27 + 1,35 0.75 _+0i34:

control.

tion of Clara cells, but others suggest that tracheal mucous cell formation is independent of this transcription factor. 112 These developmental studies were based on mice after gene manipulation; however, healthy adult mice normally do not have differentiated mucous goblet cells. In contrast, it appears that HNF-3/forkhead homolog-4 transcription factor is critical for the differentiation of ciliated cells in respiratory epithelial populations. 113'114A number of epithelial-mesenchymal interactions appear to be critical to the differentiation of the epithelium; two operate through HNF-3 and GATA6 and may serve as regulators of cell-cell communication activities. The two interleukins that appear to have the most impact on transdifferentiation of airway epithelium in proximal airways of mice from a Clara cell phenotype to a mucus cell phenotype are IL9 and ILl3.11s'116 When the ILl3 gene is attached to a promoter for the Clara cell 10 KDa protein, it actively promotes the expression of mucus cells of the proximal airways of mice. 116-118 In all these cases, the airways of transgenic mice have three characteristic features: mucus metaplasia, eosinophilic inflammation and airway hyperresponsiveness. In vivo studies have also shown that inhibition of ILl3 blocks allergen induced effects on the airways of mice. On the other hand, direct administration of ILl3 to the airways can produce the same effects as allergen. 119'12~For both IL9 and ILl3, a variety of cellular responses occurs, including elevation of eosinophils, lymphocytes, mast cells and subepithelial collagen in airway w a l l s . 115'119'120 While most of these events appear to be regulated through the ILl3 R alpha 1 subunit in combination with the IL4 receptor, they neither appear to be modulated by factors that regulate the matrix changes nor influence inflammatory cell populations. 118 Defining exactly how ILl3 produces these changes has been difficult. It is clear that for poorly differentiated airway epithelial cells in vitro, ILl3 induces a dramatically different pattern of gene expression than is observed in airway smooth muscle cells or lung fibroblasts. 121 The four major transcription factors increased were OTF2, HSP factor 4, Id-3 and NRF-1. There was elevated phosphorylation of STAT 6 and some increases in factors related to extracellular matrix production. Tracheobronchial epithelial cells, like many other epithelial cells, lose their differentiated functions upon culturing in vitro. However, the loss of differentiated functions- at least in primary tracheobronchial epithelial c u l t u r e - is transient. In repopulation studies using cultured cells, 92 undifferentiated rabbit tracheal epithelial cells maintained long term in culture are able to repopulate the grafts and form a new mucociliary epithelium. 122' 123 Epithelial cells, despite dedifferentiation in culture, apparently maintain their intrinsic differentiated potential, which is expressed if an appropriate environment is provided: hormonal requirements, vitamin A supplement and collagen gel substratum. 124-127 Based on the amino acid and carbohydrate composition analyses of the in vitro secretory products as compared with the in vivo mucin products purified from sputum and epithelial cell layer, it appears that cultured tracheal epithelial cells from a number of species are able to

Development of Airway Epithelium

secrete authentic m u c i n . 128-131 Critical to use of in vitro models for epithelial differentiation has been the development of the Whitcutt chamber to grow airway epithelial cells between air and a liquid medium interface, 132-136based on the premise that airway epithelial cells in vivo are usually located between air and a liquid interface. Using this chamber, columnarized formation of cultured epithelial cells was observed, and with further development of mucociliary differentiation in culture, 122'123 including both human and monkey tracheobronchial epithelial cells. 124'125'135'136 Scanning electron microscopy demonstrates extensive ciliary features on the culture surface, and transmission electron microscopy has demonstrated the formation of abundant mucus-secreting granules and the columnarized features with a two- to four-cell layer. The basal cell layer is compressed and resembles basal cells in vivo. Tracheobronchial epithelium is a vitamin A-targeted tissue. 137-14~ The epithelium requires vitamin A for the preservation and induction of the expression of differentiated functions. Keratinizing squamous metaplasia of mucociliary epithelium occurs with vitamin A deficiency along with a reduction in the synthesis of mucus glycoproteins. The administration of vitamin A or its synthetic derivatives (retinoids) reverses this phenomenon. Excess vitamin A can convert stratified skin epithelium in chick embryos to an epithelium containing mucus-secreting granules. 141 Vitamin A treatment enhances the proliferation of small mucus-granule cell type in primary hamster tracheal epithelial cultures. 142-144 However, vitamin A does enhance DNA synthesis of basal cells of keratinocyte cultures. 145 There is no evidence that vitamin A inhibits squamous cell proliferation. Vitamin A and its derivatives clearly play a role in the differentiation and expression of mucin genes in human tracheal bronchial epithelial cells. 146 At least four mucin genes (MUC2, MUC5 AC, MUC5 B MUC7) are retinoic - (RA) or retinol-dependent while MUCi, MUC4 and MUC8 are not. Regulation of mucin genes by retinoic acid appears to be mediated by retinoic acid receptors RAR~ and y. Two other regulators of cellular function interact closely with RA to modulate mucin genes: thyroid hormone (T3) and EGF. T 3 inhibits mucin gene expression, particularly MUC5 AC, apparently through competitive inhibition of receptor responses through the thyroid receptors by inhibiting gene transcription. While EGF is thought to stimulate mucin expression and secretion in cultured airways of some species, especially the rat, it has an inhibitory effect in human bronchial cells in culture; it is suggested that EGF's impact may in fact be retinoic aciddependent.

Regulation in bronchioles Factors regulating Clara cell differentiation are not well understood. The postnatal nature of the majority of the cytodifferentiation process in most species suggests that it is independent of the hormones associated with pregnancy and parturition. The fact that the timing varies by as much as 2-3 weeks in different species would further suggest that

the process may be under regulation of a variety of factors that act in different temporal sequences and with different levels of influence in different species. A number of mediators have been shown to stimulate cytodifferentiation of type II alveolar epithelial cell and produce architectural rearrangements of lung connective tissue elements to promote gas exchange, including corticosteroids, thyroid hormone, EGF and cAMP. 147 Whether all these mediators influence Clara cell differentiation is not known. The best studied are the glucocorticoids, especially dexamethasone. Treatment in the perinatal period retards Clara cell differentiation as evidenced by an increase in cytoplasmic glycogen and minimal alterations in organelles in both rats and mice. 148'149 Dexamethasone administered either prenatally or immediately postnatally elevates the surfactant protein messenger RNA (mRNA) levels in lungs of rats of all ages, producing this elevation in both alveolar type II cells and Clara cells. 15~ Glucocorticoid administered to pregnant rabbits appears to have a stimulatory effect on the differentiation of secretory potential in fetal Clara cells by elevating the amount of the uteroglobin-like Clara cell secretory protein. 151'152 Dexamethasone administered to pregnant rabbits also has a stimulatory effect on the pulmonary cytochrome P-450 system in fetuses, based on measurements of whole lung microsomes. 153-155 While glycogenolysis is retarded by dexamethasone treatment, glycogen, epinephrine and 8-bromo-cAMP produce a rapid drop in Clara cell glycogen content. 98 One of the factors that appears to have the most impact on Clara cell differentiation is injury during the developmental period, in which normal differentiation occurs. Normal differentiation is characterized by loss of glycogen and appearance of secretory granules, and by differentiation of Clara cells into ciliated cells, even in the absence of frank injury to either ciliated or Clara cells. Postnatal exposure to compounds that injure the respiratory system retard Clara cell differentiation. Hyperoxia during the early postnatal period inhibits differentiation. 156'157 Injury by treatment with 4-ipomeanol impedes Clara cell differentiation even for a short term after treatment is discontinued. 158 Not only are Clara cells in postnatal animals more susceptible to injury than in adults, but the expression of the P-450 system in the post-treatment period is markedly reduced. In rats, exposure to cigarette smoke of either the pregnant mother or the newborn accelerates the appearance of one cytochrome P-450 monooxygenase isozyme, CYP 1A 1, but not CYP2B. 1~ The increased P-450 expression is primarily in the Clara cell population and is not found in either alveolar type II cells or in the vascular endothelium, both targets for inducers in adult animals. Other factors besides postnatal hyperoxia, including maternal undernutrition during the last 5 days of pregnancy, retard Clara cell differentiation, but these effects appear to be reversible with time.98,156-158 There is considerable indirect evidence to suggest that a number of growth factors, including TGF-~, EGF, basic fibroblast growth factor (FGF), insulin-like growth factors,

and platelet-derived growth factor, may play roles in regulating bronchiolar epithelial differentiation. 159'16~The EGF receptor (EGFr) has been detected in bronchiolar epithelium throughout pre- and postnatal lung development in rats and humans. 161'162 EGFr has also been detected in human lung at midgestation 163'164 and has been detected in human and rat fetal lung extracts. 161'165 Both ligands of EGFr, as well as T G F - ~ and EGF, have been detected immunohistochemically in bronchiolar epithelium in a number of species. EGF is barely detectable in bronchiolar epithelium of fetal humans (first and second trimesters), but is present in postnatal human lung. 166 EGF has been reported in homogenates of lung from late fetal (21-day gestational age) and adult rats, 167 and immunoreactive protein has been detected in bronchiolar epithelium throughout fetal development in lambs and mice. 168'169 TGF-[3 has been detected in bronchiolar epithelium of mid-gestational humans. 17~ It can be extracted and mRNA can be detected in fetal rat lung homogenates. 171 Plateletderived growth factor receptor has also been detected in bronchiolar epithelium during most of the prenatal stages of lung development. 163'164 Basic F G F and its receptor are found in bronchiolar epithelium during most of fetal rat lung development. 172 Both the F G F receptor and the protein appear to colocalize in the epithelium and adjacent interstitial compartments. There is some suggestion that insulin-like growth factors are involved in aspects of epithelial development in bronchioles. 173 These growth factors may play a role in autocrine regulation because both receptors and the proteins themselves appear within the bronchiolar epithelium. They may also play a paracrine role because growth factor protein appears to be distributed to interstitial cell components, fibroblasts, and smooth muscle surrounding bronchiolar epithelium, during various stages of lung development. At present, there is no direct evidence that any of these factors influence bronchiolar epithelial maturation. There is, however, evidence that pharmacological doses of E G F alter branching morphogenesis in mice, TM enhances differentiation of alveolar type II cells in fetal rabbits, monkeys and sheep, 69'175'176 and alter the differentiation of tracheal epithelium in rhesus monkeys. 1~

REFERENCES 1. Paige RC, Wong V, Plopper CG. Long-term exposure to ozone increases acute pulmonary centriacinar injury by 1-nitronaphthalene. II. Quantitative histopathology. J. Pharmacol. Exp. Ther. 2000; 295:942-50. 2. Fanucchi MV, Murphy ME, Buckpitt AR et al. Pulmonary cytochrome P-450 monooxygenase and Clara cell differentiation in mice.Am. J. Respir. Cell Mol. Biol. 1997; 17:302-14. 3. Hyde DM, Hubbard WC, Wong V e t al. Ozone-induced acute tracheobronchial epithelial injury: relationship to granulocyte emigration in the lung. Am. J. Respir. Cell Mol. Biol. 1992; 6:481-97. 4. Paige R, Wong V, Plopper C. Dose-related airway-selective epithelial toxicity of 1-nitronaphthalene in rats. Toxicol. Appl. Pharmacol. 1997; 147:224-33.

5. Pinkerton KE, Plopper CG, Mercer RR et al. Airway branching patterns influence asbestos fiber location and the extent of tissue injury in the pulmonary parenchyma. Lab. Invest. 1986; 55(6):688-95. 6. Plopper C, Suverkropp C, Morin D etal. Relationship of cytochrome P-450 activity to Clara cell cytotoxicity. I. Histopathologic comparison of the respiratory tract of mice, rats and hamsters after parenteral administration of naphthalene. J. Pharmacol. Exp. Ther. 1992; 261:353-63. 7. Plopper C, Macklin J, Nishio S etal. Relationship of cytochrome P-450 activity to Clara cell cytotoxicity. III. Morphometric comparison of changes in the epithelial populations of terminal bronchioles and lobar bronchi in mice, hamsters, and rats after parenteral administration of naphthalene. Lab. Invest. 1992; 67:553-65. 8. Plopper CG, Chu FP, Haselton CJ etal. Dose-dependent tolerance to ozone. I. Tracheobronchial epithelial reorganization in rats after 20 months exposure.Am. J. Pathol. 1994; 144:404-20. 9. Plopper CG, Hatch GE, Wong Vet al. Relationship of inhaled ozone concentration to acute tracheobronchial epithelial injury, site-specific ozone dose, and glutathione depletion in rhesus monkeys. Am. J. Respir. Cell Mol. Biol. 1998; 19:387-99. 10. Postlethwait EM, Joad JP, Hyde DM et al. Three-dimensional mapping of ozone-induced acute cytotoxicity in tracheobronchial airways of isolated perfused rat lung. Am. J. Respir. Cell Mol. Biol. 2000; 22:191-9. 11. Van Winkle LS, Isaac JM, Plopper CG. Distribution of epidermal growth factor receptor and ligands during bronchiolar epithelial repair from naphthalene-induced Clara cell injury in the mouse.Am.J. Pathol. 1997; 151:443-59. 12. Van Winkle LS, Johnson ZA, Nishio SJ et al. Early events in naphthalene-induced acute Clara cell toxicity: comparison of membrane permeability and ultrastructure [In Process Citation]. Am. J. Respir. Cell Mol. Biol. 1999; 21:44-53. 13. West JA, Chichester CH, Buckpitt AR et al. Heterogeneity of Clara cell glutathione: a possible basis for differences in cellular responses to pulmonary cytotoxicants. Am. J. Respir. Cell Mol. Biol. 2000; 23:27-36. 14. Wilson DW, Plopper CG, Dungworth DL. The response of the macaque tracheobronchial epithelium to acute ozone injury. Am. J. Pathol. 1984; 116:193-206. 15. Plopper CG, Heidsiek JG, Weir AJ etal. Tracheobronchial epithelium in the adult rhesus monkey: a quantitative histochemical and ultrastructural study. Am. J. Anat. 1989; 184:31-40. 16. Mariassy A, Plopper C. Tracheobronchial epithelium of the sheep. I. Quantitative light-microscopic study of epithelial cell abundance, and distribution.Anat. Rec. 1983; 205:263-75. 17. Plopper C, Halsebo J, Berger W e t al. Distribution of nonciliated bronchiolar epithelial (Clara) cells in intra- and extrapulmonary airways of the rabbit. Exp. Lung. Res. 1983; 4:79-98. 18. Plopper C, St. George J, Nishio S et al. Carbohydrate cytochemistry of tracheobronchial airway epithelium of the rabbit.J. Histochem. Cytochem. 1984; 32:209-18. 19. St. George J, Plopper C, Etchison J etal. An immunocytochemical/histochemical approach to tracheobronchial mucus characterization in the rabbit. Am. Rev. Respir. Dis. 1984; 130:124-7. 20. Mariassy AT, Plopper CG. Tracheobronchial epithelium of the sheep. II. Ultrastructural and morphometric analysis of the epithelial secretory cell types.Anat. Rec. 1984; 209:523-34. 21. Wilson D, Plopper CG, Hyde DM. The tracheobronchial epithelium of the bonnet monkey (Macaca radiata): a quantitative ultrastructural study. Am. J. Anat. 1984; 171:25-40. 22. St. George JA, Cranz DL, Zicker S etal. An immunohistochemical characterization of rhesus monkey respiratory

Development of Airway Epithelium

secretions using monoclonal antibodies. Am. Rev. Respir. Dis. 1985; 132:556-63. 23. Heidsiek JG, Hyde DM, Plopper CG et al. Quantitative histochemistry of mucosubstance in tracheal epithelium of the macaque monkey.J. Histochem. Cytochem. 1987; 35:435-42. 24. Plopper C, Cranz D, Kemp L etal. Immunohistochemical demonstration of cytochrome P-450 monooxygenase in Clara cells throughout the tracheobronchial airways of the rabbit. Exp. Lung. Res. 1987; 13:59-68. 25. Mariassy AT, St. George JA, Nishio SJ et al. Tracheobronchial epithelium of the sheep. III. Carbohydrate histochemical and cytochemical characterization of secretory epithelial cells. Anat. Rec. 1988; 221:540-9. 26. Mariassy AT, Plopper CG, St. George JA et al. Tracheobronchial epithelium of the sheep. IV. Lectin histochemical characterization of secretory epithelial cells. Anat. Rec. 1988; 222:49-59. 27. Evans MJ, Plopper CG. The role of basal cells in adhesion of columnar epithelium to airway basement membrane. Am. Rev. Respir. Dis. 1988; 138:481-3. 28. Evans M, Cox RA, Shami SG et al. The role of basal cells in attachment of columnar cells to the basal lamina of the trachea.Am. J. Respir. Cell Mol. Biol. 1989; 1:463-9. 29. Dodge DE, Rucker RB, Singh Get al. Quantitative comparison of intracellular concentration and volume of Clara cell 10 kDa protein in rat bronchi and bronchioles based on laser scanning confocal microscopy. J. Histochem. Cytochem. 1993; 41:1171-83. 30. Avadhanam KP, Plopper CG, Pinkerton KE. Mapping the distribution of neuroepithelial bodies of the rat lung: a whole-mount immunohistochemical approach. Am. J. Pathol. 1997; 150:851-9. 31. Fanucchi MF, Buckpitt AR, Murphy ME et al. Development of phase II xenobiotic metabolizing enzymes in differentiating murine Clara cells. Toxicol. Appl. Pharmacol. 2000; 168:253-67. 32. Fanucchi M, Buckpitt A, Murphy ME etal. Naphthalene cytotoxicity of differentiating Clara cells in neonatal mice. Toxicol. Appl. Pharm. 1997; 144:96-104. 33. Pinkerton KE, Gallen JT, Mercer RR etal. Aerosolized fluorescent microspheres detected in the lung using confocal scanning laser microscopy. Microsc. Res. Technique 1993; 26:437-43. 34. Buckpitt A, Buonarati M, Avey L etal. Relationship of cytochrome P-450 activity to Clara cell cytotoxicity. II. Comparison of stereoselectivity of naphthalene epoxidation in lung and nasal mucosa of mouse, hamster, rat and rhesus monkey. J. Pharmacol. Exp. Ther. 1992; 261:364-72. 35. Buckpitt A, Chang A, Weir A et al. Relationship of cytochrome P-450 activity to Clara cell cytotoxicity. IV. Metabolism of naphthalene and naphthalene oxide in microdissected airways from the mouse, rat, and hamster. Mol. Pharmacol. 1995; 47:74-81. 36. Lee C, Watt KC, Chang AM et al. Site-selective differences in cytochrome P-450 isoform activities. Comparison of expression in rat and rhesus monkey lung and induction in rats. Drug Metab. Dispos. 1998; 26:396-400. 37. Paige RC, Royce FH, Plopper CG et al. Long-term exposure to ozone increases acute pulmonary centriacinar injury by 1-nitronaphthalene. I. Region-specific enzyme activity. J. Pharmacol. Exp. Ther. 2000; 295:934-41. 38. Plopper CG. Pulmonary bronchiolar epithelial cytotoxicity: microanatomical considerations. In: Gram TE (ed.), Metabolic Activation and Toxicity of Chemical Agents to Lung Tissue and Cells, UK: Pergamon Press, 1993, pp. 1-24. 39. Seaton M, Plopper C, Bond J. 1,3-Butadiene metabolism by lung airways isolated from mice and rats. Toxicology 1996; 113:314-17. 40. Watt KC, Plopper CG, Weir AJ et al. Cytochrome P-450 2El in rat tracheobronchial airways: response to ozone exposure. Toxicol. Appl. Pharmacol. 1998; 149:195-202.

41. Watt KC, Morin DM, Kurth MJ et al. Glutathione conjugation of electrophilic metabolites of 1-nitronaphthalene in rat tracheobronchial airways and liver: identification by mass spectrometry and proton nuclear magnetic resonance spectroscopy. Chem. Res. Toxicol. 1999; 12:831-9. 42. Duan X, Buckpitt AR, Plopper CG. Variation in antioxidant enzyme activities in anatomic subcompartments within rat and rhesus monkey lung. Toxicol. Appl. Pharmacol. 1993; 123:73-82. 43. Plopper CG, Duan X, Buckpitt AR etal. Dose-dependent tolerance to ozone. IV. Site-specific elevation in antioxidant enzymes in the lungs of rats exposed for 90 days or 20 months. Toxicol. Appl. Pharmacol. 1994; 127:124-31. 44. West JA, Buckpitt AR, Plopper CG. Elevated airway GSH resynthesis confers protection to Clara cells from naphthalene injury in mice made tolerant by repeated exposures. J. Pharmacol. Exp. Ther. 2000; 297:516-23. 45. Duan X, Plopper C, Brennan P etal. Rates of glutathione synthesis in lung subcompartments of mice and monkeys: possible role in species and site selective injury.J. Pharmacol. Exp. Ther. 1996; 277:1402-9. 46. Hyde DM, Miller LA, McDonald RJ etal. Neutrophils enhance clearance of necrotic epithelial cells in ozoneinduced lung injury in rhesus monkeys. Am. J. Physiol. 1999; 277:L1190-8. 47. Van Winkle LS, Buckpitt AR, Plopper CG. Maintenance of differentiated murine Clara cells in microdissected airway cultures.Am.J. Respir. Cell Mol. Biol. 1996; 14:586-98. 48. Van Winkle L, Isaac J, Plopper C. Repair of naphthaleneinjured microdissected airways in vitro. Am. J. Respir. Cell Mol. Biol. 1996; 15:1-8. 49. Plopper CG, Chang AM, Pang A et al. Use of microdissected airways to define metabolism and cytotoxicity in murine bronchiolar epithelium. Exp. Lung Res. 1991; 17:197-212. 50. Royce FR, Van Winkle LS, Yin Jet al. Comparison of regional variability in lung-specific gene expression using a novel method for RNA isolation from lung subcompartments of rats and mice.Am. J. Pathol. 1996; 148:1779-86. 51. Plopper CG, ten Have-Opbroek AAW. Anatomical and histological classification of the bronchioles. In: Epler GR (ed.), Diseases of the Bronchioles. New York: Raven Press Ltd, 1994, pp. 15-25. 52. Tyler NK, Hyde DM, Hendrickx AG et al. Cytodifferentiation of two epithelial populations of the respiratory bronchiole during fetal lung development in the rhesus monkey. Anat. Rec. 1989; 225:297-309. 53. Plopper CG, Hyde DM. Epithelial cells of bronchioles. Treatise on pulmonary Toxicology. In: Parent RA (ed.), Comparative Biology of the Normal Lung. Boca Raton, FL: CRC Press, 1992, pp. 85-92. 54. Sannes PL. Basement membrane and extracellular matrix. In: Parent RA (ed.), Comparative Biology of the Normal Lung. Boca Raton, FL: CRC Press, 1992, pp. 129-44. 55. ten Have-Opbroek AAW. The structural composition of the pulmonary acinus in the mouse. Anat. Embryol. 1986; 174:49-57. 56. ten Have-Opbroek AAW. Lung development in the mouse embryo. Exp. Lung. Res. 1991; 17:111-30. 57. ten Have-Opbroek AAW, Otto-Verberne CJM, Dubbeldam JA et al. The proximal border of the human respiratory unit, as shown by scanning and transmission electron microscopy and light microscopical cytochemistry. Anat. Rec. 1991; 229:339-54. 58. Tyler NK, Hyde DM, Hendrickx AG et al. Morphogenesis of the respiratory bronchiole in rhesus monkey lungs. Am. J. Anat. 1988; 182:215-23. 59. Smolich JJ, Stratford BF, Maloney JE et al. New features in the development of the submucosal gland of the respiratory tract.J. Anat. 1978; 127:223-38.

60. Krause W, Leeson C. The postnatal development of the respiratory system of the opossum. I. Light and scanning electron microscopy.Am. J. Anat. 1973; 137:337-56. 61. Leigh M, Gambling T, Carson J et al. Postnatal development of tracheal surface epithelium and submucosal glands in the ferret. Exp. Lung. Res. 1986; 10:153-69. 62. Plopper CG, Weir AJ, Nishio SJ et al. Tracheal submucosal gland development in the rhesus monkey, Macaca mulatta: ultrastructure and histochemistry. Anat. Embryol. 1986; 174:167-78. 63. Thurlbeck W, Benjamin B, Reid L. Development and distribution of mucous glands in the fetal human trachea. Br. J. DIS. Chest 1961; 55:54-64. 64. Bucher U, Reid LM. Development of the mucus-secreting elements in human lung. Thorax 1961; 16:219-25. 65. Tos M. Development of the tracheal glands in man. Number, density, structure, shape, and distribution of mucous glands elucidated by quantitative studies of whole mounts. Acta Pathol. Microbiol. Scand. 1966; 68 (Suppl. 185):183+. 66. Tos M. Distribution and situation of the mucous glands in the main bronchus of human fetuses. Anat. Anz. Bd. 1968; 123:481-95. 67. Lamb D, Reid LM. Acidic glycoproteins produced by the mucous cells of the bronchial submucosal glands in the fetus and child: a histochemical autoradiographic study. Br. J. DIS. Chest 1972; 66:248-53. 68. Jeffery P, Reid L. Ultrastructure of airway epithelium and submucosal glands during development. In: Hodson WA (ed.) Development of the Lung. New York: Marcel Dekker, 1977, pp. 87-134. 69. Plopper CG, St. George JA, Cardoso W e t al. Development of airway epithelium: patterns of expression for markers of differentiation. Chest 1992; 101:2S-5S. 70. Leeson TS. The development of the trachea in the rabbit, with particular reference to its fine structure. Anat. Anz. Bd. 1961; 110:214-23. 71. Kawamata S, Fujita H. Finestructural aspects of the development and aging of the tracheal epithelium of mice. Arch. Histol.Jpn. 1983; 46:355-72. 72. McDowell EM, Newkirk C, Coleman B. Development of hamster tracheal epithelium. I. A quantitative morphologic study in the fetus.Anat. Rec. 1985; 213:429-47. 73. Emura M, Mohr U. Morphological studies on the development of tracheal epithelium in the Syrian golden hamster. I. Light microscopy. Zeitschrift Versuchstierk Bd. 1975; 17:14-26. 74. Plopper C, Alley J, Weir A. Differentiation of tracheal epithelium during fetal lung maturation in the rhesus monkey, Macaca mulatta. Am. J. Anat. 1986; 175:59-72. 75. Boyden EA. The development of the lung in the pig-tail monkey (Macaca nemestrina L.). Anat. Rec. 1976; 186:15-38. 76. Lane BP, Gordon RE, Upton AC. Regeneration of rat tracheal epithelium after mechanical injury. I. The relationship between mitotic activity and cellular differentiation. Proc. Soc. Exp. Biol. Med. 1974; 145:1139-44. 77. Boren HG, Paradise LJ. Pathogenesis and Therapy of Lung Cancer. New York: Marcel Dekker, 1978; pp. 369-418. 78. Donnelly GM, Haack DG, Heird CS. Tracheal epithelium: cell kinetics and differentiation in normal rat tissue. Cell Tissue Kinet. 1982; 15:119-30. 79. Keenan KP, Combs JW, McDowell EM. Regeneration of hamster tracheal epithelium after mechanical injury. I. Focal lesions: quantitative morphologic study of cell proliferation. VirchowsArch. (Cell Pathol.) 1982; 41:193-214. 80. Evans MJ, Shami SG. Lung cell kinetics. In: Massaro D (ed.), Lung Cell Biology. New York: Marcel Dekker, 1989, pp. 1-36. 81. McDowell EM, Trump BF. Histogenesis of preneoplastic and neoplastic lesions in tracheobronchial epithelium. Survey Synth. Pathol. Res. 1983; 2:235-79.

82. Rutten AJL, Beems RB, Wilmer JWGM et al. Ciliated cells in vitamin A-deprived culture hamster tracheal epithelium to divide. In Vitro Cell Dev. Biol. 1996; 24:931-5. 83. Chopra DP. Squamous metaplasia in organ culture of vitamin A-deficient hamster trachea: cytokinetic and ultrastructural alterations. J. Natl. Cancer Inst. 1982; 69:895-905. 84. Jeffery P, Ayers M, Rogers D. The mechanisms and control of bronchial mucous cell hyperplasia. In: Chantler EN etal. (eds), Mucus in Health and Disease H. New York, 1982, pp. 399-409. 85. Evans MJ, Moiler PC. Biology of airway basal cells. Exp. Lung Res. 1991; 17:513-31. 86. Plopper CG, St. George J, Pinkerton KE. Tracheobronchial epithelium in vivo: composition, differentiation and response to hormones. In: Thomassen DG, Nettesheim P (eds), Biology, Toxicology, and Carcinogenesis of Respiratory Epithelium. 1990, pp. 6-22. 87. Hislop A, Muri DCF, Jacobsen M et al. Postnatal growth and function of the pre-acinar airways. Thorax 1972; 27:265-74. 88. Burrington JD. Tracheal growth and healing. J. Thor. Cardio Surg. 1978; 76:453-8. 89. Basbaum C, Jany B. Plasticity in the airway epithelium. Am. J. Physiol. 1990; 259:L38-46. 90. Keenan KP, Wilson TS, McDowell EM. Regeneration of hamster tracheal epithelium after mechanical injury. IV. Histochemical, immunocytochemical and ultrastructural studies. Virchows Arch. (Cell Pathol.) 1983; 43:213-40. 91. Johnson NF, Hubbs AF. Epithelial progenitor cells in the rat trachea.Am. J. Respir. Cell Mol. Biol. 1990; 3:579-85. 92. Terzaghi M, Nettesheim P, Williams ML. Repopulation of denuded tracheal grafts with normal, preneoplastic, and neoplastic epithelial cell population. Cancer Res. 1978; 38:4546-53. 93. Inayama Y, Hook GER, Brody AR et al. The differentiation potential of tracheal basal cells. Lab. Invest. 1988; 58:706-17. 94. Inayama Y, Hook GE, Brody AR et al. In vitro and in vivo growth and differentiation of clones of tracheal basal cells. Am. J. Pathol. 1989; 134(3):539-49. 95. Johnson NF, Hubbs AF, Thomassen DG. Epithelial progenitor cells in the rat respiratory tract. In: Thomassen DG, Nettesheim P (eds) Biology, Toxicology and Carcinogenesis of Respiratory Epithelium. Washington, DC: Hemisphere, 1990, pp. 88-98. 96. Plopper C, Alley J, Serabjit-Singh C et al. Cytodifferentiation of the nonciliated bronchiolar epithelial (Clara) cell during rabbit lung maturation: an ultrastructural and morphometric study.Am. J. Anat. 1983; 167:329-57. 97. ten Have-Opbroek AAW, De Vries ECP. Clara cell differentiation in the mouse: ultrastructural morphology and cytochemistry for surfactant protein A and Clara cell 10 kDa protein. Microsc. Res. Tech. 1993; 26:400-11. 98. Massaro GD. Nonciliated bronchiolar epithelial (Clara) cells. In: Massaro D (ed.), Lung Cell Biology. New York: Marcel Dekker, 1989, pp. 81-114. 99. Ito T, Newkirk C, Strum JM etal. Modulation of glycogen stores in epithelial cells during airway development in Syrian golden hamsters: a histochemical study comparing concanavalin a binding with the periodic acid-Schiff reaction. J. Histochem. Cytochem. 1990; 38:691-7. 100. Cardoso W, Stewart LG, Pinkerton KE etal. Secretory product expression during Clara cell differentiation in the rabbit and rat.Am. J. Physiol. 1993; 8:L543-52. 101. Plopper CG, Weir AJ, Morin D et al. Postnatal changes in the expression and distribution of pulmonary cytochrome P-450 monooxygenases during Clara cell differentiation in the rabbit. Mol. Pharmacol. 1993; 44:51-61. 102. Strum JM, Singh G, Katyal SL etal. Immunochemical localization of Clara cell protein by light and electron

Development of Airway Epithelium

microscopy in conducting airways of fetal and neonatal hamster lung. Anat. Rec. 1990; 227:77-86. 103. Ji CM, Plopper CG, Witschi HP et al. Exposure to sidestream cigarette smoke alters bronchiolar epithelial cell differentiation in the postnatal rat lung. Am. J. Respir. Cell Mol. Biol. 1994; 11:312-20. 104. Strum J, Ito T, Philpot R. The immunocytochemical detection of cytochrome P-450 monooxygenase in the lungs of fetal, neonatal, and adult hamsters. Am. J. Respir. Cell Mol. Biol. 1990; 2:493-501. 105. Katyal SL, Singh G, Brown WE etal. Clara cell secretory (10kDa) protein: amino acid and cDNA nucleotide sequences, and developmental expression. Prog. Respir. Res. 1990; 25:29-35. 106. Strum JM, Compton RS, Katyal SL etal. The regulated expression ofmRNA for Clara cell protein in the developing airways of the rat, as revealed by tissue in situ hybridization. Tissue Cell 1992; 24:461-71. 107. Singh G, Katyal SK. Secretory proteins of Clara cells and type II cells. In: Parent RA (ed.) Comparative Biology of the Normal Lung. Boca Raton, FL: CRC Press, 1992, pp. 93-108. 108. Singh G, Katyal SL, Wong-Chong ML. A quantitative assay for a Clara cell-specific protein and its application in the study of development of pulmonary airways in the rat. Ped. Res. 1986; 20:802-5. 109. St. George JA, Read LC, Cranz DL et al. Effect of epidermal growth factor on the fetal development of the tracheobronchial secretory apparatus in rhesus monkey. Am. J. Respir. Cell Mol. Biol. 1991; 4:95-101. 110. Bunton TE, Plopper CG. Triamcinolone-induced structural alterations in the development of the lung of the fetal rhesus macaque. Am. J. Obstet. Gynecol. 1984; 148:203-15. 111. Minoo P. Review transcriptional regulation of lung development: emergence of specificity. University of Southern California School of Medicine, Los Angeles, CA, USA. http ://respiratory-research.corn/content/I/2/109. 112. Yuan B, Li C, Kimura S etal. Inhibition of distal lung morphogenesis in Nkx2.1(-/-) embryos. Dev. Dynam. 2000; 217:180-90. 113. Chen J, Knowles H, Hebert J et al. Mutation of the mouse hepatocyte nuclear factor/forkhead homologue 4 gene results in an absence of cilia and random left-right asymmetry. J. Clin. Invest. 1988; 102:1077-82. 114. Tichelaar J, Lim L, Costa R et al. Hnf-3/forkhead homologue-4 influences lung morphogenesis and respiratory epithelial cell differentiation in vivo. Dev. Biol. 1999; 213: 405-17. 115. Temann U, Geba G, Rankin J e t al. Expression of interleukin 9 in the lungs oftransgenic mice causes airway inflammation, mast cell hyperplasia, and bronchial hyperresponsiveness. J. Exp. Med. 1998; 188:1307-20. 116. Lee J, Kaminski N, Dolganov Get al. Interleukin-13 induces dramatically different transcriptional programs in three human airway cell types: Am. J. Respir. Cell MoL Biol. 2001; 25:474-85. 117. Zhu Z etal. Pulmonary expression of interleukin-13 causes inflammation, mucus hypersecretion, subepithelial fibrosis, physiologic abnormalities and eotaxin production. J. Clin. Invest. 1999; 103:779-88. 118. Zheng T, Zhu S, Wang Z et al. Inducible targeting of IL-13 to the adult lung causes matrix metalloproteinase- and cathepsin-dependent emphysema. J. Clin. Invest. 2000; 106:1081-93. 119. Griinig, G, Warnock M, Wakil A et al. Requirement for IL-13 independently of IL-4 in experimental asthma. Science 1998; 282:2261-3. 120. Wills-Karp M, Luyimbazi J, Xu X et al. Interleukin-13: central mediator of allergic asthma. Science 282:2258-6 1. 121. Lee K, Chuang E, Giffen M etal. Molecular basis of T cell inactivation by CTLA-4' Science 1998; 282:2263-6.

122. Wu R, Groelke JW, Chang LY et al. Effects of hormones on the multiplication and differentiation of tracheal epithelial cells in culture. In: Sirbasku D, Sato GH, Pardee A (eds), Growth of Cells in Hormonally Defined Media. Cold Spring Harbor: Cold Spring Harbor Laboratory, 1982, pp. 641-656. 123. Wu R, Smith D. Continuous multiplication of rabbit tracheal epithelial cells in a defined hormone-supplemented medium. In Vitro 1982; 18:800-12. 124. Wu R. In vitro differentiation of airway epithelial cells. In: Schiff LJ (ed.), In Vitro Models of Respiratory Epithelium. Boca Raton, FL: CRC Press, 1986, pp. 1-26. 125. Robinson CB, Wu R. Culture of conducting airway epithelial cells in serum-free medium. J. Tiss. Cult. Meth. 1991; 13:95-102. 126. Lee TC, Wu R, Brody AR et al. Growth and differentiation of hamster tracheal epithelial cells in culture. Exp. Lung Res. 1983; 6:27-45. 127. Wu R, Nolan E, Turner C. Expression of tracheal differentiated functions in a serum-free hormone-supplemented medium.J. Cell. Physiol. 1985; 125:167-81. 128. Kim KC, Rearick JI, Nettesheim P e t al. Biochemical characterization of mucin secreted by hamster tracheal epithelial cells in primary culture. J. Biol. Chem. 1985; 260:4021-27. 129. Wu R, Plopper CG, Cheng PW. Mucin-like glycoprotein secreted by cultured hamster tracheal epithelial cells. Biochemical and immunological characterization. Biochem. J. 1991; 277:713-18. 130. Wu R, Martin WR, Robinson CB et al. Expression of mucin synthesis and secretion in human tracheobronchial epithelial cells grown in culture. Am. J. Respir. Cell Mol. Biol. 1990; 3:467-78. 131. Adler K, Cheng P, Kim K. Characterization of guinea pig tracheal epithelial cells maintained in biphasic organotypic culture: cellular composition and biochemical analysis of released glycoconjugates. Am. J. Resp. Cell Mol. Biol. 1990; 2:145-54. 132. Wu R, Sato GH, Whitcutt JM. Developing differentiated epithelial cell cultures: airway epithelial cells. Fundam. Appl. Toxicol. 1986; 6:580-90. 133. Adler K, Schwarz J, Whitcutt M e t al. A new chamber system for maintaining differentiated guinea pig respiratory epithelial cells between air and liquid phases. BioTech. 1987; 5:462-5. 134. Whitcutt J, Adler K, Wu R. A biphasic culture system for maintaining polarity of differentiation of cultured respiratory tract epithelial cells. In Vitro Cell Dev. Biol. 1988; 24:420-8. 135. deJong PM, Van Strekenburg MAJA, Hesseling SC etal. Ciliogenesis in human bronchial epithelial cells cultured at the air-liquid interface. Am. J. Respir. Cell Mol. Biol. 1994; 10:271-7. 136. Gray T, Guzman K, Davis C et al. Mucociliary differentiation of serially passaged normal human tracheobronchial epithelial cells.Am.J. Respir. Cell Mol. Biol. 1996; 14:104-12. 137. Wolbach SB, Howe PR. Tissue changes following deprivation of fat-soluble A vitamin.J. Exp. Med. 1926; 42:753-81. 138. Wong YC, Buck RC. An electronic microscopic study of metaplasia of the rat tracheal epithelium in vitamin A deficiency. Lab. Invest. 1971; 24:55-66. 139. Harris CC, Silverman T, Jackson R etal. Proliferation of tracheal epithelial cells in normal and vitamin A-deficient Syrian golden hamsters. J. Natl. Cancer Inst. 1973; 51:1059-62. 140. Sporn MI, Clamon GH' Dunlop NJ et al. Activity of vitamin A analogues in cell cultures of mouse epidermis and organ cultures of hamster trachea. Nature 1975; 253:47-9. 141. Fell HB, Mellanby E. Metaplasia produced in cultures of chick ectoderm by high vitamin A. J. Physiol. 1953; 119:470-88.

142. McDowell EM, Ben T, Coleman B et al. Effects of retinoic acid on the growth and morphology of hamster tracheal epithelial cells in primary culture. Virchows Arch. B Cell Pathol. 1987; 54:38-51. 143. De Luca LM, McDowell EM. Effects of vitamin A status on hamster tracheal epithelium in vivo and in vitro. Food Nutr. Bull. 1989; 11:20-24. 144. McDowell EM, DeSanti AM, Newkirk C etal. Effects of vitamin A-deficiency and inflammation on the conducting airway epithelium of Syrian golden hamsters. Virchows Arch. B Cell Pathol. 1990; 59:231-42. 145. Kopan R, Fuchs E. The use of retinoic acid to probe the relation between hyperproliferation-associated keratins and cell proliferation in normal and malignant epidermal cells.J. Cell Biol. 1989; 109:209-307. 146. Gray T, Koo J, Nettesheim. Regulation of mucous differentiation and mucin gene expression in the tracheobronchial epithelium: Toxicology 2001; 160: 35-46. 147. Smith BT. Lung maturation in the fetal rat: acceleration by injection of fibroblast-pneumonocyte factor. Science 1979; 204:1094-5. 148. Sepulveda J, Velasquez BJ. Study of the influence of NA-872 (Ambroxol) and dexamethasone on the differentiation of Clara cells in albino mice. Respiration 1982; 43:363-8. 149. Massaro D, Massaro G. Dexamethasone accelerates postnatal alveolar wall thinning and alters wall composition. Am. J. Physiol. 1986; 251 :R218-24. 150. Phelps DS, Floros J. Dexamethasone in vivo raises surfactant protein B mRNA in alveolar and bronchiolar epithelium. Am. J. Physiol. 1991; 260:L146-52. 151. Fernandez-Renau D, Lombardero M, Nieto A. Glucocorticoiddependent uteroglobin synthesis and uteroglobulin mRNA levels in rabbit lung explants cultured in vitro. Eur. J. Biochem. 1984; 144:523-7. 152. Lombardero M, Nieto A. Glucocorticoid and developmental regulation of uteroglobin synthesis in rabbit lung. Biochem. J. 1981; 200:487-94. 153. Devereux TR, Fouts JR. Effect of pregnancy or treatment with certain steroids on N,N-dimethylaniline demethylation and N-oxidation by rabbit liver or lung microsomes. Drug Metab. Disp. 1975; 3:254-8. 154. Devereux TR, Fouts JR. Effect of dexamethasone treatment on N,N-dimethylaniline demethylation and N-oxidation in pulmonary microsomes from pregnant and fetal rabbits. Biochem. Pharmacol. 1977; 27:1007-8. 155. Fouts JR, Devereux TR. Developmental aspects of hepatic and extrahepatic drug-metabolizing enzyme systems: microsomal enzymes and components in rabbit liver and lung during the first month of life. J. Pharmacol. Exp. Ther. 1972; 183:458-68. 156. Massaro GD, McCoy L, Massaro D. Development ofbronchiolar epithelium: time course of response to oxygen and recovery.Am. J. Physiol. 1988; 254:R755-60. 157. Massaro G, Olivier J, Massaro D. Brief perinatal hypoxia impairs postnatal development of the bronchiolar epithelium.Am. J. Physiol. 1989; 257:L80-5.

158. Massaro GD, McCoy L, Massaro D. Hyperoxia reversibly suppresses development of bronchiolar epithelium. Am. J. Physiol. 1986; 251:R1045-50. 159. Jetten AM. Growth and differentiation factors in tracheobronchial epithelium. Am. J. Physiol. 1991; 260:L361-73. 160. Kelley J. Cytokines of the lung. Am. Rev. Respir. Dis. 1990~ 141:765-88. 161. Strandjord TP, Clark JG, Madtes DK. Expression of TGF-ot, EGF, and EGF receptor in fetal rat lung. Lung Cell Mol. Physiol. 1994; 267:L384-9. 162. Johnson MD, Gray ME, Carpenter G etal. Ontogeny of epidermal growth factor receptor and lipocortin-1 in fetal and neonatal human lungs. Hum. Pathol. 1990; 21:182-91. 163. Han RN, Liu J, Tanswell AK etal. Ontogeny of plateletderived growth factor receptor in fetal rat lung. Microsc. Res. Tech. 1993; 26:381-8. 164. Caniggia I, Liu J, Han R etal. Fetal lung epithelial cells express receptors for platelet-derived growth factor. Am. J. Resp. Cell Mol. Biol. 1993; 9:54-63. 165. Nexo E, Kryger-Baggesen N. The receptor for epidermal growth factor is present in human fetal kidney, liver and lung. Regul. Pept. 1989; 26:1-8. 166. Stahlman MT, Orth DN, Gray ME. Immunocytochemical localization of epidermal growth factor in the developing human respiratory system and in acute and chronic lung disease in the neonate. Lab. Invest. 1989; 60:539-47. 167. Raaberg L, Seier Poulsen S, Nexo E. Epidermal growth factor in the rat lung. Histochemistry 1991; 95:471-5. 168. Johnson MD, Gray ME, Carpenter G e t al. Ontogeny of epidermal growth factor receptor/kinase and of lipocortin-1 in the ovine lung. Pediatr. Res. 1989; 25:535-41. 169. Snead M, Luo W, Oliver P e t al. Localization of epidermal growth factor precursor in tooth and lung during embryonic mouse development. Dev. Biol. 1989; 134:420-9. 170. Strandjord TP, Clark JG, Hodson WA etal. Expression of transforming growth factor alpha in mid-gestation human fetal lung.Am. J. Respir. Cell Mol. Biol. 1993; 8:266-72. 171. Kida K, Utsuyama M, Takizawa T etal. Changes in lung morphologic features and elasticity caused by streptozotocininduced diabetes mellitus in growing rats. Am. Rev. Respir. Dis. 1983; 128:125-31. 172. Han R, Liu J, Tanswell A et al. Expression of basic fibroblast growth factor and receptor: immunolocalization studies in developing rat fetal lung. Pediatr. Res. 1992; 31:435-40. 173. Stiles AD, D'Ercole AJ. The insulin-like growth factors and the lung.Am. J. Respir. Cell Mol. Biol. 1990; 3:93-100. 174. Warburton D, Seth R, Shum L e t al. Epigenetic role of epidermal growth factor expression and signalling in embryonic mouse lung morphogenesis. Dev. Biol. 1992; 149:123-33. 175. Catterton WZ, Escobedo MB, Sexson WR et al. Effect of epidermal growth factor on lung maturation in fetal rabbits. Pediatr. Res. 1979; 13:104-8. 176. Sundell HW, Gray ME, Serenius FS et al. Effects of epidermal growth factor on lung maturation in fetal lambs. Am. J. Pathol. 1980; 100:707-26.

Development of the Airway Innervation

Chapter 3

Malcolm P. Sparrow* and Markus Weichselbaum Asthma and Allergy Research Institute, Department of Medicine, University of Australia, WA, Australia .

.

.

.

Jenny Toilet and Peter ......K. McFawn Departmentof Physiology, University of Western Australia, WA, Australia

John T. Fisher Department of Physiology, Queens University, Kingston, Ontario, ..... ::

:=:::

i

C a n a d a ~.

:: i:::ii i::::

.

.

.

INTRODUCTION

The innervation and the airway smooth muscle (ASM) have recently been recognized as dominant, integral components of the developing lung: both are present in the epithelial tubules of the embryonic lung bud shortly after it evaginates from the foregut. Whereas the ASM is functionally mature shortly after it is laid down, the growth and maturation of the innervation largely follows the morphological stages of lung development. In the human lung, the primary pattern of branching of the bronchial tree is established during the pseudoglandular stage, followed by elongation of airways with increased vascularization of the lung periphery in the canalicular stage. The development within the acinus (thinning of the epithelial cells, expansion of the air space) occurs during the saccular stage. Until recently, knowledge of the overall innervation of the fetal airway was restricted because nerves were mostly detected only in thin sections by light microscopy, revealing cross-sections or short lengths of a nerve pathway that may be many centimetres in length, or by electron microscopy. Using such methods, a limited number of ganglia and nerve bundles were described in the peribronchial region of large airways of human fetal lungs *To whom correspondence should be addressed. The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

from 9 to 40 weeks gestation. 1 In recent years, the use of confocal microscopy has revealed many hundreds of ganglia, together with their connecting nerve pathways and fine branches supplying the ASM. A result of using thin sections was that studies of the functional neurophysiologic behaviour of airway afferents and efferents preceded detailed knowledge of the morphology of airway innervation. In this chapter our purpose is twofold. First, to provide an overview of the spectacular advances in the knowledge of the development of airway innervation due to new imaging technologies, and secondly, to review the functional behaviour of the efferent nerves associated with the lower airways.

ANATOMY, MORPHOLOGY AND DISTRIBUTION

This section describes recent morphological insights into the ontogeny of the pulmonary innervation in relation to the developing airways. Neural tissue is a dominant feature of the fetal lung and undergoes dramatic development during gestation. The stages of maturation have recently been graphically captured using confocal microscopy. Immunofluorescently stained whole lungs, lobes and airway segments have been scanned by optical sectioning through the entire Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

The Lung::Development,

Agingandihe

vi:ron:ment: :: :

:

:

:i

thickness of the airway wall, using markers of neural tissue in conjunction with markers for ASM and epithelial tubules. From the three-dimensional information obtained, overviews and detailed images of the network of nerves and ganglia that envelop the lung primordia have been prepared. As lung development proceeds through gestation to postnatal life, comprehensive maps of the pathways of the nerves to their target tissues have provided unique views of the airway innervation. The picture that emerges is that neural tissue and ASM are an integral part of the lung from its inception, and persist in a dynamic state throughout gestation and into postnatal life and late adulthood. We present evidence for this beginning in the embryonic lung of the mouse, then proceeding through gestation in the human and the pig lung. Origin o f t h e i n n e r v a t i o n - t h e f e t a l m o u s e lung We first describe the development of the innervation of the fetal mouse lung from embryonic days 10-14, the early pseudoglandular stage. In this period, branching morphogenesis is at its peak, and every 24 h of gestation sees a striking change in lung structure and in the maturation of the innervation that accompanies branching. In mice, two lung buds begin to evaginate from the foregut at embryonic day 10 (El0), 2 whereas in humans and most mammals the lung develops from a single lung bud. Neural crest-derived cells (NCC) are present in the foregut prior to the formation of lung buds and have been assumed to migrate into the lung where they differentiate into intrinsic pulmonary neurons. 3 This migration has recently been demonstrated in the mouse lung by immunostaining whole mounts of foregut including the lung buds and imaging them using

IR

.... :

:

::::: :::

:

: ::

:

:::: : ::: ::::::::::: ::

:

: :

:

confocal laser scanning microscopy, from El0 and thereafter (pseudoglandular stage). 4 NCC are identified with antibodies to protein gene product 9.5 (PGP 9.5; a general neural marker) and NCC-specific markers, including phox2b and p75 NTR. phox2b is a transcription factor located in NCC nuclei. 5 p75 NTR is a low-affinity trk-receptor and is present in the membranes of NCC and their nerve processes. 4-6 Neurons are also identified with an antibody to PGP 9.57 which stains mature neurons and nerve fibres but not precursors (i.e. NCC). At El0, PGP 9.5- and p75-positive nerve fibres run along the dorsal side of the foregut. Among these fibres are many migrating NCC with phox2b-positive nuclei (Fig. 3.1a) and p75NTR-positive membranes (Fig. 3.1b) with vagal processes on either side of the foregut. At this early stage, the emerging lung buds are largely free of NCC, albeit a few solitary NCC at the base of the lung buds with occasional processes directed into the bud (Fig. 3.1b, lower inset). Some NCC in the foregut have matured sufficiently to show PGP 9.5 staining, whereas the cells in the lung buds remain negative for this neuronal marker. By E l l , the neural tissue along the foregut condenses into two large nerve trunks, the vagus nerves, which stain strongly for PGP 9.5. 4 Neural processes positive for PGP 9.5 and p75 NTR reach from the vagi to the trachea and primary bronchi (Fig. 3.2a). The vagi are comprised of neural processes and many migrating NCC (Fig. 3.2b). These processes are likely to comprise both afferent fibres originating from vagal and spinal sensory ganglia, and preganglionic efferents that will ultimately synapse on NCC once they have completed migration. Neural processes from the vagi to the primary bronchi (Fig. 3.2e) and the dorsal trachea contain

b "

..

9 r

i

Ul

" r

>:.,

!t .

,

~: ~::~: c:~::::~:::~:~:::::,:~:

:.....:~:::::~E!!i84184 i Fig. 3.1. Mouse foregut with lung buds at embryonic day 10, dorsal view. (a) Confocal projection of immunostained phox2b-positive nuclei in NCCs in the foregut. Inset: enlargement of the nuclei in a single optical section. (b) Immunostaining with p75 NTR reveals membranes of NCC bodies and vagal processes running down either side of the foregut. Upper inset: enlargement of NCC showing p75-positive cell membranes in a single section. Lower inset: a confocal image of a single p75NTR-positive NCC at the base of the lung bud with projection directed into the lung.

Development

of the Airway

a

Innervation

b

. : .......

:

i!~!~i~:!i:i~ g,

~ii~...... ...~i~84 i~~,~iii!,!i~ 84

',~!,~i~=!~: ........

C .....

i

....

?

........

0

...

II

Fig. 3.2. Mouse lung at embryonic day 11. (a) A confocal projection showing a ventral view of the right upper half of an E11 mouse lung stained for nerves (black) with the protein gene product 9.5 (PGP 9.5). This also stained the undifferentiated epithelium of the tubules and growing end buds (grey). The ASM that covers the tubules is stained with e~-actin (dark grey). The carina lies at the top of the figure. The left vagus (V) sends out nerve processes to the ASM covering the left lobar bronchus. Some extend towards the mesenchymal cap. Inset: videomicrograph ventral side. (b) A single optical section through the vagus nerve shows that it contains NCC and many axons running between them (stained with an antibody to p75 NTRwhich is positive for cell membranes and axons). (c) Nerve fibres going from the vagus into the lung (see (a)) comprise processes and NCC (stained for p75NTR).

migrating NCC. Many NCC are present on the dorsal trachea, located over the trachealis muscle and some are on the ventral surface of the proximal primary bronchi, in the process of aggregating into large ganglia. By El2 the lobular organization of the lung is complete, with one large left lobe and four smaller right lobes. A large nerve plexus is present on the ventral side of the lung on the hilum (Fig. 3.3a) which originates from the vagus 4 and is comprised of nerve fibres and large ganglia-like clusters of NCC, with numerous cells in each cluster. From these ganglia, nerves positive for PGP 9.5 (Fig. 3.3b) and p75 NTR (Fig. 3.3e) extend along the bronchi, following the smooth muscle-covered tubules. NCC also migrate along these nerve tracts but lag behind the

growth of the nerve axons (Fig. 3.3e). Superimposed confocal projections of both the neural tissue and the ASM reveal their close relationship. By El3 the neuronal precursors lying over the dorsal trachea have matured to form a PGP 9.5-positive network of thin nerve trunks interconnected by small ganglia, giving fine fibres that penetrate the smooth muscle layer. By E14 this plexus is more extensive comprising larger ganglia and more numerous thick nerve trunks with multiple connections to the vagi (Fig. 3.4a). Small nerves from the ganglia branch into many fine varicose fibres that run along the smooth muscle bundles. The ganglia vary greatly in size, and many large ganglia contain over 100 cell bodies positive for PGP9.5 (Fig. 3.4a) and their nuclei positive for phox2b

~~The ~k~ngil ~DeVelOpmenL~ :~Agi:ng ~ n d i h e E n ~ i m n m e n t ~

a

b

i :~: ~i~

i: 84 ........~: ........~

~ ,~ ~

Fig. 3.3. Mouse lung at embryonic day 12. (a) A confocal projection (ventral view) showing the Iobular organization (the accessory lobe and the vagi have been removed) with the epithelial tubules in longitudinal section. The first two laterals of the left lobe reveal the end buds in the process of dividing. The undifferentiated epithelium of the tubules, and particularly their end buds are immunoreactive to PGP 9.5 (grey-black). PGP 9.5 diffusely stains ganglia (black) connected by nerve trunks and fibres in the ventral hilum (long arrow). (b) PGP 9.5-positive nerve fibres issue from the large ganglion (long arrow) at the base of the left pulmonary bronchi and reach along the left lobar bronchus (short arrows) and along some of the laterals but no PGP 9.5-positive cells and ganglia are present along the tubules. (c) The large ganglion at the base of the left lobar bronchus (long arrow) contains many NCCs with phox2b-positive nuclei (black) and p75NTR-positive membranes (grey). The NCC migrate along the p75 NTR nerve fibres (grey, short arrows) that grow along the lobar bronchus and laterals. The cells lag behind the growth of fibres; the majority have only reached as far as the first lateral and a few as far as the second lateral. ((a) and (b) can be viewed in colour, see Ref.4, courtesy American Association of Anatomists.)

(Fig. 3.4b). The axons in the nerve bundles connecting the ganglia stain strongly for GFRtxl, the receptor for Glialderived neurotrophic factor (see below). The innervation from the vagus to the main ganglia lying on the dorsal trachea, ventral hilum and left lobe are shown schematically in Fig. 3.4e. During this early pseudoglandular phase, most nerves follow the smooth muscle-covered tubules, but some nerves course through the mesenchyme toward the lung cap, where they form varicose terminal arborizations by E13. 4 Among the first neurotransmitters to appear in the foregut is CGRP at El2. s By El3, nNOS can be demonstrated in the lung by NADPH-diaphorase activity in nerves associated with the airways and blood vessels. At El5, immunostaining reveals the presence of nNOS in neurons and fibres on the trachea, and from the hilum to the bronchioles. 9 Glial-derived neurotrophic factor (GDNF) has been identified as the most important neurotrophic factor in the development of the enteric nervous system 1~ and there is increasing evidence to suggest that G D N F is of similar importance during lung development. 11 In the gut of mice lacking G D N F or RET (receptor for GDNF), all neurons below the oesophagus and proximal stomach are absent 12 but it is not known whether neurons of the lung are affected. In cultured explants of left lung lobes at El2, neurons survive and display proliferation, differentiation and continued migration along the developing smooth musclecovered tubules. 11 In the presence of serum, a characteristic of these explants is the formation of a layer of tx-actin-positive cells (possibly smooth muscle precursors) that grows out from the lung periphery and attracts nerves that grow onto

this layer. When cultured in GDNF-supplemented medium, the amount of neural tissue on this layer increases 14-fold. The neural tissue consists of a high-density network of nerve trunks and large ganglia and comprises many PGP 9.5-positive cells, indicating that migration, proliferation and differentiation of neuronal precursors as well as neurite extension have taken place as a result of stimulation by GDNF. This suggests that GDNF is a chemoattractant to both nerves and NCC. GDNF-impregnated beads attract nerves growing out from cultured lung explants and in some instances NCC surround the treated beads. The membranes and nerve processes of the NCC are positive for the GDNFreceptor, GFRtxl (Fig. 3.4b), suggesting that nerves and NCC are guided by GDNF. The presence of GDNF-mRNA has been demonstrated in the mesenchyme adjacent to the fetal mouse epithelial tubules 13 possibly in the smooth muscle, which thus may play an important role to attract nerve fibres and migrating NCC.

MAPPING THE FETAL PIG AND

INNERVATION: HUMAN LUNG

The rapid development seen in mice during the pseudoglandular stage from days El0 to E14 contrasts with that of large mammals where the equivalent time period lasts from 3 to 8 weeks in pig and 5 to 17 weeks in human. 14'15 In mice at El4 and subsequently, the application of confocal microscopy becomes more difficult. The signal emission is reduced at increasing depth of scanning, which is a consequence of

Development

o f the A i r w a y

Innervation

C

I.vag ga. ga.

br.

Fig. 3.4. Mouse trachea at embryonic day 14. (a) PGP 9.5-positive (black) network of ganglia connected by thick bundles to the vagus (lower right). Nerves from ganglia spread over smooth muscle on the surface of trachea (upper part of panel). (b) Ganglia with phox2b-positive nuclei (white) and nerve trunks staining for GFR0d (grey) lying over the dorsal trachea. (c) Scheme showing the innervation from the vagus to ganglia lying on the dorsal trachea and ventral hilum. Main nerve trunks to the lobes arise from the latter. Oblique ventral view. tr, trachea; fo, foregut; ga, ganglia; I.vag, left vagus; br, bronchus.

the increased tissue thickness and density, and the associated decrease of antibody penetration. These problems can be overcome by removal of the lung cap, mesenchyme and pulmonary vascular tissue leaving the bronchial tree fully exposed, which is only feasible in larger mammals. 16-18 Thus, the entire bronchial tree can be progressively scanned with the confocal microscope at high resolution. Using this approach montages of near-complete bronchial trees in the pseudoglandular stage about 6 m m in length from fetal pigs 14'16 (Fig. 3.5), and smaller lengths of subsegmental airways from fetal humans 17 have been assembled. These clearly display the organization of nerves and ganglia and their relationship to the ASM, the glands and blood vessels; fine detail is also shown at selected sites. 16 Thus, the development of the innervation from the embryonic lung bud to postnatal life is revealed. The structural characteristics and distribution of the nerves are similar in the three species (mouse, pig and human) at comparable developmental stages, and likewise the ASM. The muscle bundles are orientated around the airways perpendicular to their long axis from the trachea to the base of the epithelial buds, and this arrangement persists into

postnatal life. The innervation of the porcine and human bronchial tree from the adventitia to the epithelium has been reported from early gestation to postnatal life. 16-19

Pseudoglandular stage The main characteristics of this stage are: (i)

Chains of forming ganglia connected by thick nerve trunks to each other and to the vagus lying over the ASM of the dorsal trachea and the ventral surface of the hilum. (ii) In general, two thick main nerve trunks extend from the hilum along each airway to the growing tips. These lie above the ASM supported by the mesenchyme. In the fetal pig at 5.5 weeks gestation, proximal trunks (~-50ktm in diam.) run about 40-60~tm above the ASM, progressively thinning distally over a length of 4 mm to --20 ktm in diam. and 15-20 ktm from the ASM. The nerve trunks terminate as thin bundles in the collar of ASM that surrounds the epithelial buds. 16'2~ All along the length of the trunks branches descend towards the smooth muscle, and break up into small

T he: L:un :Deve !o p m:en:i' :::Ag:[n g and he: En:~i::i~:n:~en

a

iliiiii:i~:84:

:::

::

:

:: ::

:

~ S ~: ::

?

~:! ,~.'~

1 rnrn ",I

Fig. 3.5. (a) Montage showing dorsal view of the bronchial tree of fetal pig lung at 5.8 weeks gestation (pseudoglandular stage) with nerves and ganglia stained for synaptic vesicle protein (SV2) and for ASM with smooth muscle myosin. It has been optically sectioned to show the outline of the airway wall. Nerve trunks run down the length of the airways to the epithelial buds. Arrow shows a large ganglion. Ganglia were seen at airway bifurcations and at the branching points of the nerve trunks. An immature ganglion (boxed region) is shown in (b). (b) Two major nerve trunks stained with SV2 give rise to a fine network of varicose processes overlying the ASM. At this stage the varicose fibres are randomly distributed on and in the smooth muscle, some located within 1 pm from the muscle cells. The accumulation of cell bodies (arrow) is a precursor ganglion present at the bifurcation point of the airway. The cell profiles in the ganglion can be distinguished by the SV2-positive nerve fibres lying around them. ((a) and (b) from Ref.16, courtesy American Thoracic Society.) (c) The innervation and ASM in the developing airways of a fetal human lung at 58 days of gestation (pseudoglandular stage). The field shows branching epithelial tubules in the periphery of a lobe. Nerves and ganglia are stained for PGP 9.5 and form a network overlying the ASM stained for 0~-actin. The circumferential arrangement of the muscle bundles around the epithelial tubule is faintly seen in each of the above, where it is perpendicular to the long axis of the tubule. (See Color plate 3.)

Development of the Airway lnnervation

bundles. From these, fine varicose fibres issue that spread over the muscle ending in arborizations --1 ktm from muscle cells, suggesting functional innervation. At this stage the varicose fibres are randomly distributed on and in the smooth muscle (Fig. 3.5b) but later become oriented along the smooth muscle bundles. 16'18 (iii) Immature ganglia are present along the main trunks from which nerve branches radiate out to connect to many other smaller ganglia that form a network covering the airway wall. Fig. 3.5c shows this innervation in the distal airways of a fetal human lung at 7.5 weeks gestation. The mean distance between ganglia TM is 64_+ 18 ktm (n-87), very similar to the pig (70ktm at comparable gestation). Ganglia also lie at most airway branch points and give rise to smaller trunks that follow the airways as they proceed distally. Proximal ganglia are large (e.g. > 300 cell bodies at 5.5 weeks gestation) whereas distal ganglia are small and ultimately comprise a few neurons. Individual neurons within the ganglia show different intensities of staining with PGP 9.5, indicating variance in their type or maturity. (iv) PGP 9.5 diffusely stains nerve trunks, but many cell profiles of Schwann cells remain unstained (revealed using an antibody to the Schwann cell marker S-100). Staining for synaptic vesicle protein 2 (SV2), a component of the membranes of the vesicles in the varicosities, reveals individual varicose fibres in the nerve trunks indicating that vesicle traffic is prolific at this stage of development. 16'2~This abundance of SV2-positive fibres decreases with ongoing maturation; by postnatal life, varicose fibres are restricted to the distal nerve bundles and the fine fibres that lie on and in the ASM. Staining for neurofilament sharply defines a small proportion of individual fibres in a trunk; these can be traced along the tubules where several terminate in the collar of smooth muscle that surrounds the base of the epithelial bud. 2~ The low proportion of neurofilament-positive fibres in the nerve trunks may reflect the level of maturity of these nerves, since the proportion of neurofilament-positive neural tissue increases as gestation progresses. TM

Canalieular stage With airway growth there is increasing spatial separation of the ganglia. The large ganglia lying on the central airways that form nodes at nerve junctions undergo a fourfold increase in separation to --254 ktm. The ganglia vary greatly in s i z e - large ones are 1201.tin at their greatest width and contain as many as 200 neurons of--11 t.tm diameter. Many of those lying on the trunks gradually become displaced laterally to become attached by a stem, with nerves radiating out from them over the airway. TMThe bronchial vasculature becomes more prominent with arterioles running adjacent to the trunks and around the ganglia. Nerve fibres penetrate the submucosal glands. By midterm, ganglia have condensed and become compact and spherical. Fig. 3.6a shows a montage of nerve tracts in the subsegmental airways of

an 18-week fetal human lung. 17 Large nerve trunks run the entire length of airways reducing in diameter from 45 to 9) resulted in a slow increase in force that amounted to approximately 30-40% of the maximal response in this fibre. Force (vertical bar) and time (horizontal bar) calibrations as indicated. (Tracing from McFawn and Fisher, unpublished observations.)

:

ROK may be particularly important in the newborn where significant contractile effects of ct-adrenergic agonists are present. 184-186

ACKNOWLEDGEMENTS Supported by the Raine Medical Foundation of Western Australia, the National Health and Medical Research Council of Australia and a bequest from the late Annie Phillips (M. Sparrow), by the Canadian Institutes of Health Research (CIHR) and the Ontario Thoracic Society (J. Fisher), and a joint Fellowship from the Canadian Thoracic Society/ Glaxo/CIHR (P. Mc Fawn).

REFERENCES 1. Loosli CG, Hung K-S. Development of pulmonary innervation. In Hodson WA (ed.), Development of the Lung, Vol. 6. New York: Marcel Dekker. 1977, pp. 269-309. 2. Spooner BS, Wessells NK. Mammalian lung development: interactions in primordium formation and bronchial morphogenesis.J. Exp. Zool. 1970; 175:445-54. 3. Dey RD, Hung K-S. Development of innervation in the lung. In McDonald JA (ed.), Lung Growth and Development, Vol. 100. New York: Marcel Dekker. 1997, pp. 244-65. 4. Toilet J, Everett AW, Sparrow MP. Spatial and temporal distribution of nerves, ganglia, and smooth muscle during the early pseudoglandular stage of fetal mouse lung development. Dev. Dyn. 2001; 221:48-60. 5. Young HM, Ciampoli D, Hsuan J etal. Expression of ret-, p75(NTR)-, phox2a-, phox2b-, and tyrosine hydroxylaseimmunoreactivity by undifferentiated neural crest-derived cells and different classes of enteric neurons in the embryonic mouse gut. Dev. Dyn. 1996; 216:137-52. 6. Chalazonitis A, Rothman TP, Chen J etal. Age-dependent differences in the effects of GDNF and NT-3 on the development of neurons and glia from neural crest-derived precursors immunoselected from the fetal rat gut: expression of GFRalpha-1 in vitro and in vivo. Dev. Biol. 1998; 204:385-406. 7. Thompson RJ, Doran JF, Jackson P e t al. PGP 9.5 - a new marker for vertebrate neurons and neuroendocrine cells. Brain Res. 1983; 278:224-8. 8. Tharakan T, Kirchgessner AL, Baxi LV et al. Appearance of neuropeptides and NADPH-diaphorase during development of the enteropancreatic innervation. Brain Res. Dev. Brain Res. 1995; 84:26-38. 9. Guembe L, Villaro AC. Histochemical demonstration of neuronal nitric oxide synthase during development of mouse respiratory tract. Am. J. Respir. Cell Mol. Biol. 1999; 20:342-51. 10. Young HM, Hearn CJ, Farlie PG et al. GDNF is a chemoattractant for enteric neural cells. Dev. Biol. 2001; 229:503-16. 11. Toilet J, Everett AE, Sparrow MP. Development of neural tissue and airway smooth muscle in fetal mouse lung explants: a role for GDNF in lung innervation. Am. J. Respir. Cell Mol. Biol. 2002; 26:420-9. 12. Durbec P, Marcos-Gutierrez CV, Kilkenny C etal. GDNF signalling through the ret receptor tyrosine kinase. Nature 1996; 381:789-93. 13. Towers PR, Woolf AS, Hardman P. Glial cell line-derived neurotrophic factor stimulates ureteric bud outgrowth and enhances survival of ureteric bud cells in vitro. Exp. Nephrol. 1998; 6:337-51.

Development of the Airway Innervation

14. Weichselbaum M, Sparrow MP. A confocal microscopic study of the formation of ganglia in the airways of fetal pig lung. Am. J. Respir. Cell Mol. Biol. 1999; 21:607-20. 15. Burri PH. Structural aspects of prenatal and postnatal development and growth of the lung. In McDonald JA (ed.), Lung Growth and Development, Vol. 100. New York: Marcel Dekker. 1997, pp. 1-35. 16. Weichselbaum M, Everett AW, Sparrow MP. Mapping the innervation of the bronchial tree in fetal and postnatal pig lung using antibodies to PGP 9.5 and SV2. Am. J. Respir. Cell Mol. Biol. 1996; 15:703-10. 17. Sparrow MP, Weichselbaum M, McCray PB. Development of the innervation and airway smooth muscle in human fetal lung.Am. J. Respir. Cell Mol. Biol. 1999; 20:550-60. 18. Weichselbaum M. The structure and distribution of the innervation of the developing lung: confocal microscope study. PhD Thesis. Nedlands, Australia: University of Western Australia, Department of Physiology, 2001. 19. Lamb JP, Sparrow MP. Three-dimensional mapping of the sensory innervation with substance P in porcine bronchial mucosa: comparison with human airways. Am. J. Respir. Crit. Care Med. 2002; 166:1269-1281. 20. Sparrow MP, Warwick SP, Everett AW. Innervation and function of the distal airways in the developing bronchial tree of fetal pig lung.Am.J. Respir. Cell Mol. Biol. 1995; 13:518-25. 21. Ten Have-Opbroek AA. The development of the lung in mammals: an analysis of concepts and findings. Am. J. Anat. 1981; 162:201-19. 22. Cadieux A, Springall DR, Mulderry PK etal. Occurrence, distribution and ontogeny of CGRP immunoreactivity in the rat lower respiratory tract: effect of capsaicin treatment and surgical denervations. Neuroscience 1986; 19:605-27. 23. Butcher L. Acetylcholinesterase histochemistry. In: Hoekfeld ABaT (ed.), Handbook of Chemical Neuroanatomy, Vol. 1. Amsterdam: Elsevier. 1983, pp. 1-49. 24. Schemann M, Sann H, Schaaf C et al. Identification of cholinergic neurons in enteric nervous system by antibodies against choline acetyltransferase. Am. J. Physiol. 1993; 265:G1005-9. 25. Lee LY, Pisarri TE. Afferent properties and reflex functions of bronchopulmonary C-fibers. Respir. Physiol. 2001; 125:47-65. 26. Sparrow MP, Weichselbaum M. Structure and function of the adventitial and mucosal nerve plexuses of the bronchial tree in the developing lung. Clin. Exp. Pharmacol. Physiol. 1997; 24:261-8. 27. Undem BJ, Myers AC. Autonomic ganglia. In: Barnes PJ (ed.), Autonomic Control of the Respiratory System. United Kingdom: Harwood Academic Publications. 1997, pp. 87-118. 28. Agostoni E, Chinnock M, Daly MB etal. Functional and histological studies of the vagus nerve and its branches to the heart, lungs and abdominal nerve viscera in the cat. J. Physiol. London 1957; 135:182-205. 29. Warburton D, Zhao J, Berberich MA et al. Molecular embryology of the lung: then, now, and in the future. Am. J. Physiol. 1999; 276:L697-704. 30. Sparrow MP, Warwick SP, Mitchell HW. Foetal airway motor tone in prenatal lung development of the pig. Eur. Respir. J. 1994; 7:1416-24. 31. McCray PB Jr. Spontaneous contractility of human fetal airway smooth muscle. Am. J. Respir. Cell Mol. Biol. 1993; 8:573-80. 32. Schittny JC, Miserocchi G, Sparrow MP. Spontaneous peristaltic airway contractions propel lung liquid through the bronchial tree of intact and fetal lung explants. Am. J. Respir. Cell Mol. Biol. 2000; 23:11-18. 33. Baluk P, Nadel JA, McDonald DM. Substance P-immunoreactive sensory axons in the rat respiratory tract: a quantitative study of their distribution and role in neurogenic inflammation.J. Comp. Neurol. 1992; 319:586-98.

49

34. Larsell O. The ganglia, plexus and nerve-terminations of the mammalian lung and pleura pulmonis. J. Comp. Neurol. 1922; 35:97-132. 35. Sparrow AK, Sparrow MP. The spatial relationship of CGRP nerves in the mucosal nerve plexus to the bronchial circulation in the pig lung. Eur. Resp. J. 1996; 9:290s. 36. Ricco MM, Kummer W, Biglari B etal. Interganglionic segregation of distinct vagal afferent fibre phenotypes in guinea-pig airways. J. Physiol. 1996; 496:521-30. 37. Lundberg JM, Hokfelt T, Martling CR etal. Substance Pimmunoreactive sensory nerves in the lower respiratory tract of various mammals including man. Cell Tissue Res. 1984; 235:251-61. 38. Barnes PJ. Neurogenic inflammation in the airways. Respir. Physiol. 2001; 125:145-54. 39. Larsell O, Dow RS. The innervation of the human lung. Am. J. Anat. 1993; 52:125-46. 40. Li ZS, Fox-Threlkeld JE, Furness JB. Innervation of intestinal arteries by axons with immunoreactivity for the vesicular acetylcholine transporter (VAChT). J. Anat. 1998; 192:107-17. 41. Black JL. Innervation of airway smooth muscle. In: Barnes PJ (ed.), Autonomic Control of the Respiratory System. United Kingdom: Harwood Academic Publications. 1997, pp. 185-200. 42. Rogers DF. Motor control of airway goblet cells and glands. Respir. Physiol. 2001; 125:129-44. 43. Undem BJ, Myers AC. Neural regulation of the immune response. In: Busse WM, Holgate ST (eds), Asthma and Rhinitis, 2nd edition. Oxford: Blackwell Scientific. 2000, pp. 927-44. 44. Rogers DF, Barnes PJ. Neural control of the airway vasculature. In: Barnes PJ (ed.),Autonomic Control of the Respiratory System. United Kingdom: Harwood Academic Publications. 1997, pp. 185-200. 45. Barnes PJ. Neuromodulation in airways. In: Barnes PJ (ed.), Autonomic Control of the Respiratory System. United Kingdom: Harwood Academic Publications. 1997, pp. 139-84. 46. Barnes PJ. Airway neuropeptides. In: Busse WM, Holgate ST (eds), Asthma and Rhinitis, 2nd edition. Oxford: Blackwell Scientific. 2000, pp. 891-908. 47. Honjin R. On the ganglia and nerves of the lower respiratory tract of the mouse.J. Morph. 1954; 95:263-87. 48. Honjin R. On the nerve supply of the mouse with special reference to the structure of the peripheral vegetative system.J. Comp. Neurol. 1956; 105:587-625. 49. Ebina M, Yaegashi H, Takashi T et al. Distribution of smooth muscles along the bronchial tree: a morphometric study of ordinary autopsy lungs. Am. Rev. Respir. Dis. 1990; 141:1322-6. 50. Sparrow MP, Lamb JP. Ontogeny of airway smooth muscle: structure, innervation, myogenesis and function in the fetal lung. Respir. PhysioL Neurobiol. 2003; in press. 51. Sorokin SP, Hoyt RF. Neuroepithelial bodies and solitary small granule-cells. In: Massaro D (ed.), Lung Cell Biology. New York: Marcel Dekker. 1989, pp. 91-344. 52. Scheuermann DW. Comparative histology of pulmonary neuroendocrine cell system in mammalian lungs. Microsc. Res. Tech. 1997; 37:31-42. 53. Adriaensen D, Scheuermann DW. Neuroendocrine cells and nerves of the lung.Anat. Rec. 1993; 236:70-85. 54. Johnson EW, Eller PM, Jafek BW. Protein gene product 9.5like and calbindin-like immunoreactivity in the nasal respiratory mucosa of perinatal humans. Anat. Rec. 1997; 247:38-45. 55. Stahlman MT, Gray ME. Ontogeny of neuroendocrine cells in human fetal lung. I. An electron microscopic study. Lab. Invest. 1984; 51:449-63.

50

he Lung. Development,:::: Aging and:the

56. Stahlman MT, Gray ME. Immunogold EM localization of neurochemicals in human pulmonary neuroendocrine cells. Microsc. Res. Tech. 1997; 37:77-91. 57. Polak JM, Becker KL, Cutz E et al. Lung endocrine cell markers, peptides, and amines.Anat. Rec. 1993; 236:169-71. 58. Lauweryns JM, Cokelaere M, Deleersynder M e t a l . Intrapulmonary neuro-epithelial bodies in newborn rabbits. Influence of hypoxia, hyperoxia, hypercapnia, nicotine, reserpine, L-DOPA and 5-HTP. Cell Tissue Res. 1977; 182:425-40. 59. Youngson C, Nurse C, Yeger H et al. Oxygen sensing in airway chemoreceptors. Nature 1993; 365:153-5. 60. Reynolds SD, Giangreco A, Power JH etal. Neuroepithelial bodies of pulmonary airways serve as a reservoir of progenitor cells capable of epithelial regeneration. Am. J. Pathol. 2000; 156:269-78. 61. Lauweryns JM, Peuskens JC, Cokelaere M. Argyrophil, fluorescent and granulated (peptide and amine producing?) AFG cells in human infant bronchial epithelium. Light and electron microscopic studies. Life. Sci. 1970; 9:1417-29. 62. van Lommel A, Lauweryns JM. Neuroepithelial bodies in the Fawn Hooded rat lung: morphological and neuroanatomical evidence for a sensory innervation.J. Anat. 1993; 183:553-66. 63. Adriaensen D, Timmermans JP, Brouns I etal. Pulmonary intraepithelial vagal nodose afferent nerve terminals are confined to neuroepithelial bodies: an anterograde tracing and confocal microscopy study in adult rats. Cell Tissue Res. 1998; 293:395-405. 64. Weichselbaum M, Sparrow MP, Thompson PJ et al. A confocal microscopy study of pulmonary neuroendocrine cells in human adult airway epithelium. In: Conference Proceedings of the Thoracic Society of Australia and New Zealand (Cairns) 2002. 65. Gosney JR, Sissons MC, O'Malley JA. Quantitative study of endocrine cells immunoreactive for calcitonin in the normal adult human lung. Thorax 1985; 40:866-9. 66. Coleridge HM, Coleridge JC. Reflexes evoked from the tracheobronchial tree and lungs. In: Cherniack NS, Widdicombe JG (eds), Handbook of Physiology, Section 3: The Respiratory System, Vol. II: Control of Breathing Part 1. Washington, DC: American Physiological Society. 1986, pp. 395-429. 67. Waldron MA, Fisher JT. Neural control of airway smooth muscle in the newborn. In: Haddad G, Farber JP (eds), Developmental Neurobiology of Breathing. New York: Marcel Dekker. 1991, pp. 483-518. 68. Merritt TA, Northway WH, Boynton BR. Bronchopulmonary Dysplasia. Boston: Blackwell Scientific Publications, 1985. 69. Northway WH. Bronchopulmonary dysplasia. Biomed. Pharmacother. 1991; 45: 323. 70. Motoyama EK, Fort MD, Klesh KW et al. Early onset of airway reactivity in premature infants with bronchopulmonary dysplasia.Am. Rev. Respir. Dis. 1987; 136:50-7. 71. Kao LC, Warburton D, Platzker ACG et al. Effect of isoproterenol inhalation on airway resistance in chronic bronchopulmonary dysplasia. Pediatrics 1984; 73:509-14. 72. Brundage KL, Mohsini KG, Froese ABet al. Bronchodilator response to ipratropium bromide in infants with bronchopulmonary dysplasia.Am. Rev. Respir. Dis. 1990; 142:1137-42. 73. Denjean A, Guimaraes H, Migdal M e t a l . Dose-related bronchodilator response to aerosolized salbtamol (albuterol) in ventilator-dependent infants.J. Pediatr. 1992; 120:974-9. 74. Sollmann T, Gilbert AJ. Microscopic observations ofbronchiolar reactions.J. Pharmacol. Exp. Ther. 1937; 61:272-85. 75. Spina D, Shah S, Harrison S. Modulation of sensory nerve function in the airways. Trends Pharmacol. Sci. 1998; 19: 460-6. 76. McCray Jr. PB. Spontaneous contractility of human fetal airway smooth muscle.Am. J. Respir. Cell Mol. Biol. 1993; 8:573-80.

:::::....

:

77. Sparrow MP, Mitchell HW. Contraction of smooth muscle of pig airway tissues from before birth to maturity. J. Appl. Physiol. 1990; 68:468-77. 78. Mitchell HW, Sparrow MP, Tagliaferri RP. Inhibitory and excitatory responses to field stimulation in fetal and adult pig airway. Pediatr. Res. 1990; 28:69-74. 79. Booth RJ, Sparrow MP, Mitchell HW. Early maturation of force production in pig tracheal smooth muscle during fetal development. Am. J. Respir. Cell Mol. Biol. 1992; 7:590-7. 80. Sparrow MP, Warwick SP, Mitchell HW. Fetal airway motor tone in prenatal lung development of the pig. Eur. Respir. J. 1994; 7:1416-24. 81. Sparrow MP, Warwick SP, Everett AW. Innervation and function of the distal airways in the developing bronchial tree of fetal pig lung. Am. J. Respir. Cell Mol. Biol. 1995; 13:518-25. 82. McFawn PK, Mitchell HW. Bronchial compliance and wall structure during development of the immature human and pig lung. Eur. Respir. J. 1997; 10:27-34. 83. McFawn PK, Mitchell HW. Effect of transmural pressure on preloads and collapse of immature bronchi. Eur. Respir. J. 1997; 10:322-9. 84. Richards IS, Kulkarni A, Brooks SM. Human fetal tracheal smooth muscle produces spontaneous electromechanical oscillations that are Caz+ dependent and cholinergically potentiated. Dev. Pharmacol. Ther. 1991; 16:22-8. 85. Liu M, Skinner SJ, Xu Jet al. Stimulation of fetal rat lung cell proliferation in vitro by mechanical stretch. Am. J. Physiol. 1992; 263:L376-83. 86. Liu M, Qin Y, Liu J e t al. Mechanical strain induces pp60src activation and translocation to cytoskeleton in fetal rat lung cells.J. Biol. Chem. 1996; 271:7066-71. 87. Liu M, Tanswell AK, Post M. Mechanical force-induced signal transduction in lung cells. Am. J. Physiol. 1999; 277:L667-83. 88. Liu M, Post M. Invited review: mechanochemical signal transduction in the fetal lung. J. Appl. Physiol. 2000; 89:2078-84. 89. Weichselbaum M, Everett AW, Sparrow MP. Mapping the innervation of the bronchial tree in fetal and postnatal pig lung using antibodies to PGP 9.5 and SV2. Am. J. Respir. Cell Mol. Biol. 1996; 15:703-10. 90. Sparrow MP, Weichselbaum M. Structure and function of the adventitial and mucosal nerve plexuses of the bronchial tree in the developing lung. Clin. Exp. Pharmacol. Physiol. 1997; 24:261-8. 91. Weichselbaum M, Sparrow MP. A confocal microscopic study of the formation of ganglia in the airways of fetal pig lung. Am. J. Respir. Cell Mol. Biol. 1999; 21:607-20. 92. Wong KA, Bano A, Rigaux A et al. Pulmonary vagal innervation is required to establish adequate alveolar ventilation in the newborn lamb.J. Appl. Physiol. 1998; 85:849-59. 93. Hasan SU, Lalani S, Remmers JE. Significance of vagal innervation in perinatal breathing and gas exchange. Respir. Physiol. 2000; 119:133-41. 94. Schwieler GH, Douglas JS, Bouhuys A. Postnatal development of autonomic efferent innervation in the rabbit. Am. J. Physiol. 1970; 219:391-7. 95. Waldron MA, Fisher JT. Differential effects of CO z and hypoxia on bronchomotor tone in the newborn dog. Respir. Physiol. 1988; 72:271-82. 96. Fisher JT, Brundage KL, Waldron MA et al. Vagal cholinergic innervation of the airways in newborn cat and dog. J. Appl. Physiol. 1990; 69:1525-31. 97. Tepper RS. Maturation affects the maximal pulmonary response to methacholine in rabbits. Pediatr. Pulmonol. 1993; 16:48-53.

Development of the Airway Innervation

98. Tepper RS, Gunst SJ, Doerschuk CM et al. Effect of transpulmonary pressure on airway closure in immature and mature rabbits.J. Appl. Physiol. 1995; 78:505-12. 99. Tepper RS, Du T, Styhler A et al. Increased maximal pulmonary response to methacholine and airway smooth muscle in immature compared with mature rabbits. Am. J. Respir. Crit. Care Med. 1995; 151:836-40. 100. Hayashi S, Toda N. Age-related alterations in the response of rabbit tracheal smooth muscle to agents. J. Pharmacol. Exp. Ther. 1980; 214:675-81. 101. Duncan PG, Douglas JS. Influences of gender and maturation on responses of guinea-pig airway tissues to LTD4. Eur. J. Pharmacol. 1985; 112:423-7. 102. Panitch HB, Allen JL, Ryan JP etal. A comparison of preterm and adult airway smooth muscle mechanics. J. Appl. Physiol. 1989; 66:1760-5. 103. Murphy TM, Mitchell RW, Blake JS etal. Expression of airway contractile properties and acetylcholinesterase activity in swine.J. Appl. Physiol. 1989; 67:174-80. 104. Mitchell RW, Murphy TM, Kelly E etal. Maturation of acetylcholinesterase expression in tracheal smooth muscle contraction.Am. J. Physiol. 1990; 259:L130-5. 105. Murphy TM, Mitchell RW, Phillips IJ. Ontogenic expression of acetylcholinesterase activity in trachealis of young swine. Am. J. Physiol. Lung Cell Mol. Physiol. 1991; 261:L322-6. 106. Sauder RA, McNicol KJ, Stecenko AA. Effect of age on lung mechanics and airway reactivity in lambs. J. Appl. Physiol. 1986; 61:2074-80. 107. Ikeda K, Mitchell RW, Guest KA et al. Ontogeny of shortening velocity in porcine trachealis. Am. J. Physiol. 1992; 262:L280-5. 108. Murphy TM, Mitchell RW, Halayko A et al. Effect of maturational changes in myosin content and morphometry on airway smooth muscle contraction. Am. J. Physiol. 1991; 260:L471-80. 109. Fisher JT. Airway smooth muscle contraction at birth: in vivo versus in vitro comparisons to the adult. Can. J. Physiol. Pharmacol. 1992; 70:590-6. 110. Stevens EL, Uyehara CFT, Southgate WM et al. Furosemide differentially relaxes airway and vascular smooth muscle in fetal, newborn, and adult guinea pigs. Am. Rev. Respir. D/s. 1992; 146:1192-7. 111. Southgate WM, Pichoff BE, Stevens EL etal. Ontogeny of epithelial modulation of airway smooth muscle function in the guinea pig. Pediatr. Pulmonol. 1993; 15:105-10. 112. Fayon M, Ben-Jebria A, Elleau Cet al. Human airway smooth muscle responsiveness in neonatal lung specimens. Am. J. Physiol. 1994; 267:L180-6. 113. Penn RB, Wolfson MR, Shaffer TH. Effect of tracheal smooth muscle tone on collapsibility of immature airways. J. Appl. Physiol. 1988; 65:863-9. 114. Mitchell HW, McFawn PK, Sparrow MP. Increased narrowing of bronchial segments from immature pigs. Eur. Resp. J. 1992; 5:207-12. 115. Bhutani VK, Koslo RJ, Shaffer TH. The effects of tracheal smooth muscle tone on neonatal airway collapsibility. Pediatr. Res. 1986; 20:492-5. 116. Bhutani VK, Rubenstein SD, Shaffer TH. Pressure-volume relationships of tracheae in fetal, newborn, and adult rabbits. Respir. Physiol. 1981; 43:221-31. 117. Bhutani VK, Rubenstein SD, Shaffer TH. Pressure induced deformation in immature airways. Pediatr. Res. 1981; 15:829-32. 118. Fisher JT, Haxhiu MA, Martin RJ. Regulation of lower airway function. In: Polin RA, Fox WW (eds), Fetal and Neonatal Physiology. Philadelphia, PA: W.B. Saunders. 1998, pp. 1060-70.

119. Fisher JT, Brundage KL, Anderson JW. Cardiopulmonary actions of muscarinic receptor subtypes in the newborn dog. Can. jT. Physiol. Pharmacol. 1995; 74:603-13. 120. Fisher JT, Mortola JP. Statics of the respiratory system in newborn mammals. Respir. Physiol. 1980; 41:155-72. 121. Fisher JT, Mortola JP. Statics of the respiratory system and growth: an experimental and allometric approach. Am..7. Physiol. 1981; 241:R336-41. 122. Mortola JP. Dynamics of breathing in newborn mammals. Physiol. Rev. 1987; 67:187-243. 123. Mortola JP. Respiratory physiology of newborn mammals: a comparative perspective. Baltimore: The Johns Hopkins University Press, 2001. 124. Fisher JT, McFawn PK, Allen MA etal. Unpublished observations. 125. Anderson JW, Fisher JT. Capsaicin-induced reflex bronchoconstriction in the newborn. Respir. Physiol. 1993; 93:13-27. 126. Nault MA, Vincent SG, Fisher JT. Mechanisms of capsaicinand lactic acid-induced bronchoconstriction in the newborn dog.J. Physiol. (Lond.) 1999; 515:567-78. 127. Marantz MJ, Vincent SG, Fisher JT. Role of vagal C-fiber afferents in the bronchomotor response to lactic acid in the newborn dog. J. Appl. Physiol. 2001; 60:2311-18. 128. Richardson CA, Herbert DA, Mitchell RA. Modulation of pulmonary stretch receptors and airway resistance by parasympathetic efferents. J. Appl. Physiol. 1984; 57:1842-9. 129. Nadel JA, Widdicombe JG. Effect of changes in blood gas tensions and carotid sinus pressure on tracheal volume and total lung resistance to airflow. J. Physiol. 1962; 163:13-33. 130. Green M, Widdicombe JG. The effect of ventilation of dogs with different gas mixtures on airway calibre and lung mechanics.J. Physiol. (Lond.) 1966; 186:363-81. 131. Fisher JT, Waldron MA, Armstrong CJ. Effects ofhypoxia on lung mechanics in the newborn cat. Can. J. Physiol. Pharmacol. 1987; 65:1234-8. 132. Mortola JP, Tenney SM. Effects of hyperoxia on ventilatory and metabolic rates of newborn mice. Respir. Physiol. 1986; 63:267-74. 133. Mortola JP, Rezzonico R. Metabolic and ventilatory rates in newborn kittens during acute hypoxia. Respir. Physiol. 1988; 73:55-68. 134. Caterina MJ, Schumacher MA, Tominaga M et al. The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 1997; 389:816-24. 135. Lee LY, Pisarri TE. Afferent properties and reflex functions of bronchopulmonary C-fibers. Respir. Physiol. 2001; 125:47-65. 136. Lee LY, Morton RF, Lundberg JM. Pulmonary chemoreflexes elicited by intravenous injection of lactic acid in anesthetized rats. J. Appl. Physiol. 1996; 81:2349-57. 137. Diaz V, Dorion D, Kianicka Iet al. Vagal afferents and active upper airway closure during pulmonary edema in lambs. J. Appl. Physiol. 1999; 86:1561-9. 138. Diaz V, Dorion D, Renolleau S e t al. Effects of capsaicin pretreatment on expiratory laryngeal closure during pulmonary edema in lambs. J. Appl. Physiol. 1999; 86:1570-7. 139. Hosey MM. Diversity of structure, signalling and regulation within the family of muscarinic cholinergic receptors. FASEBJ. 1992; 6:845-52. 140. Hulme EC, Birdsall NJM, Buckley NJ. Muscarinic receptor subtypes. Annu. Rev. Pharmacol. Toxicol. 1990; 30:633-73. 141. Wess J. Molecular biology of muscarinic acetylcholine receptors. Crit. Rev. Neurobiol. 1996; 10:69-99. 142. Pitcher JA, Freedman NJ, Lefkowitz RJ. G protein-coupled receptor kinases. Annu. Rev. Biochem. 1998; 67:653-92. 143. Gomeza J, Zhang L, Kostenis E etal. Enhancement of D1 dopamine receptor-mediated locomotor stimulation in

M(4) muscarinic acetylcholine receptor knockout mice. Proc. Natl. Acad. Sci. USA 1999; 96:10483-8. 144. Gomeza J, Shannon H, Kostenis E et al. Pronounced pharmacologic deficits in M2 muscarinic acetylcholine receptor knockout mice. Proc. Natl. Acad. Sci. USA 1999; 96:1692-7. 145. Shapiro MS, Loose MD, Hamilton SE etal. Assignment of muscarinic receptor subtypes mediating G-protein modulation of Ca(2+) channels by using knockout mice. Proc. Natl. Acad. Sci. USA 1999; 96:10899-904. 146. Stengel PW, Gomeza J, Wess J etal. M(2) and M(4) receptor knockout mice: muscarinic receptor function in cardiac and smooth muscle in vitro. J. Pharmacol. Exp. Ther. 2000; 292:877-85. 147. Fisher JT, Froese AB, Brundage KL. Bases physiologiques de l'utilisation des antagonistes muscariniques dans les dysplasies bronchopulmonaires. Arch. Pediatr. 1995; 2 (Suppl. 2):163S-71S. 148. Julia-Serda G, Molfino NA, Chapman KR et al. Heterogeneous airway tone in asthmatic subjects. J. Appl. Physiol. 1992; 73:2328-32. 149. Molfino NA, Slutsky AS, Hoffstein Vet al. Changes in crosssectional airway areas induced by methacholine, histamine, and LTC4 in asthmatic subjects. Am. Rev. Respir. Dis. 1992; 146:514-80. 150. Molfino NA, Slutsky AS, Julia-Serda G et al. Assessment of airway tone in asthma: comparison between double lung transplant patients and healthy subjects. Am. Rev. Respir. DIS. 1993; 148:1238-43. 151. Haxhiu-Poskurica B, Ernsberger P, Haxhiu MA et al. Development of cholinergic innervation and muscarinic receptor subtypes in piglet trachea. Am. J. Physiol. 1993; 264: L606-14. 152. Maclagan J, Barnes PJ. Muscarinic pharmacology of the airways. TIPS 1989; 10 (Suppl. Dec.):88-92. 153. Barnes PJ. Muscarinic receptor subtypes: implications for therapy.Agents Actions 1993; 43 (Suppl.):243-52. 154. Barnes PJ. Muscarinic receptor subtypes in airways. Life Sci. 1993; 52:521-7. 155. Lazareno S, Buckley NJ, Roberts FF. Characterization of muscarinic M 4 binding sites in rabbit lung, chicken heart, and NG108-15 cells. Mol. Pharmacol. 1990; 38:805-15. 156. D6rje F, Levey AI, Brann MR. Immunological detection of muscarinic receptor subtype proteins (ml-m5) in rabbit peripheral tissues. Mol. Pharmacol. 1991; 40:459-62. 157. Mak JCW, Haddad E-B, Buckley NJ etal. Visualization of muscarinic m 4 mRNA and M 4 receptor subtype in rabbit lung. Life Sci. 1993; 53:1501-8. 158. Maclagan J, Fryer AD, Faulkner D. Identification of M 1 muscarinic receptors in pulmonary sympathetic nerves in the guinea-pig by use of pirenzepine. Br. J. Pharmacol. 1989; 97:499-505. 159. Beck KC, Vettermann J, Flavahan NA et al. Muscarinic M1 receptors mediate the increase in pulmonary resistance during vagus nerve stimulation in dogs. Am. Rev. Respir. DIS. 1987; 136:1135-9. 160. Lammers J, Minette P, McCusker M etal. The role of pirenzepine-sensitive (M1) muscarinic receptors in vagally mediated bronchoconstriction in humans. Am. Rev. Respir. Dis. 1989; 139:446-9. 161. Maclagan J, Faulkner D. Effect of pirenzepine and gallamine on cardiac and pulmonary muscarinic receptors in the rabbit. Br. J. Pharmacol. 1989; 97:506-12. 162. Eltze M, Galvan M. Involvement of muscarinic M z and M3, but not of M 1 and M 4 receptors in vagally stimulated contractions of rabbit bronchus/trachea. Pulmonary Pharmacol. 1994; 7:109-20. 163. Fryer AD, Maclagan J. Muscarinic inhibitory receptors in pulmonary parasympathetic nerves in the guinea-pig. Br. J. Pharmacol. 1984; 83:973-8.

164. Blaber LC, Fryer AD, Maclagan J. Neuronal muscarinic receptors attenuate vagally induced contraction of feline bronchial smooth muscle. Br. J. Pharmacol. 1985; 86:723-8. 165. Faulkner D, Fryer AD, Maclagan J. Postganglionic muscarinic inhibitory receptors in pulmonary parasympathetic nerves in the guinea-pig. Br. J. Pharmacol. 1986; 88:181-7. 166. Ito Y, Yoshitomi T. Autoregulation of acetylcholine release from vagus nerve terminals through activation of muscarinic receptors in the dog trachea. Br. J. Pharmacol. 1988; 93:636-46. 167. Minette PA, Barnes PJ. Prejunctional inhibitory muscarinic receptors on cholinergic nerves in human and guinea pig airways.J. Appl. Physiol. 1988; 64:2532-7. 168. Watson N, Barnes PJ, Maclagan J. Actions of methocitramine, a muscarinic M2 receptor antagonist, on muscarinic and nicotinic cholinoceptors in guinea-pig airways in vivo and in vitro. Br. J. Pharmacol. 1992; 105:107-12. 169. Fryer AD, Jacoby DB. Parainfluenza virus infection damages inhibitory M z muscarinic receptors on pulmonary parasympathetic nerves in the guinea pig. Br. J. Pharmacol. 1991; 102:267-71. 170. Schultheis AH, Bassett DJP, Fryer AD. Ozone-induced airway hyperresponsiveness and loss of neuronal M2 muscarinic receptor function.J. Appl. Physiol. 1994; 76:1088-97. 171. Gambone LM, Elbon CL, Fryer AD. Ozone-induced loss of neuronal M z muscarinic receptor function is prevented by cyclophosphamide.J. Appl. Physiol. 1994; 77:1492-9. 172. Janssen LJ, Daniel EE. Pre- and postjunctional muscarinic receptors in canine bronchi. Am. J. Physiol. (Lung Cell Mol. Physiol.) 1990; 259:L304-14. 173. Gross NJ. Ipratropium bromide. N. Engl. J. Med. 1988; 319: 486-94. 174. DeTroyer A, Yernault JC, Rodenstein D. Effects of vagal blockade on lung mechanics in normal man. J. Appl. Physiol. 1979; 46:217-26. 175. Somlyo AP, Somlyo AV. Signal transduction and regulation in smooth muscle. Nature 1994; 372:231-6. 176. Somlyo AP, Somlyo AV. From pharmacomechanical coupling to G-proteins and myosin phosphatase [published erratum appears in Acta Physiol. Scand. 1999; 165(4):423]. Acta. Physiol. Scand. 1998; 164:437-48. 177. Somlyo AP, Himpens B. Cell calcium and its regulation in smooth muscle. FASEB J. 1989; 3:2266-76. 178. Walsh MP. Regulation of vascular smooth muscle tone. Can. J. Physiol. Pharmacol. 1994; 72:919-36. 179. Kitazawa T, Gaylinn BD, Denney GH etal. G-proteinmediated Ca2+ sensitization of smooth muscle contraction through myosin light chain phosphorylation. J. Biol. Chem. 1991; 266:1708-15. 180. Masuo M, Reardon S, Ikebe M et al. A novel mechanism for the CaZ+-sensitizing effect of protein kinase C on vascular smooth muscle: inhibition of myosin light chain phosphatase. J. Gen. Physiol. 1994; 104:265-86. 181. Kitazawa T, Somlyo AP. Desensitization and muscarinic re-sensitization of force and myosin light chain phosphorylation to cytoplasmic Ca2+ in smooth muscle. Biochem. Biophys. Res. Commun. 1990; 172:1291-7. 182. Gerthoffer WT. Regulation of the contractile element of airway smooth muscle. Am. J. Physiol. 1991; 261 :L 15-28. 183. Gunst SJ, Gerthoffer WT, al Hassani MH. Ca2+ sensitivity of contractile activation during muscarinic stimulation of tracheal muscle. Am. J. Physiol. 1992; 263:C1258-65. 184. Pandya KH. Postnatal developmental changes in adrenergic receptor responses of the dog tracheal muscle. Arch. Int. Pharmacodyn. Ther. 1977; 230:53-64. 185. Watanabe H, Vincent SG, Fisher JT. Alpha adrenergic receptor contractile responses in the newborn dog. Can. J. Physiol. Pharmacol. 1994; 72 (Suppl. 1):483.

Development of the Airway Innervation

186. Watanabe H, Fisher JT. Alpha adrenergic receptors mediate contractile responses to catecholamines in neonatal canine airway smooth muscle. Am. J. Respir. Crit. Care Med. 1995; 151 :A438 [Abstract]. 187. Nauli SM, Ally A, Zhang L et al. Maturation attenuates the effects of cGMP on contraction, [Ca2+]i and Ca2+ sensitivity in ovine basilar arteries. Gen. Pharmacol. 2000; 35:107-18. 188. Macara IG, Lounsbury KM, Richards SA etal. The Ras superfamily of GTPases. FASEB J. 1996; 10:625-30. 189. van Eyk JE, Arrell DK, Foster DB et al. Different molecular mechanisms for Rho family GTPase-dependent, Ca2+independent contraction of smooth muscle. J. Biol. Chem. 1998; 273:23433-9. 190. Schmitz AA, Govek EE, Bottner B etal. Rho GTPases: signaling, migration, and invasion. Exp. Cell Res. 2000; 261:1-12. 191. Bishop AL, Hall A. Rho GTPases and their effector proteins. Biochem. J. 2000; 348 (Pt 2):241-55. 192. Manser E, Leung T, Salihuddin H etal. A brain serine/ threonine protein kinase activated by Cdc42 and Racl. Nature 1994; 367(6458):40-6.

193. Foster DB, Shen LH, Kelly Jet al. Phosphorylation of caldesmon by p21-activated kinase. Implications for the Ca(2+) sensitivity of smooth muscle contraction. J. Biol. Chem. 2000; 275:1959-65. 194. Kubota Y, Nomura M, Kamm KE etal. GTPgammaSdependent regulation of smooth muscle contractile elements. Am. J. Physiol. (Cell Physiol.) 1992; 262:C405-10. 195. Uehata M, Ishizaki T, Satoh H et al. Calcium sensitization of smooth muscle mediated by a Rho-associated protein kinase in hypertension. Nature 1997; 389(6654) :990-4. 196. McFawn PK, Shen L, Vincent SG et al. Calcium-independent contraction and sensitisation of airway smooth muscle by p21-activated protein kinase. Am. J. Physiol. (Lung Cell Mol. PhysioL) 2003; 284:L863-70. 197. Dechert MA, Holder JM, Gerthoffer WT. p21-activated kinase 1 participates in tracheal smooth muscle cell migration by signaling to p38 Mapk. Am. J. Physiol. Cell Physiol. 2001; 281:C123-32.

Chapter 4

Development of Aiveoli Stephen E. McGowan Department of Internal Medicine, Veterans Affairs Research Service, University of Iowa, iowa City, IA, USA

Jeanne M. Snyder* Department of Anatomy and Cell Biology, University of Iowa College of Medicine, Iowa City, IA, USA

INTRODUCTION

The respiratory system consists of the trachea, the conducting airways and the gas-exchange portion of the lung, which includes the respiratory bronchioles, alveolar ducts, alveolar sacs and alveoli. 1 Alveoli are tiny, thin-walled sacs that facilitate the exchange of gases between capillaries in the alveolar wall and air brought into the lung during inspiration. The alveolar surface area available for gas exchange is enormous, approximately 100 m 2 in an adult human, and represents the contribution of approximately 300• 106 alveoli. 1-3 The structure of an individual lung alveolus is deceptively simple. The alveolar wall in the adult lung consists of a narrow connective tissue core that contains fibroblasts, myofibroblasts and capillary endothelial cells plus extracellular matrix (ECM) components, most importantly, elastin. The alveolar epithelium is made up of two cell types, namely alveolar type I cells and alveolar type II cells. Alveolar type I cells are thin, flattened cells that, together with the capillary endothelial cell and the fused basal laminae of the capillary and epithelium, form the actual gas-blood exchange barrier. Alveolar type I cells cover about 90% of the alveolar surface area. 3 The alveolar type II cell is a roughly cuboidal cell, frequently located in the corner of an alveolus, that occupies less than 10% of the alveolar surface area. The alveolar type II cell is a stem cell for renewal of the damaged alveolar epithelium since alveolar type I cells do not divide. 4 Alveolar type II cells secrete pulmonary surfactant, a lipoprotein substance that spreads on the alveolar aqueous *To whom correspondence should be addressed. The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

lining layer and reduces its surface tension 5 (see Chapter 10). Adequate amounts of properly functioning pulmonary surfactant are required for normal lung function. 5 In the human, the formation of pulmonary alveoli begins during late gestation and continues after birth. 1-3 In other species, alveolarization is either predominantly postnatal (i.e. in rats and mice) or predominantly prenatal (i.e. in guinea pigs and rabbits). 6 Regardless of the timing, alveolarization seems to occur via a roughly similar process in all vertebrates. An impairment of alveolarization has been implicated in the pathogenesis of bronchopulmonary dysplasia (BPD), a disease that affects prematurely born human infants. 7 Damage to existing alveoli, with a resulting decrease in gasexchange surface area, is involved in the pathogenesis of emphysema. 8 Thus, a better understanding of the cellular and biochemical events involved in alveolarization, as well as the factors that regulate the formation of new alveoli, could lead to improved treatments for several serious pulmonary conditions. During the past 10 years, a clearer understanding of many of the factors that regulate alveolarization has emerged, in large part the result of genetic studies performed in mice. Alveolarization is a relatively late event in lung development, occurring after branching morphogenesis has laid down the conducting airway system. However, some of the regulatory factors that control alveolarization are also involved in lung bud formation and branching morphogenesis. Therefore, it is useful to review briefly the regulation of the early events in the embryogenesis of the lung in order to gain greater insight into the processes involved in the formation of alveoli in the developing lung. Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

The Lung:: D~w:.~iopment, A.oin~, ,~nd the Environmer~t :

O V E R V I E W OF EARLY L U N G DEVELOPMENT

STAGES OF FETAL L U N G DEVELOPMENT

Lung bud formation

Pseudoglandular phase

The lung begins as a diverticulum from the embryonic foregut, at about 4 weeks post-fertilization in the human. 1 The foregut and its diverticula (such as the lung) are lined with epithelial cells that are derived from the endoderm germ layer. The lung diverticulum is covered with splanchnic mesoderm that gives rise to the connective tissue components of the lung (cartilage, smooth muscle, blood vessels, etc.). Factors that initiate the formation of the lung diverticulum remain obscure but signals between splanchnic mesoderm and foregut endoderm may initiate lung bud formation. 9 Studies in rodents suggest that transcription factors downstream from the Sonic hedgehog signaling cascade (Gli-2, -3 and Foxfl)regulate the formation of the lung diverticulum. 1~ There is also compelling evidence of a role for fibroblast growth factor 10 (FGF-10) and its receptor (FGFR2) in lung diverticulum formation. 12'13 The lung diverticulum displays aspects of the right/left symmetry, such that lung buds in the human initially form three diverticula in the right lung bud and two in the left that correspond to the lobes of the adult lungs. The control of right/left asymmetry has been studied by transgene studies in mice, which also have an asymmetric lung structure. 9 Several transcription factors, including lefty-1, lefty-2, nodal and Pitx-2, are probably involved in determining lung asymmetry. 9'14 In addition retinoic acid receptors (RARs), which are also nuclear transcription factors, may be involved, since mice in which the RAR~ and RAR[3 genes have both been deleted are missing the left lung. 15

This phase of lung development, characterized by repeated branching, occurs in the human from about 6-16 weeks of gestation. 1 It is thought that almost all the pulmonary conducting airways are created during this process, about 20-22 orders of airways in the human. 19 Differentiation of the epithelium of the conducting airways also commences during this early phase of lung development. 1 The most distal aspects of the branching duct system will eventually be remodeled to give rise to the alveolar region of the lung. Epithelial cells in the distal region remain undifferentiated as tall columnar cells with no specialized features other than large pools of intracellular glycogen (Fig. 4.1). 20

Branehing morphogenesis Early lung development is characterized by branching morphogenesis, in which the endoderm-lined ducts of the lung buds undergo dichotomous branching giving rise to the primary bronchi, then the secondary (lobular) bronchi, the tertiary (segmental) bronchi and so forth. 16 Branching morphogenesis involves the stabilization of the linear portion of the distal ducts, the creation of a cleft region at the rounded tips of the ducts and growth on either side of the cleft, a process that results in branching of the terminal portion of the duct. The regulation of branching morphogenesis involves epithelial-mesenchymal interactions via ECM components and growth factors. 16 In the developing lung, growth factors such as epidermal growth factor (EGF), platelet-derived growth factor (PDGF), insulin-like growth factors I and II (IGF-I and IGF-II), hepatic growth factor (HGF), vascular endothelial growth factor (VEGF) as well as retinoic acid have been shown to regulate branching morphogenesis. 16'17 Other mediators of branching morphogenesis in the developing lung include transcription factors such as HNF-3[3, HFH-4 and TTF-1.18

Canalicular phase Remodeling of the distal portions of the branching duct system to form gas-exchange units begins during this stage, which in the human lasts from 16 to 24 weeks of gestation. 1 This stage is characterized by increasing numbers of capillaries between the terminal ducts and the beginning of the differentiation of the presumptive alveolar epithelium. 2~ Some of the distal epithelial cells become more cuboidal and begin to synthesize and release surfactant; these type II cells are characterized by decreased intracellular glycogen pools, microvilli on their apical surface, and the appearance of lamellar bodies. 2~ At the same time, capillaries in the interstitium induce the overlying epithelial cells to flatten and differentiate into alveolar type I cells. Factors that regulate angiogenesis in the developing lung include VEGF and its receptors as well as FGFs and their receptors. 21 Type I and type II alveolar epithelial cells (AECs) differentiate in response to a number of regulatory factors 22 (see also Chapter 9). Thus, some aspects of alveolar development, in particular AEC differentiation, begin during the canalicular stage of lung development. However, in the human, true alveoli do not begin to form until much closer to birth. 1

Saccular phase During this phase, which occurs from about 24 weeks of gestation to term in the human, the terminal portions of the duct system (the terminal sacs) lengthen and may undergo further branching to give rise to alveolar ducts and alveolar sacs. 1 In the human, it is estimated that 15-18% of true alveoli form late in gestation, but the bulk are formed after birth. 1 Premature infants whose lungs are in the late canalicular/early saccular stage of lung development at the time of birth can survive without alveoli, in response to surfactant therapy and ventilation, although they are at risk of developing BPD. 7 The regulation of alveolarization is poorly understood, probably because there are no easily measured end points. Investigators who study alveolarization have relied primarily

Development

of Alveoli

on morphological techniques such as measurements of alveolar number and dimensions and the surface area available for gas exchange. 6 While there are well-defined markers for AEC differentiation, primarily those associated with type II cell differentiation and surfactant production, there are few biochemical markers for alveolarization per se, with the possible exception of elastin production. 23

C H R O N O L O G Y OF M O R P H O L O G I C CHANGES DURING ALVEOLARIZATION The process of alveolarization has been described in detail in rats and humans. 1 During the late canalicular and terminal saccular stages of prenatal lung development, when the conducting airways have stopped branching and are enlarging at their distal termini, there is a progressive loss of the mesenchymal cells that separate capillaries from the epithelium lining the future air spaces. This yields a rudimentary gas-exchange surface that can support respiration. Shortly after birth, in both rats and humans, the surface area of the air-blood interface begins to increase markedly as the terminal saccules become alveolar ducts and these in turn give rise to alveolar sacs and the alveoli.

Secondary septation Alveolar formation in the developing lung has been divided into several phases" the first is termed secondary septation

i!ii!iiiiiiiiiii!iiiiiiiii! i~!i~i~i~i!i!i!il!il!~i!i!i!~!

Fig. 4.1. Light micrographs of methylene blue-stained, epoxy sections of human fetal lung tissue at different stages of development (550x): (A) Lung tissue from a 14-week-gestational-age fetus. Lung tissue in the pseudoglandular stage of lung development is characterized by abundant connective tissue and a few branching ducts that are lined by a columnar epithelium. There are very few capillaries in the tissue and none are associated with the ductal epithelium. (B) Lung tissue from a 24-week-gestational-age fetus. At the canalicular stage of development, the ducts are more numerous and their epithelial cells are cuboidal. Capillaries are more abundant than in previous stages and are closely associated with the ductal epithelium. Differentiated type II cells are first observed at this stage of development in the human fetus. (C) Lung tissue from a 40-week-gestational-age fetus. In the terminal sac stage of lung development, the number of capillaries is greatly increased and the relative amount of connective tissue greatly decreased when compared to previous stages of lung development. (Reproduced with permission from Mallampalli RK, Acarregui MJ, Snyder JM. Differentiation of the alveolar epithelium in the fetal lung. In: McDonald JA (ed.), Lung Growth and Development. New York: Marcel Dekker, 1997, pp. 119-62.)

new alveolar septa by septation of the terminal sacs. Septation is initiated by the protrusion of secondary crests from primary septa. The primary septa, which are the walls of the terminal sacs, are comprised of a central core that contains fibroblasts and other connective tissue components, surrounded on each side by capillaries and the epithelium of the terminal s a c . 24 These same components are also found in the secondary crest. Capillaries are interconnected in the primary septa, but the two capillary layers formed in the secondary septa during septation initially have few interconnections. On postnatal days 2-4 in the rat, fibroblasts proliferate at the site of origin of the secondary crest, and this causes the secondary septum to lengthen and project perpendicularly into the alveolar sac. 25 The secondary crests are usually adjacent to elastic fibers and arise where the capillaries in the primary septum can be folded u p . 1'26 The tips of the secondary crests are occupied by myofibroblastic cells that are located next to elastic fibers. 27 After postnatal day 4 in the rat, fibroblast proliferation decreases in the proximal aspect of the secondary septum, but persists at a higher level in the distal septal t i p s y The increase in the height of the secondary septae is accompanied by an increase in the number of lamellar bodies in type II AECs and a corresponding increase in surfactant secretion. 1 Levels of surfactant produced in the lung apparently increase in parallel with increasing surface area in the distal portion of the developing lung. 28

::T:he Lung: Developmenti: Aging and: the

Growth Alveolarization continues through the third postnatal week in rats and until 2 years in humans. 1 The lung then enters the third phase of alveolar formation, a growth phase in which the gas-exchange surface area increases to the 0.7th power of lung volume, a factor consistent with an isotropic expansion of the alveoli. 24 This process is accompanied by an increase in the volume of air space in the lung, at the expense of alveolar septal tissue mass. During the growth phase of alveolarization, the volume density of alveolar type I and type II epithelial cells increases roughly in proportion to the increase in lung volume. 1 In mammals, this phase of lung development usually ends prior to the termination of the growth of long bones and increase in lean body mass. However, in mice and rats, remodeling in the sub-pleural regions of the lung may continue throughout life. 3~

that alveolar septation starts in utero. 35 Postnatal alveolar septation occurs at a more constant rate in humans than in rats; children do not undergo the rapid burst of septal outgrowth that occurs after birth in rats. In children, septation ends between 2 and 5 years of age. 1 In humans, peripheral lung desmosine can be detected during the final 10 weeks of gestation, but the major, ---&fold increase in its concentration occurs during the first 2 years of life, during alveolar secondary septal formation. 36 During this same period, lung hydroxyproline, a marker for collagen, increases ---3-fold when adjusted to dry lung weight. 33 Despite the differences in the kinetics and duration of the alveolar phases in the two species, the proportional enlargement of the alveolar surface area after birth is very similar in rats (21.4-fold) and humans (20.5-fold). 35 Some mammals, such as the rabbit and guinea pig, develop the majority of their alveoli prior to birth and the majority of the internal surface area of the lung is acquired prenatally. 37 Unlike humans, monkeys develop nearly all their alveoli prior to birth (26.2 x 106cm -2 in newborns compared to 26.6 x 106 cm -2 in the adult). 3s Similarly, in sheep, alveolar septation primarily occurs prenatally. 39'4~Secondary septa in the sheep fetus double in number during the final quarter of gestation and alveolarization is accompanied by both thinning of the alveolar septa and maturation of capillaries, events which occur primarily postnatally in humans and rats. Alveolarization in the sheep fetus is accompanied by a 20-30% increase in parenchymal elastin and an approximately 10% increase in parenchymal collagen over the same interval. 41'42

INTERSPECIES COMPARISON OF ALVEOLARIZATION

DEVELOPMENT OF THE ALVEOLAR EPITHELIUM

Rats and mice

The two types of AECs, the type I and type II cells, 2~ both arise from the endoderm-derived epithelial cells that line the distal portion of the branching duct system in the developing lung (Fig. 4.2). The differentiation of AECs commences in the human fetus prior to the formation of true alveoli, which is primarily a postnatal event.

Remodeling of the alveolar wall The second phase of alveolarization is marked by further lengthening and thinning of the secondary septae, primarily via the loss of interstitial mesenchymal cells and extensive capillary remodeling. 1 During this remodeling phase, the original dual capillary system becomes a single capillary system in which the capillaries are interconnected through a process termed intussusceptive microvascular growth. 29 This involves the fusion of adjacent capillaries by interconnecting endothelium-lined tissue pillars. During the period of exuberant new septal formation, the surface area of the lung increases to the 1.6th power of lung volume. 24

At birth, the gas-exchanging regions in rats, mice (and humans) consist primarily of immature terminal saccules with some secondary septa. In rats, alveolar septation occurs during the first 3 postnatal weeks, followed by a period in which the alveolar surface increases through the enlargement of pre-existing alveoli, without the formation of new alveoli. 1 Elastin is a critical structural protein in the primary and secondary septa. Tropoelastin (TE, the soluble precursor of elastin) mRNA is present in the lung during the pseudoglandular stage, but only in the walls of airways and blood vessels. 23 Elastin synthesis in the primary alveolar septa begins during the canalicular stage of lung development. 31 Alveolar septation is accompanied by a ---4-6-fold increase in parenchymal desmosine (a molecule that is unique to elastin and is a marker for elastin fiber deposition) and a --~6-fold increase in hydroxyproline residues, which are primarily found in collagen. 32'33 Postnatal elastin accumulation has also been studied in mice, in which the most abundant deposition of cross-linked elastin occurs between postnatal days 9 and 20. 34 In humans, the lungs at birth contain ---50x 106 alveoli (---18% of the alveolar number found in adults), suggesting

Alveolar epithelium stem cells Alveolar type II cells can divide and give rise to new type II cells or differentiate into type I cells. 43-45 There is little evidence that type I cells can divide; therefore, most investigators consider the type I cell to be a terminally differentiated cell type. When the lung epithelium is injured, alveolar type II cells divide and cover the injured area and this is followed by the gradual reappearance of type I cells. 4'44'45 Some investigators have hypothesized that during fetal lung development, undifferentiated epithelial cells first differentiate into alveolar type II cells, followed by the differentiation of some of these cells into alveolar type I cells.46'47 When isolated type II cells are maintained in vitro on tissue culture plastic surfaces under particular physicochemical conditions, they rapidly take on the phenotypic characteristics of alveolar type I cells. 48'49

Development

Fig. 4.2. Electron micrograph of lung tissue from a human newborn. The alveolar type II cell is filled with lamellar bodies. The thin cytoplasm of an alveolar type I cell covers the basement membrane that is shared with a capillary endothelial cell. R, red blood cell; A, alveolar lumen; C, capillary lumen; Ib, lamellar body; mv, microvilli; thick arrow, type I cell cytoplasm; arrowhead, endothelial cell cytoplasm. Bar equals 1 l~m. (Reproduced with permission from Mallampalli RK, Acarregui MJ, Snyder JM. Differentiation of the alveolar epithelium in the fetal lung. In: McDonald JA (ed.), Lung Growth and Development. New York: Marcel Dekker, 1997, pp. 119-62.)

Regulation of alveolar epithelial cell differentiation Numerous factors have been shown to regulate the differentiation of the alveolar epithelium in the fetal lung. As reported in several reviews 9'18'22'5~ these factors include transcription factors, hormones, growth factors, regulatory agents such as cyclic AMP, neurotransmitters and physical factors such as stretch. Many AEC differentiation factors also regulate the process of alveolarization itself and the role of some of these factors in the formation of alveoli will be discussed at the end of this chapter. Epithelial-mesenchymal interactions are well-established mediators of lung development and are also likely to be specifically involved in alveolarization. 16'53 The composition of the ECM beneath the alveolar epithelium has dramatic effects on the differentiated state of AECs) 4 In particular, laminin substrata seem to promote the retention of alveolar type II cell differentiation in vitro. 55 In addition, the growth of the pulmonary vasculature, in particular the capillary network of the septa, influences the differentiation state of the overlying epithelium, probably promoting an alveolar type I cell phenotype in the cells that directly overlie capillaries. TM Finally, it has been reported that alveolar type II cells are localized preferentially over cables of elastin in the alveolar wall. 57

Differentiation of the alveolar type I cell The alveolar type I cell is an important component of the air-blood barrier, as it overlies capillaries in the alveolar wall and comprises most of its surface area (-90%). In the past, studies of the structure/function of the type I cells

of Alveoli

were difficult because very few biochemical markers characteristic of this cell type were known. A few markers for the type I cell had been proposed, for example, binding to certain lectins. 5s'59 However, several investigators have recently described proteins that are predominantly expressed in the lung in type I AECs. 49'5s Tl-t~ and HT1 are proteins that were originally identified based on monoclonal antibodies that recognized type I cell-specific proteins. 49'6~ Tl-o~ is the best-characterized marker and is a plasma membrane protein. 6~ It is expressed only in alveolar type I cells in the lung, but is also present in epithelial cells of the ciliary body in the eye and of the choroid plexus of the brain, sites of expression suggestive that Tl-o~ is involved in fluid transport. 6~ Tl-ct is induced in type II cells undergoing a phenotypic transition to a type I-like appearance during culture in vitro. 49 Another important class of proteins expressed in type I AECs are the water channels, aquaporins. 62 Aquaporins are a family of water channels that are expressed ubiquitously; however, in the distal portion of the human lung, aquaporins 4 and 5 seem to be relatively restricted to the alveolar type I cell while aquaporin 3 is expressed in the alveolar type II cell. 63 Other antigens that are either relatively specific to the type I cell or else are highly enriched in this cell type include an intercellular adhesion molecule, ICAM-I, which may be involved in establishing and maintaining the flattened phenotype of the alveolar type I cell. 64 Gap junctions are thought to exist between adjacent alveolar type I and II cells in the alveolar wall. 65 The expression of relatively unique complexes of connexins have recently been described in type I and type II AECs. 66 Finally, several recent reports have documented that type I cells are characterized by abundant caveoli and the expression of caveolin 1 protein while alveolar type II cells possess few caveoli and express little caveolin 1 protein. 67 Caveoli are plasma membrane structures that may be important in mediating the transport of materials across a cell and also may concentrate signaling mediators. 68

Differentiation of alveolar type H cells Although type II AECs, which produce pulmonary surfactant, 5 occupy only a small portion of the alveolar surface area (---7%),3 there are about twice as many of them as type I cells. 69 Type II cell differentiation begins as early as 24 weeks gestation in the human (and about day E18 in the mouse fetus). 1'18 Surfactantphospholipids Surfactant is comprised of about 80% phospholipids, 10% cholesterol and 10% protein. 7~ The most abundant class of phospholipids in pulmonary surfactant is phosphatidylcholine, in particular dipalmitoylphosphatidylcholine (DPPC), which is the primary surface tension lowering component in surfactant. 7~ The next most abundant phospholipid is phosphatidylglycerol (PG), an anionic phospholipid. Interestingly, the surfactant initially produced by the fetal type II cell contains another anionic phospholipid, phosphatidylinositol (PI), rather

T:he::Lung:: Development, Aging and the Environment .....

than PG. 7~ As gestation proceeds, the relative amount of PI decreases while that of PG increases. 7~ Various other phospholipids are also present in characteristic amounts in pulmonary surfactant. 7~ Cholesterol is also present, but its functional significance is not clear. 72 The synthesis of surfactant phospholipids in differentiating alveolar type II cells is accompanied by several morphologic changes in the type II cell (Fig. 4.3). 20 First, lamellar bodies, which are the intracellular storage form of pulmonary surfactant, appear in the cytoplasm. 2~ Secreted lamellar bodies undergo a structural transformation to form tubular myelin, a surfactant intermediate which is thought to give rise to the monolayer of surfactant that lines the alveolar aqueous lining layer and reduces its surface tension. 2~ Surfactant phospholipids are recycled by the type II cell. 5 Another characteristic change in differentiating type II AECs is the disappearance of the large glycogen pools that are characteristic of the tall columnar undifferentiated epithelial cells in the developing lung (Fig. 4.4). 2~ It is thought that the intracellular glycogen in presumptive type II cells is broken down to provide the metabolic substrates necessary for the initial burst of surfactant phospholipid synthesis. 7~ Fatty acids used for surfactant phospholipid synthesis are probably synthesized in the alveolar type II cell. 73 Cholesterol is predominantly taken up by the type II

:

::

cell via the binding and internalization of VLDL particles from the serum. 72'74 VLDL could potentially also supply some of the fatty acids needed for phospholipid synthesis and, in fact, have been shown to increase phospholipid synthesis. 75 Premature human newborns frequently do not have adequate numbers of differentiated type II AECs and thus do not produce sufficient amounts of pulmonary surfactant, resulting in the respiratory distress syndrome ( R D S ) . 76'77 During the 1980s, supplementation with synthetic surfactants was introduced as a treatment for RDS 78 and this was followed by new generations of surfactants, many derived from natural surfactants. 79 Surfactant therapy has dramatically decreased the incidence of neonatal RDS, although some premature infants do not respond well to this therapy. 77'78

Surfactantproteins

The surfactant-associated proteins are an important component of pulmonary surfactant8~ four have been identified to date, i.e. SP-A, SP-B, SP-C and SP-D. 8~ All are expressed in type II AECs and all are developmentally regulated. 8~ SPoA: The most abundant surfactant protein is SP-A, a -35-kDa glycoprotein, sl Several reviews concerning the structure and function of this interesting protein have been published in recent years.82--86SP-A is a member of the collectin family of proteins, calcium-dependent C-type lectins that are involved in innate host defense against pathogens. 83'84 SP-A also blocks the inhibitory effects of serum proteins on surfactant surface tension lowering properties, s7'88 SP-A genedeleted mice have normal lung structure and function but are more susceptible to infection with several pulmonary pathogens than intact, wild-type mice. 89 In the human, SP-A is also expressed in submucosal glands of the conducting airways and, in smaller amounts, in the gastrointestinal tract. 90-92

.... ~~

Fig. 4.3. Electron micrographs of lamellar bodies and tubular myelin in rabbit fetal lung explants. (A) Secreted lamellar bodies and tubular myelin (arrows) observed within a lumen in a cultured explant. (B) A lamellar body observed within a differentiated type II cell. The lamellae are surrounded by a perilamellar membrane. Bars equal 0.1 l.tm. (Reproduced with permission from Snyder JM. The biology of the surfactant proteins. In: Bourbon J (ed.), Pulmonary Surfactant: Biochemical, Functional, Regulatory and Clinical Concepts, Boca Raton, FL: CRC Press, 1991, pp. 105-21.)

SP-B: SP-B is a small (-6.5 kDa), extremely hydrophobic surfactant protein that facilitates the spreading of surfactant proteins on the alveolar surface. 93 Recent studies have also implicated SP-B as being involved in lung antioxidant function. 94 SP-B is synthesized as a high molecular weight precursor that is cleaved to the active molecule within the lamellar body or its precursor. 95 SP-B gene-deleted mice die immediately after birth due to respiratory distress. 96 In the human, genetic mutations in the SP-B gene cause congenital alveolar proteinosis, a lethal condition in human newborns that can only be treated by lung transplantation. 97 The deficiency in SP-B production in these patients is also associated with a defect in the intracellular processing of the SP-C precursor protein. 97 In the human, SP-B is expressed in alveolar type II cells and in Clara cells of the conducting airways. 93 SP-C: SP-C is another low molecular weight ( - 5 kDa), extremely hydrophobic surfactant protein. 93 SP-C genedeleted mice have almost no phenotypic differences from wild-type mice other than a tendency of their surfactant to have atypical biophysical properties. 98 Interestingly, individuals with mutations in their SP-C genes and with

Development

of A l v e o l i

Fig. 4.4. Electron micrographs of rabbit fetal lung tissue. (A) Electron micrograph of a fetal lung epithelial cell at day 19 of gestation. The undifferentiated tall columnar cells contain well-defined glycogen pools. (B) Electron micrograph of a fetal lung epithelial cell at day 28 of gestation. The cells contain lamellar bodies, multivesicular bodies and no well-defined glycogen pools. Bars equal 1 l~m. (Reproduced with permission from Snyder JM, Magliato SA. An ultrastructural morphometric analysis of rabbit fetal lung type II cell differentiation in vivo. Anat. Rec. 1991; 229:73-85.)

reduced levels of SP-C protein were found to have interstitial lung fibrosis. 99 In the human, SP-C is expressed only in alveolar type II cells of the lung. 93 SP-D: SP-D is the most recently characterized surfactantassociated protein. 1~176 It is an ---43-kDa glycoprotein that is a member of the collectin protein family and, as such, is involved in host defense. 83-s5 SP-D gene-deleted mice have abnormalities in surfactant metabolism and accumulate large amounts of surfactant in their alveoli 1~ but the surface tension lowering properties of their surfactant appear to be normal. 1~ These mice go on to develop emphysema by about 1 year of age by an unknown mechanism. 1~ No human mutations in SP-D have yet been described. 97 SP-D is expressed in alveolar type II cells but is also widely distributed in mucosal surfaces throughout the body in humans. 85

DEVELOPMENT

OF THE A L V E O L A R

INTERSTITIUM

During fetal lung development, the interstitial mesenchyme situated between the branching ducts plays an important inductive role in airway and terminal sac formation. During late gestation, the mesenchyme gradually becomes more attenuated as there is a progressively closer apposition of

the epithelium of the terminal sacs and the vasculature. During alveolarization, the interstitium must assume the function of providing structural support for the gas-exchange unit, which postnatally is under the phasic mechanical stress of respiratory movements. Interstitial fibroblasts The interstitial fibroblast (IF) is the major synthetic cell in the interstitium. It is thought to produce much of the ECM in the alveolar interstitium and to provide metabolic substrates to the epithelium. During alveolarization, four functions characterize IF development: proliferation, migration, synthesis of ECM components and apoptosis. The IF is not synchronized in these functions, and since the IF population is heterogenous, some IF may be proliferating while others are migrating or involved in synthesis of ECM. Two populations of IF have been described based on the presence or absence of lipid droplets within their cytoplasm. 1~ These have been termed lipid interstitial cells (LICs) and non-lipid interstitial cells (NLICs; see below).

Fibroblast proliferation Autoradiography with 3H-thymidine has been used to characterize the proliferation of IF in postnatal rat lungs. 25'1~ IF proliferation peaks at postnatal day 4, then progressively

Th e Lungi:: bevel opm:e nti Ag::i nga nd: :the: Envir on m:e:nt::: ::: :: :::: : :::: :: ...... :: ....... ..... : : declines until day 13 with very little proliferation occurring after this time. 25 LIC proliferation declines earlier than that of NLIC so that by postnatal day 11, LIC proliferation has ceased while the labeling index of NLIC is still 2%. 104 Since LICs are frequently located at the base of the elongating secondary septa, IF proliferation is sustained for a longer period in the more distal portions of the developing lung. Under normal circumstances, IF proliferation ceases after alveolarization has been completed, but is more sustained if the developing postnatal lungs have been exposed to glucocorticoids. 1~ IF proliferation can be induced by lung injury and is an important component of the fibrotic response of the lung to injurious agents such as bleomycin and during inflammatory interstitial lung diseases.106,1~

Fibroblast migration Little is known about the migration of IF in the alveolar septa during its elongation. It is clear that cellular movement must occur as the secondary septa elongate as cellular proliferation slows before elongation is complete. Therefore, the distal septal cells must migrate and do not arise solely from cellular division at their ultimate destination. At the free (and most distal) margin of the alveolar wall, in the interstitium, is a ring of myofibroblasts, cells that contain contractile actin-myosin filament arrays arranged in parallel with the long axis of the cell. These cells differentiate late in the alveolarization process and are thought to be able to vary the dimensions of the alveolus via their contraction. 26 The septal myofibroblasts that form the alveolar septal ring are present in the attenuated secondary septa at the beginning of IF migration and maintain their position at the septal tip throughout the elongation process. 27 Cellular migration is difficult to study in tissues in vivo and most of the available information about this process has been derived from studies conducted in vitro. Embryonic and neonatal lung fibroblasts migrate towards a gradient of chemotactic molecules such as fibronectin, fibrin and elastin proteolytic fragments. 1~176 IFs that have assumed a myofibroblastic phenotype migrate in larger numbers and further in response to fibronectin than nonmyofibroblastic lung fibroblasts. 1~ Components of the elastic fiber have been shown to stimulate fibroblast migration and may also contribute to IF migration during septal elongation. 1~ Elastin degradation products may be important in the response of the lung alveolus to proteolytic injury or in the alveolar remodeling that occurs after the neonatal lung is exposed to hyperoxic conditions. 111 PDGFs and their receptors are critical for myofibroblast migration and the expression of PDGF receptors tx and are increased in myofibroblasts. 112'113 PDGF receptor ~containing myofibroblasts fail to populate the tips of the secondary septa in PDGF-~ null mice; this may result from the failure of these cells to migrate in the septa. 113 The phenotype of these mice includes a failure of alveolarization resulting in an emphysematous lung.

::: : : :: :...................

::: ::

Structural proteins in the alveolar wall Pulmonary IFs are the major producers of the structural proteins present in the alveolar wall interstitium. One of these, elastin, provides resilience during phasic respiration and contributes to lung recoil. It is a major component of elastic fibers, structures that also contain fibrillins, microfibrilassociated glycoproteins, emelin, lysyl oxidase and fibulins. Elastin is produced by both the LICs and the NLICs (see below), and the developmental regulation of the elastin gene has been studied in detail. 23 The structural collagens, types I, III and IX are also produced by IF and provide tensile strength to the alveolar wall. IFs produce proteoglycans which help provide a hydrated environment in which the elastic fiber retains its elastomeric nature. 114 The glycosaminoglycans hyaluronic acid, chondroitin sulfate and heparin sulfate contribute to the stiffness of the postnatal alveolar walls and changes in their contents explain agerelated differences in the mechanical properties of the distal lung. 115 The LIC accumulates lipids during late prenatal and early postnatal life that is largely dispersed by the third postnatal week. 116

Elastin The developmental regulation of elastin synthesis in the lung has been most extensively studied in mice and rats and, in these species, occurs primarily postnatally. Early morphologic studies established that extensive elastin accumulation in the alveoli occurs between postnatal days 4 and 20. 34'117'118 Amorphous elastic fbers are deposited adjacent to both LIC and NLIC in the interstitium and are crosslinked extracellularly, enzymatically by lysyl oxidase and non-enzymatically through aldol condensation. 1~ There is a close temporal correlation between the increase in elastic fiber length and the volume density of the interstitium of alveoli during postnatal days 4-20 in the rat. These morphologic findings have been corroborated by biochemical studies that demonstrate that TE production is maximal during postnatal days 7-12, and that the peak in TE production precedes the maximal desmosine and cross-linked elastin accumulation during postnatal days 10-20.120'121 Inhibition of elastin cross-linking by inhibiting lysyl oxidase (the ratelimiting enzyme) during alveolarization markedly alters alveolar structure and decreases the gas-exchange surface area of the lung. 122 Together these observations suggest that critical regulatory events occur after birth to initiate and terminate elastin synthesis. TE is synthesized during the saccular phase and surrounds the terminal sacs. 123 In mice bearing a null deletion of the elastin gene, the absence of elastin influences airway branching during the final day in utero. 124'125 The elastinnull mouse also demonstrates a decrease in the initiation of secondary crests during gestational day 18.5 to postnatal day 0.5 and a decrease in terminal sac air space units at birth. 125 The initiation of elastin synthesis in the terminal respiratory units normally results from an abrupt increase in TE transcription between days El8 and E21 in the rat, when it reaches its peak. After declining at postnatal day 2,

Development of Alveoli

TE transcription increases again at postnatal day 9.120'121 The postnatal increase in TE gene transcription is coordinated with a decrease in IF proliferation and may result from a reduction in the effects of suppressive growth factors such as FGF-2.126'127 Alveolar levels of TE mRNA reach a maximum around postnatal days 9-12 in the rat and then decline by postnatal day 15.120'128This postnatal increase in TE expression is regulated both transcriptionally and posttranscriptionally. 129'13~An increase in TE mRNA stability contributes to the postnatal increase in TE mRNA levels; the decline in TE mRNA after postnatal day 12 is solely accounted for by a decline in TE mRNA stability. 129'131The factors that are responsible for the increase in TE mRNA stability in the lung have not been identified, but studies using cultured lung fibroblasts indicate that TGF-13s can enhance TE mRNA stability by inhibiting the binding of a destabilizing protein to an element in the exon 30 region of TE mRNA. TM The destabilizing protein is more abundant in adult rat IF and an increase in the levels of this protein likely contributes to the reduction in TE mRNA stability that occurs after postnatal day 12.

The elastic fiber TE is rapidly exported from IF and associates with microfibril elements. This interaction is critically dependent on the association of basic amino acid residues in the carboxy terminus of the protein with microfibril-associated glycoproteins and with the amino terminal region of fibrillins-1 and -2.132 TE monomers also associate with one another prior to cross-linking by a process termed coacervation. Coacervation is driven by hydrophobic interactions involving TE exon 26 and is promoted by sulfated glycosaminoglycans which interact with the positively charged carboxy-terminal lysines o n T E . 133'134 These protein-protein and protein-glycosaminoglycan interactions are accompanied by an enhancement in lysyl oxidase-mediated crosslinking. 135 From postnatal days 14 to 21, alveolar septal elastin undergoes cross-linking which confers increased chemical and proteolytic stability and contributes to the extraordinary longevity of elastin in the elastic fiber. TM The pulmonary IF also contributes to the production of the microfibril, which is an essential component of the elastic fiber. Both fibrillins-1 and -2 are expressed in the adult lung with fibrillin-2 primarily present prenatally. 137'138 Postnatally, fibrillin-1 predominates and makes a significant contribution to the maintenance of the elastic fiber network. A duplication in the coding region of fibrillin-1 occurs in the tight skin mouse and this destabilizes microfibrils and leads to a reduction of elastic fibers and a decrease in the formation of alveolar septa starting 4 weeks after birth. 139

THE L I P I D I N T E R S T I T I A L CELL Based on their lipid contents, pulmonary interstitial cells (interstitial fibroblasts) are divided into two populations,

the lipid-droplet laden LICs and the NLICs which lack lipid droplets and are located more peripherally in the alveolar septum. The initial morphologic description of these cells emphasized the abundant lipid droplets, high glycogen content and localization of LIC to the central region of the alveolar septum (Fig. 4.5). 1~ Some of the lipid droplets in LIC are surrounded by glycogen deposits. 14~ Subsequent studies established that LIC contain contractile filaments similar to those observed in myofibroblasts. 119'14~ These contractile filaments are generally more dense than those present in the NLIC and are oriented with their long axis either perpendicular or oblique to the plasma membrane. 1~ Elastic fibers are found adjacent to both NLIC and LIC, and the density of the intracellular contractile filaments is greatest where cells contact elastic and collagen fibers. 26 LIC and NLIC independently undergo cell division during alveolarization, and neither cell type appears to be a precursor for the other. TM LICs are first evident in rat lungs at day El6. The triglyceride content of whole rat lung tissue increases 3-fold between days El7 and El9.141 Lung triglyceride content increases another 2.5-fold between day E21 and postnatal day 1, and then peaks during the second postnatal week. 141 The abundance of LIC in the lung follows the same timecourse. 116 The lipid droplets of LIC contain primarily neutral lipids (---86%) with phospholipids comprising the remaining 14%.116 Like adipocytes, LICs express lipoprotein lipase, fatty acid transporter and intracellular lipid binding proteins and are able to accumulate neutral lipids when purified triglycerides are added to the culture medium. 142'143 They also contain LDL and VLDL receptors whose expression increases during the postnatal period. TM LICs contain vimentin and desmin, both intermediate filaments, and m-smooth muscle actin (~SMA). The number of LIC decreases prior to weaning, a result, in part, of a decrease in cell proliferation and a decrease in plasma lipids. 1~ LICs are present at the base of the alveolar septae in adult rats and their number can be increased by the administration of retinyl palmitate. 14~ Similarities exist between the contractile filaments in neonatal LIC and those found in the contractile interstitial cell (CIC) present in the adult lung. 119 Based on the ultrastructure of the contractile filaments, several investigators have speculated that the LIC may be the neonatal equivalent of the CIC. 26 LIC lose their lipid droplets when cultured in the presence of fetal bovine serum. 143 They also assume more myofibroblastic characteristics and abundantly express ~-SMA, desmin, as well as vimentin. 146 However, they maintain their neutral lipid content if they are cultured in triglyceride-rich, neonatal rat serum or are exposed to activators of the peroxisome proliferator-activated receptors, which are involved in the induction of lipid storage by adipocytes. 146 During lung development, mesenchymal cells lie in close apposition to epithelium and play a central role in the growth and differentiation of epithelial cells into alveolar type II cells, the site of pulmonary surfactant synthesis. 16'147

The Lung: Development,

Aging and:the Environment:

:

: :

:

:

:

:

:

:

:

:

:

:

:

: i:

:

:::::::

: ::

:

:

::

::

:

:

ALV

NLIC

Typ e

II

LIC

ALV

Fig. 4.5. Lipid-filled interstitial fibroblasts (LICs) as seen in micrographs of rat lung from late fetal developmental stage. L = lipid, Cap =capillary, ALV = alveolar lumen, NLIC = non-lipid-filled interstitial fibroblast, type II =alveolar type II cell. (Reproduced with permission from Vaccaro C, Brody JS. Ultrastructure of developing alveoli. I. The role of the interstitial fibroblast. Anat. Rec. 1978; 192:467-79.)

In the rat, LICs are first evident during the canalicular phase, with triglyceride content maximal just prior to the appearance of lamellar bodies in neighboring type II cells. 148 Triglycerides of fibroblast origin are used for surfactant phospholipid synthesis by type II cells in culture. 149 The mechanism of transfer of neutral lipids from the LIC to type II cells has not been fully characterized, but like adipocytes, the LICs contain lipase, which may de-esterify the lipids prior to their export as fatty acids. 14z It is also possible that lipid droplets are transferred directly from one cell to the other, since intercellular communications have been observed between type II cells and LIC cells in neonatal rat lung. 15~ In addition to triglycerides, the LICs accumulate retinyl esters. 151'152The retinoid content (primarily retinyl palmitate) of the fetal rat lung increases markedly after day El5, peaks at day El8 and then decreases 4-fold by day E21.153 There is a decline in retinyl ester content of LIC from day El9 to postnatal day 2 in the rat. 151 Retinol and retinoic acid (RA) in the lung increase concurrently by postnatal day 2. Retinol remains elevated until postnatal day 8 and then declines, while RA falls abruptly after day 2.151 The timing of these events has led to the hypothesis that retinoids (and in particular RA) may be important in alveolarization. 151'152'154 In whole lung tissue and in isolated LIC, the steady-state levels of RAR-mRNA appear to be the highest in the early neonatal period. 151'155 In lung tissue, the levels

of the mRNAs for RARo~, RARI3 and RARy increase significantly at birth. 155'156 In isolated LIC, RARI3 increases -4-fold while RARy increases --8-fold between day El8 and postnatal day 2.151 Endogenous retinoids increase TE expression in explant cultures of lung obtained from rats on gestational day 19 while exogenous RA increases TE expression in cultured LIC. 157 Mice that have a null deletion of RARy, and are lacking one allele of RXRt~, have diminished levels of TE mRNA in their LIC at postnatal day 10 and contain less elastin in their lungs at postnatal day 28.158 This is accompanied by a decrease in alveolar number and the gas-exchange surface area.

ALVEOLAR MYOFIBROBLASTS The term alveolar myofibroblast (AMF) has generally been applied to pulmonary IFs that express uSMA, an actin isoform that is most commonly associated with smooth muscle cells (Fig. 4.6). The AMFs have been further sub-classified based on their intermediate filament profile (vimentin and/or desmin) and the presence or absence of myosin heavy chains. 159 The AMF is of functional importance during lung development and disease. During lung development, it is present in the primary septa where secondary septal buds form. 118'125 It is responsible for producing the elastin that localizes at the tip of the elongating septum and

Development of Alveoli

!

9

A

i:

9

~ A :~

Fig. 4.6. Alveolar myofibroblasts in rat lung. (a) An alveolar myofibroblast (MF) is located at the junction of three alveolar septa. The microfilaments end in a dense body situated against the alveolar basement membrane (arrows). A, alveoli; C, capillaries. (b) Note the abundance of actin filaments parallel to the capillary basement membrane (arrows). Cytoplasmic processes of the alveolar myofibroblast extend into the thick portion of the air-blood barrier (arrowheads). cf=collagen fibers. On panel A, bar=5 ~m and on panel B, bar= 1 lLtm. (Reproduced with permission from Kapanci Y, Gabbiani G. Contractile cells in pulmonary alveolar tissue. In: Crystal R, West JB, Weibel ER etal. (eds), The Lung: Scientific Foundations. Philadelphia, PA: Lippincott-Raven, 1997, pp. 697-708.)

forms the alveolar contractile ring. In rats, the AMFs are most abundant during secondary septal formation when TE expression is at its highest; it is likely to be the major source of elastin synthesis in the pulmonary interstitium. During septal elongation, the AMF proliferates in a PDGFA-dependent process that is a requirement for secondary septal formation. 16~Mice bearing a deletion of the PDGF-A gene fail to undergo secondary septation, their primitive alveolar walls lack elastic fibers and the mice die during the first 2 weeks of life. 160 Other defects, such as a failure in AMF migration, may also contribute to failed septation in PDGF-A null mice. 113 It has also been shown that PDGFR-dependent lung myofibroblast proliferation involves signaling through the p38 MAP kinase pathway. 1~ The cellular precursor of the AMF has not been identified. In other organs, myofibroblasts arise from either primitive stem cells, pleuripotent cells such as neural crest cells or fibroblasts. 161 Based on the morphology of the subcortical cytoskeleton, some investigators have proposed that the LIC is the precursor of the AMF. 119When LICs are cultured in vitro to confluence, they express more ~SMA and also express myosin heavy chains. 146 Autocrine and paracrine factors stimulate tzSMA production by cultured lung fibroblasts. These include TGF-[3s, PDGFs, IGFs, interleukin-4, and, in inflammatory states, interleukin-1.161 The effects of TGF-[3s, secreted by myofibroblasts themselves, have been most extensively studied. 161 TGF-[31 stimulates t~SMA gene transcription through jun-kinase in lung myofibroblasts, with the effect mediated through a TGF-[3-responsive element in the proximal portion of the ~zSMA promoter. 162 The TGF-[3-responsive elements in the

~SMA promoter differ in myofibroblasts compared to smooth muscle and endothelial cells. 162 IL-I[3 stimulates inducible NO synthase in cultured lung fibroblasts, resulting in increased NO production, which reduces t~SMA mRNA and protein. 163 IL-I[3 and other inflammatory mediators may be involved in interstitial pulmonary fibrosis since in response to these mediators interstitial myofibroblasts (also termed contractile interstitial cells) proliferate and produce more ECM proteins, most notably the interstitial collagens. TM

M E C H A N I S M S OF T H I N N I N G OF THE ALVEOLAR SEPTUM Following expansion of the IF population, movement of IF into the secondary septa, and the greatest period of TE synthesis, which in the rat ends around postnatal day 13, there is a 20% loss in the number of IF; this loss results from IF apoptosis. 165 At postnatal day 16, there is an abrupt increase in IF apoptosis that is characterized by an increase in chromatin condensation and DNA strand breaks, as well as an increase in the levels of Bax, a pro-apoptotic protein, and a decrease in the expression of the anti-apoptotic protein Bcl-2.165 This apoptotic phase is transient and ends after postnatal day 19. While the number of IF decreases, the mass of the ECM does not increase and assumes a larger proportion of the volume of the interstitium. The reduction in IFs is accompanied by an increase in capillary surface area, which improves the efficiency of gas exchange. 1 During septal thinning, there is a fusion of the paired

The: Lung:Developmenti:Aging and the Environment :::

capillary loops in each septum by a process which most likely involves the formation of connecting pillars. 1 After breaching the interstitium between the two loops, these pillars then increase in diameter and decrease in height. Thus the pillars assume the shape of disks which increase capillary surface area, while reducing the need for extensive new endothelial cell proliferation. As the alveolar walls thin, the pores of Kohn form and facilitate the merging of air sacs. 166 While the lung volume of humans and rats increases 23-fold between birth and adulthood, capillary volume increases about 35-fold, a finding which is consistent with an increase in the interconnecting capillary meshwork.

DEVELOPMENTAL DEFECTS IN A L V E O L A R I Z A T I O N Studies of developmental defects in alveolar formation have taken two general approaches. The first has been to study BPD in human infants or spontaneous mutations in mice that result in defects in alveolar formation. The second approach has been to target specific genes in mice and to study the effects of their deletion or over-expression on alveolar formation. The second approach has yielded considerable information about the roles of specific regulatory factors such as PDGF-A and PDGF receptor o~, the RARs, VEGF and glucocorticoids. The phenotypes of mice with gene deletions that affect these agents will be discussed in more detail below.

~:: ::::

:::: :: ::

:

:: :

This developmental defect results from a spontaneous mutation in which a portion of the coding region of the fibrillin-1 gene is duplicated. 139 When both alleles contain this duplication, the affected mice die in utero. The heterozygotes survive, but have morphologic abnormalities: in their skin and lungs. The fibrillin-1 gene duplication results in the synthesis of a longer fibrillin protein that does not polymerize normally and results in the formation of microfibrils which are more sensitive to proteolytic degradation. 17~The abnormal microfibrils also fail to support normal elastic fiber formation and thus elastin levels are reduced in the lung parenchyma of tight-skin mice at 4 weeks. 171 In contrast, collagen deposition is increased in the lung and skin of the affected mice. The pallid mouse also spontaneously develops enlarged and fewer lung alveoli, but at 8 months of age, much later than in the tight-skin mouse. The pallid mutation maps to the same chromosome as the Tsk mutation but appears to result from a different genetic abnormality. 172 The lungs of pallid mice contain less elastin at 12 months of age and this is correlated with an increase in extracellular elastase in the pulmonary interstitium. 172 These mice may also have lower levels of circulating o~l-antiprotease, which along with the increase in elastase lead to an increase in elastin degradation. 169 Therefore, the alveolar enlargement in the pallid mouse most likely results from alveolar wall destruction rather than a defect in alveolar wall formation.

REGULATION

OF A L V E O L A R I Z A T I O N

Bronchopulmonary dysplasia BPD occurs in infants who are born prematurely and who require mechanical ventilation to treat RDS; the accompanying oxygen toxicity and pressure/volume-related lung injury was previously thought to play a major role in the development of BPD. However, following the use of surfactant replacement therapy, reduced ventilation trauma and lower O 2 levels, it has become clear that BPD is characterized by abnormal alveolarization. Infants dying of BPD have fewer and larger alveoli, with more attenuated alveolar walls and a paucity of alveolar capillaries. 7 It is hypothesized that this results from decreased alveolarization as well as injury to the existing primary septa. Multiple factors are thought to contribute to the defective alveolar formation that occurs in BPD, including prenatal or antenatal administration of glucocorticoids, nutritional deficiencies, inflammation, as well as the effects of hyperoxia and mechanical ventilation. 77 Studies in prematurely delivered baboons and lambs that received exogenous surfactant and were ventilated using conditions that minimized mechanical trauma and hyperoxia revealed that they had fewer and enlarged alveoli in their lungs, with a reduction in alveolar capillary surface area. 167'168The abnormalities in BPD have recently been reviewed in detail. 77

Tight-skin and pallid mice The tight-skin mouse is characterized by fewer and enlarged lung alveoli, apparent by 4 weeks after birth. 169

Glucocorticoids Glucocorticoids are thought to play an important role in normal lung development. 173 In many species, including the human, there is an increase in fetal serum glucocorticoid levels towards the end of gestation that correlates well with the structural and functional maturation of the fetal l u n g . 174 Liggins and H o w i e 175'176 w e r e the first to report, in the sheep and then in humans, that exogenous glucocorticoids, administered to the mother, could accelerate fetal lung development and reduce the incidence of RDS in premature infants. Transgenic mice in which the corticotrophin-releasing hormone (CRH) gene was deleted have low levels of glucocorticoids and delayed lung development. 177 Likewise, mice in which the glucocorticoid receptor gene was deleted also exhibit delayed lung development and usually die shortly after birth. 178 Since neonatal RDS is principally caused by surfactant deficiency, much attention has focused on glucocorticoid regulation of type II AEC differentiation and surfactant production. In general, glucocorticoids increase both surfactant phospholipid and surfactant protein levels. 173 The lungs of glucocorticoid-treated fetuses have thinner septal walls due to decreased connective tissue. The levels of elastin in alveolar walls are increased in glucocorticoidtreated lungs. 41'123 These structural changes may facilitate gas-exchange in the lung. However, another, well-documented

Development of Alveoli

effect of glucocorticoids on lung development is an inhibition of alveolarization. 179'18~ Septation of the terminal sac into alveoli may be impaired in glucocorticoid-treated fetuses and newborns. 179'18~Glucocorticoids may act indirectly via promoting the production of certain growth factors and their receptors which in turn stimulate premature lung differentiation at the cost of further lung growth and development. 173 The negative effect of glucocorticoids on alveolarization apparently persists into adulthood. 179 Since glucocorticoids are used prenatally to prevent RDS and are also often used postnatally to avoid the development of BPD, there is concern about the appropriateness, doses used, and duration of glucocorticoid treatment for neonatal lung disease. TM

VEGF VEGF is a growth factor that regulates endothelial cell proliferation and differentiation. 182 VEGF is present in distal epithelial cells of the human fetal lung. 183 VEGF is also expressed by alveolar II cells in adult lung tissue after oxygen injury. TM VEGF binds to at least two receptors, VEGFR1 (Fit-l) and VEGFR2 (KDR). 182 K D R is present in human fetal lung tissue. 185 The formation and growth of the capillary bed in the fetal lung parallels the formation of primary and secondary septa in the lung. 1 In premature infants and baboons that develop BPD, decreased numbers of capillaries have been observed in distal lung tissue. 186'187 Recently, two groups have reported that BPD in human infants is associated with alterations in VEGF and its receptors. 187'188 VEGF gene-deleted mice as well as those in which the VEGFR genes are deleted die early in development as a result of disrupted vasculogenesis. 182 In mice which produced excess VEGF in distal lung epithelial cells, pulmonary blood vessels were more abundant; however, alveolar development was defective. 189 The 188 amino acid isoform of VEGF (VEGF 188) is abundant in the lung and accounts for a major portion of the perinatal increase in pulmonary VEGF mRNA. 19~ Cells at the epithelial surface of the distal air sacs are responsible for much of the expression of this VEGF isoform.

PDGF The PDGFs are a family of growth factors that bind two receptors, PDGF-R~ and PDGF-R~. TM PDGF-A is expressed in the developing lung in the epithelium of the distal lung tubules as well as in the connective tissue. 192 PDGF-Rt~ is also expressed in the developing lung. 192'193 Mice in which the PDGF-A gene is deleted die soon after birth due to respiratory distress. 16~The lungs of wild type and PDGF-A null mice are indistinguishable prior to birth and during early postnatal life. 160'194 However, during the postnatal period of rapid alveolarization, lungs in the PDGF-A gene deleted mice do not form alveoli. 16~ Alveolar myofibroblasts which are PDGF-R~ positive do not proliferate in the PDGF-A null mice nor do they migrate into secondary septa. 113 In addition, elastin is not formed in the secondary septa in PDGF-A-null mice. 113 These findings confirm the

67

critical role of myofibroblasts in the formation of secondary septa in the postnatal lung. 113'16~Mice in which the PDGFR~ is deleted die before birth due to multiple abnormalities. 195 The lungs of these animals are small and up until the time of death, at about day El6, appear to undergo branching morphogenesis in a manner similar to lungs in wild type mice. TM Thus the PDGF-A-PDGF-Rt~ axis is unique in that it appears to be a specific regulator of alveolarization in the postnatal lung.

RETINOIC ACID Retinoic acid and its receptors have been shown to be involved in almost every aspect of lung development and in the maintenance of lung differentiation. 196 All-trans retinoic acid is the major, biologically active metabolite of retinol (vitamin A). 196 All-trans retinoic acid binds to RARs and to retinoid X receptors (RXRs). 197 Another metabolite of alltrans retinoic acid, 9-cis retinoic acid, binds primarily to the R X R s . 197 There are three isoforms of RARs, ct, ~ and y and, likewise, three RXR isoforms, t~, fl and Y.197 Retinol is taken up in the digestive tract and is stored in the body as retinyl esters. The fetal lung stores retinyl esters in fibroblasts during lung development. 153 Coordinate with AEC differentiation, lung stores of retinyl esters decline, suggesting that lung fibroblasts may release retinol locally, which can then be metabolized to retinoic acid in retinoid-sensitive tissues such as the adjacent lung epithelium. 198 The lung epithelium expresses all isoforms of the RARs. 156 Deletion of the RARy gene leads to an impairment of alveolarization in mice. 158 Simultaneous deletion of one RXRt~ gene (a R A R 7 - / - , RXR +/- genotype) exacerbates this effect. 158 Thus RARy is probably involved in the formation of alveoli in postnatal lung. Deletion of the RAR[3 gene has been reported to increase the number of alveoli, decrease their size and increase the rate of alveolar formation in affected mice. 199 Interestingly, there was no effect of the RAR[3 gene deletion on the gas-exchange surface area of the lung. 199 In a recent study, we found that deletion of the RAR[3 gene results in impaired alveolarization as reflected by increased alveolar size, decreased gas-exchange surface area and a decrease in respiratory function (Fig. 4.7). 200 The first evidence for a role of RA in alveolarization was presented in two studies by Massaro et al. 154'2~ In the first study, it was shown that administration of all-tram R A to newborn rats increased the number of alveoli, decreased the size of the alveoli and increased the surface area of the lungs. TM Glucocorticoids inhibit alveolarization in the newborn rat, and it was shown that RA administration could overcome this inhibition. 154'2~ In another study performed in adult rats, RA stimulated the formation of alveoli in animals previously treated with elastase to produce an emphysematous phenotype. TM Clinical evidence supports a role for retinoic acid in lung development: preterm infants are frequently deficient in retinol and low

: The Lung::: Development,: Aging andthe:Environ:ment

!

::: ::: : :::

:

ACKNOWLEDGEMENTS !i~!~i!il!ii~i!!i!iii~i!i!'~i~i~:84184 i,:~i) ~i~:~:!::i~:~ i:

,ii!iii:)~i

|/

The authors wish to thank Jean Gardner for preparing the manuscript. The authors' work is supported by grants from the National Institutes of Health HL62861, HL53430, DERC DK-25295, Department of Veterans Affairs Research Service, and a grant from the March of Dimes Birth Defects Foundation.

REFERENCES

k Fig. 4.7. Van Gieson elastic tissue stain of lung tissue at 28 days. Lungs from R A R - / - and wild-type mice were fixed at an inflation pressure of 20 cm and paraffin sections were prepared. Sections were stained using the van Gieson elastin stain and photographed. Lung tissues from wild-type mice were stained simultaneously so that the incubation times were equivalent for the two sources of tissue. Panel A is from a representative wild-type mouse and panel B from a representative RARy-/- mouse. Arrows point to elastic fibers in alveolar septa, airway = (a), blood vessel = (b). (Reproduced with permission from McGowan S, Jackson SK, Jenkins-Moore M e t al. Mice bearing deletions of retinoic acid receptors demonstrate reduced lung elastin and alveolar numbers. Am. J. Respir. Cell Mol. Biol. 2000; 23:162-7.)

levels of plasma retinol have been correlated with an increased risk of developing BPD in this population. 2~ Furthermore, clinical trials have shown that treatment of premature infants with retinoic acid can reduce the incidence of BPD and decrease mortality. 2~176

CONCLUSIONS The formation of alveoli in the developing lung is necessary for normal lung function. When alveolarization is impaired, as in BPD, respiratory insufficiency and chronic lung disease ensue. When alveoli are destroyed in adult lung, as in emphysema, they do not regenerate. As a result of fundamental studies such as those reviewed in this chapter, we now know more about alveolarization and its regulation than ever before. We know that the myofibroblast in the wall of the primary septa probably orchestrates the growth of the secondary septa and the deposition of elastin that result in the formation of alveolar walls. We also know that PDGF-A and its receptor as well as retinoic acid and its receptors regulate this process. Other important mediators of alveolarization will be revealed by further study. Hopefully, future insights about alveolarization will lead to new treatment approaches for lung diseases in which alveolarization is impaired. Specifically, strategies to enable reactivation of alveolarization in the adult emphysematous lung and in newborns with BPD are greatly needed.

1. Burri PH. Structural aspects of prenatal and postnatal development and growth of the lung. In: Lung Growth and Development. New York: Marcel Decker, 1997, pp. 1-35. 2. Langston C, Kida K, Reed Met al. Human lung growth in late gestation and in the neonate. Am. Rev. Respir. Dis. 1984; 129:607-13. 3. Crapo JD, Barry BE, Gehr P etal. Cell number and cell characteristics of the normal human lung. Am. Rev. Respir. D/s. 1982; 126:332-7. 4. Evans MJ, Cabral LJ, Stephens RJ etal. Renewal of alveolar epithelium in the rat following exposure to NO 2. Am. J. Pathol. 1973; 70:175-98. 5. Wright JR, Clements JA. Metabolism and turnover of lung surfactant.Am. Rev. Respir. Dis. 1987; 136:426-44. 6. Massaro GD, Massaro D. Formation of pulmonary alveoli and gas-exchange surface area: quantitation and regulation. Annu. Rev. Physiol. 1996; 58:73-92. 7. Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am. J. Respir. Crit. Care Med. 2001; 163:1723-9. 8. Walter R, Gottlieb DJ, O'Connor GT. Environmental and genetic risk factors and gene-environment interactions in the pathogenesis of chronic obstructive lung disease. Environ. Health Perspect. 2000; 108 (Suppl. 4):733-42. 9. Cardoso WV. Molecular regulation of lung development. Annu. Rev. Physiol. 2001; 63:471-94. 10. Motoyama J, Liu J, Mo R et al. Essential function of Gli2 and Gli3 in the formation of lung, trachea and oesophagus. Nature Genet. 1998; 20:54-7. 11. Mahlapuu M, Enerback S, Carlsson P. Haploinsufficiency of the forkhead gene Foxfl, a target for Sonic hedgehog signaling, causes lung and foregut malformations. Dev. Suppl. 2001; 128:2397-406. 12. Min H, Danilenko DM, Scully SA etal. Fgf-10 is required for both limb and lung development and exhibits striking functional similarity to Drosophila branchless. Genes Dev. 1998; 12:3156-61. 13. Arman E, Haffner-Krausz R, Gorivodsky M etal. Fgfr2 is required for limb outgrowth and lung-branching morphogenesis. Proc. Nat. Acad. Sci. USA 1999; 96:11895-9. 14. Meno C, Shimono A, Saijoh Y etal. Lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 1998; 94:287-97. 15. Mendelsohn C, Lohnes D, Decimo D et al. Function of the retinoic acid receptors (RARs) during development. II. Multiple abnormalities at various stages of organogenesis in RAR double mutants. Development 1994; 120:2749-71. 16. Shannon JM, Deterding RR. Epithelial-mesenchymal interactions in lung development. In: McDonald JA (ed.), Lung Growth and Development. New York: Marcel Dekker, 1997, pp. 81-118. 17. Keyzer R, Post M. Lung branching morphogenesis: role of growth factors and extracellular matrix. In Gaultier C, Bourbon JR, Post M (eds), Lung Development. New York: Oxford University Press, 1999, pp. 1-27.

Development of Alveoli

18. Costa RH, Kalinichenko VV, Lim L. Transcription factors in mouse lung development and function. Am. J. Physiol. 2001; 280:L823-38. 19. Kitaoka H, Burri PH, Weibel ER. Development of the human fetal airway tree: analysis of the numerical density of airway endtips.Anat. Rec. 1996; 244:207-13. 20. Mallampalli RK, Acarregui MJ, Snyder JM. Differentiation of the alveolar epithelium in the fetal lung. In: McDonald JA (ed.), Lung Growth and Development. New York: Marcel Dekker, 1997, pp. 119-62. 21. Morrell NW, Weiser CM, Stenmark KR. Development of the pulmonary vasculature. In: Gaultier C, Bourbon JR, Post M (eds), Lung Development. New York, NY: Oxford University Press, 1999, pp. 152-95. 22. Mendelson CR. Endocrinology of the Lung. Totowa, NJ: Humana Press. 23. Mariani TJ, Pierce RA. Development of lung elastic matrix. In: Gaultier C, Bourbon JR, Post M (eds), Lung Development. New York, NY: Oxford University Press, 1999, pp. 28-45. 24. Burri PH, Dbaly J, Weibel ER. The postnatal growth of the rat lung. I. Morphometry. Anat. Rec. 1974; 178:711-30. 25. Kauffman SL, Burri PH, Weibel ER. The postnatal growth of the rat lung. II. Autoradiography.Anat. Rec. 1974; 180:63-76. 26. Adler KB, Low RB, Leslie KO etal. Contractile cells in normal and fibrotic lung. Lab. Invest. 1989; 60:473-85. 27. Mitchell JJ, Reynolds SE, Leslie KO et al. Smooth muscle cell markers in developing rat lung. Am. J. Respir. Cell Mol. Biol. 1990; 3:515-23. 28. Vidic B, Burri PH. Morphometric analysis of the remodeling of the rat pulmonary epithelium during early postnatal development.Anat. Rec. 1983; 207:317-24. 29. Burri PH, Tarek MR. A novel mechanism of capillary growth in the rat pulmonary microcirculation. Anat. Rec. 1990; 228:35-45. 30. Massaro GD, Massaro D. Postnatal lung growth: evidence that the gas-exchange region grows fastest at the periphery. Am. J. Physiol. 1993; 265:L319-22. 31. Noguchi A, Samaha H. Developmental changes in tropoelastin gene expression in the rat lung studied by in situ hybridization.Am. J. Respir. Cell Mol. Biol. 1991; 5:571-8. 32. Nardell EA, Brody JS. Determinants of mechanical properties of rat lung during postnatal development. J. Appl. Physiol. 1982; 53:140-8. 33. Cherukupalli K, Larson JE, Puterman M e t a l . Comparative biochemistry of gestational and postnatal lung growth and development in the rat and human. Pediatr. Pulmonol. 1997; 24:12-21. 34. Goncalves C, Barros J, Honouio A etal. Quantification of elastin from the mouse lung during postnatal development. Exp. Lung Res. 2001; 27:533-45. 35. Zeltner TB, Caduff JH, Gehr P et al. The postnatal development and growth of the human lung. I. Morphometry. Respir. Physiol. 1987; 67:247-67. 36. Desai R, Wigglesworth JS, Aber V. Assessment of elastin maturation by radioimmunoassay of desmosine in the developing human lung. Early Human Dev. 1988; 16:61-71. 37. Lin Y, Lechner AJ. Development of alveolar septa and cellular maturation within the perinatal lung. Am. J. Respir. Cell Mol. Biol. 1991; 4:59-64. 38. Hislop A, Howard S, Fairweather DV. Morphometric studies on the structural development of the lung in Macaca fascicularis during fetal and postnatal life.J. Anat. 1984; 138:95-112. 39. Docimo SG, Crone RK, Davies P e t al. Pulmonary development in the fetal lamb: morphometric study of the alveolar phase. Anat. Rec. 1991; 229:495-8. 40. Davies P, Reid L, Lister G et al. Postnatal growth of the sheep lung: a morphometric study.Anat. Rec. 1988; 220:281-6. 41. Willet KE, McMenamin P, Pinkerton KE et al. Lung morphometry and collagen and elastin content: changes during

42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56.

57. 58.

59. 60.

61. 62. 63. 64.

normal development and after prenatal hormone exposure in sheep. Pediatr. Res. 1999; 45:615-25. Schellenberg JC, Liggins GC. Elastin and collagen in the fetal sheep lung. I. Ontogenesis. Pediatr. Res. 1987; 22:335-8. Uhal BD. Cell cycle kinetics in the alveolar epithelium. Am. J. Physiol. 1997; 272:L1031-45. Kauffman SL. Cell proliferation in the mammalian lung. Int. Rev. Exp. Pathol. 1980; 22:131-91. Evans MJ, Cabral LJ, Stephens RJ etal. Transformation of alveolar type 2 cells to type 1 cells following exposure to NO 2. Exp. Mol. Pathol. 1975; 22:142-50. Mercurio AR, Rhodin JA. An electron microscopic study on the type I pneumocyte in the cat: differentiation. Am. J. Anat. 1976; 146:255-71. Mercurio AR, Rhodin JA. An electron microscopic study on the type I pneumocyte in the cat: pre-natal morphogenesis. J. Morphol. 1978; 156:141-55. Reynolds LJ, McElroy M, Richards RJ. Density and substrata are important in lung type II cell transdifferentiation in vitro. Int. J. Biochem. Cell Biol. 1999; 31:951-60. Borok Z, Danto SI., Lubman RL et al. Modulation of tlalpha expression with alveolar epithelial cell phenotype in vitro. Am. J. Physiol. 1998; 275:L155-64. Warburton D, Zhao J, Berberich MA et al. Molecular embryology of the lung: then, now, and in the future. Am. J. Physiol. 1999; 276:L697-704. Hackett BP, Bingle CD, Gitlin JD. Mechanisms of gene expression and cell fate determination in the developing pulmonary epithelium. Annu. Rev. Physiol. 1996; 58:51-71. Bourbon JR. Gene expression in development. In: Gaultier C, Bourbon JR, Post M (eds), Lung Development. New York, NY:Oxford University Press, 1999, pp. 77-121. Wright C, Strauss S, Toole K etal. Composition of the pulmonary interstitium during normal development of the human fetus. Pediatr. Dev. Pathol. 1999; 2:424-31. Roman J. Fibronectin and fibronectin receptors in lung development. Exp. Lung Res. 1997; 23:147-59. Schuger L. Laminins in lung development. Exp. Lung Res. 1997; 23:119-29. Burri PH. Lung development and pulmonary angiogenesis. In: Gaultier C, Bourbon JR, Post M (eds), Lung Development. New York, NY:Oxford University Press, 1999, pp. 122-51. Honda T, Ishida K, Hayama M e t al. Type II pneumocytes are preferentially located along thick elastic fibers forming the framework of human alveoli.Anat. Rec. 2000; 258:34-8. Dobbs LG, Williams MC, Brandt AE. Changes in biochemical characteristics and pattern of lectin binding of alveolar type II cells with time in culture. Biochim. Biophys. Acta 1985; 846:155-66. Joyce-Brady MF, Brody JS. Ontogeny of pulmonary alveolar epithelial markers of differentiation. Dev. Biol. 1990; 137:331-48. Williams MC, Cao Y, Hinds A et al. T1 alpha protein is developmentally regulated and expressed by alveolar type I cells, choroid plexus, and ciliary epithelia of adult rats. Am. J. Respir. Cell Mol. Biol. 1996; 14:577-85. Newman V, Gonzalez RF, Matthay MA et al. A novel alveolar type I cell-specific biochemical marker of human acute lung injury. Am. J. Respir. Crit. Care Med. 2000; 161:990-5. Verkman AS, Matthay MA, Song Y. Aquaporin water channels and lung physiology.Am. J. Physiol. 2000; 278:L867-79. Kreda SM, Gynn MC, Fenstermacher DA etal. Expression and localization of epithelial aquaporins in the adult human lung.Am.J. Respir. Cell Mol. Biol. 2001; 24:224-34. Attar MA, Bailie MB, Christensen PJ etal. Induction of ICAM-1 expression on alveolar epithelial cells during lung development in rats and humans. Exp. Lung Res. 1999; 25:245-59.

70

The::Lun : Development,:Agingand

65. Abraham V, Chou ML, George P etal. Heterocellular gap junctional communication between alveolar epithelial cells. Am. J. Physiol. 2001; 280:L1085-93. 66. Isakson BE, Lubman RL, Seedorf GJ etal. Modulation of pulmonary alveolar type II cell phenotype and communication by extracellular matrix and KGF. Am. J. Physiol. 2001; 281 :C1291-9. 67. Newman GR, Campbell L, von Ruhland C etal. Caveolin and its cellular and subcellular immunolocalisation in lung alveolar epithelium: implications for alveolar epithelial type I cell function. Cell Tissue Res. 1999; 295:111-20. 68. Gumbleton M. Caveolae as potential macromolecule trafficking compartments within alveolar epithelium. Adv. Drug Delivery Rev. 2001; 49:281-300. 69. Crapo JD, Young SL, Fram EK etal. Morphometric characteristics of cells in the alveolar region of mammalian lungs. Am. Rev. Respir. Dis. 1983; 128:$42-6. 70. Veldhuizen R, Nag K, Orgeig S et al. The role of lipids in pulmonary surfactant. Biochim. Biophys. Acta 1998; 1408:90-108. 71. Rooney SA. Phospholipid composition, biosynthesis and secretion. In: Parent RA (ed.), Comparative Biology of the Normal Lung. Boca Raton, FL:CRC Press, 1991, pp. 511-44. 72. Orgeig S, Daniels CB. The roles of cholesterol in pulmonary surfactant: insights from comparative and evolutionary studies. Comp. Biochem. Physiol. 2001; 129:75-89. 73. Wagle S, Bui A, Ballard PL etal. Hormonal regulation and cellular localization of fatty acid synthase in human fetal lung.Am. J. Physiol. 1999; 277:L381-90. 74. Guthmann F, Harrach-Ruprecht B, Looman AC etal. Interaction of lipoproteins with type II pneumocytes in vitro: morphological studies, uptake kinetics and secretion rate of cholesterol. Fur. J. Cell Biol. 1997; 74:197-207. 75. Mallampalli RK, Salome RG, Bowen SL et al. Very low density lipoproteins stimulate surfactant lipid synthesis in vitro. J. Clin. Invest. 99:2020-9. 76. Avery ME, Mead J. Surface properties in relation to atelectasis and hyaline membrane disease.Am.J. DisabledChild 1959; 97:517-23. 77. Jobe AH, Ikegami M. Lung development and function in preterm infants in the surfactant treatment era. Ann. Rev. Physiol. 2000; 62:825-46. 78. Jobe AH. Pulmonary surfactant therapy. New Engl. J. Med. 1993; 328:861-8. 79. Halliday HL. Natural vs synthetic surfactants in neonatal respiratory distress syndrome. Drugs 1996; 51:226-37. 80. Creuwels LA, van Golde LM, Haagsman HP. The pulmonary surfactant system: biochemical and clinical aspects. Lung 1997; 175:1-39. 81. Khubchandani KR, Snyder JM. Surfactant protein A (SP-A): the alveolus and beyond. FASEBJ. 2001; 15:59-69. 82. Hawgood S, Poulain FR. The pulmonary collectins and surfactant metabolism. Annu. Rev. Physiol. 2001; 63:495-519. 83. Crouch E, Wright JR. Surfactant proteins a and d and pulmonary host defense.Annu. Rev. Physiol. 2001; 63:521-54. 84. Lawson PR, Reid KB. The roles of surfactant proteins A and D in innate immunity. Immunol. Rev. 2000; 173:66-78. 85. van Rozendaal BA, van Golde LM, Haagsman HP. Localization and functions of SP-A and SP-D at mucosal surfaces. Pediatr. Pathol. Mol. Med. 2001; 20:319-39. 86. McCormack FX. Functional mapping of surfactant protein A. Pediatr. Pathol. Mol. Med. 2001; 20:293-318. 87. Cockshutt AM, Weitz J, Possmayer F. Pulmonary surfactantassociated protein A enhances the surface activity of lipid extract surfactant and reverses inhibition by blood proteins in vitro. Biochemistry 1990; 29:8424-9. 88. Ikegami M, Elhalwagi BM, Palaniyar N et al. The collagen-like region of surfactant protein A (SP-A) is required for correction of surfactant structural and functional defects in the SP-A null mouse.J. Biol. Chem. 2001; 276:38542-8.

the:Environment

:::::::::::::::::: ::: ::

89. Korfhagen TR, LeVine AM, Whitsett JA. Surfactant protein A (SP-A) gene targeted mice. Biochim. Biophys. Acta 1998; 1408:296-302. 90. Rubio S, Lacaze-Masmonteil T, Chailley-Heu B etal. Pulmonary surfactant protein A (SP-A) is expressed by epithelial cells of small and large intestine. J. Biol. Chem. 1995; 270:12162-9. 91. Goss KL, Kumar AR, Snyder JM. SP-A2 gene expression in human fetal lung airways. Am. J. Respir. Cell Mol. Biol. 1998; 19:613-21. 92. Khubchandani KR, Goss KL, Engelhardt JF etal. In situ hybridization of SP-A mRNA in adult human conducting airways. Pediatr. Pathol. Mol. Med. 2001; 20:349-66. 93. Weaver TE, Conkright JJ. Function of surfactant proteins B and C.Annu. Rev. Physiol. 2001; 63:555-78. 94. Tokieda K, Iwamoto HS, Bachurski C etal. Surfactant protein-B-deficient mice are susceptible to hyperoxic lung injury.Am. J. Respir. Cell Mol. Biol. 1999; 21:463-72. 95. Korimilli A, Gonzales LW, Guttentag SH. Intracellular localization of processing events in human surfactant protein B biosynthesis.J. Biol. Chem. 2000; 275:8672-9. 96. Clark JC, Wert SE, Bachurski CJ et al. Targeted disruption of the surfactant protein B gene disrupts surfactant homeostasis, causing respiratory failure in newborn mice. Proc. Nat. Acad. Sci. USA 1995; 92:7794-8. 97. Cole FS, Hamvas A, Nogee LM. Genetic disorders of neonatal respiratory function. Pediatr. Res. 2001; 50:157-62. 98. Glasser SW, Burhans MS, Korfhagen TR etal. Altered stability of pulmonary surfactant in SP-C-deficient mice. Proc. Nat. Acad. Sci. USA 2001; 98:6366-71. 99. Nogee LM, Dunbar AE, III, Wert SEet al. A mutation in the surfactant protein C gene associated with familial interstitial lung disease. New Eng. J. Med. 2001; 344:573-9. 100. Crouch EC. Structure, biologic properties, and expression of surfactant protein D (SP-D). Biochim. Biophys. Acta 1998; 1408:278-89. 101. Botas C, Poulain F, Akiyama J etal. Altered surfactant homeostasis and alveolar type II cell morphology in mice lacking surfactant protein D. Proc. Nat. Acad. Sci. USA 1998; 95:11869-74. 102. Wert SE, Yoshida M, LeVine AM et al. Increased metalloproteinase activity, oxidant production, and emphysema in surfactant protein D gene-inactivated mice. Proc. Nat. Acad. Sci. USA 2000; 97:5972-7. 103. Vaccaro C, Brody JS. Ultrastructure of developing alveoli. I. The role of the interstitial fibroblast. Anat. Rec. 1978; 192:467-79. 104. Brody JS, Kaplan NB. Proliferation of alveolar interstitial cells during postnatal lung growth: evidence for two distinct populations of pulmonary fibroblasts. Am. Rev. Respir. Dis. 1983; 127:763-70. 105. Tschanz TA, Damke BM, Burri PH. Influence of postnatally administered glucocorticoids on rat lung growth. Biol. Neonate 1995; 68:229-45. 106. Adler KB, Callahan LM, Evans JN. Cellular alterations in the alveolar wall in bleomycin-induced pulmonary fibrosis in rats: an ultrastructural morphometric study. Am. Rev. Respir. Dis. 1986; 133:1043-8. 107. Lindroos PM, Wang YZ, Rice AB etal. Regulation of PDGFR-alpha in rat pulmonary myofibroblasts by staurosporine. Am. J. Physiol. 2001; 280:L354-62. 108. Kawamoto M, Matsunami T, Ertl RF et al. Selective migration of alpha-smooth muscle actin-positive myofibroblasts toward fibronectin in the Boyden's blindwell chamber. Clin. Sci. 1997; 93:355-62. 109. Senior RM, Griffin GL, Mecham RP. Chemotactic responses of fibroblasts to tropoelastin and elastin-derived peptides. J. Clin. Invest. 1982; 70:614-18.

Development of Alveoli

110. Knapp DM, Helou EF, Tranquillo RT. A fibrin or collagen gel assay for tissue cell chemotaxis: assessment of fibroblast chemotaxis to GRGDSP. Exp. Cell Res. 1999; 247:543-53. 111. Bruce MC, Schuyler M, Martin RJ et al. Risk factors for the degradation of lung elastic fibers in the ventilated neonate: implications for impaired lung development in bronchopulmonary dysplasia. Am. Rev. Respir. Dis. 1992; 146:204-12. 112. Osornio-Vargas AR, Goodell AL, Hernandez-Rodriguez NA etal. Platelet-derived growth factor (PDGF)-AA, -AB, and-BB induce differential chemotaxis of early-passage rat lung fibroblasts in vitro. Am. J. Respir. Cell Mol. Biol. 1995; 12:33-40. 113. Lindahl P, Karlsson L, Hellstrom M et al. Alveogenesis failure in PDGF-A-deficient mice is coupled to lack of distal spreading of alveolar smooth muscle cell progenitors during lung development.Development 1997; 124:3943-53. 114. Pasquali-Ronchetti I, Baccarani-Contri M. Elastic fiber during development and aging. Microsc. Res. Tech. 1997; 38:428-35. 115. Tanaka R, A1-Jamal R, Ludwig MS. Maturational changes in extracellular matrix and lung tissue mechanics. J. Appl. Physiol. 2001; 91:2314-21. 116. Maksvytis HJ, Vaccaro C, Brody JS. Isolation and characterization of the lipid-containing interstitial cell from the developing rat lung. Lab. Invest. 1981; 45:248-59. 117. Bruce MC, Lo PY. A morphometric quantitation of developmental changes in elastic fibers in rat lung parenchyma: variability with lung region and postnatal age. J. Lab. Clin. Med. 1991; 117:226-33. 118. Burri PH. The postnatal growth of the rat lung. 3. Morphology. Anat. Rec. 1974; 180:77-98. 119. Kaplan NB, Grant MM, Brody JS. The lipid interstitial cell of the pulmonary alveolus: age and species differences. Am. Rev. Respir. Dis. 1985; 132:1307-12. 120. Bruce MC. Developmental changes in tropoelastin mRNA levels in rat lung: evaluation by in situ hybridization. Am. J. Respir. Cell Mol. Biol. 1991; 5:344-50. 121. Myers B, Dubick M, Last JA etal. Elastin synthesis during perinatal lung development in the rat. Biochim. Biophy. Acta 1983; 761:17-22. 122. Kida K, Thurlbeck WM. Lack of recovery of lung structure and function after the administration of beta-aminopropionitrile in the postnatal period. Am. Rev. Respir. Dis. 1980; 122:467-75. 123. Pierce RA, Mariencheck WI, Sandefur S etal. Glucocorticoids upregulate tropoelastin expression during late stages of fetal lung development. Am. J. Physiol. 1995; 268:L491-500. 124. Li DY, Brooke B, Davis EC etal. Elastin is an essential determinant of arterial morphogenesis. Nature 1998; 393:276-80. 125. Wendel DP, Taylor DG, Albertine KH etal. Impaired distal airway development in mice lacking elastin. Am. J. Respir. Cell Mol. Biol. 2000; 23:320-6. 126. Brettell LM, McGowan SE. Basic fibroblast growth factor decreases elastin production by neonatal rat lung fibroblasts.Am.J. Respir. Cell Mol. Biol. 1994; 10:306-15. 127. Rich CB, Fontanilla MR, Nugent M etal. Basic fibroblast growth factor decreases elastin gene transcription through an AP1/cAMP-response element hybrid site in the distal promoter.J. Biol. Chem. 1999; 274:33433-9. 128. Noguchi A, Firsching K, Kursar JD etal. Developmental changes of tropoelastin synthesis by rat pulmonary fibroblasts and effects of dexamethasone. Pediatr. Res. 1990; 28:379-82. 129. Swee MH, Parks WC, Pierce RA. Developmental regulation of elastin production: expression of tropoelastin pre-mRNA

persists after downregulation of steady-state mRNA levels. J. Biol. Chem. 1995; 270:14899-906. 130. Bruce MC, Bruce EN, Janiga K et al. Hyperoxic exposure of developing rat lung decreases tropoelastin mRNA levels that rebound postexposure.Am. J. Physiol. 1993; 265:L293-300. 131. Zhang M, Pierce RA, Wachi H et al. An open reading frame element mediates post-transcriptional regulation of tropoelastin and responsiveness to transforming growth factor betal. Mol. Cell. Biol. 1999; 19:7314-26. 132. Trask TM, Trask BC, Ritty TM etal. Interaction of tropoelastin with the amino-terminal domains of fibrillin-1 and fibrillin-2 suggests a role for the fibrillins in elastic fiber assembly.J. Biol. Chem. 2000; 275:24400-6. 133. Jensen SA, Vrhovski B, Weiss AS. Domain 26 of tropoelastin plays a dominant role in association by coacervation. J. Biol. Chem. 2000; 275:28449-54. 134. Wu WJ, Vrhovski B, Weiss AS. Glycosaminoglycans mediate the coacervation of human tropoelastin through dominant charge interactions involving lysine side chains. J. Biol. Chem. 1999; 274:21719-24. 135. Stone PJ, Morris SM, Griffin S etal. Building elastin: incorporation of recombinant human tropoelastin into extracellular matrices using nonelastogenic rat-1 fibroblasts as a source for lysyl oxidase. Am. J. Respir. Cell Mol. Biol. 2001; 24:733-9. 136. Shapiro SD, Endicott SK, Province MA etal. Marked longevity of human lung parenchymal elastic fibers deduced from prevalence of D-aspartate and nuclear weapons-related radiocarbon. J. Clin. Invest. 1991; 87:1828-34. 137. Ramirez F, Pereira L. The fibrillins. Int. J. Biochem. Cell Biol. 1999; 31:255-9. 138. Zhang H, Hu W, Ramirez F. Developmental expression of fibrillin genes suggests heterogeneity of extracellular microfibrils.J. Cell Biol. 1995; 129:1165-76. 139. Siracusa LD, McGrath R, Fisher JK etal. The mouse tight skin (Tsk) phenotype is not dependent on the presence of mature T and B lymphocytes. Mammalian Genome 1998; 9:907-9. 140. Spit BJ. Induction of lipid droplets in fibroblasts of the hamster lung by a diet high in vitamin A. Exp. Lung Res. 1983; 4:247-57. 141. Tordet C, Marin L, Dameron F. Pulmonary di- and -triacylglycerols during the perinatal development of the rat. Experientia 1981; 37:333-4. 142. Chen H, Jackson S, Doro M etal. Perinatal expression of genes that may participate in lipid metabolism by lipidladen lung fibroblasts.J. Lipid Res. 1998; 39:2483-92. 143. Maksvytis HJ, Niles RM, Simanovsky L etal. In vitro characteristics of the lipid-filled interstitial cell associated with postnatal lung growth: evidence for fibroblast heterogeneity. J. Cell. Physiol. 1984; 118:113-23. 144. McGowan SE, Doro MM, Jackson S. Expression of lipoprotein receptor and apolipoprotein E genes by perinatal rat lipid-laden pulmonary fibroblasts. Exp. Lung Res. 2001; 27:47-63. 145. Okabe T, Yorifuji H, Yamada E et al. Isolation and characterization of vitamin-A-storing lung cells. Exp. Cell Res. 1984; 154:125-35. 146. McGowan SE, Jackson SK, Doro MM etal. Peroxisome proliferators alter lipid acquisition and elastin gene expression in neonatal rat lung fibroblasts. Am. J. Physiol. 1997; 273:L1249-57. 147. Grobstein C. Mechanisms of organogenic tissue interaction. Nat. Cancer Inst. Monogr. 1967; 26:279-99. 148. Sorokin S, Padykula HA, Herman E. Comparative histochemical patterns in developing mammalian lungs. Dev. Biol. 1959; 1:125-51. 149. Torday J, Hua J, Slavin R. Metabolism and fate of neutral lipids of fetal lung fibroblast origin. Biochim. Biophys. Acta 1995; 1254:198-206.

::TheLung: Developmentl Aging and the Environment::

150. Grant MM, Cutts NR, Brody JS. Alterations in lung basement membrane during fetal growth and type 2 cell development. Dev. Biol. 1983; 97:173-83. 151. McGowan SE, Harvey CS, Jackson SK. Retinoids, retinoic acid receptors, and cytoplasmic retinoid binding proteins in perinatal rat lung fibroblasts. Am. J. Physiol. 1995; 269:L463-72. 152. Chytil F. The lungs and vitamin A. Am. J. Physiol. 1992; 262:L517-27. 153. Shenai JP, Chytil F. Vitamin A storage in lungs during perinatal development in the rat. Biol. Neonate 1990; 57: 126-32. 154. Massaro GD, Massaro D. Postnatal treatment with retinoic acid increases the number of pulmonary alveoli in rats. Am. J. Physiol. 1996; 270:L305-10. 155. Grummer MA, Zachman RD. Postnatal rat lung retinoic acid receptor (RAR) mRNA expression and effects of dexamethasone on RAR beta mRNA. Pediatr. Pulmonol. 1995; 20:234-40. 156. Grummer MA, Thet LA, Zachman RD. Expression of retinoic acid receptor genes in fetal and newborn rat lung. Pediatr. Pulmonol. 1994; 17: 234-8. 157. McGowan SE, Doro MM, Jackson SK. Endogenous retinoids increase perinatal elastin gene expression in rat lung fibroblasts and fetal explants. Am. J. Physiol. 1997; 273:L410-6. 158. McGowan S, Jackson SK, Jenkins-Moore M et al. Mice bearing deletions of retinoic acid receptors demonstrate reduced lung elastin and alveolar numbers. Am. J. Respir. Cell Mol. Biol. 2000; 23:162-7. 159. Leslie KO, Mitchell J, Low R. Lung myofibroblasts. Cell Motilility Cytoskeleton 1992; 22:92-8. 160. Bostrom H, Willetts K, Pekny M e t al. PDGF-A signaling is a critical event in lung alveolar myofibroblast development and alveogenesis. Cell 1996; 85:863-73. 161. Powell DW, Mifflin RC, Valentich JD et al. Myofibroblasts. I. Paracrine cells important in health and disease. Am. J. Physiol. 1999; 277:C1-9. 162. Roy SG, Nozaki Y, Phan SH. Regulation of alpha-smooth muscle actin gene expression in myofibroblast differentiation from rat lung fibroblasts. Int. J. Biochem. Cell Biol. 2001; 33:723-34. 163. Zhang HY, Phan SH. Inhibition of myofibroblast apoptosis by transforming growth factor beta(l). Am. J. Respir. Cell Mol. Biol. 1999; 21:658-65. 164. Hashimoto S, Gon Y, Takeshita I e t al. Transforming growth factor-betal induces phenotypic modulation of human lung fibroblasts to myofibroblast through a c-Jun-NH2terminal kinase-dependent pathway. Am. J. Respir. Crit. Care Med. 2001; 163:152-7. 165. Bruce MC, Honaker CE, Cross RJ. Lung fibroblasts undergo apoptosis following alveolarization. Am. J. Respir. Cell Mol. Biol. 1999; 20:228-36. 166. Weiss MJ, Burri PH. Formation of interalveolar pores in the rat lung. Anat. Rec. 1996; 244:481-9. 167. Coalson JJ, Winter V, deLemos RA. Decreased alveolarization in baboon survivors with bronchopulmonary dysplasia. Am. J. Respir. Crit. Care Med. 1995; 152:640-6. 168. Albertine KH, Jones GP, Starcher BC etal. Chronic lung injury in preterm lambs: disordered respiratory tract development.Am. J. Respir. Crit. Care Med. 1999; 159:945-58. 169. Martorana PA, Brand T, Gardi C et al. The pallid mouse, a model of genetic alpha 1-antitrypsin deficiency. Lab. Invest. 1993; 68:233-41. 170. Kielty CM, Raghunath M, Siracusa LD et al. The Tight skin mouse: demonstration of mutant fibrillin-1 production and assembly into abnormal microfibrils. J. Cell Biol. 1998; 140:1159-66. 171. Gardi C, Martorana PA, de Santi MM et al. A biochemical and morphological investigation of the early development

:: :

: :::::

:

of genetic emphysema in tight-skin mice. Exp. Mol. Pathol. 1989; 50:398-410. 172. de Santi MM, Martorana PA, Cavarra E et al. Pallid mice with genetic emphysema: neutrophil elastase burden and elastin loss occur without alteration in the bronchoalveolar lavage cell population.Lab. Invest. 1995; 73:40-7. 173. Bolt RJ, van Weissenbruch MM, Lafeber HN etal. Glucocorticoids and lung development in the fetus and preterm infant. Pediatr. Pulmonol. 2001; 32:76-91. 174. Ballard PL. Hormones and Lung Maturation. Berlin; Germany: Springer-Verlag, 1986. 175. Liggins GC. Premature delivery of fetal lambs infused with glucocorticoids. J. Endocrinol. 1969; 45:515-23. 176. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics, 1972; 50:515-25. 177. Muglia LJ, Bae DS, Brown TT etal. Proliferation and differentiation defects during lung development in corticotropin-releasing hormone-defcient mice. Am. J. Respir. Cell Mol. Biol. 1995; 20:181-8. 178. Cole TJ, Blendy JA, Monaghan AP et al. Targeted disruption of the glucocorticoid receptor gene blocks adrenergic chromaffin cell development and severely retards lung maturation. Genes Dev. 1995; 9:1608-21. 179. Massaro D, Teich N, Maxwell Set al. Postnatal development of alveoli: regulation and evidence for a critical period in rats.J. Clin. Invest. 1985; 76:1297-305. 180. Massaro D, Massaro GD. Dexamethasone accelerates postnatal alveolar wall thinning and alters wall composition. Am. J. Physiol. 1986; 251 :R218-24. 181. Jobe AH, Ikegami M. Prevention of bronchopulmonary dysplasia. Curr. Opin. Pediatr. 2001; 13:124-9. 182. Ferrara N. Vascular endothelial growth factor and the regulation of angiogenesis. Recent Progr. Hormone Res. 2000; 55:15-35 (discussion 35-6). 183. Shifren JL, Doldi N, Ferrara N etal. In the human fetus, vascular endothelial growth factor is expressed in epithelial cells and myocytes, but not vascular endothelium: implications for mode of action. J. Clin. Endocrinol. Metabol. 1994; 79:316-22. 184. Maniscalco WM, Watkins RH, Finkelstein JN etal. Vascular endothelial growth factor mRNA increases in alveolar epithelial cells during recovery from oxygen injury. Am. J. Respir. Cell Mol. Biol. 1995; 13:377-86. 185. Brown KR, England KM, Goss KL etal. VEGF induces airway epithelial cell proliferation in human fetal lung in vitro.Am. J. Physiol. 2001; 281:L1001-10. 186. Coalson JJ, Winter VT, Siler-Khodr T et al. Neonatal chronic lung disease in extremely immature baboons. Am. J. Respir. Crit. Care Med. 1999; 160:1333-46. 187. Bhatt AJ, Pryhuber GS, Huyck H et al. Disrupted pulmonary vasculature and decreased vascular endothelial growth factor, Fit-l, and TIE-2 in human infants dying with bronchopulmonary dysplasia. Am. J. Respir. Crit. Care Med. 2001; 164:1971-80. 188. Lassus P, Turanlahti M, Heikkila Pet al. Pulmonary vascular endothelial growth factor and Flt-1 in fetuses, in acute and chronic lung disease, and in persistent pulmonary hypertension of the newborn. Am. J. Respir. Crit. Care Med. 164:1981-7. 189. Zeng X, Weft SE, Federici R et al. VEGF enhances pulmonary vasculogenesis and disrupts lung morphogenesis in vivo. Dev. Dyn. 1998; 211:215-27. 190. Ng YS, Rohan R, Sunday ME et al. Differential expression of VEGF isoforms in mouse during development and in the adult. Dev. Dyn. 2001; 220:112-21. 191. Betsholtz C, Karlsson L, Lindahl P. Developmental roles of platelet-derived growth factors. Bioessays 2001; 23:494-507.

Develoi)ment of Alveoli

192. Han RN, Mawdsley C, Souza P e t al. Platelet-derived growth factors and growth-related genes in rat lung. III. Immunolocalization during fetal development. Pediatr. Res. 1992; 31:323-9. 193. Han RN, Liu J, Tanswell AK etal. Ontogeny of plateletderived growth factor receptor in fetal rat lung. Microsc. Res. Tech. 1993; 26:381-8. 194. Bostrom H, Gritli-Linde A, Betsholty C. PDGF-A/PDGF alpha-receptor signaling is required for lung growth and the formation of alveoli but not for early lung branching morphogenesis. Dev. Dyn. 2002; 223:155-62. 195. Soriano P. The PDGF alpha receptor is required for neural crest cell development and for normal patterning of the somites.Development 1997; 124:2691-700. 196. Chytil F. Retinoids in lung development. FASEB J. 1996; 10:986-92. 197. Chambon P. The molecular and genetic dissection of the retinoid signaling pathway. Recent Progr. Hormone Res. 1995; 50:317-32. 198. Geevarghese SK, Chytil F. Depletion of retinyl esters in the lungs coincides with lung prenatal morphological maturation. Biochem. Biophys. Res. Commun. 1994; 200: 529-35. 199. Massaro GD, Massaro D, Chan WY etal. Retinoic acid receptor-beta: an endogenous inhibitor of the perinatal

200. 201. 202. 203. 204.

205.

formation of pulmonary alveoli. Physiol. Genomics 2000; 4:51-7. Snyder JM, Jenkins-Moore M, Jackson SK et al. Alveolarization in retinoic acid receptor-b deficient mice. Pediatr. Res. (in preparation). Massaro GD, Massaro D. Retinoic acid treatment abrogates elastase-induced pulmonary emphysema in rats. Nature Med. 1997; 3:675-7. Massaro GD, Massaro D. Formation of alveoli in rats: postnatal effect of prenatal dexamethasone. Am. J. Physiol. 1992; 263:L37-41. Shenai JP, Kennedy KA, Chytil F etal. Clinical trial of vitamin A supplementation in infants susceptible to bronchopulmonary dysplasia.J. Pediatr. 1987; 111:269-77. Kennedy KA, Stoll BJ, Ehrenkranz RA etal. Vitamin A to prevent bronchopulmonary dysplasia in very-low-birthweight infants: has the dose been too low? The NICHD Neonatal Research Network. Early Human Dev. 1997; 49:19-31. Tyson JE, Wright LL, Oh W e t al. Vitamin A supplementation for extremely-low-birth-weight infants. National Institute of Child Health and Human Development Neonatal Research Network. New Engl. J. Med. 1999; 340:1962-8.

Development of the Pulmonary Basement Membrane Zone

Chapter

5

Michael J. Evans* Department of Anatomy, Physiology and Cell Biology, School of Veterinary Medicine, Center for Comparative Respiratory Biology and Medicine, University of California, Davis, CA, USA

Philip L. Sannes ) ~Department ofMolecular Biomedical Sciences, College of Veterinary :~ / :Medicine,~North Carolina State University, Raleigh, NC, U S A i~ii

i

i

STRUCTURE AND COMPOSITION OF THE B A S E M E N T M E M B R A N E Z O N E The basement membrane zone (BMZ) is a mat-like sheet of specialized extracellular matrix (ECM) that serves as a complex interface between epithelia, peripheral nerves or muscle cells and their surrounding tissue microenvironments. In a recent review, the morphology and composition of the BMZ and its functional significance in pulmonary development were presented. 1 In this chapter, we will discuss the structure, composition and development of the BMZ. The term basement membrane was originally derived from observations of tissues with light microscopy. In these preparations, the epithelial and endothelial basement membrane appears as a distinct layer beneath the cells. With transmission electron microscopy, the basement membrane appears as threecomponent layers: the lamina lucida, the lamina densa and the lamina reticularis. Together they make up the basal lamina. When studying the molecular structure of the basal lamina (BL) it is commonly referred to as the BMZ. In the lung, epithelial and endothelial cells, smooth muscle cells and nerves have a BMZ. In smooth muscle cells and nerves there is no basal region and the BMZ around these cells are referred to as the external lamina. However, both the BMZ and external lamina are the same functionally and molecu*To whom correspondence should be addressed. The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

larly and we will use the term BMZ to include the external lamina of nerves and smooth muscle cells. 2 The lamina lucida seen with transmission electron microscopy is a clear area between the cells and the lamina densa. This is thought to be a fixation artifact morphologically, but it appears to actually function as the region of attachment between adjacent cells and lamina densa. It contains cell adhesion molecules and anchoring filaments of laminins 5, 6 and 10. 3 The lamina densa is a sheet of connective tissue made up of type IV collagen, laminin, entactin/nidogen and proteoglycans. This region of the BMZ has been studied extensively and is commonly referred to as the basal lamina, basement membrane or true basement membrane. It is a biologically conserved structure that is virtually the same in all animals. On the ECM side of the lamina densa is the lamina reticularis. This is a region of attachment between the lamina densa and the ECM. 4-6 The lamina reticularis is variable in its distribution, thickness and composition. It is not apparent in all tissues; however, it is well developed under multilayered epithelium. The lamina reticularis is especially pronounced under the respiratory epithelium of large conducting airways, where it may be several microns thick. In the lung the lamina reticularis is thicker in the larger airways and becomes smaller as it extends into the small airways and alveoli. In alveoli, with fused BMZ, it is not physically present but still exists as a region of interaction of the lamina densa with the ECM. Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

~The L u n g - D e v e l o p m e n t , Aging and the E n v i r o n m e n t ]

Table 5.1.

.

:

Characteristics

.

:

of the basement

!/(i~i:i~ii~il~i!i 84184 Basement membrane i~i~(lightlmicroscopy)

i

.

.

: ....

.

]

.

.

.

.

.

]

;

membrane. ~

i

i

~

~

~

~

~ i8484 ~ ~i]i84184 i i )~i

Basal lamina ~ Basement membrane izone~ (electron microscopy) ~ ~(molecular structure)

~~

i-!,, Cellular interface

Lamina lucida

Collagen (XVll) Laminin (5, 6 and 10) Integrins (~6, IM, a, [3)

Lamina densa

Collagen (IV) Laminin (1) Entacti n/nidogen Proteoglycans (perlecan, bamacan, agrin, collagen XVlII) Stored growth factors (FGF-2)

Lamina reticularis

Collagen (I, III, V, Vl & VII) Proteoglycans (perlecan, bamacan, collagen XVlII) Stored growth factors (FGF-2)

Cellular-matrix Interface

Basement membrane

Matrix interface

The reason for the differences in the thickness of the lamina reticularis is not known. With transmission electron microscopy, the lamina reticularis is made up of numerous collagen fibrils. Immunohistochemical studies have shown that the collagen fibrils consist of types I, III, V, VI and VII collagen. 2 Collagen types I, III and V form heterogeneous fibers that account for the thickness of the lamina reticularis. Anchoring fibrils of type VII collagen loop through strands of collagen fibers in the lamina reticularis and then reattach to the lamina densa. 7 In addition proteoglycans are considered to be a structural component of the lamina reticularis as in the lamina densa. There are three proteoglycans that are considered to be an integral component of both the lamina densa and lamina reticularis. These proteoglycans (perlecan, agrin and bamacan) are specifically classified as BMZ proteoglycans. 8 In addition type XVIII collagen has recently been shown to be a BMZ heparan sulfate proteoglycan 9 and type XV collagen has been shown to be chondroitin/dermatin sulfate BMZ proteoglycan. 1~ In a study of airway whole mounts, the fiat surface of the BMZ was visualized with scanning electron microscopy and fluorescent microscopy. The lamina lucida is not visible in these preparations; however, when the epithelium is removed the lamina densa appears as a smooth, dense layer. 11 Fluorescent microscopy and scanning electron microscopy revealed that the structural proteins of the lamina reticularis are not randomly arranged, but instead appear as a mat of large fibers oriented along the longitudinal axis of the airway. Smaller fibers are cross-linked with the larger fibers to complete this structure. 12 Other small fibers are oriented around the large

fibers and an amorphous material covered individual fibers. The large fibers oriented along the longitudinal axis of the airway are consistent with prior descriptions of fibers comprised of collagen types I, III and V with some small fibers encircling the large fibers that may be collagen type VI. The cross-linking fibers are made up of elastin and probably elastin-associated microfibrils. 12'13 This study demonstrates that the structural proteins of the lamina reticularis are arranged as fibers in a highly organized manner comparable to type IV collagen and laminin in the lamina densa. A break down of the structure and composition of the BMZ is given in Table 5.1.

FUNCTIONS MEMBRANE

OF THE ZONE

BASEMENT

The BMZ has several functions. It serves as a barrier, binds growth factors, hormones and ions, and is involved with cellular adhesion, electrical charge and cell-cell and cellmatrix communication. 1'2'4'5 The lamina lucida is the interface between a layer of cells and the lamina densa. It exists as a zone of molecules attaching the layer of cells to the lamina densa. Other than a role in attachment, the molecules in this zone are important in signaling between the matrix and the adjacent layer of cells. As mentioned previously, the lamina densa is a sheet of connective tissue made up of type IV collagens, laminins, entactin/nidogen and proteoglycans. One of the main functions of the lamina densa is to provide separation between cells and the ECM. It also provides a surface for cells to migrate

Development of the Basement Membrane

and differentiate into distinct phenotypes, 1'14 which helps to determine tissue shape, stability and architecture. The lamina densa also influences important cellular activities, including adhesion spreading polarization locomotion, chemotaxis and proliferation. These functions are accomplished through binding of the cells to specific components of the lamina densa, namely collagen type IV, laminins, entactin/nidogen and proteoglycans. The interface between the lamina densa and the ECM is the lamina reticularis. It is comprised mainly of types I, III and V collagen fibers that are arranged as a mat of large fibers oriented along the longitudinal axis of the airway. Smaller fibers are cross-linked with the larger fibers to complete this structure. 12 Within and around the collagen fibrils are BMZ proteoglycans (perlecan, agrin, bamacan, type XVIII collagen and type XV collagen). Both the lamina densa and lamina reticularis store growth factors, hormones and ions, and are involved with cellular adhesion, electrical charge and cell-cell communication. The predominant BMZ proteoglycan is perlecan and the predominant stored growth factor is basic fibroblast growth factor (FGF-2). 8 FGF-2 is a ubiquitous multifunctional growth factor that plays roles during development and as a regulator of growth and differentiation in adult tissues. 15 FGF-2 is stored in the BMZ through binding with perlecan. Perlecan is a heparan sulfate proteoglycan that is an intrinsic constituent of the BMZ. FGF-2 can be released from perlecan in response to various conditions and become a signaling molecule. Thus perlecan functions as a regulator of FGF-2 transport allowing for rapid responses to local environmental conditions. 16 In the lung, FGF-2 is stored in the BMZ of airway epithelium, alveolar epithelium and endothelium of developing and adult rats. x7'18 Fibroblast growth factor receptor-1 (FGFR-1) is the primary receptor for FGF-2.17 Agrin is also a heparan sulfate proteoglycan that is most prominent in neuromuscular and glomerular basement membranes. It has also been reported in the alveolar region of the lung. 19'2~BMZs from these three tissues are similar in that they are fused. The function of agrin in the alveoli is not known. It binds to ~-dystroglycan and thus anchors the BMZ to alveolar cells and may be involved with signal transduction. It may also be involved with charge, selective filtration processes, and storage of various growth factors. It has also been related to inhibition of proteases. Agrin is not present in the BMZ of airway epithelium or smooth muscle cells. The heparan sulfate proteoglycan, type XVIII collagen, is found in alveolar epithelial and endothelial BMZs and in airway epithelial BMZs. It was not reported in smooth muscle BMZs. 21 The purpose of type XVIII in the lung is not clear. Bamacan is a chondroitin sulfate proteoglycan. Its function is also not known; however, its late appearance developmentally suggests a role in BMZ stability. 22 The recently described chondroitin sulfate proteoglycan, type XV collagen, is not found in lung BMZs. 21 The functional consequences of these collective molecules are profound by forming the basis for all cell-cell, cell growth factor interactions. Their composition defines epithelial differentiation and plasticity. 23'24

BMZ D E V E L O P M E N T The lamina densa region of the BMZ is composed of collagen IV, laminins, proteoglycans and entactin/nidogen. Collagen IV, laminin and entactin/nidogen are expressed with BMZs at early stages of lung development and persist throughout life. Entactin/nidogen is thought to play a crucial role in BMZ formation due to its ability to form complexes with type IV collagen, laminins and perlecan. Both collagen IV and laminins spontaneously polymerize into separate twodimensional networks. BMZs are thought to form by linking these two networks together with entactin/nidogen. 25 The BMZ proteoglycans are incorporated into this structure through binding with collagen IV, laminin and entactin/ nidogen. Biological diversity is maintained through different isoforms of the molecules making up the lamina densa. For example, there are six tx-chains of collagen type IV and at least three molecular forms found in various BMZs. One molecular form of collagen type IV contains the t~3, ~4 and ct5 chains and is found in the BMZ of neuromuscular, glomerular and pulmonary alveolar tissues. These three tissues are structurally similar in that they contain adjacent BMZs that appear to be fused. 26 Most other BMZs express collagen type IV ~1 and t~2 chains. 26-28 Laminins also contribute to the diversity of the lamina densa. They are a family of heterotrimeric glycoproteins containing an t~-, [3- and y-chain. Currently five t~-, three [3-, and three y-chains have been described that make up 15 laminin isoforms. Laminin-t~l and-~2 are present in fetal lungs and laminin-3, -t~, -4, t~ and -tx5 in both fetal and adult lungs. The three laminin [3-chains and laminin-y1 and -y2 are found in both fetal and adult lungs. 29-31 Recently, an isoform of entactin/nidogen was discovered and named entactin-2/nidogen-2. 32 The different BMZ proteoglycans (perlecan, agrin and bamacan, type XVIII collagen and type XV collagen) and their degree of sulfation represent diversity of the proteoglycan component of both the lamina densa and the lamina reticularis. The degree of sulfation was recently shown to be an important determinant of cell function in the lung. 33 The degree of sulfation is known to vary beneath cells lining the alveoli and different levels of the tracheo-bronchial tree. It also varies between the various regions of the BMZ. The highest levels of sulfation are found in the lamina reticularis of the trachea, bronchi and large bronchioles and the lowest levels in the lamina densa. 34 In the rat, chondroitin sulfate proteoglycans are widely distributed in the lung as early as 14 days before birth. 1 Their expression in the BMZ remains strong until day 7 after birth, after which their localization progressively becomes highly focal and limited to airways and septal regions of alveolar ducts. They are absent from the remaining BMZ regions of alveolar structures. Heparan sulfate proteoglycans are strongly expressed in these same regions from birth to adulthood. 22 Of the heparan sulfate proteoglycans, perlecan has been studied most extensively. In the

The:

Lung::Development, :

Aging

and:the::: Environment:::::

. . .: . : . :. . . . :. .

rat, perlecan mRNA is first detected in the fetus (day 19), and is abundantly expressed for the remainder of fetal and postnatal growth. It remains expressed at low levels throughout adult life. 35 Whereas the lamina densa is present at all stages of development, the lamina reticularis in rats develops postnatally. 36 Instead of collagen IV, collagens I, III and V are major components of the fibers making up the lamina reticularis. Collagen I is not expressed with BMZs during fetal lung development. Collagen III expression is light and discontinuous in epithelial BMZ. 37 Although collagens I and III are not expressed during the early stages of fetal development, collagen type V is expressed at the early stages. 37 Collagen type V is associated with determining the diameter of collagen fibrils and its early appearance indicates an important role in fiber formation. The above observations are in agreement with findings of Mariani et al. 3s who showed that, in the mouse, -- 11,000 genes are expressed throughout the morphological stages of lung development. Of these genes, they focused on a subset of ECM genes associated with development of the BMZ (collagen types III and IV) and described their pattern of expression. They found that collagen IV was expressed at early stages of lung development and collagen Ill not until much later. Cells responsible for synthesizing BMZ collagens, proteoglycans, laminins and entactin/nidogen are the fibroblasts beneath the BMZ and the epithelial and smooth muscles on the other side of the BMZ. 39-41 It has been shown that F G F signaling is required for BMZ formation. 42 Also important in development of the BMZ is tissue transglutaminase. It covalently and irreversibly cross-links ECM proteins. It is hypothesized that tissue transglutaminase prevents or delays remodeling of the BMZ and stabilizes other extracellular components during development. 43

SUMMARY In summary, development of the BMZ involves continuous growth of the lamina densa as the tissue grows. The type IV collagen and laminin are thought to spontaneously polymerize into networks that are joined together by entactin/ nidogen. Proteoglycans are then added to this structure. This is an ongoing process and the cells involved are the adjacent epithelial or smooth muscle cells and the fibroblast layer on the other side of the BMZ. The lamina reticularis is formed postnatally, probably by the same cell types as the lamina densa; however, in contrast to the lamina densa, it does not stay at the same width, but increases in width. The reason for the variable width of the lamina reticularis is not known. Understanding this process and its significance may be very important in answering questions about diseases associated with thickening of the BMZ such as asthma. However, very little research has been done on the lamina reticularis. Other important questions to be answered concern (1) the significance of the various isoforms of collagen and

:

: ::

::

::::: :::

:::: ::::::: ::: ::::::::::::::

: :: :

:::

:::: :: ::: ::::::::

laminin in lung function, (2) the significance of proteoglycan sulfation in cellular control, (3) sequestering of growth factors in the BMZ, and (4) changes in BMZ composition during morphogenesis. Future research on the BMZs of the lung will be crucial in developing a better understanding of these processes.

ACKNOWLEDGEMENT Supported by NIH grants ES00628, ES04311, ES06700, ES05707, HL44497.

REFERENCES 1. Sannes PL, Wang J. Basement membranes and pulmonary development. Exp. Lung Res. 1997; 23:101-8. 2. Merker HJ. Morphology of the basement membrane. Microsc. Res. Tech. 1994; 28:95-124. 3. Aumailley M, Rousselle P. Laminins of the dermo-epidermal junction. Matrix Biol. 1999; 18:19-28. 4. Adachi EHI, Hayashi T. Basement-membrane stromal relationships: interactions between collagen fibrils and the lamina densa. Inter. Rev. Cytol. 1997; 173:73-156. 5. Erickson AC, Couchman JR. Still more complexity in mammalian basement membranes. J. Histochem. Cytochem. 2000; 48:1291-306. 6. Yurchenco PD, O'Rear JJ. Basal lamina assembly. Curt. Opin. Cell Biol. 1994; 6:674-81. 7. Nievers MG, Schaapveld RQ, Sonnenberg A. Biology and function ofhemidesmosomes. Matrix Biol. 1999; 18:5-17. 8. Iozzo RV. Matrix proteoglycans: from molecular design to cellular function. Annu. Rev. Biochem. 1998; 67:609-52. 9. Halfter W, Dong S, Schurer Bet al. Collagen XVIII is a basement membrane heparan sulfate proteoglycan. J. Biol. Chem. 1998; 273:25404-12. 10. Li D, Clark CC, Myers JC. Basement membrane zone type XV collagen is a disulfide-bonded chondroitin sulfate proteoglycan in human tissues and cultured cells. J. Biol. Chem. 2000; 275:22339-47. 11. Evans MJ, Burke AS, Cox RA et al. In situ preparation of rat tracheal basal cells. Tissue Cell 1993; 25:639-44. 12. Evans MJ, Van Winkle LS, Fanucchi MV et al. Three-dimensional organization of the lamina reticularis in the rat tracheal basement membrane zone. Am. J. Respir. Cell Mol. Biol. 2000; 22:393-7. 13. Kluge M, Mann K, Dziadek Met al. Characterization of a novel calcium-binding 90-kDa glycoprotein (BM-90) shared by basement membranes and serum. Eur. J. Biochem. 1990; 193:651-9. 14. Yang Y, Palmer KC, Relan N et al. Role of laminin polymerization at the epithelial-mesenchymal interface in bronchial myogenesis. Development 1998; 125:2621-9. 15. Bikfalvi A, Klein S, Pintucci G etal. Biological roles of fibroblast growth factor-2. Endocrinol. Rev. 1997; 18:26-45. 16. Dowd CJ, Cooney CL, Nugent MA. Heparan sulfate mediates bFGF transport through basement membrane by diffusion with rapid reversible binding. J. Biol. Chem. 1999; 274:5236-44. 17. Powell PP, Wang CC, Horinouchi H et al. Differential expression of fibroblast growth factor receptors 1 to 4 and ligand genes in late fetal and early postnatal rat lung. Am. J. Respir. Cell Mol. Biol. 1998; 19:563-72. 18. Sannes PL, Burch KK, Khosla J. Immunohistochemical localization of epidermal growth factor and acidic and basic

Development of the Basement Membrane

fibroblast growth factors in postnatal developing and adult rat lungs.Am.J. Respir. Cell Mol. Biol. 1992; 7:230-7. 19. Groffen AJ, Buskens CA, van Kuppevelt TH etal. Primary structure and high expression of human agrin in basement membranes of adult lung and kidney. Eur. J. Biochem. 1998; 254:123-8. 20. Groffen AJ, Ruegg MA, Dijkman H etal. Agrin is a major heparan sulfate proteoglycan in the human glomerular basement membrane.J. Histochem. Cytochem. 1998; 46:19-27. 21. Tomono Y, Naito I, Ando K. Epitope-defined monoclonal antibodies against multiplexin collagens demonstrate that type XV and XVIII collagens are expressed in specialized basement membranes. Cell Struct. Funct. 2002; 27:9-20. 22. Sannes PL, Burch KK, Khosla J e t al. Immunohistochemical localization of chondroitin sulfate, chondroitin sulfate proteoglycan, heparan sulfate proteoglycan, entactin, and laminin in basement membranes of postnatal developing and adult rat lungs. Am. J. Respir. Cell Mol. Biol. 1993; 8:245-51. 23. Shannon JM, Nielsen LD, Gebb SA et al. Mesenchyme specifies epithelial differentiation in reciprocal recombinants of embryonic lung and trachea. Dev. Dyn. 1998; 212:482-94. 24. Demayo F, Minoo P, Plopper CG etal. Mesenchymal-epithelial interactions in lung development and repair: are modeling and remodeling the same process? Am. J. Physiol. Lung Cell Mol. Physiol. 2002; 283 :L510-17. 25. Yurchenco PD, O'Rear JJ. Basement membrane assembly. Meth. Enzymol. 1994; 245:489-518. 26. Sanes JR, Engvall E, Butkowski R etal. Molecular heterogeneity of basal laminae: isoforms oflaminin and collagen IV at the neuromuscular junction and elsewhere. J. Cell Biol. 1990; 111:1685-99. 27. Sado Y, Kagawa M, Naito I et al. Organization and expression of basement membrane collagen IV genes and their roles in human disorders. J. Biochem. (Tokyo) 1998; 123:767-76. 28. Hudson BG, Reeders ST, Tryggvason K. Type IV collagen: structure, gene organization, and role in human diseases: molecular basis of Goodpasture and Alport syndromes and diffuse leiomyomatosis. J. Biol. Chem. 1993; 268:26033-6. 29. Miner JH, Patton BL, Lentz SI etal. The laminin alpha chains: expression, developmental transitions, and chromosomal locations of alphal-5, identification of heterotrimeric laminins 8-11, and cloning of a novel alpha3 isoform. J. Cell Biol. 1997; 137:685-701. 30. Pierce RA, Griffin GL, Mudd MS et al. Expression of laminin alpha3, alpha4, and alpha5 chains by alveolar epithelial

31. 32. 33. 34. 35.

36.

37. 38.

39. 40. 41. 42.

43.

cells and fibroblasts. Am. J. Respir. Cell Mol. Biol. 1998; 19:237-44. Pierce RA, Griffin GL, Miner JH et al. Expression patterns of laminin alphal and alpha5 in human lung during development.Am. J. Respir. Cell Mol. Biol. 2000; 23:742-7. Kohfeldt E, Sasaki T, Gohring Wet al. Nidogen-2: a new basement membrane protein with diverse binding properties. J. Mol. Biol. 1998; 282:99-109. Sannes PL, Khosla J, Peters BP. Biosynthesis of sulfated extracellular matrices by alveolar type II cells increases with time in culture.Am. J. Physiol. 1997; 273:L840-7. Khosla J, Correa MT, Sannes PL. Heterogeneity of sulfated microdomains within basement membranes of pulmonary airway epithelium. Am. J. Respir. Cell Mol. Biol. 1994; 10:462-9. Belknap JK, Weiser-Evans MC, Grieshaber SS etal. Relationship between perlecan and tropoelastin gene expression and cell replication in the developing rat pulmonary vasculature. Am. J. Respir. Cell Mol. Biol. 1999; 20:24-34. Evans MJ, Fanucchi MV, Van Winkle LS etal. Fibroblast growth factor-2 during postnatal development of the tracheal basement membrane zone. Am. J. Physiol. Lung Cell Mol. Physiol. 2002; 283:L1263-70. Wright C, Strauss S, Toole K etal. Composition of the pulmonary interstitium during normal development of the human fetus. Pediatr. Dev. Pathol. 1999; 2:424-31. Mariani TJ, Reed JJ, Shapiro SD. Expression profiling of the developing mouse lung: insights into the establishment of the extracellular matrix. Am. J. Respir. Cell Mol. Biol. 2002; 26:541-8. Evans MJ, Guha SC, Cox RA etal. Attenuated fibroblast sheath around the basement membrane zone in the trachea. Am. J. Respir. Cell Mol. Biol. 1993; 8:188-92. Evans MJ, Van Winkle LS, Fanucchi MV et al. The attenuated fibroblast sheath of the respiratory tract epithelial-mesenchymal trophic unit.Am. J. Respir. Cell Mol. Biol. 1999; 21:655-7. Holgate ST, Davies DE, Lackie PM etal. Epithelialmesenchymal interactions in the pathogenesis of asthma. J. Allergy Clin. Immunol. 2000; 105:193-204. Li X, Chen Y, Scheele S etal. Fibroblast growth factor signaling and basement membrane assembly are connected during epithelial morphogenesis of the embryoid body. J. Cell Biol. 2001; 153:811-22. Schittny JC, Paulsson M, Vallan C et al. Protein cross-linking mediated by tissue transglutaminase correlates with the maturation of extracellular matrices during lung development.Am. J. Respir. Cell Mol. Biol. 1997; 17:334-43,

Chapter

Development of the Pulmonary Vasculature

6

Rosemary Jones* Harvard Medical School and Department of Anesthesia and Critical Care, Massachusetts General Hospital, Boston, MA, USA

Lynne M. Reid Department of Pathology, Harvard Medical School, Children's Hospital, Boston, MA, USA

INTRODUCTION

The vascular units of the lung form in close coordination with airways and alveoli. While many of the genes initiating lung morphogenesis, the determination of left-right asymmetry and laterality, and the regulation of airway branching have been identified, 1-s less is known about genes regulating vascularization. Emerging data indicate a role for positive and negative factors, and that the formation of pulmonary vascular units may direct epithelial morphogenesis. 2-4 As our understanding of the role of 'instructive' and 'permissive' genes in the developing lung increases (genomics), so will our need to understand the modulation of cell phenotype by correctly and incorrectly assembled proteins (proteomics). Lung vascular growth is achieved by expansion of existing structures and by change in existing templates. As vascular networks increase in size and three-dimensional complexity, growth is accompanied by regression of unneeded units until these are appropriate to the stage of lung development. In this way, vascular systems formed in utero, in the postnatal period, and in early childhood, develop, enlarge and are remodeled; in late childhood and in the young adult, they continue to enlarge and remodel until thoracic growth i s complete. Lung vessels develop by vasculogenesis and by angiogenesis, the latter including sprout formation, splitting or intusussceptive (i.e. of-itself) microvascular growth, and simple expansion. The addition of mural cells provides wall support and as the developing units increase in size, vessels form. *To whom correspondence should be addressed. The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

The growth of the small vessels from capillaries eventually gives rise to all large vessels within the lung. These mechanisms combine to form the lung's pulmonary and bronchial arterial systems, and double venous systems. Little is yet understood of the way spatial and fractal (tree-like) dimensions of developing units are determined within the confines of mesenchymal or connective tissue spaces, although familial patterns of branching in the human lung indicate that, at least for central vessels, there is a genetic component. 6 The pulmonary arteries supply capillaries in the intraacinar region and pleura (except at the hilum) and drain to pulmonary veins. While arteries run centrally, the veins are distributed at the periphery of lung units, the most distal lying at the edge of acini. Venous tributaries arise from alveolar walls, alveolar ducts, bronchial walls, pleura and connective tissue sheaths, and drain to axial vessels that increase in size towards the hilum. The wall structure of vessels proximal to the acinus reflects their role as conduits of de-oxygenated or oxygenated blood, the structure of the small thin-walled intra-acinar vessels and capillaries of the alveolar-capillary membrane their role as gas-exchanging vessels. The bronchial arteries supply the airway mucosa and peri-hilar structures; they divide with the bronchi sending one submucosal and one peri-bronchial branch along each airway wall to form communicating arcades. 7-9 When extrapulmonary, they drain to pulmonary veins and to the right side of the heart and via the azygos system; when intrapulmonary they form a network that anastomoses with the capillaries of the pulmonary arterial bed and drain via pulmonary veins to the left side of the heart. They provide nutrients to large airway and vascular structures. Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

Between the hilum and the start of the acinus, the simple endothelial channels of the lymphatics lie within connective tissue sheaths. Liquid and plasma proteins move through the matrix to the lymphatic plexus surrounding terminal airways and drain centrally through the lung's lymphatic channels to the systemic circulation. 1~ To maintain adequate function, both at rest and on exercise, each of these interconnecting systems must develop appropriately. This chapter reviews current concepts of the formation of vascular beds in the normal lung at different stages of growth and maturation, comments briefly on aberrant growth that results in a still adequately functioning lung and summarizes the little known of the lung vasculature in aging. Finally, we highlight the presence of signaling systems that, triggered by the ambient oxygen tension, change vascular density in the adult lung.

CELLULAR BASIS OF VESSEL MORPHOGENESIS Currently, a major area of study in developmental biology and in disease, particularly as it relates to new anti-angiogenic therapy in tumor growth, concerns blood vessel formation and regression. Most data relating to intercellular signaling pathways and cell assembly into capillaries and vessels arise from these studies. 12-19The cell-cell interactions and molecular mechanisms of cell survival and loss they reveal, and ones of cell proliferation, migration, recruitment and differentiation, highlight processes likely involved in normal lung vessel growth. Vessel formation first requires endothelial cells to coalesce into a tube, for their contiguous membranes to fuse, and for junctional complexes to form. 2~ Networks of large and small vessels evolve as the initial vascular webs remodel into channels. Further wall formation requires signaling between endothelial cells and mural cell precursors. As the number of wall cells increases, the formation of elastic laminae is critical for their organization into intima (endothelial cells), media (smooth muscle cells, SMCs) and adventitia (fibroblasts) to establish a fully mature vessel wall. As new growth proceeds (by cell proliferation), vascular density is adjusted by release of endothelial and mural cell contacts, and, as patterns of blood flow change, by regression (by cell apoptosis or loss) of unperfused channels (vascular pruning). Homotypic and heterotypic contacts between neighboring cells and matrix are important in determining cell phenotype, and in regulating tissue growth and organization, by interaction with growth factors and vaso-mediators 21-2s (see also Chapter 7). Adhesion receptors at the endothelial cell surface (transmembrane glycoproteins that include members of the integrin, cadherin, immunoglobulin, selectin and proteoglycan superfamilies) interact with those of adjacent cells or with matrix proteins such as collagens, fibronectins, laminins and proteoglycans forming fibrils or other macromolecular arrays. 29-31 Cytoplasmic plaque proteins further link the cell membrane receptors to the cytoskeleton, transducing

signals from the cell surface and regulating receptor function. 29-31 Nuclear DNA and its associated protein-scaffold connect to matrix components through the cytoskeleton, contributing to communication between intracellular and extracellular environments. 32

FORMATION AND GROWTH OF E N D O T H E L I A L CHANNELS

Vasculogenesis In vasculogenesis, a primitive vascular network assembles from local aggregates of progenitor cells. 13'33'34Mesenchymal cells migrate and differentiate in situ into angioblasts (i.e. cells committed to an endothelial lineage) or hemangioblasts (blood cell precursors). They form sinusoidal nests of cells and spaces (blood islands). These evolve into capillary-like structures as the angioblasts form primitive vascular networks enclosing blood cells (Fig. 6.1a). The intercellular spaces forming the lumen of the first channels arise by loss of vesicles from the apical membranes of mesenchymal cells. By thinning their processes and reforming of their apical membrane, the mesenchymal cells become endothelial-like cells (Fig. 6.1a). As these channels give rise to branches, further growth is thought to occur by angiogenesis.

Angiogenesis (sprouting) In the most widely recognized form of angiogenesis, pre-existing capillaries and small vessels form capillary-like sprouts (Fig. 6.1b). Sprout formation starts with wall destabilization. Focal degradation of the endothelial basement membrane and surrounding matrix by proteolytic enzymes is followed by the extension of endothelial pseudopodia through the gap to form a spur. Endothelial cell migration in the direction of the growth spur forms a sprout. 35'36The delicate microspikes at the leading edge of migrating endothelial cells forming the tip of the sprout lack basement membrane. Rather than migrating singly, they move as a shifting sheet of cells. 35 Typically, proliferation of cells lying behind the growing tip increases the length of the sprout. 36 Sprouts can develop in the absence of endothelial cell proliferation (e.g. as in inflammation) but an increasing cell population is needed for sustained growth. 36'37While still connected at their origin to a patent vessel or capillary, sprouts continue to elongate and to branch or fuse to an adjacent sprout at their blind-end to form a new loop. Development of a slitlike lumen followed by the entry of plasma and blood cells completes the formation of a continuous channel.

Angiogenesis (intussusceptive microvascular growth/splitting) Elegant studies by Burri and co-workers 38'39 reveal the structural basis of the division of capillary units by intussusceptive microvascular growth. Areas of contact develop between opposing endothelial cell membranes to form inter-endothelial bridges (Fig. 6.1c). These reorganize into endothelial junctions as the central region of the endothelial

Fig. 6.1. (a) Vasculogenesis: mesenchymal cells (Mc, top) differentiate to angioblasts (Ab), i.e., cells committed to an endothelial lineage, and hemangioblasts (Hb), i.e., primitive blood cells. These organize to form blood islands (center). As angioblasts assume an endothelial cell (Ec) phenotype and continue to coalesce and organize into cellular channels, they form primitive capillaries enclosing red blood cells (RBC) (bottom). (Reproduced with permission from Sadler TW. Embryonic period (third to eighth week). In: Gardner JN (ed.), Langman's Medical Embryology, 6th edition, Baltimore, MD: Williams and Wilkins. 1990, pp. 61-84.)(b) Angiogenesis (1)- sprout formation: endothelial cells and their processes, released from the constraint of a basement membrane, and in response to a signal triggering growth, invade the surrounding tissue in the direction of the growth spurt (top) to form cell-lined channels (sprouts). These remain connected to the parent channel at their origin. Sprout formation is followed or accompanied by the development of a lumen and the connection of adjacent sprouts to establish a contiguous network (bottom). (Reproduced with permission from Schoefl GI. Electron microscopic observations on the regeneration of blood vessels after injury. Ann. NY Acad. Sci. 1964; 116:789-802.) (c) Angiogenesis (2)- intussusceptive microvascular growth: diagram illustrating structural events, based on morphological data obtained from serial sections and electron microscopy. Opposing endothelial cells form a capillary inter-endothelial bridge and the area of contact is sealed and organized into interendothelial junctions (top left and right). The endothelium then thins centrally (bottom left, arrows) and gives way (bottom right, open arrows) to invading components of the interstitium (bottom right, arrow) which divide the capillary channel. (Reproduced with permission from Burri PH, Tarek MR. A novel mechanism of capillary growth in the rat pulmonary microcirculation. Anat. Rec. 1990; 228:35-45.) (d) Angiogenesis (3) - expansion: growth of an existing vascular bed occurs, by the addition of wall cells and increase in lumen diameter and segment length. (Reproduced with permission from Ref.35)

layer is perforated by an invading connective tissue post (1-2.5 ~tm in diameter). The post is subsequently stabilized by the inclusion of myofibroblast and pericyte processes and by the formation of collagen fibrils. 39 With growth, posts evolve into pillars (--2.5 l.tm in diameter) which result in the formation of a mature mesh.

Angiogenesis (expansion) Slow expansion of a formed vascular network by increase in the diameter or length of existing units is achieved by endothelial and mural cell proliferation in the absence, or in excess, of cell loss (Fig. 6.1d). 35

DEVELOPMENT OF M U R A L CELLS Origin and role of mural cells Mesenchymal cells are recruited to an endothelial channel as peri-endothelial cells (pericytes and SMCs), stabilizing the new structures, and with adventitial fibroblasts, regulating its response to vaso-mediators. 15-19'4~ In capillary and microvessel walls, particularly in post-capillary venules, peri-endothelial cells (Fig. 6.2) develop into pericytes. 43-45 In arterioles adjacent to capillaries, they develop into intermediate cells with a phenotype midway between a pericyte and SMC. 44-46 Pericytes and endothelial cells have different l i n e a g e s - pericytes develop from cells derived from the mesoectoderm, endothelial cells from the mesodermal lateral plate. 47 In vessels, where transmural pressure is higher than in capillaries, the peri-endothelial cells express a SMC phenotype. Emerging data on the origin of these cells 48-5~ indicate that embryonic endothelial cells can become mesenchymal cells that express smooth muscle proteins and so are also a possible source of these cells. 51 Adventitial fibroblasts align circumferentially to form the outermost layer of vessel walls in which SMCs develop. Their number and that of the SMCs determine oxygen diffusion, which is restricted once the perivascular tissue cuff is 100 ~tm thick. Evidence that interstitial fibroblasts are recruited as mural cells in mature vascular beds, where they differentiate into pericytes and SMCs, comes from studies of dorsal and mesenteric capillaries 44'52'53 and vascular remodeling in lung. 54-58

Fig. 6.2. Development of perivascular cells around capillary sprouts (based on intravital video recordings and electron micrographs of the mesenteric microcirculation of young rats). Small endothelial extensions or buds (1) penetrate the basal lamina (fine stippling) and evolve into cellular protrusions (2), develop into endothelial spurs and short sprouts (3) and gradually lengthen into full sprouts (4). Fibroblasts settle down on the adluminal walls of sprouts and transform into pericytes by sharing the endothelial cell basement membrane. Loss of plasma, blood cells and platelets into the interstitial tissue during sprout formation is prevented by pericytes temporarily assuming an umbrella shape. Endothelial cells can contribute to the length of forming sprouts (regardless of proliferation) by organelle streaming and re-organization of their cytoplasm into long extensions. (Reproduced with permission from Rhodin JAG, Fujita H. Capillary growth in the mesentery of normal young rats: intravital video and electron microscope analyses. J. Submicrosc. Cytol. Pathol. 1989; 21:1-34.)

Smooth muscle cells and elastic laminae Pericytes Unlike SMCs, pericytes share the endothelial basement membrane and lack extensive filament networks, dense bodies or attachment plaques. 44'45'53'59Their processes often extend between membrane leaflets to contact the endothelial cell, and rare gap junctions allow nucleotides to pass between the cells.53'59 During sprout formation, pericytes prevent plasma from escaping into interstitial tissue (Fig. 6.2). Possessing cyclic GMP-dependent kinase, actin, desmin, vimentin, u-tropomyosin and myosin, pericytes contract or relax in response to vaso-mediators. 6~Their ability to express proteins typical of contractile smooth muscle such as smooth muscle myosin heavy chain (SM-MHC), u-smooth muscle actin (u-SMA) or desmin (see below) varies greatly within a vascular bed. 61-63

The network of contractile and cytoskeletal filaments occupying the cytoplasm of differentiated SMCs (Fig. 6.3a,b) confers tensile strength and the ability to contract. 64'65 In larger vessels small bundles of collagen and collagen fibrils are present between cells. In all but the smallest venules, SMCs are surrounded by basement membrane and characterized by extensive filaments, fusiform dense bodies and attachment plaques (Fig. 6.3). To proliferate or migrate, they readily de-differentiate from a 'contractile' to a 'synthetic' phenotype by disassembling this network, and can revert to a contractile phenotype by its re-assembly. The proteins required for this (see Fig. 6.3) appear in sequence in the cells of developing vessels. 41'66-73 Thus u-SMA expression is followed by expression of calponin, h-caldesmon, u-tropomyosin and

Fig. 6.3. (a) Arrangement of the SMC contractile and cytoskeletal filament lattice (top) and organization of its structural components (bottom). Oblique, face-polar, smooth muscle (SM)-myosin filaments (14-16 nm diameter) cross-bridge to c~-SM-actin filaments (4-6 nm) and anchor to the cytoskeleton at dense bodies - ovoid structures consisting of ~-actinin and [3-actin - to form the contractile apparatus. Whether these attach to the cell membrane is not established. Longitudinal intermediate filaments (7-11 nm) composed of desmin (a SMC specific protein) or vimentin, and a cytoplasmic domain of [3-actin and filamin (an actin cross-linking protein), form the cell cytoskeleton. These filaments also anchor to the contractile apparatus at dense bodies - linking it to the cell's supporting structure to give the cell tensile strength; they also link the contractile apparatus to the plasmalemmal membrane and to elastic components of the extracellular matrix via peripherally located attachment plaques, i.e., submembranous structures (0.2-0.5 nm) containing o~-actinin, filamin, metavinculin or vinculin, which anchor at the cell membrane via proteins such as plectin. Other intermediate filaments traverse the network to provide further support by anchoring to the dense bodies and attachment plaques. Attachment plaques are separated by membrane regions rich in caveolae and characterized by transmembrane receptors or integrins that link components of the cytoskeleton to the extracellular matrix. Dense bodies and attachment plaques are considered the functional equivalent of Z-bands in striated muscle. (Adapted from Small JV, North AJ. Architecture of the smooth muscle cell. In: Schwartz SM, Mecham RP (eds), The Vascular Smooth Muscle Cell. Molecular and Biological Responses to the Extracellular Matrix. San Diego, CA: Academic Press. 1995, pp. 169-88 with permission.) (b) Example of o~-SMA immunoreactive sites identified by 10 nm gold particles decorating the filaments of a SMC developing in the wall of an alveolar vessel (31 lLtm in diameter). Unicryl section (80 nm) of rat lung after 28 days at FiO 2 0.87, using the protein A-gold technique with a monoclonal antibody to c~-SMA (1:400 dilution, Sigma) followed by uranyl acetate and lead citrate staining. The cell lies surrounded by matrix with the lumen and endothelium to the left. Filaments typically develop along the adluminal cell margin. The gold particles and filaments (at arrowhead) are shown at higher magnification in the inset. Bars = 1 and 0.1 ~tm.

metavinculin, while the SM-MHC SM1 isoform appears in immature cells during the fetal period and the SM-MHC SM2 isoform, metavinculin and ~-tropomyosin appear in differentiated cells after birth. Molecular screening techniques are beginning to identify developmentally regulated genes expressed by immature cells. 74 The number of SMCs is normally proportional to blood flow and transmural pressure. 75 The exception is in small 'resistance' vessels where wall thickness is normally high for lumen size. Their cell processes penetrate their surrounding basal lamina to

form lateral contacts and penetrate the basal lamina of adjacent endothelial cells in junctional complexes (myoendothelial junctions). 44 As vessels approach capillaries, the number of SMCs decreases, endothelial contacts increase, and it is endothelial cell processes that penetrate the basal lamina to contact SMCs. 76 Differentiated SMCs become separated from adjacent endothelial cells and from fibroblasts by internal and external laminae which form along the outer and inner margins of the SMC layer to establish the vessel media. In large vessels

additional laminae further divide the cells into multiple layers; close to capillaries, where SMCs are absent from the wall, a single lamina separates endothelium from surrounding connective tissue. The regulation of smooth muscle growth by elastin is demonstrated by obstructive intimal hyperplasia and death of mice lacking the elastin gene. 77 Each vascular cell type is elastogenic (endothelial cell, SMC and fibroblast) but their relative contributions to lamina formation is unknown. Current understanding of the process of lamina assembly derives from data for other tissue sites. 78-9~ The presence of endostatin (an inhibitor of endothelial cell proliferation) within matrix and elastic laminae of large vessels91 could restrict sprouting from the wall.

Fibroblasts Fibroblasts typically express vimentin, actin isoforms and non-muscle myosin. Under certain conditions, they express two SMC-specific proteins, ot-SMA and desmin, 92 but this is uncommon in normal vessels. Their processes are characterized by microfilaments (4--6nm) and intermediate filaments, pinocytotic vesicles and extensive rough endoplasmic reticulum:

basement membrane is absent, and collagen fibers and fibrils form along the adluminal and abluminal cell margins.

CELL-CELL INTERACTIONS AND SIGNALING

Vessel formation and wall and network remodeling are regulated by paracrine signals between the receptor tyrosine kinases (RTKs) of endothelial and peri-endothelial ceils and growth factors (ligands) (Fig. 6.4). 14-19 Close apposition to endothelial basement membrane triggers cell differentiation, likely in response to endothelial expression of TGF-13, and inhibits endothelial cell movement and proliferation. 17 The ligand-receptor signals required for vascular growth and regression (molecules that appear to modulate rather than induce the interactions required for wall formation), are still being actively analyzed. Ligands of interest include vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), fibroblast growth factor (FGF) and the angiopoietins (Angl and Ang2). Their effect is fine-tuned

Fig. 6.4. A model for regulation of the vascular endothelium as demonstrated by the prototypical angiogenesis factor VEGF and the class of angiogenic regulators, Angl and Ang2. All three ligands bind to RTKs that have similar cytoplasmic signaling domains. Yet their downstream signals elicit distinctive cellular responses. Only VEGF binding to the VEGF-R2 (Flkl) sends a classic proliferative signal. When first activated in embryogenesis, this interaction induces the birth and proliferation of endothelial cells. In contrast, VEGF binding to VEGF-R1 (Fit1) elicits endothelial cell-cell interactions and capillary tube formation, a process that closely follows proliferation and migration of endothelial cells. Angl binding to the receptor tunica interna endothelial cell kinase-2 (Tie2) RTK recruits and likely maintains association of peri-endothelial support cells (pericytes, smooth muscle cells, myocardiocytes), thus solidifying and stabilizing a newly formed blood vessel. Ang2 although highly homologous to Angl, does not activate the Tie2 RTK; rather, it binds and blocks kinase activation in endothelial cells. The Ang2 negative signal causes vessel structures to become loosened, reducing endothelial cell contacts with matrix and disassociating peri-endothelial support cells. This loosening appears to render the endothelial cells more accessible and responsive to the angiogenic inducer VEGF (and likely to other inducers). (Reproduced with permission from Hanahan D. Signaling vascular morphogenesis and maintenance. Science 1997; 277:48-50.)

by the number and sub-type of receptors expressed by the target cell population and by availability of ligand from producer cells. Most data derive from gene inactivation studies; as most deletions are embryonic lethal, however, less is known of their role in vessel formation beyond birth. Other molecules, such as extravasated plasma proteins and inflammatory mediators such as prostaglandins, tumor necrosis factor-m, interleukins (ILs) and nitric oxide (NO), also induce angiogenesis in vir 18'19'93'94 but their role in vascular growth and remodeling in the lung is largely unknown.

VEGF to prevent endothelial cells from detaching and undergoing apoptosis, but during postnatal life endothelial cells become VEGF-independent. How endothelial cells commit to arterial or venous channels within a forming vascular bed is an intriguing question. It appears that their location within a network is determined by expression of the ligand Eprin B2 and of its receptor eph 4 (Fig. 6.5). Arterio-venous relationships are likely established and maintained by the expression of these molecules. 1~176

Endothelial channel formation- VEGF/VEGFR

and Ang/Tie

Growth factors such as the FGF family first induce angioblast and hemangioblast formation in the mesoderm. VEGF is then needed to maintain angioblast differentiation and survival. Alternative splicing produces homodimeric isoforms VEGF121, VEGF165, VEGF189 and VEGF206. Of these VEGF121 and VEGF165 act as survival factors for endothelial ceils of immature vessels. VEGF signals controlling angiogenesis come from adjacent cells, stimulating endothelial cell RTKs that include VEGF R-1 (Flt-1), VEGFR-2 (Flk-1/KDR) and VEGFR-3 (Flt-4). Assembly of angioblasts into vascular channels requires their expression of VEGFR-195-99 and after they differentiate to endothelial ceils they express VEGFR-2. VEGF binding to VEGFR-2 results in proliferation while its binding to VEGFR-1 results in tubule formation and sprouting (see Fig. 6.4). Endothelial cells develop in VEGF null mice but the vessels are malformed. VEGFR-1 null mice show similar changes while VEGFR-2 null mice fail to develop endothelial cells, possibly because hemangioblasts fail to differentiate. VEGF also stabilizes developing vessel walls by accelerating the development of peri-endothelial cells. 1~176176 Immature vessels that lack peri-endothelial ceils need

PDGF isoforms play important roles in both lung vascular and alveolar development. 1~176 While four genes are identified (A, B, C and D), most is known of PDGF-A and PDGF-B in relation to vessel formation. Dimers of two homologous polypeptide PDGF chains, a secreted A-chain and cell-associated B-chain (the c-sis homologue of the v-sis oncogene), dimerize via disulfide bonds to form functional in vivo isoforms (PDGF-AA, -AB or -BB), which in turn dimerize RTKs, PDGFR-cx and PDGFR-[3, on the cell surface. The striking absence of peri-endothelial cells, specifically pericytes, in PDGF-B gene deficient mice results in capillary dilatation, microaneurysms, and vascular leak and hemorrhage; the absence of the PDGFR-]3 produces a similar response. Absence of the PDGF-A gene, on the other hand, results in an inability to form alveolar structures, in part, because SMC progenitors fail to spread. 1~ Both isoforms promote mesenchymal cell proliferation, PDGF-B initiating progression through the cell cycle and PDGF-A (a competence factor) requiring a further signal (e.g. IL-1) to induce mitogenesis. 113 PDGF-B alone induces chemotaxis. 1~ It is suggested that SMCs use PDGF-B to enter the cell cycle,

Mural cell d e v e l o p m e n t - PDGF/PDGFR

Fig. 6.5. Presumed distribution of ephrin-B2 and Eph-B4 - which are defined as a ligand/receptor pair that identify arterial and venous endothelial cells, respectively, in a capillary plexus. 1~ These are presumed to interact between opposing arterial and venous endothelial cells in a "cis" manner (left). During remodeling of the primary plexus (right) by interdigitation, branching and differential growth of vascular segments, they remain localized to arterial and venous units ("cis" interactions) but may also interact at the interdigitating surfaces of large vessels ("trans" interactions). (Reproduced with permission from Yancopoulos GD, Klagsbrun M, Folkman J. Vasculogenesis, angiogenesis and growth factors: ephrins enter the fray at the border. Cell 1998; 93:661-4.)

and/or to suppress differentiation, and to stimulate selfreplication via synthesis of PDGF-A. 11s'116 Actin re-organization and membrane ruffling (essential for cell migration) are induced by PDGF-AB and PDGF-B, and therefore via the PDGF-[3 receptor. PDGF-B also enhances wall stabilization by inducing VEGF expression in peri-endothelial cells. 117 Tiel and Tie2 (Tek) form a second family of RTKs expressed by endothelial cells. Angl, the major physiological ligand for Tie2, is expressed by mesenchymal cells (see Fig. 6.4). As yet no ligand is identified for Tiel, which modulates transcapillary fluid exchange, and Ang2 is a negative Tie2 ligand (see Fig. 6.4). 15-17'118-123 Tie2 expression further modulates VEGF activity and is required for sprout formation. 119'120'124 While Tie2 null mice have normal numbers of endothelial cells, these assemble into immature channels that lack branching networks and the presence of large and small vessels. Angl-deficient mice die with similar vascular defects to Tie2-deficient animals. Angl together with VEGF enhances vascular density, whereas Ang2 and VEGF produce longer sprouts indicating a role for Ang2 in vessel formation in addition to one in vessel regression. TM

EMBRYONIC DEVELOPMENT

AND

FETAL

VASCULAR

Overview" In the human embryo, the formation of vascular networks starts with the onset of organogenesis in the fourth week of gestation. 34 As the lung anlage develops from the foregut, it is vascularized by ingrowth of a vascular plexus derived from the heart. The main pulmonary artery develops from migrating angioblasts. By the sixth week of gestation, the adult pattern of central vascular and airway structures consisting of lobar and segmental branches is present. The pre-acinar vessels develop hand-in-hand with airways while the intra-acinar vessels develop later within distal alveolar structures. 125-135 The precise timing of the

three recognized stages of fetal lung morphogenesis 42'136 the pseudoglandular, the canalicular and the saccular varies with species, as does the degree of distal lung maturation achieved at birth. 137 The main features of vascular beds forming in normal lung are pre-acinar and intra-acinar vessel number, vessel size and wall structure. Pre-acinar vessels run with a bronchus (Br), bronchiolus (B) or terminal bronchiolus (TB), intra-acinar ones with a respiratory bronchiolus (RB) or alveolar duct (AD), or are found within the alveolar wall (AW). The branching pattern and the size of large vessels can be assessed on angiograms lzs and additional details of development obtained from serial reconstruction of vessels in tissue. Where numerous elastic laminae envelop the SMCs, as in the main pulmonary artery and its large branches, these are 'elastic' arteries. More peripherally, as the number of laminae decreases, the arteries are 'transitional'. The arteries become 'muscular' with a wall consisting simply of SMCs between internal and external laminae. Eventually, the muscle layer thins to a few cells and forms a spiral in 'partially muscular' arteries. The wall then consists of endothelium and a single lamina as the muscle layer disappears in 'non-muscular' arteries. By the 20th week, the full number of pre-acinar pulmonary vessels is present in each segment. During fetal life there is an increase in vessel size, length and diameter but no change in the density. 12s-13~ The vessels at the hilum grow faster than at the periphery so that the gradient of diameter against length increases with fetal age.

Early vessel development Overview: Studies of early lung development in both the human and mouse reveal details of the cellular activity in vessel formation. 5~ A branching network evolves by angiogenesis from the central arterial and venous trunks and as these expand in diameter and length offshoots grow by irregular dichotomous branching (Fig. 6.6a-c). Distal

Fig. 6.6. Pulmonary arterial Mercox casts; peripheral vessels steadily increase in density in the lung within the 72-h period illustrated (a-c). (a) 12 day mouse fetus. S=systemic; R =right; L =left; PA= pulmonary artery. (b) 14-day mouse fetus; peripheral vessels steadily increase in density within the 72-h period illustrated. R=right; L=left; PA=pulmonary artery; PV=pulmonary vein; CL=cardiac lobe. (c) 15-day mouse fetus. R = right; L = left; PA = pulmonary artery; PV = pulmonary vein. (Reproduced with permission from deMello DE, Sawyer D, Galvin N etal. Early fetal development of lung vasculature. Am. J. Respir. Cell Mol. Biol. 1997; 16:568-81 .)

vessel and capillary networks develop by vasculogenesis (Fig. 6.7a,b) and then fuse with the proximal networks (see below); 139 they have also been reported to develop by sprouting from the main vascular plexus. 5~ While mesenchymal cells represent an important source of peri-endothelial precursors, 41'1~176 these cells are also reported to develop from bronchial SMCs and endothelial cells. 5~ Despite their different origin, these cells express the same smooth muscle proteins. 5~ Serial sections of human embryos show that blood lakes form first, being present in the primitive mesenchyme surrounding the lung bud at the neck (at 32-44 days). 14~As the first (5-6) airway branches form at --50.5 days, such lakes are abundant in the subpleural mesenchyme. At this stage, the pulmonary artery accompanies airways only to the third or fourth generation. The first connection between developing distal and proximal networks appears between

peripheral lakes and a thin-walled hilar vein, venous drainage thus being established before the pulmonary artery supply. At ---56.5 days, the branching of the pulmonary artery, a thick-walled blind-ended tube, lags behind airway branching by 2-3 generations. At 12-14 weeks, an extensive capillary network surrounds distal airway buds although well separated from them by the subpleural mesenchyme. By 22-23 weeks, the capillary network closely approaches the alveolar epithelium and the pulmonary artery now accompanies each airway branch. Conventional and supernumerary arteries and veins In the human fetal lung, as in the child and adult, the number of arterial and venous branches exceeds that of the airways. 12s-13~ Branches of the main pulmonary artery running alongside airways are termed 'conventional' arteries: numerous additional branches arising from the axial artery

Fig. 6.7. (a) Electron photomicrographs of a 9-day fetal mouse thorax showing (top) densely packed mesenchymal cells containing cytoplasmic vesicles and (bottom) intercellular spaces appearing between mesenchymal cells surrounding the developing lung bud. Intercellular spaces appear to result from the discharge of intra-cytoplasmic vesicles leaving a ruptured cell membrane while mesenchymal cells around intercellular spaces show membrane continuity. Bars= 10 pm (top) and 5 pm (bottom). (Reproduced with permission from deMello DE, Sawyer D, Galvin Net al. Early fetal development of lung vasculature. Am. J. Respir. Cell Mol. Biol. 1997; 16:568-81 .) (b) Electron photomicrographs of a 10-day mouse fetus showing (top) thinned mesenchymal cells which appear endothelial-like hematopoietic precursor cells are present in some spaces (bottom). Bars= 5 pro. (Reproduced with permission from Ref.139)

to the TB, i.e. pre-acinar and passing directly to adjacent respiratory tissue to supply the capillary bed are termed 'supernumerary' arteries (Fig. 6.8). For example, in the posterior basal segment of the left lung of a 19-week fetus 128 (length 16 mm), the 25 bronchial branches present to the level of the TB are accompanied by 21 conventional arteries and 58 supernumerary arteries. The number of airways and conventional artery branches is similar from 18 weeks gestation onwards. In the fetus the veins arise from the saccular respiratory region, pleura, connective tissue septae and airway walls. Conventional and supernumerary veins appear together, developing progressively from hilum to periphery. The axial veins running from the periphery to the hilum receive drainage from both conventional and supernumerary tributaries. From 20 weeks of gestation, the number of pre-acinar conventional veins is within the adult range for intrasegmental airway number. While the number of conventional veins in a segment equals that of the airways and conventional arteries, there are more supernumerary veins than supernumerary arteries. Many small arteries and veins develop in late fetal life and continue to increase with age; their density per unit area of lung increases as respiratory bronchioles and saccules appear.

Vessel wall structure Depending on gestational age and their proximal to distal location, the wall structure of fetal pulmonary arteries is either elastic, transitional, muscular, partially muscular or non-muscular. 128 ' 129 From 19 weeks of gestation, an elastic structure is present to the same level in axial and conventional arteries, and in smaller vessels than in the adult. Because of high tensile strength and the ability to maintain vessel

Fig. 6.8. Diagram of a broncho-arterial bundle illustrating conventional versus supernumerary artery branches. Conventional arteries arise at acute angles to the main axis and supply the respiratory region at the end of the axial pathway. Supernumerary arteries are short and have varying diameters. They arise at right angles to the main axis and supply adjacent air spaces (shaded areas), ultimately providing collateral circulation to a respiratory unit via the backdoor.

patency, these arteries can be regarded as supportive. Phenotypically distinct populations of SMCs in the wall of large developing pulmonary arteries raise the possibility of different lineages. TM In small vessels, when present, the SMCs are immature. The pulmonary veins are relatively free of muscle in the fetus. At 20 weeks gestation, for example, no muscle is present even in the largest veins. By 28 weeks, muscle fibers are seen within vein walls but a continuous muscle layer is present only at term. 129'138 As in arteries, the SMCs extend into veins as small as 80 ktm but unlike arteries the veins are thin-walled and no elastic laminae are present. 129'138 The walls of bronchial arteries and bronchial veins are typical of systemic vessels, being muscular and relatively thicker (for vessel size) than pulmonary arteries. 138

Distribution of intra-acinar vessels A population count of small vessels identified by size and wall structure offers a useful way to establish that the growth pattern is normal. 128 For example, in the normal human lung at term, all arteries above 150ktin in diameter are muscular: below this, they are partially muscular, the smallest being --75ktm. The largest non-muscular arteries are --90~tm in diameter, and below 60ktm all are non-muscular. This relationship between vessel size and wall structure is similar at all fetal ages. 128 While the size of the smallest muscular artery, and the size range for partially muscular and nonmuscular arteries, is the same in the fetal and adult lung, the wall thickness of an artery (of given diameter) is twice as thick in the fetus as in the adult, and muscular arteries fail to extend into the acinus. Even at birth little muscle is found in the walls of arteries beyond the terminal bronchiole (Fig. 6.9).

Fig. 6.9. The development of muscle in the wall of human intra-acinar arteries. In the fetus, no muscle is seen within the acinus. It gradually extends with age but reaches vessels in walls of alveoli only in the adult (19 years). (Reproduced with permission from Hislop A, Reid L. Pulmonary arterial development during childhood: branching pattern and structure. Thorax 1973; 28:129-35.)

Development of bronchial arteries and veins As cartilage plates form, bronchial arteries grow along the airway wall to supply the walls of vessels of the bronchopulmonary sheath. 145'146Bronchial arteries and veins gradually increase in size and number and at term are found as far into lung as bronchioli. The term 'pre-capillary anastomosis' is used to describe a channel that is pre-capillary in position and larger than a capillary in size. A small number of these connections between bronchial arteries and pulmonary arteries (35-100 ktm in diameter) are present at all fetal ages. In the late fetal stage, pre-capillary anastomoses between pulmonary and bronchial arteries are found within the bronchial wall but none are larger than 15 ktm.

POSTNATAL VASCULAR DEVELOPMENT A N D G R O W T H Overview: At 36 weeks to term of human gestation, the preacinar pattern of arteries and veins is complete. The lung now can support air breathing but structurally it is not the adult lung in miniature. It responds by a burst of vascular growth as existing units continue to expand in diameter and length, and many new intra-acinar units are added with formation of the gas-exchange surface. 42'129'130'131'136'138 In the first 4 months, as alveoli form and increase in size, the number of arteries per unit area of lung and density of capillary networks increase. 39 These vessels are thought to form and grow by angiogenesis, and the complexity of capillary networks to increase by intussusceptive microvascular growth. 39 At birth the respiratory saccules (primary septae) are supplied by small vessels and a double capillary system. Within 2 weeks, as secondary septae form and enclose the interstitial tissue between the two capillary layers, these fuse into a single network. In the first months of postnatal life, the diameter of proximal intra-acinar vessels increases more than distal ones reflecting the burst of small vessels developing at the lung periphery. After 18 months, the number of new vessels forming slows along with alveolar growth. 13~ Between 4 months and 4 years, the number of arteries (up to ---200 ~tm) per unit area of lung increases greatly. While the ratio of intra-acinar arteries to alveoli remains similar throughout childhood, the concentration of arteries per unit area of lung falls after 5 years of age as alveoli increase in size. 13~

Conventional and supernumerary arteries and veins The veins grow at the same time as airways and arteries. While the pre-acinar drainage pattern is complete half-way through fetal life, the intra-acinar pattern develops during childhood. Both conventional and supernumerary vessels continue to develop in the postnatal lung, conventional arteries accompanying new airways appearing up to 18 months and new supernumerary arteries to 8 years of age. 13~ Conventional veins, like the axial veins, run in their own connective tissue sheath. They enter the axial vein at an

acute angle, are of similar size, and lie at some distance from the capillary bed they drain. 129 Supernumerary veins drain the lung tissue immediately around the axial vein. Some have no collagen sheath but pass directly through the main vein sheath to the axial vein: others receive post-capillary tributaries and are surrounded by a collagenous sheath continuous with the sheath of the axial vein. 129 Along the axial pathway, they are equivalent in number to airway generations and the arteries accompanying them. Each type may be found along the length of the axial vein. Both conventional and supernumerary veins become more frequent towards the periphery.

Vessel wall structure During childhood, the number of large arteries with an elastic or transitional wall remains constant. Between 4 and 10 months, vessels increase in size but muscle development lags. By 10-11 years, muscle extends further distally and is present in alveolar duct vessels but not alveolar wall ones (see Fig. 6.9). The presence of a high population of thin-walled arteries within the acinus may provide children with an advantage since arteries as large as 200 ~tm in diameter that have virtually a capillary wall can contribute to oxygen transfer. The wall structure of veins is more developed in children; post-capillary veins consist only of endothelium but larger vessels have internal and external laminae and occasional SMCs. In larger veins there is a continuous muscle coat but even in the largest still no definite elastic lamina is present. 129

Distribution of intra-acinar vessels Based on the changes described above, the distribution of intra-acinar vessels shifts significantly in childhood with more partially muscular and non-muscular arteries than muscular ones present in the fetal or adult lung.

Bronchial artery to pulmonary artery connections By 10 weeks after birth pre-capillary anastomoses (formed in the fetal period) are obliterated by fibrin and muscle; pulmonary and bronchial arteries then communicate with the pulmonary veins only through their capillary bed. Pre-capillary vessels are present between the pulmonary and bronchial arteries but normally these are not functional; they have the potential, however, to open if a block occurs. In congenital heart defects, for example, these systems adapt to the altered hemodynamic state. The persistence of pre-capillary fetal anastomoses as well as selective opening of capillaries could both contribute to communication between the large vessels of different venous and arterial systems in the newborn and young lung.

VASCULAR GROWTH AND R E O R G A N I Z A T I O N IN THE A D U L T Overview" Angiograms demonstrate the branching pattern of central and peripheral pulmonary arteries and veins in

the adult lung, and density of peripheral vessels. 135 The branching pattern can be defined by counting the number of branches either as generations or by order. 143'147'148 A useful convention is to consider a segmental airway as the first g e n e r a t i o n - the trachea and lobar branches are counted separately. For example, the inferior lingular segment commonly has about 28 generations, the apical lower lobes may have as few as 15. Vessel branching parallels that of the airway in the above features. While endothelial cell turnover is extremely low in vessels of the adult lung, new vessels continue to form from pre-existing ones by angiogenesis in line with lung growth. Circulating endothelial and bone marrow derived 'stem cells' have been identified in the adult where they are thought to contribute to vessel formation. 18 The lobule represents a group of 3-5 acini clumped at the end of an airway, be it subpleural or deep in the lung. The acinus of the adult human lung is --1 cm in diameter or 1 ml in volume, each acinus consisting of many alveoli. The extensive alveolar surface of the adult lung (-~70m 2) is composed of---300 x 106 alveoli each with many small vessel and capillary segments. The density of the capillary bed increases 4-fold between birth and adulthood, when the capillary endothelial surface area is --- 126 m2. 39'149-151

Conventional and supernumerary arteries and veins The supernumerary arteries of the adult lung remain more numerous than conventional ones, both absolutely and relatively. They have a frequency ratio of 2.5-3.4:1 over the length of the pre-acinar segment and form 20-45% of the total cross-sectional area of side branches. In cattle, it has been shown that at the origin of each supernumerary artery and the parent conventional artery the wall is organized into a V-shaped musculo-elastic cushion. Beginning as a funnelshaped channel on the hilum side of each supernumerary artery it forms into a baffle that projects over the lumen and may regulate blood flow. 152 Supernumerary arteries also appear to have their own vaso-regulatory pathways; their response to vasoconstrictors is greater than for conventional arteries, and the response to NO is different. 153 Vascular wall structure The wall of the main pulmonary artery has more than 7 elastic laminae in elastic segments and 4-7 laminae in transitional o n e s . 143 Elastic arteries are >3200~tm diam., transitional ones are 3200-2000 ~m, and muscular ones are , "rm

E w e no. 1 7 0 2 9 9

40

9

140 days

9

11

m

o

250

8 30 A

9

9

9

9

9

.s ~

t q. 9176

vE

20

E

200 ee9 eeo o. eeee

o "o "5 ._o"

10

I-

0

I

I

I

I

I

I

I

1

2

3

4

5

6

7

Hours after birth Fig. 8.2. Filling of human lungs with air after birth. The linear regression of lung volume against time is statistically significant. However, the line fitted to the data is the best least-squares exponential approach to an asymptote, with an initial increase in volume of 75 ml/h. From Milner et al. 92

How long it takes for these epithelial transport processes to remove the bulk of the lung liquid is not precisely known. However, as shown in Fig. 8.2, the lungs of human newborns fill with air at an initial rate of--75 ml/h. 92 In contrast, liquid removal from the lungs of newly born sheep is at --25 ml/h for a total lung volume of---250 ml. 93 Thus, there are clearly species differences in the rate of clearance of liquid after birth, but in general it seems that it may take several hours for complete removal of lung liquid. All studies have shown that liquid removal at birth is inhibited by amiloride. 94-96 In newborn rat lung, blockers of Na/H exchange or Na-glucose cotransport were without effect on liquid removal. 95 In fetal sheep, the removal of lung liquid immediately after birth is associated with a surge in plasma epinephrine content 93 (Fig. 8.3). Further, there is a progressive increase in sensitivity to epinephrine between 122 and 142 days of gestation. Thus, at 122 days, infusion of epinephrine merely caused a slight inhibition of liquid secretion. At 133 days, epinephrine induced a small level of absorption with the critical concentration for epinephrine being ---0.5 ng/ml. 93 Close to term, at 142 days, the magnitude of the switch from secretion to absorption was considerably greater than at 133 days, and the critical concentration of epinephrine had declined to 0.08 ng/ml. The beta-adrenergic agent, isoproterenol, had similar effects on lung liquid movement as epinephrine, and the actions of both agents were inhibited by propranolol (a beta-adrenergic blocker); norepinephrine was without effect on lung liquid secretion. 97 To respond to epinephrine it appears that the pulmonary epithelium must first be primed with triiodothyronine (T 3) and hydrocortisone (HC), circulating concentrations of which progressively increase during ovine gestation. 98'99 Thyroidectomy of fetal sheep completely abolishes the actions of epinephrine on lung liquid absorption, 1~176 and infusion of T 3 restores responsiveness. 1~ However, fetal infusion of T 3 did not change the age at which responsiveness

100 o

,,.-2.

%

ee~ ~

== 150 =, 01"

w

9

o o Jr

,,.-'-,..% ~6

9

0

o

o,p

tv

o ~d,# 9 eee ~

o

100

9

o

0.1

~E

o o

13..

o o

o

7~o 660 s60 ".do a60 a60 ~6-0 Time before delivery (min)

6

~6o

0.01

After delivery

Fig. 8.3. Changes in lung liquid movement and maternal plasma epinephrine levels at birth in the sheep. (Reproduced with permission from Brown MJ, Olver RE, Ramsden CA et al. Effects of adrenaline and of spontaneous labour on the secretion and absorption of lung liquid in the fetal lamb. J. Physiol. 1983; 344:137-52.) See text for details.

to epinephrine first developed (i.e. ---122 days). But, if both T 3 and HC were infused (to produce plasma levels in the upper part of the range seen just before full term), then responsiveness to epinephrine could be induced in younger fetuses (116-120 days) than normal. 1~ Work with explants of fetal rat lung or trachea indicated that the effects of hormones on lung liquid absorption were mediated predominantly by actions on alveolar rather than airway epithelium. 1~ Thus, cultures of rat lung and trachea were initiated at 14 or 16 days of gestation, respectively. They were exposed to various hormones or blocking agents for varying periods (depending on hormone). Then on day 8 of culture, wet weights (a measure of liquid secretion) and TEP were measured. A combination of T3, HC, and terbutaline reduced the weight of lung buds relative to control, but was without effect on tracheal buds. Results on both untreated tracheas and lung buds were consistent with the presence of active C1 secretion. Thus, TEP was stimulated by terbutaline and inhibited by bumetanide. Further, pretreatment with bumetanide abolished the effects of terbutaline. However, pretreatment of lung buds for 2 days with HC + T 3 produced results consistent with the presence of active Na absorption. Thus, terbutaline now increased TEP in the presence of bumetanide, and this increase was blocked by amiloride. In contrast, in tracheal buds exposed to HC + T3, the terbutaline-induced increase in TEP was blocked only by bumetanide, not amiloride. Thus, only in the lung buds did the combination of T3, HC, and beta-adrenergic agent convert C1 secretion to Na absorption. In rabbits, treatment with an irreversible beta-agonist inhibited surfactant production but had no effect on lung water content at or after birth. TM In sheep, it has also been reported that propranolol does not prevent absorption of

liquid at birth. 1~ Of course, it is possible that occupancy of only a small fraction of the total number of receptors is needed for the effects of epinephrine. Nevertheless, these results do suggest that neurohumoral agents other than betaadrenergic agents could be involved in lung liquid clearance. In fact, arginine vasopressin (AVP), prolactin, and epidermal growth factor have all been shown to influence liquid secretion in the fetal lung. AVP, for instance, at concentrations within the range seen during labor, inhibited liquid secretion in > 140-day fetal sheep by -~80%, an effect that occurred without change in circulating epinephrine levels. 1~ Unphysiologically high concentrations of prolactin stimulate lung liquid production in fetal goats. 1~ EGF slows lung liquid secretion in fetal sheep even in the presence of beta-blockade. 1~ PGE 2 and atrial natriuretic factor have also been shown to reduce tracheal liquid production in fetal lambs. 48'1~ Lung liquid absorption is stimulated by labor. In 1-7day-old human newborns, the average thoracic gas volume of babies delivered vaginally was 32.7_+1.7 ml/kg, but in those delivered by caesarian section it was significantly lower (19.7_+ 1.4 ml/kg). 92 Likewise, in rabbits delivered by caesarian, the lungs contain more water than those delivered vaginally. 11~ Thus it appears that labor in some way stimulates removal of lung liquid. It is uncertain whether this is a mechanical effect or mediated by increased levels of epinephrine or other hormones. Fetal and postnatal O 2 tensions are very different: 3 and 14%, respectively. Carbon dioxide tensions also differ: ---8% in the fetus and 5% in the adult. To determine whether these changes contributed to neonatal lung liquid absorption, lung explants from 14-, 20-, and 22-day fetal rats and 2-day neonatal rats were exposed to fetal or adult PO 2 and PCO2 .112 The wet-to-dry weight ratios of 14-day fetal or 2-day neonatal explants were independent of gas tensions. However, the water contents of 20- and 22-day explants were markedly increased by fetal as opposed to adult gas tensions. Thus, in the perinatal period, fetal gas tensions favored liquid secretion. 112 The effects of gas tension have also been investigated in primary cultures of fetal distal lung epithelium. Within 4 h of switching primary cultures of fetal lung distal epithelium from 3 to 21% O2, there was a significant decline (--30%) in TER. 113Resistance then recovered, but amiloridesensitive Isc was increased by N2-fold at 18 h and ---8-fold at 48 h after the switch. The findings suggest that the initial decline in TER was associated with an increase in H20 permeability that in the intact lung would facilitate H20 removal down oncotic and hydrostatic pressure gradients. However, water removal by active Na absorption was more important later after birth. Other studies have confirmed that the effects of changing PO 2 are quite slow, with Isc being significantly elevated 24 h but not 6 h after a switch from fetal to adult oxygen tensions. 114'115 Thus, changes in PO 2 are more important at maintaining liquid absorption in the adult than in removing lung liquid at birth. Some of the developmental changes in ion and water transport across pulmonary epithelium may be effected by changes in the extracellular matrix. Thus, when isolated

fetal pneumocytes were grown on matrix derived from mixed lung cells at the canalicular phase, they reverted to an immature phenotype with less amiloride-sensitive Na absorption and more C1 secretion than when grown on a variety of other substrates. 116

MECHANISMS SWITCH

OF THE P E R I N A T A L

Hydrostatic pressures may contribute to lung liquid absorption at birth. The first breaths are associated with an increase in permeability to non-electrolytes that could in theory correspond to a large increase in hydraulic conductivity. 117 Also, the characteristic end-expiratory pauses (expiratory grunting) that occur between the newborn's first few breaths will create positive lung pressures 118 that should also promote liquid absorption. Further enhancing hydrostatic absorption of water is a drop in interstitial hydrostatic pressure with air inflation. 119'12~ An increase in water permeability should also increase movement of water down the oncotic gradient between lung liquid and interstitium, lung liquid being essentially protein-free. 32 An increase in plasma protein concentration in the last few days before birth increases the oncotic driving force. 119'12~ Stretch of the bladder epithelium results in insertion of ENaC into the apical membrane 121'122 and a dramatic increase in active absorption of Na. This effect has not been looked for in alveolar or airway epithelium. However, the first breaths presumably stretch the lung epithelium, and it has been reported that liquid expansion of the lung of fetal goats causes a decrease in liquid secretion or induces absorption. 123 By contrast, Vejlstrup et al. TM concluded that in rabbits "pulmonary inflation renders active liquid clearance ineffective". In sheep fetuses of 135-141 days gestation, epinephrine converted a secretion of 8 ml/h to an absorption of 16 ml/h. 96 Amiloride, added in the continued presence of epinephrine, returned liquid movement to a secretion of 5ml/h. 96 Ussing's flux ratio equation was applied to Na and C1 fuxes under baseline conditions, after administration of epinephrine and after epinephrine plus amiloride. Driving forces (mV) for Na under the three conditions were 0.6, -8.6, and 0.0mV, respectively. Thus, Na was passively distributed under baseline conditions (i.e. in equilibrium with TEP), out of equilibrium (i.e. actively absorbed) after epinephrine, and passively distributed in the presence of both epinephrine and amiloride. For C1, the corresponding driving forces were 21, 2.2, and 18 mV. Thus, epinephrine not only stimulated active absorption of Na, but also inhibited active C1 secretion. This suggests that the same cells perform Na absorption and C1 secretion, and by opening Na channels in the apical membrane, epinephrine will depolarize this membrane and inhibit C1 secretion. In contrast, induction of C1 secretion by amiloride is presumably due to a hyperpolarization caused by block of Na channels. 125 Interestingly, the epithelium can still secrete C1 in the presence of epinephrine indicating

that there have been no major changes in the numbers of C1 channels or basolateral NaK2C1 exchangers; the inhibition of C1 transport by epinephrine is solely due to an unfavorable change in the apical membrane potential difference. In the longer term, however, it seems likely that the transport proteins involved in C1 secretion (apical membrane C1 channels and the NaK2C1 cotransporter) will decline in numbers, whereas numbers of ENaC will increase. The mechanisms underlying the transition from C1 secretion to Na absorption at birth have been studied with isolated dispersed type II cells from the rabbit. Levels of ouabain-sensitive 86Rb uptake in fetal, neonatal, and adult pneumocytes were in the ratio 1:3:12, respectively. 126 Studies, in which ouabain-sensitive Rb uptakes were compared to pump numbers (determined from [3H]-ouabain binding), 127 showed that during the transition from fetus to newborn there was no change in the numbers of pump sites, but that turnover of individual pumps increased -4-fold. By contrast, in moving from newborn to adult, there was no change in turnover rate, but a 5-fold increase in pump density (from - 2 5 0 to ~-1000 pumps per l.tmz ofbasolateral membrane). The ouabain-sensitive uptake of Rb is driven by entry of Na. Thus, there must be a dramatic increase in the rate of Na entry at birth that increases the rate of Na-K-ATPase turnover presumably by elevating [Na]i. Evidence, discussed below, suggests that this increase in Na entry reflects greater numbers or higher levels of activation of ENaC. The increase in pump numbers between neonatal and adult cells may reflect the effects of chronically elevated [Na]i, a condition that leads to synthesis and membrane insertion of further Na-K-ATPase units in other cell types. 12s The sensitivity of 86Rb uptake to amiloride and loop diuretics has not been studied thoroughly. Nevertheless, 86Rb uptake into adult pneumocytes is inhibited -25% by amiloride, as also is uptake of 22Na.126 Bumetanide, however, also has small effects on Rb uptake in adult pneumocytes. 126'129 Clearly, adult cells may retain the capacity for at least some active secretion of CI. Also consistent with this conclusion is the demonstration of an apical membrane CI conductance in pneumocyte cultures from adult rats. 63'13~Detailed comparison of the effects of bumetanide and amiloride on ouabain-sensitive 86Rb uptake into fetal, newborn, and adult pneumocytes could determine whether the apparatus for CI secretion is dismantled at the same time that Na transport is increased. Rats also show changes in Na-K-ATPase levels in the neonatal period, but the timing is little different from the rabbit. Thus, Na-K-ATPase alpha 1 subunit increased from fetal day 17 to fetal days 20-22 and then declined in the early postnatal period. 65 Direct measurement of Na-KATPase activity showed a 2.6-fold increase between days 17 and 20-22. Also in rat, the expression of ENaC subunits has been studied during development, fetuses being harvested at 17-22 days of gestation. TM Alpha-ENaC was first detectable at 19 days and progressively increased in utero. Beta- and gamma-ENaC were not detected until 21 and 22 days. TM Comparing the timing of expression of Na-K-ATPase and

ENaC suggested that the latter may be more important in the perinatal changes in lung liquid transport. Measurements of apical membrane Na conductance (GNa) and basolateral Na-K-ATPase activity have been made in cultures of distal lung epithelial cells from fetal rats following a switch from 3 to 14% 02 .114'115 To study GNa, the basolateral membrane was permeabilized with nystatin, and gradients in [Na] were imposed across the remaining apical membrane. Conversely, to study Na-K-ATPase activity in the basolateral membrane, the apical membrane was permeabilized and ouabain-sensitive currents across the intact basolateral membrane were measured. Short-circuit current in intact cell sheets increased 6-24 h after the change in O 2 tension, and was associated with increased Na pump capacity. Levels of GNa and of alpha-ENaC promoter activity were raised at 24 h, but increased further by 48 h. Thus, increased levels of ENaC were seemingly a consequence rather than a cause of the increased Isc seen at 6 h. Absorption of lung liquid at birth may involve changes in permeability to water as well as changes in ion transport. Thus, in the alveolar epithelium of rats, expression of mRNA for aquaporin 4 increased 8-fold during the 2 days immediately before birth. 6 In contrast, levels of AQP-4 mRNA in brains and kidneys did not change. Further, switching from 4 to 21% O 2, a change known to participate in the neonatal switch from CI secretion to Na absorption, increased AQP-4 expression.

CONCLUSION In the fetus, active secretion of CI across both airway and alveolar epithelia drives water into the lung, and the resulting elevation of intraluminal hydrostatic pressure is essential important for normal lung development. At birth, lung liquid is removed with a half-time of an hour or so. Of several factors involved in this liquid absorption, quantitatively the most important is a switch in the primary active ion transport process operating across alveolar epithelium from C1 secretion to Na absorption. This switch is effected by the interplay of a number of hormones of which the ultimate is usually epinephrine. Changes in arterial O 2 tension and in the extracellular matrix help maintain liquid absorption. The perinatal increase in Na absorption is initially due to insertion of ENaC into the apical membrane, followed later by an increase in the numbers of Na-K-ATPase on the basolateral membrane. The perinatal decrease in active C1 secretion is probably mainly due to a change in apical membrane potential difference consequent on increased Na conductance.

REFERENCES 1. Welsh MJ. Electrolyte transport by airway epithelia. Physiol. Rev. 1987; 67:1143-84. 2. Strang LB. Fetal lung liquid: secretion and reabsorption. Physiol. Rev. 1991; 71:991-1016.

3. Matalon S. Mechanisms and regulation of ion transport in adult mammalian alveolar type II pneumocytes. Am. J. Physiol. 1991; 261 :C727-38. 4. Koefoed-Johnsen V, Ussing HH. The nature of the frog skin potential. Acta Physiol. Scand. 1958; 42:298-308. 5. Matalon S, O'Brodovich H. Sodium channels in alveolar epithelial cells: molecular characterization, biophysical properties, and physiological significance. Annu. Rev. Physiol. 1999; 61:627-61. 6. Ruddy MK, Drazen JM, Pitkanen OM etal. Modulation of aquaporin 4 and the amiloride-inhibitable sodium channel in perinatal rat lung epithelial cells. Am. J. Physiol. 1998; 274:L1066-72. 7. Pitk~inen OM, Smith D, O'Brodovich H etal. Expression of alpha-, beta-, and gamma-hENaC mRNA in the human nasal, bronchial, and distal lung epithelium. Am. J. Respir. Crit. Care Med. 2001; 163:273-6. 8. Farman N, Talbot CR, Boucher R etal. Noncoordinated expression of alpha-, beta-, and gamma-subunit mRNAs of epithelial Na § channel along rat respiratory tract. Am. J. Physiol. 1997; 272:C131-41. 9. Gaillard D, Hinnrasky J, Coscoy S et al. Early expression of beta- and gamma-subunits of epithelial sodium channel during human airway development. Am. J. Physiol. 2000; 278:L177-84. 10. Hummler E, Barker P, Gatzy J et al. Early death due to defective neonatal lung liquid clearance in alpha-ENaC-deficient mice. Nat. Genet. 1996; 12:325-28. 11. Orser BA, Bertlik M, Fedorko L et al. Cation selective channel in fetal alveolar type II epithelium. Biochim. Biophys. Acta. 1991; 1094:19-26. 12. Marunaka Y. Amiloride-blockable Ca2§ Na § permeant channels in the fetal distal lung epithelium. Pflugers Arch. 1996; 431:748-56. 13. Anderson M, Welsh M. Calcium and cAMP activate different chloride channels in the apical membrane of normal and cystic fibrosis epithelia. Proc. Natl. Acad. Sci. USA 1991; 88:6003-7. 14. Ding C, Potter ED, Qiu Wet al. Cloning and widespread distribution of the rat rod-type cyclic nucleotide-gated cation channel.Am. J. Physiol. 1997; 272:C1335-44. 15. Widdicombe JH, Welsh MJ. Ion transport by dog tracheal epithelium. Fed. Proc. 1980; 39:3062-66. 16. Frizzell RA, Field M, Schultz SG. Sodium-coupled chloride transport by epithelial tissues.Am. J. Physiol. 1979; 236:F1-8. 17. Cassin S, Gause G, Perks AM. The effects of bumetanide and furosemide on lung liquid secretion in fetal sheep. Proc. Soc. Exp. Biol. Med. 1986; 181:427-31. 18. Thorn J, Perks AM. The effects of furosemide and bumetanide on lung liquid production by in vitro lungs from fetal guinea pigs. Can. J. Physiol. Pharmacol. 1990; 68:1131-5. 19. Anderson MP, Sheppard DN, Berger HA etal. Chloride channels in the apical membrane of normal and cystic fibrosis airway and intestinal epithelia. Am. J. Physiol. 1992; 263:L1-14. 20. Hanrahan JW, Tabcharani JA, Grygoczyk R. Patch clamp studies of apical membrane chloride channel. In: Dodge JA, Brock JH, Widdicombe JH (eds), Current Topics in Cystic Fibrosis, Vol. 1. Chichester: John Wiley & Sons, 1993, pp. 93-137. 21. Basset G, Crone C, Saumon G. Fluid absorption by rat lung in situ: pathways for sodium entry in the luminal membrane of alveolar epithelium.J. Physiol. 1987; 384:325-45. 22. Saumon G, Basset G, Bouchonnet F et al. Cellular effects of beta-adrenergic and of cAMP stimulation on potassium transport in rat alveolar epithelium. Pflugers Arch. 1989; 414:340-5. 23. Finkbeiner WE, Widdicombe JH. Control of nasal airway secretions, ion transport, and water movement. In: Parent

RA (ed.), Treatise on Pulmonary Toxicology, Vol. 1. Comparative Biology of the Normal Lung. Boca Raton, FL: CRC Press, 1992, pp. 633-57. 24. Olver R, Davis B, Marin M e t al. Active transport of Na§ and C1- across the canine tracheal epithelium in vitro. Am. Rev. Respir. Dis. 1975; 112:811-15. 25. Nord EP, Brown SE, Crandall ED. C1-/HCO3- exchange modulates intracellular pH in rat type II alveolar epithelial cells. J. Biol. Chem. 1988; 263:5599-606. 26. Nielson DW, Goerke J, Clements JA. Alveolar subphase pH in the lungs of anesthetized rabbits. Proc. Natl. Acad. Sci. 1981; 78:7119-23. 27. Gatto LA. pH of mucus in rat trachea. J. Appl. Physiol. 1981; 50:1224-6. 28. Kyle H, Ward JPT, Widdicombe JG. Control of pH of airway surface liquid of the ferret trachea in vitro. J. Appl. Physiol. 1990; 68:135-41. 29. Lubman RL, Danto SI, Crandall ED. Evidence for active H § secretion by rat alveolar epithelial cells. Am. J. Physiol. 1989; 257:L438-45. 30. Fischer H, Widdicombe JH, Illek B. Acid secretion and proton conductance in human airway epithelium. Am. J. Physiol. 2001; 282:C736-43. 31. Adams FH, Fujiwara T, Rowsham G. The nature and origin of the fluid in fetal lamb lung.J. Pediatr. 1963; 63:881-8. 32. Adamson TM, Boyd RD, Platt HS et al. Composition of alveolar liquid in the foetal lamb.J. Physiol. 1969; 204:159-68. 33. Cotton CU, Lawson EE, Boucher RC et al. Bioelectric properties and ion transport of airways excised from adult and fetal sheep.J. Appl. Physiol. 1983; 55:1542-49. 34. Jost A, Policard A. Contribution experimental a l'etude du development prenatal du poumon chez le lapin. Arch. Anat. Microsc. Morphol. Exp. 1948; 37:323-32. 35. Carmel JA, Friedman F, Adams FH. Tracheal ligation and lung development.Am.J. Dis. Child. 1965; 109:452-6. 36. Adams FH, Desilets DT, Towers B. Control of flow of fetal lung fluid at the laryngeal outlet. Respir. Physiol. 1967; 2:302-9. 37. Alcorn D, Adamson TM, Lambert TF etal. Morphological effects of chronic tracheal ligation and drainage in the fetal lamb lung.J. Anat. 1977; 123:649-60. 38. Moessinger AC, Harding R, Adamson TM et al. Role of lung fluid volume in growth and maturation of the fetal sheep lung.J. Clin. Invest. 1990; 86:1270-7. 39. Olver RE, Strang LB. Ion fluxes across the pulmonary epithelium and the secretion of lung liquid in the foetal lamb. J. Physiol. 1974; 241:327-57. 40. Widdicombe JH, Welsh MJ. Anion selectivity of the chloridetransport process in dog tracheal epithelium. Am. J. Physiol. 1980; 239:Cl12-17. 41. McAteer JA, Cavanagh TJ, Evan AP. Submersion culture of the intact fetal lung. In Vitro 1983; 19:210-18. 42. Krochmal EM, Ballard ST, Yankaskas JR etal. Volume and ion transport by fetal rat alveolar and tracheal epithelia in submersion culture.Am. J. Physiol. 1989; 256:F397-407. 43. McCray PB, Jr., Welsh MJ. Developing fetal alveolar epithelial cells secrete fluid in primary culture. Am. J. Physiol. 1991; 260:L494-500. 44. McCray PB, Jr., Reenstra WW, Louie E et al. Expression of CFTR and presence of cAMP-mediated fluid secretion in human fetal lung.Am. J. Physiol. 1992; 262:L472-81. 45. Graeff RW, Wang G, McCray PB, Jr. KGF and FGF-10 stimulate liquid secretion in human fetal lung. Pediatr. Res. 1999; 46:523-9. 46. Barker PM, Gatzy JT. Effects of adenosine, ATP, and UTP on chloride secretion by epithelia explanted from fetal rat lung. Pediatr. Res. 1998; 43:652-9. 47. McCray PB, Jr., Bettencourt JD. Prostaglandins stimulate fluid secretion in human fetal lung.J. Dev. Physiol. 1993; 19:29-36.

48. Castro R, Ervin MG, Ross MG et al. Ovine fetal lung fluid response to atrial natriuretic factor. Am. J. Obstet. Gynecol. 1989; 161:1337-43. 49. Zhou L, Graeff RW, McCray PB, Jr et al. Keratinocyte growth factor stimulates CFTR-independent fluid secretion in the fetal lung in vitro.Am. J. Physiol. 1996; 271 :L987-94. 50. Dobbs LG. Isolation and culture of alveolar type II cells. Am. J. Physiol. 1990; 258:L134-47. 51. Ussing HH, Zerahn K. Active transport of sodium as the source of electric current in short-circuited isolated frog skin. Acta Physiol. Scand. 1951; 23:110-27. 52. O'Brodovich H, Rafii B, Post M. Bioelectric properties of fetal alveolar epithelial monolayers. Am. J. Physiol. 1990; 258:L201-6. 53. Rao AK, Cott GR. Ontogeny of ion transport across fetal pulmonary epithelial cells in monolayer culture. Am. J. Physiol. 1991; 261:L178-87. 54. Barker PM, Stiles AD, Boucher RC et al. Bioelectric properties of cultured epithelial monolayers from distal lung of 18-day fetal rat.Am. J. Physiol. 1992; 262:L628-36. 55. Barker PM, Boucher RC, Yankaskas JR. Bioelectric properties of cultured monolayers from epithelium of distal human fetal lung.Am. J. Physiol. 1995; 268:L270-7. 56. McCray PB, Jr., Bettencourt JD, Bastacky J et al. Expression of CFTR and a cAMP-stimulated chloride secretory current in cultured human fetal alveolar epithelial cells. Am. J. Respir. Cell. Mol. Biol. 1993; 9:578-85. 57. Tessier GJ, Lester GD, Langham MR. etal. Ion transport properties of fetal sheep alveolar epithelial cells in monolayer culture.Am. J. Physiol. 1996; 270:L1008-16. 58. Widdicombe JH, Coleman DL, Finkbeiner WE. et al. Primary cultures of the dog's tracheal epithelium: fine structure, fluid and electrolyte transport. Cell Tissue Res. 1987; 247:94-103. 59. Widdicombe JH, Coleman DL, Finkbeiner WE et al. Electrical properties of monolayers cultured from cells of human tracheal mucosa.J. Appl. Physiol. 1985; 58:1729-35. 60. Yamaya M, Finkbeiner WE, Chun SY etal. Differentiated structure and function of cultures from human tracheal epithelium.Am. J. Physiol. 1992; 262:L713-24. 61. Kondo M, Finkbeiner WE, Widdicombe JH. A simple technique for culture of highly differentiated cells from dog tracheal epithelium. Am. J. Physiol. 1991; 261 :L 106-17. 62. Dobbs LG, Pian MS, Maglio M e t al. Maintenance of the differentiated type II cell phenotype by culture with an apical air surface.Am.J. Physiol. 1997; 273:L347-54. 63. Jiang X, Ingbar DH, O'Grady SM. Adrenergic regulation of ion transport across adult alveolar epithelial cells: effects on C1- channel activation and transport function in cultures with an apical air interface. J. Membr. Biol. 2001; 181:195-204. 64. Schneeberger EE, McCarthy KM. Cytochemical localization of Na+-K+-ATPase in rat type II pneumocytes. J. Appl. Physiol. 1986; 60:1584-9. 65. Ingbar DH, Weeks CB, Gilmore-Hebert Met al. Developmental regulation of Na, K-ATPase in rat lung. Am. J. Physiol. 1996; 270:L619-29. 66. Johnson MD, Widdicombe JH, Allen L e t al. Alveolar epithelial type I cells actively transport sodium and are likely to play a role in mediating lung liquid homeostasis. Proc. Natl. Acad. Sci. 2002; 99:1966-71. 67. Cotton CU, Boucher RC, Gatzy JT. Bioelectric properties and ion transport across excised canine fetal and neonatal airways.J. Appl. Physiol. 1988; 65:2367-75. 68. Zeitlin PL, Loughlin GM, Guggino WB. Ion transport in cultured fetal and adult rabbit epithelia. Am. J. Physiol. 1988; 254:C691-8. 69. Matthay MA, Folkesson HG, Verkman AS. Salt and water transport across alveolar and distal airway epithelia in the adult lung.Am. J. Physiol. 1996; 270:L487-503.

70. Sakuma T, Pittet JF, Jayr C et al. Alveolar liquid and protein clearance in the absence of blood flow or ventilation in sheep.J. Appl. Physiol. 1993; 74:176-85. 71. Sakuma T, Okaniwa G, Nakada T et al. Alveolar fluid clearance in the resected human lung. Am. J. Respir. Crit. Care Med. 1994; 150:305-10. 72. Garat C, Rezaiguia S, Meignan M et al. Alveolar endotoxin increases alveolar liquid clearance in rats. J. Appl. Physiol. 1995; 79:2021-8. 73. Rezaiguia S, Garat C, Delclaux C et al. Acute bacterial pneumonia in rats increases alveolar epithelial fluid clearance by a tumor necrosis factor-alpha-dependent mechanism. J. Clin. Invest. 1997; 99:325-35. 74. Fukuda N, Jayr C, Lazrak A et al. Mechanisms of TNF-alpha stimulation of amiloride-sensitive sodium transport across alveolar epithelium.Am. J. Physiol. 2001; 280:L1258-65. 75. Basset G, Crone C, Saumon G. Significance of active ion transport in transalveolar water absorption: a study on isolated rat lung.J. Physiol. 1987; 384:311-24. 76. Norlin A, Lu LN, Guggino SE et al. Contribution of amilorideinsensitive pathways to alveolar fluid clearance in adult rats. J. Appl. Physiol. 2001; 90:1489-96. 77. Junor RW, Benjamin AR, Alexandrou D et al. Lack of a role for cyclic nucleotide gated cation channels in lung liquid absorption in fetal sheep.J. Physiol. 2000; 523:493-502. 78. Mason RJ, Williams MC, Widdicombe JH et al. Transepithelial transport by pulmonary alveolar type II cells in primary culture. Proc. Natl. Acad. Sci. USA 1982; 79:6033-7. 79. Kim KJ, Cheek JM, Crandall ED. Contribution of active Na+ and CI- fluxes to net ion transport by alveolar epithelium. Respir. Physiol. 1991; 85:245-56. 80. Goodman BE, Fleischer RD, Crandall ED. Evidence for active Na + transport by cultured monolayers of pulmonary alveolar epithelial cells.Am. J. Physiol. 1983; 245:C78-83. 81. Kondo M, Finkbeiner WE, Widdicombe JH. Cultures of bovine tracheal epithelium with differentiated ultrastructure and ion transport. In Vitro 1993; 29A:19-24. 82. Uyekubo SN, Fischer H, Maminishkis A et al. Cyclic AMPdependent absorption of chloride across airway epithelium. Am. J. Physiol. 1998; 275:L1219-27. 83. Kitterman JA, Ballard PL, Clements JA et al. Tracheal fluid in fetal lambs: spontaneous decrease prior to birth. J. Appl. Physiol. 1979; 47:985-9. 84. Dickson KA, Maloney JE, Berger PJ. Decline in lung liquid volume before labor in fetal lambs. J. Appl. Physiol. 1986; 61:2266-72. 85. Bland RD. Dynamics of pulmonary water before and after birth. Acta Paediatr. Scand. Suppl. 1983; 305:12-20. 86. Perks AM, Dore JJ, Dyer R et al. Fluid production by in vitro lungs from fetal guinea pigs. Can. J. Physiol. Pharmacol. 1990; 68:505-13. 87. Dickson KA, Harding R. Decline in lung liquid volume and secretion rate during oligohydramnios in fetal sheep. J. Appl. Physiol. 1989; 67:2401-7. 88. Cheng JB, Goldfien A, Ballard PL et al. Glucocorticoids increase pulmonary beta-adrenergic receptors in fetal rabbit. Endocrinology 1980; 107:1646-8. 89. Warburton D, Parton L, Buckley S e t al. Beta-receptors and surface active material flux in fetal lamb lung: female advantage. J. Appl. Physiol. 1987; 63:828-33. 90. Whitsett JA, Manton MA, Darovec-Beckerman C et al. Betaadrenergic receptors in the developing rabbit lung. Am. J. Physiol. 1981; 240:E351-7. 91. Karlberg P, Adams FH, Beubelle F etal. Alteration of the infant's thorax during vaginal delivery. Acta Obstet. Gynecol. Scand. 1962; 41:223-9. 92. Milner AD, Saunders RA, Hopkin IE. Effects of delivery by caesarean section on lung mechanics and lung volume in the human neonate.Arch. Dis. Child. 1978; 53:545-8.

93. Brown MJ, Olver RE, Ramsden CA et al. Effects of adrenaline and of spontaneous labour on the secretion and absorption of lung liquid in the fetal lamb. J. Physiol. 1983; 344:137-52. 94. O'Brodovich H, Hannam V, Seear Met al. Amiloride impairs lung water clearance in newborn guinea pigs. J. Appl. Physiol. 1990; 68:1758-62. 95. O'Brodovich H, Hannam V, Rafii B. Sodium channel but neither Na(+)-H + nor Na-glucose symport inhibitors slow neonatal lung water clearance. Am. J. Respir. Cell. Mol. Biol. 1991; 5:377-84. 96. Olver RE, Ramsden CA, Strang LB et al. The role of amilorideblockable sodium transport in adrenaline-induced lung liquid reabsorption in the fetal lamb. J. Physiol. 1986; 376:321-40. 97. Waiters DV, Olver RE. The role of catecholamines in lung liquid absorption at birth. Pediatr. Res. 1978; 12:239-42. 98. Nathanielsz PW, Comline RS, Silver M e t al. Cortisol metabolism in the fetal and neonatal sheep. J. Reprod. Fertil. 1972; (Suppl. 16):39-59. 99. Fraser M, Liggins GC. Thyroid hormone kinetics during late pregnancy in the ovine fetus. J. Dev. Physiol. 1988; 10:461-71. 100. Barker PM, Brown MJ, Ramsden CA et al. The effect of thyroidectomy in the fetal sheep on lung liquid reabsorption induced by adrenaline or cyclic AMP. J. Physiol. 1988; 407:373-83. 101. Barker PM, Markiewicz M, Parker KA etal. Synergistic action of triiodothyronine and hydrocortisone on epinephrine-induced reabsorption of fetal lung liquid. Pediatr. Res. 1990; 27:588-91. 102. Barker PM, Waiters DV, Markiewicz Met al. Development of the lung liquid reabsorptive mechanism in fetal sheep: synergism of triiodothyronine and hydrocortisone. J. Physiol. 1991; 433:435-49. 103. Krochmal-Mokrzan EM, Barker PM, Gatzy JT. Effects of hormones on potential difference and liquid balance across explants from proximal and distal fetal rat lung. J. Physiol. 1993; 463:647-65. 104. McDonald JV, Jr., Gonzales LW, Ballard PL etal. Lung beta-adrenoreceptor blockade affects perinatal surfactant release but not lung water. J. Appl. Physiol. 1986; 60:1727-33. 105. Bland RD, Nielson DW. Developmental changes in lung epithelial ion transport and liquid movement. Annu. Rev. Physiol. 1992; 54:373-94. 106. Wallace MJ, Hooper SB, Harding R. Regulation of lung liquid secretion by arginine vasopressin in fetal sheep. Am. J. Physiol. 1990; 258:R104-11. 107. Cassin S, Perks AM. Studies of factors which stimulate lung fluid secretion in fetal goats. J. Dev. Physiol. 1982; 4:311-25. 108. Kennedy KA, Wilton P, Mellander M e t al. Effect of epidermal growth factor on lung liquid secretion in fetal sheep. J. Dev. Physiol. 1986; 8:421-33. 109. Kitterman JA. Fetal lung development. J. Dev. Physiol. 1984; 6:67-82. 110. Bland RD, Bressack MA, McMillan DD. Labor decreases the lung water content of newborn rabbits. Am. J. Obstet. Gynecol. 1979; 135:364-7. 111. Bland RD, McMillan DD, Bressack MA etal. Clearance of liquid from lungs of newborn rabbits. J. Appl. Physiol. 1980; 49:171-7.

112. Barker PM, Gatzy JT. Effect of gas composition on liquid secretion by explants of distal lung of fetal rat in submersion culture.Am. J. Physiol. 1993; 265:L512-17. 113. Pitk~inen O, Transwell AK, Downey G. et al. Increased PO z alters the bioelectric properties of fetal distal lung epithelium.Am. J. Physiol. 1996; 270:L1060-6. 114. Baines DL, Ramminger SJ, Collett A etal. Oxygen-evoked Na § transport in rat fetal distal lung epithelial cells. J. Physiol. 2001; 532:105-13. 115. Ramminger SJ, Baines DL, Olver RE etal. The effects ofPO e upon transepithelial ion transport in fetal rat distal lung epithelial cells.ft. Physiol. 2000; 524:539-47. 116. Pitk~inen OM, Tanswell AK, O'Brodovich HM. Fetal lung cell-derived matrix alters distal lung epithelial ion transport. Am. J. Physiol. 1995; 268:L762-71. 117. Egan EA, Olver RE, Strang LB. Changes in non-electrolyte permeability of alveoli and the absorption of lung liquid at the start of breathing in the lamb. J. Physiol. 1975; 244:161-79. 118. Mortola JP. Dynamics of breathing in newborn mammals. Physiol. Rev. 1987; 67:187-243. 119. Fike CD, Lai-Fook SJ, Bland RD. Alveolar liquid pressures in newborn and adult rabbit lungs. J. Appl. Physiol. 1988; 64:1629-35. 120. Raj JU. Alveolar liquid pressure measured by micropuncture in isolated lungs of mature and immature fetal rabbits. J. Clin. lnvest. 1987; 79:1579-88. 121. Loo DDF, Lewis SA, Ifshin MS etal. Turnover, membrane insertion and degradation of sodium channels in rabbit urinary bladder. Science 1983; 221:1288-90. 122. Lewis SA, de Moura JLC. Incorporation of cytoplasmic vesicles into apical membrane of mammalian urinary bladder. Nature 1982; 297:685-8. 123. Perks AM, Cassin S. The rate of production of lung liquid in fetal goats, and the effect of expansion of the lungs. J. Dev. Physiol. 1985; 7:149-60. 124. Vejlstrup NG, Boyd CA, Dorrington KL. Effect of lung inflation on active and passive liquid clearance from in vivo rabbit lung.Am. J. Physiol. 1994; 267:L482-7. 125. Boucher RC, Willumsen NJ, Knowles MR etal. Na § and C1absorption in respiratory epithelia: the role of apical and basolateral membranes. In: Mastella G, Quinton P (eds), Cellular and Molecular Basis of Cystic Fibrosis. San Francisco: San Francisco Press, 1988, pp. 107-14. 126. Bland RD, Boyd CA. Cation transport in lung epithelial cells derived from fetal, newborn, and adult rabbits. J. Appl. Physiol. 1986; 61:507-15. 127. Chapman DC, Widdicombe JH, Bland RD. Developmental Am. differences in rabbit lung epithelial Na§247 J. Physiol. 1990; 259:L481-7. 128. Lamb JF. Regulation of sodium pump abundance in animal cells. Prog. Clin. Biol. Res. 1988; 273:361-8. 129. Kemp PJ, Roberts GC, Boyd CA. Identification and properties of pathways for K § transport in guinea-pig and rat alveolar epithelial type II cells.ft. Physiol. 1994; 476:79-88. 130. Jiang X, Ingbar DH, O'Grady SM. Adrenergic stimulation of Na § transport across alveolar epithelial cells involves activation of apical C1-channels. Am. J. Physiol. 1998; 275:C1610-20. 131. Tchepichev S, Ueda J, Canessa C etal. Lung epithelial Na channel subunits are differentially regulated during development and by steroids. Am. J. Physiol. 1995; 269:C805-12.

In humans, the ability of the lung to exchange respiratory gases is dependent upon many unique structural, physiological and biochemical features which must have developed by the time of birth for the fetus to survive the transition to extra-uterine life. The lung must have developed an intricate tree-like airway structure to conduct air to and from the respiratory gas-exchange surface. The gas-exchange regions must, collectively, have a large surface area which is closely apposed to a rich vascular network to enhance gas diffusion between air and blood. The mechanics of the respiratory system must be sufficiently mature to readily allow lung expansion during inspiration and prevent lung collapse during expiration. Many of these features are present or have begun to develop in the lung by the time of birth, despite the lung playing no role in gas exchange before birth. Thus, at birth, the lung must immediately assume a role that it has not performed before and, in most cases, they smoothly and efficiently take over the role of gas exchange at birth. However, in some pregnancies fetal lung development is compromised, leading to respiratory insufficiency, a major cause of neonatal morbidity and mortality. Inadequate fetal lung development can result from either an insufficient period of in utero development, due to premature birth, or to inappropriate development due to disruption of developmental processes. Because appropriate lung development during fetal life is essential for the successful transition to extra-uterine life, it is important to understand the factors that control prenatal lung development. One of the major differences between fetal and neonatal lungs is that, during fetal life, the future airways of the lung are filled with liquid. This liquid is produced by the lung; it *To whom correspondence should be addressed. The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

leaves the lung by flowing out of the trachea 1 and is either swallowed or enters the amniotic sac. 2 It is now evident that fetal lung liquid is secreted across the pulmonary epithelium into the future airspace due to an osmotic gradient established by the net movement of CI- in the same direction. 3 This Cl- gradient is thought to be generated by the Na/K ATPase pump located on the basolateral surface of the epithelial cells which creates the free-energy for Cl- to enter the cell, coupled with Na +, against its electrochemical gradient. 3'4 Cl- is thought to exit the cell down its electrochemical gradient across the apical membrane into the lung lumen through specific Cl- channels (see Chapter 8). It is now recognized that fetal lung liquid plays an integral role in the development of the lung before birth by maintaining the lungs in a constantly distended state. Indeed, the lungs are not collapsed during fetal life but are maintained in a state of expansion that is greater than the end-expiratory volume of the postnatal, air-filled, lung 5'6 (see Fig. 14.1). The decrease in basal lung volume at birth results from an increase in lung recoil due to the creation of surface tension upon the entry of air into the lungs. By maintaining the fetal lungs in a distended state, fetal lung liquid is thought to be essential for fetal lung growth and maturation by acting as an internal "splint" for lung tissue. 5-7 As a result, much attention has focused on how altering the basal degree of lung expansion affects fetal lung growth and development, as discussed in detail below. Similarly, the effect of phasic alterations in lung expansion, as result of fetal breathing movements (FBM), has also been extensively studied and is detailed below. Together, the basal degree of lung expansion and FBM are considered to be important physical factors that influence fetal lung growth and development. However, it is clear that circulating Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

endocrine factors as well as paracrine growth factors can mediate, potentiate, integrate and regulate the lung growth in response to these physical factors.

ROLE OF PHYSICAL FACTORS IN R E G U L A T I N G FETAL L U N G GROWTH It is often stated that the lung is either collapsed during fetal life or is maintained at the same degree of expansion as after birth. However, it is now apparent that healthy fetal lungs are expanded to a greater degree before birth than after birth (see Chapter 14). Although there has been some controversy in the literature as to the exact volume of liquid retained in the future airways during the later stages of gestation, data from fetal sheep show that the volume is 35-45ml/kg 6'8-1~ which is considerably higher than the end-expiratory lung volume in the neonate 6 (25-30ml/kg; see Fig. 14.1); some studies have reported volumes as high as 60 ml/kg in individual ovine fetuses. 1~The discrepancies in fetal lung liquid volumes reported in the literature are probably because the fetus actively participates in maintaining its lung volume (see below) and that the physical environment of the fetus also influences the volume of lung liquid. Thus, values that have been reported from dead (e.g. histological measurements), anesthetized (and usually exteriorized) or paralysed fetuses will underestimate lung luminal volumes because lung liquid is rapidly lost following death, anesthesia and paralysis. 1~ Similarly, measurements of lung volumes in chronically catheterized fetuses (usually fetal sheep) are questionable unless it was verified that a sufficient volume of amniotic fluid was present at the time of measurement and that animals were not in labour. 8'12

following a reduction in volume is by a decrease in lung liquid efflux. 16

Role of the fetal upper airway and the trans-pulmonary pressure gradient in regulating fetal lung liquid volumes The rate of liquid efflux from the fetal lung is dependent upon the pressure gradient between the lung lumen and amniotic sac (trans-pulmonary pressure) as well as the resistance to liquid efflux through the upper airway, predominantly the glottis. 1'17 During apnea, the intra-luminal pressure within the fetal lungs is 1-2 mmHg above amniotic sac pressure is which is the driving force for liquid to leave the lungs (see Fig. 9.1). This pressure gradient is created by the inherent recoil of lung tissue and is maintained, during periods of apnea, by active adduction of the glottis which provides a high resistance to liquid movement through the tracheal7; this is analogous to constricting the neck of an inflated rubber balloon (Fig. 9.1). Thus, during apnea, liquid

ADnea

lung

~

~

........

\

High r e s i s t a n c e

-"k ............... (-2I,,mrnHg/rnI/m,n)

e ,ux.

"""".ecr..on ...........I /

~ _,, :'qK---

Distending pressure

9

i

.

/".-"74

larynx

/ Fetal breathinq movements

R E G U L A T I O N OF THE BASAL DEGREE OF L U N G EXPANSION IN THE FETUS The basal degree of fetal lung expansion is determined by the volume of liquid retained within the future airways. As this profoundly influences fetal lung development, it is important to understand the factors that regulate lung liquid volumes in the fetus. In theory, lung liquid volume should be controlled by a balance between the rate at which it is secreted and the rate at which it flows out of the lungs via the trachea. In practice, however, the volume of fetal lung liquid (in the absence of labour) is controlled by its rate of efflux from the trachea, as alterations in the rate of fetal lung liquid secretion simply cause corresponding changes in the efflux of liquid, resulting in little change in lung liquid volume; this has been clearly demonstrated following reductions in fetal lung liquid secretion. 13 Similarly, although prolonged reductions in fetal lung expansion increase fetal lung liquid secretion rates in vivo, 11'14'15 the principal mechanism for restoring fetal lung liquid volume

lung liquid ............. secretion

~ !

'~~\~

i,.=

]j ]

.

i ,v

Low r e s i s t a n c e

"\

(-3 mmHgimltmin) .

...................... ~

Lung liquid effiux

/ ..................... .... " ' ~

-

....

,anX

tntra-tuminal pressure = amnioticsac pressure

Fig. 9.1. Diagram showing the function of the upper airway in regulating the efflux of lung liquid during periods of apnea (upper panel) and fetal breathing movements (lower panel). During periods of apnea, the resistance to lung liquid efflux through the upper airway is increased. As a result, liquid tends to accumulate within the future airways causing the lungs to expand, generating a trans-pulmonary pressure gradient of 1-2 mmHg (intra-luminal > amniotic sac pressure). During periods of FBM, the resistance of the upper airway decreases and accumulated liquid leaves the lungs at an increased rate. As a result, the trans-pulmonary pressure gradient decreases towards 0mmHg during FBM. Reproduced from Harding and Hooper, 6 with additional data from Harding et al. 17

tends to accumulate within the fetal lungs. However, during FBM the resistance to liquid efflux through the upper airway is actively reduced, 17 due to phasic abduction of the glottis, which increases lung liquid efflux 17 (Fig. 9.1). Thus, during FBM, despite contraction of the diaphragm, there is a net loss of liquid from the lungs which is 2-3 times greater than the loss during intervening apneic periods. 1'19 Liquid can, however, enter the fetal lungs during episodes of vigorous FBM, which may be related to changes in upper airway function1 or to changes in trans-pulmonary pressure. The trans-pulmonary pressure gradient of the fetus is predominantly generated by the intrinsic recoil of lung tissue, but because the chest wall in the fetus is highly compliant, 2~ it is also influenced by external factors like abdominal pressure. Thus, increases in abdominal pressure, which could occur due to changes in fetal posture (e.g. trunk flexion), increase intra-luminal pressures, which increase the trans-pulmonary pressure gradient leading to a loss of lung liquid 12 (Fig. 9.1). These changes in fetal posture could arise from fetal movements or could be imposed on the fetus by limited intra-uterine space. 12For example, if the mechanical buffering effect of amniotic fluid is lost, as in oligohydramnios, the uterus may compress the fetus thereby increasing trunk flexion (Fig. 9.2). 12'21 This increases (-2-fold) fetal abdominal pressure, which increases the trans-pulmonary pressure gradient leading to increased lung liquid efflux (see Fig. 9.3); 12 a similar scenario may explain the loss of large quantities of lung liquid during early labour, s An increase in lung compliance late in gestation, associated with increased circulating corticosteroid levels, will greatly increase the volume change caused by even small changes in trans-pulmonary pressure. Thus, as fluid space within the intra-uterine compartment can become increasingly reduced in late gestation, particularly in multiple pregnancies, it is not surprising that different researchers have found a wide variety of lung liquid volumes late in gestation.

4 03 03 r

3

o

T

..,E 2 O 03 o

_r 0

. . . . . . . . . . .

16

.r

E

12

x

.-g_ _--~

....

T

...I

0

.......

Control

Oligohydramnios

Recovery

Fig. 9.3. The increase in fetal tracheal pressure and lung liquid efflux associated with non-labour uterine contractions during a control period, 48h of oligohydramnios and a recovery period. These data demonstrate that, if intra-uterine volume is limited, non-labour uterine contractions increase fetal abdominal (data not shown) and tracheal pressures, resulting in an increase in liquid efflux from the fetal lungs. Data obtained from Harding et al. 12

Role of the chest wall in maintaining fetal lung liquid volumes 1.0 or) .m

-o 0.8

......T

w r" .m t~ m

0.6

-o 0.4 o o,..,,. u

E 0.2 o z 0.0

,,,I. . . . . . . . . . . . . . .

Control

!

Oligohydramnios

Fig. 9.2. The effect of oligohydramnios induced by drainage of amniotic fluid on the degree of spinal cord flexion (measured as a normalized spinal radius) in fetal sheep. A smaller normalized spinal radius is indicative of a greater curvature of the spine. Data obtained from Harding et al. 12

After birth the stiffness of the chest wall plays an important role in maintaining end-expiratory lung volume by opposing lung recoil and preventing lung collapse. The tendency of the lung to collapse away from the chest wall generates a negative intra-pleural pressure o f - 5 cmH20. In contrast, the fetal lung is maintained in a distended state by the retention of liquid within the future airways, resulting in a distending pressure of 1-2 mmHg during apnea. TMAs a result, at rest, the intra-pleural pressure is essentially zero. 18 This indicates that the chest wall plays little if any mechanical role in maintaining lung volume at rest, although the dome of the diaphragm extends into the chest, as in adults, indicating that its position is influenced by lung recoil and/or abdominal pressure. Consequently, it is possible that the growth and shape of the chest wall is influenced by the volume of the fetal lung. 21 For example, infants with severe pulmonary hypoplasia characteristically have bell-shaped chests which could result from a combination of the small

size of the lung as well as its reduced compliance and increased recoil. The effects of diaphragmatic contractions on fetal lung liquid volumes are discussed below.

E V I D E N C E FOR THE ROLE OF L U N G E X P A N S I O N IN FETAL L U N G GROWTH AND DEVELOPMENT Much experimental and clinical evidence has accumulated to indicate that growth and maturation of the fetal lung are critically dependent upon the degree to which it is expanded by liquid. Sustained reductions in lung expansion retard the growth and structural maturation of the fetal lung, whereas these processes are accelerated by sustained increases in lung expansion.

Clinical evidence A wide variety of disorders result in fetal pulmonary hypoplasia including oligohydramnios, congenital diaphragmatic hernias (CDH), space-occupying lesions like pulmonary cysts, tumours and pleural effusions, as well as a number of fetal muscular-skeletal deformities. Although these disorders are diverse in nature, they all share a common mechanism by which they induce fetal lung hypoplasia, namely a prolonged reduction in the degree of fetal lung expansion. 21 Oligohydramnios is a relatively common clinical problem occurring in approximately 10% of all pregnancies. It can result from the loss of amniotic fluid due to premature rupture of the membranes or from inadequate production of amniotic fluid due to urinary tract disorders, including bilateral renal agenesis, renal dysplasia as well as disruptions to urine outflow into the amniotic sac due to agenesis or stenosis to the ureters, urethra or urethral valve. 21 The severity of the lung growth deficit varies considerably between individuals depending upon a number of factors, particularly the gestational age at onset and the duration of exposure. 22 Depending upon these factors, the resulting pulmonary hypoplasia can be lethal within hours of birth in its most severe form, but can be sub-clinical during the neonatal period in less severe forms possibly contributing to neonatal respiratory distress. The mechanism by which oligohydramnios causes a reduction in fetal lung liquid volume is via an increase in the trans-pulmonary pressure gradient. 12 In the absence of amniotic fluid, the intra-uterine space is markedly reduced and the uterine wall compresses the fetus causing exaggerated flexion of the fetal trunk (Fig. 9.2). 12'21 This leads to an increase in abdominal pressure which results in an increase in the trans-pulmonary pressure gradient and the loss of lung liquid (Fig. 9.3). 12'23 The degree of compression imposed by the uterus can be so severe that it causes marked facial and limb disorders. 24 CDH is less common than oligohydramnios, but can result in very severe pulmonary hypoplasia. The mortality rates associated with this disorder vary widely between studies, arguably due to differences in the inclusion/exclusion of

subjects, but are usually reported as > 50~ 25'26 CDH is sometimes associated with other fetal malformations and occurs due to failure of the diaphragm ligament to close, thereby failing to separate the chest from the abdomen during embryonic development. 27 The hernias can either be unilateral (both left- or right-sided) or bilateral and allow abdominal contents to migrate into the thorax, thereby preventing the lung from expanding; in its most severe form, the liver can herniate into the chest. 27 It is not clear whether the abdominal contents compress the lung or simply occupy thoracic space, thereby preventing the lung from expanding and occupying that space. However, the latter mechanism is clearly consistent with the mechanism by which turnouts, cysts and intra-pleural fluid accumulation cause pulmonary hypoplasia. A variety of muscular-skeletal disorders can also result in severe pulmonary hypoplasia. 21 Although the precise mechanisms by which these disorders induce pulmonary hypoplasia will depend on the type of disorder, the mechanism likely involves a reduction in lung expansion. That is, any disorder which reduces the ability of the fetus to defend its lung volume will likely result in pulmonary hypoplasia. For instance, diaphragmatic activity and glottic adduction play important roles in maintaining lung expansion during fetal development 12 and, therefore, interfering with these activities can impact upon fetal lung liquid volumes (see

Fig. 9.4).

100

E.-.

80

V

O

> ~

~8

60

._g~ C:~

~///2

40 / /

,_1

2O //_///,,

Control fetus

No FBM

No FBM + No UAR

Lungs ex fetus

Fig. 9.4. The influence of fetal muscular activity on defending the volume of liquid retained within the future airways. Compared with control fetuses (solid column), the inhibition of fetal breathing movements (FBM) by either fetal spinal cord transection TMor selective blockade of the phrenic nerves, 11 causes an -25% decrease in fetal lung liquid volume (open column), demonstrating the importance of fetal diaphragmatic activity in maintaining the volume of fetal lung liquid. If upper airway resistance (UAR) is eliminated (by by-passing the upper airway) in addition to the inhibition of FBM, the volume of lung liquid is reduced further, demonstrating the independent effect that the upper fetal airway has in maintaining fetal lung liquid volumes. The further reduction in the volume of lung liquid following the removal of the lungs from the fetus demonstrates the contribution that the fetal chest wall makes to maintaining fetal lung expansion. Diagram reproduced from Harding and Hooper. 6

Experimental evidence Most experimental evidence indicating that alterations in fetal lung expansion regulate the growth and development of the fetal lungs is derived from experiments in which the fetal trachea has been obstructed or the lungs drained of liquid. Fetal tracheal obstruction was first performed to demonstrate that fetal lung liquid is produced by the fetal lung and is not inhaled amniotic fluid. 28'29 Following tracheal ligation, the lungs accumulated liquid within the future airways, causing them to overexpand, but it was also noted that the lungs were larger and structurally more mature. 29 Subsequent experiments demonstrated that prolonged periods of tracheal ligation in fetal sheep increased fetal lung weights, whereas prolonged periods of lung deflation, caused by lung liquid drainage, reduced fetal lung weights. 7 It was also noted that reductions in fetal lung expansion increased the density of type-II cells. 7 In fetal sheep it has been demonstrated that the lung growth response to increased lung expansion is dependent on local factors within the stretched tissue, rather than circulating factors. 3~ Prolonged ligation of the left bronchus caused overexpansion and increased growth of the left lung, whereas the fight lung, which remained at a control level of expansion, remained at the same size as a control right lung. 3~ More recent studies have advanced this concept by showing that, although alteration in lung expansion at the local level is the primary determinant of lung tissue growth, circulating endocrine factors can influence the relationship between lung expansion and lung growth. For example, the lung growth response to an increase in lung expansion depends upon growth hormone both in the fetus (following tracheal obstruction) 31 and after birth (following hemipneumonectomy). 32Furthermore, exogenous cortisol enhances the fetal lung growth response to an increase in lung expansion, most probably via an increase in lung compliance. 33 Prolonged reductions in fetal lung expansion reduce lung growth rates, and the extent of the growth rate reduction appears to depend on the decrease in lung expansion. A 25% reduction in lung expansion causes a 25% reduction in lung DNA content, TM whereas total lung deflation, due to lung liquid drainage, causes lung growth to cease.7'15This cessation in growth results in severe pulmonary hypoplasia within 3-4 weeks. 7'15'34 Understanding the mechanisms by which sustained changes in fetal lung expansion alter lung growth is important because these mechanisms are likely to involve those responsible for regulating normal lung growth. In the case of increased lung expansion induced by tracheal obstruction, the growth-promoting processes are not continuously active, but show a time dependency which may depend on how the lung expands. 35 A detectable increase in lung DNA content can be measured within 2 days of tracheal obstruction in fetal sheep 36 and the increase in growth is completed within 7 days, resulting in an almost doubling (-70% increase) in DNA content. 35'37However, the rate of lung cell proliferation is not uniform throughout this period (Fig. 9.5) and appears to differ in different species and at different stages of lung

Fig. 9.5. Diagram showing how the extracellular matrix (ECM) receptors (integrins) mechanically couple the intracellular microfilaments with the ECM, to form a structural continuum. Proteins associated with the intracellular domain are associated with numerous intracellular signalling pathways 72 which may mediate the response to alterations in fetal lung expansion. Diagram reconstructed from Ingber etal. 6z and Rubin et al. 72

development. In fetal sheep, increased DNA synthesis rates peak within 2 days of obstructing the fetal trachea during the alveolar stage of lung development, 35 whereas this same period corresponds to a "stagnation in lung growth" during the pseudoglandular stage of lung development in fetal rabbits. 38 These differences most probably reflect the different stage of lung development at the time of obstruction rather than species differences. Indeed, no stagnation of lung growth was subsequently observed in older fetal rabbits following tracheal obstruction. 39 It is now known that the type and rate of lung growth induced by an increase in lung expansion differs at different stages of lung development. Although the rate of accelerated lung growth induced by tracheal obstruction is less in fetuses at the late pseudoglandular to early canalicular stages of lung development than during the alveolar stage, the eventual increase in lung size and DNA content is greater in younger fetuses. 36'4~The slower rate of accelerated growth is thought to be due to a lower rate of lung expansion in the younger fetuses, which have stiffer lungs, 36 whereas the greater final increase in lung size following tracheal obstruction in younger fetuses is thought to be due to a more compliant chest wall which allows the lungs to expand to a greater degree. 4~ However, a greater degree of lung expansion may also be responsible for the hydropic state that has been observed in fetal humans 41'42and sheep 4~when tracheal obstruction is performed during pseudoglandular/ canalicular stage of lung development; the hydrops is probably due to a restriction in venous return or the expanded lungs constraining the heart. Furthermore, the characteristics of the lung growth induced at the different stages of lung development is very different. During the alveolar stage, tracheal obstruction causes the proliferation of most major cell types in the lung, 43 but during the late pseudoglandular/ early canalicular stage it predominantly causes proliferation of mesenchymal cells. 4~ As a result, the inter-airway tissue distances markedly increase resulting in a large increase in the percentage of space occupied by tissue. 4~In contrast, these tissue distances are reduced when tracheal obstruction is performed during the alveolar stages of lung development,

resulting in marked reductions in the percentage of space occupied by tissue. 43 Thus, care should be taken when comparing the effects of increased lung expansion between species and at different stages of development. During the alveolar stage of lung development, the mechanisms responsible for the acceleration in lung growth caused by tracheal obstruction are maximally activated within 2 days. 35 Using a modified version of the left main bronchus ligation model, 3~ a recent study used a differential gene analysis technique to identify genes activated and suppressed during this maximally activated growth phase. 44 The first differentially expressed gene identified was calmodulin-2 which could be an important intracellular mediator of the proliferation induced by an increase in lung expansion. The effect of alterations in fetal lung expansion on the fetal lung are not restricted to growth. Increases in fetal lung expansion also accelerate structural maturation of the lung in a variety of species, causing a reduction in the percentage of tissue space due to a reduction in inter-alveolar tissue as well as an increase in alveolar number and surface a r e a . 7'38'39'43'45 In contrast, reductions in fetal lung expansion greatly retard structural development of the lung, resulting in marked increases in inter-alveolar tissue space and reduced alveolar development, particularly alveolar number. 7 In addition, alterations in fetal lung expansion have a profound impact on alveolar epithelial cell (AEC) differentiation. Increases in fetal lung expansion induce type-II AECs to differentiate, via an intermediate cell type, into type-I AECs such that within 10 days of obstructing the fetal trachea, 5%. 55 These experiments have shown that simulated FBMs in vitro stimulate lung cell proliferation, causing a time-dependent increase in DNA synthesis. 56 The proliferation is thought to be mediated by increased synthesis of platelet-derived growth factor (PDGF) as the cellular proliferation can be inhibited by antisense oligonucleotides for PDGF-B. 57 PDGF is thought to act via both PDGF alpha and beta receptors to induce prenatal lung growth, 5s most probably by activating phospholipase C and PKC intracellular pathways. 56 Although these experiments have provided useful information on the transduction pathways by which mechanical stimuli initiate lung cell proliferation, the stimulus does not accurately simulate FBM in vivo. Indeed, in vivo, individual breathing movements are essentially iso-volumic and, therefore, the percentage length change experienced by a cell with each FBM is negligible. This is primarily because the fetal chest wall is very compliant, fetal lung liquid is very viscous compared with air and, due to its bulk, has a large inertia. Thus, although activation of the diaphragm causes a reduction in intra-thoracic pressure, very little liquid is inhaled because other sections of the chest wall are simultaneously drawn in; 59 indeed liquid has to be present within the pharynx before any liquid can be inhaled. 17 As a result, tidal volume in the late-gestation fetus is < 1% of resting lung volume, but is much higher immediately after birth (---20%). As the resistance to liquid movement through the upper airway during FBM is 3-4 mmHg/(ml/min) in fetal sheep, 17 the pressures required to move a volume equivalent to 5% of resting lung volume would be much greater than intratracheal pressure fluctuations during FBM. 18 In vivo evidence The in vivo evidence supporting the concept that FBMs play an important role in fetal lung development involves studies in which their thoracic components have been eliminated. Sectioning the phrenic nerves in the fetus causes a reduction in fetal lung growth, 6~ but also causes atrophy of the diaphragm resulting in its upward displacement into the thorax, resulting in a reduction in lung expansion. Paralysis of the fetal diaphragm, without causing its atrophy,

can be achieved by sectioning the spinal cord (between C1 and C2) above the level of the phrenic motoneurones. Although this procedure results in a reduction in fetal lung growth, 14'62'63 this decrease is likely due to an associated reduction in lung expansion. 14 Indeed, the reduction in lung expansion (--25%) equates to the reduction in lung DNA content (--25%) and is caused by blocking activation of the diaphragm without blocking activation of the glottis; TM the glottis is innervated by the recurrent laryngeal nerves which are not affected by spinal cord transection. The net effect is increased loss of lung liquid and a reduction in lung expansion due to continued phasic dilation of the larynx (Fig. 9.4), TMas diaphragmatic contractions inhibit the efflux of lung liquid during FBM episodes. Similarly, reversible pharmacological blockade of both phrenic nerves increases lung liquid efflux and leads to a reduction in lung liquid volume which is quickly restored after the phrenic nerve blockade is removed. 11 Thus, it is not surprising that total fetal paralysis leads to increased lung liquid efflux, a reduction in lung liquid volume 11 and a decrease in lung growth. 64 Taken together, these data provide strong evidence that fetal muscular activity, whether it is active glottic adduction during apnea, or activation of the diaphragm during FBM, plays a critical role in maintaining lung liquid volumes and hence the mean level of fetal lung expansion. However, at present there is no in vivo evidence to suggest that phasic stretch of the lung during FBM is an important determinant of fetal lung growth, other than by defending lung liquid volumes. Even the finding that bilateral thoracoplasty induces lung hypoplasia can also be explained by a decrease in fetal lung expansion. 65

MECHANO-TRANSDUCTION MECHANISMS An understanding of the transduction pathways by which mechanical forces are translated into chemical stimuli is important as all cells, tissues and whole organs of the body are subjected to mechanical forces in vivo. These forces include shear stress, strain, stretch and compression and can result from gravity, fluid flow, intracellular tensile forces (e.g. muscle contractions), ambulatory body movements, as well as from changes resulting from the expansion of organs such as the bowel, stomach, uterus, heart, bladder and lung. It has been proposed that cells exist in a state of isometric tension that is generated by the intracellular contractile filaments. 66 Thus, externally applied forces are thought to be imposed on a pre-existing force equilibrium, causing changes in cell shape and intracellular structural fibre alignment until the force equilibrium is re-established. 66'67 It has been recognized for many years that a variety of mechanical forces play an important role in cellular growth and differentiation and is a critical regulator of threedimensional tissue structure, 67 particularly in the lung. 5'68 Thus, they are an important pathway by which cells interact with and respond to their environment. 66'67'69 In general, the response of a cell to a mechanical force is to reduce the

impact of that force on the cell. For example, an attached cell exposed to a flow of fluid across it will act to reduce the shear force by orientating the longitudinal axis of the cell in the direction of flow and adopting a streamlined shape. 7~ Cells exposed to increased strain will align their intracellular structural fibres along the principal direction of strain and eventually recruit, synthesize and align new intracellular structural fibres also along this plane. 71 In addition, a cell exposed to strain may synthesize ECM components that form an extracellular framework to help resist the load. 71 Other cellular responses to strain include cellular proliferation and differentiation 67'68which can lead to marked changes in the size, structure and function of an organ.

The role of the ECM and "outside-in" cell signalling The discovery of cell-surface receptors that bind to a variety of ECM proteins has greatly advanced our understanding of mechano-transduction mechanisms. For example, the "integrin" family of trans-membrane proteins cluster at focal adhesion sites and bind to a specific sequence, arg-gly-asp (RGD), that is common in many ECM proteins. 72 The intracellular domains of these ECM receptors are mechanically linked to fibrillar-actin bundles, via a variety of cytoskeletal-associated proteins (e.g. talin, vinculin, paxillin) and are closely associated with a number of protein kinases. 72 As actin bundles form a major component of the intracellular structural scaffolding, it is clear that the intracellular and extracellular components are mechanically coupled, via ECM receptors, to form a structural continuum (Fig. 9.5). It is via these couplings that mechanical forces can be detected and translated into intracellular chemical signals. 67'72'73 The intracellular signalling pathways are less well defined, although they are thought to include stretch-activated ion channels, activation of intracellular second messenger systems and the direct activation of RNA polymerases and DNA synthetic enzymes via changes in nuclear shape. Indeed, the structural continuum between a cell and its surrounding ECM includes the nucleus, which is connected to the cell surface via intermediate filaments. Thus, mechanical forces that distort cell shape will also alter the shape of the nucleus, which can influence gene transcription and DNA synthetic machinery via pathways that are currently not understood. 67'69 Nuclear DNA is attached to the structural scaffolding of the nucleus, the spindle microtubules, via kinetochore proteins located at the centromeric regions of chromosomes. It is not difficult to envisage, therefore, that mechanical forces may influence DNA replication and gene transcription via direct manipulation of DNA and associated enzymatic pathways. The development of integrin-mediated cell adhesions also results in the relocation of pre-existing mRNA molecules and protein translation machinery to the focal adhesion to induce a rapid enhancement of protein translation at the site of cell attachment. TM

Potential cytoplasmic second messenger systems In addition to effects mediated by the cytoskeleton, focal adhesion sites also include numerous signalling molecules

that are activated or inactivated by activation of ECM receptors (e.g. integrins). The signalling molecules associated with focal adhesion sites include focal adhesion kinase (FAK), which is an intracellular tyrosine kinase that binds to a number of other signalling and structural proteins, including PI-3-kinase, Src, Grb2 and pl30Cas (Fig. 9.6). Activation of PI-3-K leads to activation of the inositol triphosphate pathway, PKA and PKC as well as enhancing calcium signalling. Activation of the Src family ofintracellular tyrosine kinases leads to the phosphorylation of a number proteins, leading to the recruitment and activation of downstream signalling molecules such as PLC-y, the Rho-like GTPases and the ERK and JNK signalling cascades (Fig. 9.6). Indeed, there is extensive cross-talk between integrin activation and the growth factor activated ERK signalling pathways that generally leads to enhanced and sustained ERK activation (Fig. 9.6). 75-77 This includes numerous mechanisms leading to the activation of Grb2 which, through its association with Sos, leads to the activation of Ras and the subsequent activation of Raf, MEK and ERK proteins and eventually to altered gene transcription. Similarly, Raf can be activated by a number of mechanisms independent of Ras. Apart from these intracellular signalling pathways, integrin clustering can also lead to the direct activation of a number of growth factor receptors in the absence of the growth factor ligand. 7s'79

Potential role of growth factors It is often assumed that any stimulus that induces cellular proliferation must be driven by growth factors, but the roles of growth factors in mediating expansion-induced fetal lung growth, at least in vivo, are unclear. Although mechanical forces are potent stimuli of cell behaviour in their own right, it is important to recognize that the application of mechanical forces in vivo (e.g. due to alterations in lung expansion) will manifest as a variety of different mechanical stimuli on different cells depending upon their location within the three dimensional structure of the lung. For instance, in response to an increase in lung expansion, alveolar epithelial cells, lying on the internal surface of the terminal gasexchange regions, will be exposed to stretch as will some capillary endothelial cells, particularly those whose basement membranes are fused with the epithelial basement membrane. On the other hand, many cells will experience compression due to terminal airway expansion. It is of interest, therefore, that an increase in fetal lung expansion induces proliferation of most major cell types within the lung and induces a coordinated growth response across the entire portion of stretched lung during the alveolar stage of lung development. 43 Thus, cells located in positions that expose them to mechanical stimuli like stretch may release growth factors which potentiate and integrate the response in adjacent non-activated cells. Similarly, remodelling of the ECM may release extracellularly bound growth factors which act to potentiate and integrate the growth response.

ROLE OF G R O W T H FACTORS IN FETAL L U N G DEVELOPMENT The role of growth factors during embryonic lung morphogenesis has been described in Chapter 1. This section wilt briefly cover the roles of the numerous growth factors and their receptors that are expressed in the lung during the later stages of lung development and that are likely to play roles in the finely coordinated development of the lung, including airway growth and branching, alveolarization, ECM remodelling, epithelial cell differentiation and angiogenesis. Particular emphasis is placed on those factors that may play a role in transducing the growth response of the lung to alterations in lung expansion/stretch. There is much evidence that the PDGF A and B and their receptors PDGFR-tx and -[3, are involved in the earlymid stages of lung development. PDGF-A and B mRNA levels increase 8~ in parallel with cell proliferation rates 82 during the pseudoglandular and canalicular stages of lung development, and cell proliferation within lung explants is inhibited by PDGF-A and B antisense oligonucleotides as well as PDGF neutralizing antibodies. 57'83 Tight regulation of PDGF-A is important as the PDGF-A knockout mouse exhibits a complete failure of alveolar septation, sl whereas overexpression of PDGF-A causes marked mesenchymal cell proliferation within the lungs, which fail to progress to the saccular stage of development, s4 Both PDGF-A s5 and B 86 have sheer stress-responsive elements (SSRE) in their promoter regions suggesting their expression levels can be influenced by exposure of cells to physical forces. This is supported by studies showing that cyclic stretch of cultured lung cells from the pseudoglandular and canalicular stages of lung development induces PDGF-B and PDGFR-[3 expression as well as cell proliferation, s7 This stretch-induced cell proliferation is abolished by PDGF-B and PDGFR-~ antisense oligonucleotides, PDGF-BB neutralizing antibodies and a PDGFR inhibitor. s7 In contrast to its role during the pseudoglandular and canalicular stages of lung development, PDGF-B expression is reduced at the time when DNA synthesis rates are maximally elevated in response to increased lung expansion during the alveolar stage of lung development (M. Wallace and S. Hooper, unpublished observations). This reduction in PDGF-B mRNA may reflect an acceleration of the normal decline in PDGF-B during the saccular and alveolar stages of development sI'82 and suggests that the factors controlling lung cell proliferation vary during the different stages of lung development. Vascular endothelial growth factor (VEGF) exerts potent mitogenic effects on endothelial cells via the VEGF receptor 2 (VEGFR2). ss VEGF and VEGFR2 mRNA increase as the lung matures, s9'9~which coincides with the increase in cross-sectional area of the pulmonary vascular bed. 9~ However, VEGF and VEGFR2 protein levels peak during the canalicular stage of lung development and decline thereafter. 92 VEGF protein is localized to the basement membrane beneath distal airway epithelial cells 93 suggesting a role in directing vascular growth to regions destined for gas exchange.

Fig. 9.6. Interactions between integrin pathways and growth factor-stimulated pathways. There are at least 18 0~ and 8 J3 integrin subunits 94 that form heterodimers between an o~ subunit and a 13subunit. These subunits consist of a large extracellular domain which binds to specific ECM molecules, a trans-membrane domain and a short intracellular domain that can interact with a large number of cytoskeletal, adapter and signalling molecules. Activation of integrins can lead to the potentiation of the effects of many growth factors via extensive interactions between signalling molecules activated by integrins and the primary growth factor-activated pathway (ERK pathway; see Fig. 9.7 and above - black arrows). Some of the pathways activated by integrins that potentiate the ERK pathway are shown in this diagram and are denoted by grey arrows, e.g. Src binds to and phosphorylates FAK generating a binding site for Grb2 which, through its association with Sos, leads to the activation of Ras. The adaptor protein Shc also binds to activated FAK and similarly binds Grb2. Src can phosphorylate p130Cas generating a binding site for Crk which, via C3G and Rap-l, also leads to activation of Raf. PI-3-K can similarly activate Raf via Rac and PAK or by the modulation of Sos. Integrins can also lead to activation of the ERK pathway in an FAK-independent manner via the association of the integrin-associated protein Caveolin and the Src family kinase Fyn which recruits and activates Shc. PKC is also activated in focal adhesions and can directly activate Raf. The p21 activated kinases (PAK) are also activated by integrin clustering via a variety of mechanisms (only one is shown above) and these kinases can phosphorylate and activate both Raf and MEK 75 and may enhance nuclear translocation of Erk.95 Integrins can also activate a number of growth factor receptors in the absence of the appropriate growth factor ligand. 79'95

Fig. 9.7. Diagram showing the ERK pathway. The extracellular binding of growth factors to their tyrosine kinase receptor (RTK) induces receptor dimerization and autophosphorylation on specific tyrosine residues in the cytoplasmic domains. The phosphorylated tyrosines (pTyr) act as binding sites for proteins containing pTyr recognition motifs (SH2 domains), which then activate a variety of intracellular signalling cascades. The pathway shown is the ERK pathway that commonly leads to cell division or differentiation (black arrows), as well as other signalling molecules that may be activated and may potentiate the ERK pathway (grey arrows). RTK autophosphorylation allows Grb2, an SH2-containing protein, to bind to the RTK, linking it to the guanine nucleotide exchange factor Sos, thus exchanging GTP for GDP on the small GTP binding protein Ras. Ras recruits Raf (a MKKK) to the cell membrane where it is activated. Raf then phosphorylates and activates MKKs (MEK1, MEK2) which phosphorylate and activate MAPKs (ERK1 = p44MAPK, ERK2 =p42MAPK). The activated MAPKs form dimers and can either directly phosphorylate and activate targets in the cell cytoplasm or they can translocate to the cell nucleus where they phosphorylate a variety of transcription factors leading to altered gene expression and subsequent cell division, differentiation or modified cell function.

Vascular development is often compromised in hypoplastic fetal lungs,92'96'97an effect which can be reversed by increased lung expansion. 96'97 As VEGF expression is upregulated by phasic stretch of lung cells,98'99 it has been suggested that VEGF was responsible for the endothelial cell proliferation induced by increased lung expansion. 43 However, a more recent study has been unable to confirm this suggestion (M. Wallace and S. Hooper, unpublished observations). The insulin-like growth factors (IGFs) I and II act as potent mitogens via the type-I IGF receptor 1~176 (IGF1R) and their bioactivity is modulated by the IGF-binding proteins 1~ (IGF-BPs). IGF1 is predominantly expressed during the saccular and alveolar stages of lung development, corresponding with a peak in IGF1R mRNA; 1~176 insufficiency of either leads to perinatal respiratory failure, possibly due to poorly formed alveoli. 1~176 IGF-II is predominantly expressed in the pseudoglandular and canalicular stages of lung development, with levels decreasing as the lung matures. 1~176 The IGF-BPs 2, 3, 4 and 5 are distributed spatially and temporally during lung development, 1~176176 suggesting that they may regulate the IGFs in a cell-specific manner. IGF-I and II mRNA levels are increased and decreased following increases and decreases in lung expansion, respectively, 14'37'1~ suggesting that they may mediate the changes in lung growth induced by altering lung expansion. However, this is unlikely, at least for IGF-II, as its expression does not increase until after the cell proliferation induced by increased lung expansion has ceased (unpublished observations). Transforming growth factor (TGF)-I3 inhibits fetal, newborn and adult type-II cell proliferation, surfactant synthesis and surfactant protein gene expression in culture. 1~ Type-II cell-specific expression of TGF-I31 has been shown to inhibit epithelial cell differentiation and halt lung development at the pseudoglandular stage of lung development in transgenic mice; 112 this was attributed to reduced lung expansion. However, although increased fetal lung expansion reduces the proportion of type-II alveolar epithelial cells, by inducing differentiation into the type-I cell phenotype, this did not correlate with increased TGF-I31 mRNA and bioactive TGF-[3 protein levels (unpublished observations). TGF-I~2 mRNA and protein are elevated after 4 weeks of increased lung expansion to reverse lung hypoplasia; 113 however, by that time it is likely that growth and structural alterations have returned to control levels. In contrast to TGF-I3, TGF-~ and epidermal growth factor (EGF) induce proliferation of isolated type-II cells from newborn rabbits 11~ and EGF treatment accelerates lung development in fetal monkeys. 114'115 The EGF receptor (EGFR) is activated by both TGF-~ and EGF, and gene deletion of this receptor prevents the attenuation of mesenchymal tissue between adjacent airways and reduces surfactant protein levels in fetal mice; these mice commonly die of respiratory failure postnatally. 116'117 The fibroblast growth factor (FGF) family consists of over 20 growth factors that bind to four FGF receptors (FGFR). 118 In fetal mice, the double knockout of FGFR3 and 4 prevents the formation of alveoli suggesting that these two receptors

act co-operatively to induce alveolarization. 119 The effects of exogenous FGF1 (acidic FGF) and FGF2 (basic FGF) on fetal lung growth differ between normal fetal mice and fetal mice with nitrofen-induced lung hypoplasia. 12~As hypoplastic lungs are also developmentally immature, this suggests that the effects of FGF1 and FGF2 differ at different stages of lung development or that prior nitrofen exposure interferes with FGF signalling pathways. The lungs of knockout mice for FGF7 (keratinocyte growth factor; KGF) appear to be normal, 121 although it has been proposed that KGF both enhances 122'123 and inhibits TM growth of the developing lung, while maturing type-II cells 123'125 and promoting this phenotype. 126 Both FGF2 and KGF mRNA are increased by cyclic stretch of postnatal lung cells. 99 Parathyroid hormone-related protein (PTHrP) and PTH/ PTHrP receptor knockout mice die at birth from respiratory failure, indicating that this protein is important for fetal lung development. Although the primary defect in the PTHrP knockout is pulmonary hypoplasia, 127 death was attributed to non-distensibility of the ribcage 128 and to reduced lung liquid clearance. 127PTHrP also induces surfactant protein and surfactant phospholipid biosynthesis 129 in lung explants from fetal mice. Furthermore, increased lung expansion 13~ and stretch of fetal lung cells in culture, TM increase PTHrP expression and receptor binding and enhance its effects on surfactant phospholipid synthesis. TM All the above growth factors act to induce cell proliferation and/or differentiation via the activation of intracellular cascades. The receptors for all of these growth factors (except TGF-13, a serine/threonine kinase receptor and PTHrP, a G-protein coupled receptor) belong to the transmembrane receptor tyrosine kinase family (RTK). The respective growth factors bind to the extracellular region, inducing receptor dimerization and autophosphorylation on specific tyrosine residues in the cytoplasmic domains. The pTyr act as binding sites for proteins containing pTyr recognition motifs (SH2 domains), which then activate a variety of intracellular signalling cascades. The most well-described pathway linking RTK activation to cell proliferation and differentiation is the extracellular signal-regulated (ERK) type of mitogenactivated protein kinase (MAPK) pathway. The core unit of this pathway consists of three successive tiers of phosphorylation (Fig 9.7). The MAPK (ERK) proteins are phosphorylated and activated by MAPK kinases (MKK also known as MEK=MAPK/ERK kinase), which themselves are phosphorylated by MKK kinases (MKKK). Although the specific components of the pathway vary depending on the stimuli and receptor activated, the most commonly described pathway involves the RTK binding of the SH2 domain-containing protein Grb2. This links the RTK to the guanine nucleotide exchange factor Sos, thus exchanging GTP for GDP on the small GTP binding protein Ras. Ras recruits Raf (a MKKK) to the cell membrane where it is activated. Raf then phosphorylates and activates MKKs (MEK1, MEK2) which phosphorylate and activate MAPKs (ERK1 = p44MAPK, ERK2 =p42MAPK). The activated MAPKs form dimers and can either directly phosphorylate

and activate targets in the cell cytoplasm or they can translocate to the cell nucleus where they phosphorylate a variety of transcription factors leading to altered gene expression (Fig. 9.7). The above schema (Fig 9.7) is greatly simplified and described in a linear format; it is more correctly viewed as a complex network of interactions, the balance of which will determine a cell's response. For example, activation of RTKs can also lead to activation of other pathways (e.g. the phosphoinositide-3 kinase pathway, PI3K), some components of which can influence the MAPK cascade. MAPK cascades can also be influenced by a variety of stimuli other than growth factors including cellular and oxidative stresses, inflammatory cytokines, G-protein stimulated cascades and integrin and ion channel activation. However, some degree of pathway specificity is determined by the MAPK family member activated; currently there are at least 12 known MAPKs, 7 MKKs and 14 MKKKs. 132 Thus, an enormous diversity of responses can be elicited by these stimuli, including changes in gene transcription, cell metabolism, proliferation, migration, differentiation, survival, apoptosis and inflammation. In mouse lung explants, the addition of an MEK inhibitor has been shown to reduce epithelial cell proliferation and to inhibit branching morphogenesis, 133suggesting that activation of the ERK pathway is necessary for normal lung development. ERK1 activity was also reduced in nitrofen-induced lung hypoplasia TM and increased following tracheal ligation. 135

ROLE OF E N D O C R I N E A N D O T H E R C I R C U L A T I N G FACTORS IN FETAL LUNG DEVELOPMENT Apart from the growth factors listed above, a number of endocrine factors have been implicated in regulating fetal lung growth and development. Much attention has focused on a variety of factors, particularly in the search for potential therapies for lung immaturity and lung hypoplasia. Corticosteroids have received much attention following the discovery that they are critical for lung maturation. However, the precise roles that most of these factors play in lung growth and development in vivo are still unclear.

The role of corticosteroids in fetal lung growth and development The pre-parturient increase in circulating corticosteroids is known to induce maturational changes in a variety of organ systems, which facilitates the transition to extra-uterine life. In the case of the lung, the maturational changes induced by corticosteroids are essential for the independent survival of the newborn, 136'137 although the precise mechanisms involved remain largely unknown. The reported effects of corticosteroids on lung development include those on (1) growth, (2) tissue remodelling, (3) type-II AECs and the surfactant system and (4) the reabsorption of lung liquid.

It is often assumed that corticosteroids induce lung maturation at the expense of lung growth, but the in vivo data are contradictory, probably a reflection of species differences as well as differences in dose, number of doses and route of administration. Most studies have used synthetic glucocorticoids (betamethasone and dexamethasone), which have a 30-40-fold greater bioactivity than cortisol. When administered to the mother, betamethasone 138'139 causes a decrease in both fetal body and lung growth. 140'141However, when administered directly to the fetus, an even greater dose of betamethasone does not affect fetal body or lung growth. 142 Similarly, physiological doses of cortisol, infused directly into the fetus to simulate the pre-parturient increase in fetal plasma cortisol levels, induce structural maturation of the lung without affecting either fetal lung or body growth. 33'143 These data suggest that the reported effect of maternally administered betamethasone on fetal lung growth may be mediated via an effect on the placenta. Furthermore, they indicate that the endogenous increase in fetal plasma cortisol concentrations does not induce lung maturation at the expense of lung growth. The principal action of corticosteroids on lung development is an effect on lung structure which greatly improves lung mechanics postnatally. TM In particular, corticosteroids markedly reduce inter-alveolar wall thickness leading to a reduction in percent tissue space and a marked increase in potential lung air-volume. Although this is arguably the major pulmonary effect of corticosteroids, the mechanisms involved remain unknown. TM Corticosteroids can also affect alveolarization, although again this affect may be dose- or species-dependent as betamethasone administered to rats has been shown to arrest alveolarization, T M whereas physiological doses of cortisol administered to fetal sheep increases alveolar number (R. Boland and S. Hooper, unpublished observations). In addition, the effect of corticosteroids on increasing lung compliance also interacts with the relationship between lung expansion and lung growth. A corticosteroidmediated increase in lung compliance is likely responsible for the high lung liquid volumes that have been observed in late gestation fetal sheep 5'6 and for the greater increase in lung expansion and lung growth in cortisol-infused fetuses following tracheal obstruction. 33 Although it was originally proposed that the beneficial effects of corticosteroids on lung maturation was due to an effect on surfactant and type-II AECs, recent studies in transgenic mice indicate that this concept needs re-evaluation. 147 Expression of the surfactant-associated proteins, SP-A, -B, -C and -D is commonly synonymously associated with type-II cell differentiation and has been used as an indicator of type-II AEC maturation. The glucocorticoid regulation of SP-B and SP-C in vitro supports a role for corticosteroid regulation of these genes and the type-II cell phenotype in vivo. 148 However, the effect of corticosteroids on surfactant and surfactant protein gene expression is complex and also appears to be dose-dependent in vitro, although the predominant in vivo effect (with supra-physiological doses) is the stimulation of surfactant components. TM As anticipated,

glucocorticoid receptor deficient mice die soon after birth with lungs that are structurally very immature. 147 However, these mice have normal surfactant protein expression levels 147 and, surprisingly, they have a greater proportion of type-II AECs than controls (T. Cole and S. Hooper, unpublished observations). These data suggest that, at least in mice, type-II cell differentiation and surfactant protein expression are independent of glucocorticoid receptor activation. An additional, often overlooked, beneficial effect of corticosteroids on the fetal lung is an increase in the lung's ability to reabsorb lung liquid. 143 Lung liquid re-absorption results from a reversal of the osmotic gradient across the pulmonary epithelium via the opening of amiloride-blockable Na +channels located on the apical surface of AECs. 4 This mechanism is thought to be activated by adrenaline during labour and is responsible for clearing significant volumes of lung liquid from the airways at birth, but only develops late in gestation. 149 Indeed, adrenaline-induced re-absorption of fetal lung liquid increases markedly near term due to the increase in endogenous corticosteroids at this time. 143'15~ Thus, the lungs of infants born preterm, who have not been exposed to prenatal corticosteroids, are likely to be incapable of re-absorbing liquid from the airways, which impairs their respiratory function.

The role of growth hormone in fetal lung growth and development Although circulating growth hormone (GH) concentrations are very high in the fetus and decrease rapidly at birth, fetal growth is thought to be independent of GH. TM Indeed, many fetal tissues contain GH receptors 152(GHR) particularly the lung, 153 but these receptor systems are thought to be immature, making fetal GH largely inactive. 153 GH receptor expression is complex; alternate leader exon usage of the 5'UTR for the GHR confers tissue-specific regulation, TM whereas alternate mRNA splicing between exons 8 and 9 can produce a short inactive form of the GHR. 155 Near term, cortisol is thought to increase GHR expression and activate the adult version of the alternately spliced version of the receptor, which may explain how postnatal growth becomes GH-dependent. However, a recent study has shown that the lung growth response to an increase in fetal lung expansion is GH-dependent. In the absence of GH (due to hypophysectomy), the initial growth response to tracheal obstruction was abolished whereas the re-infusion of GH restored the growth response. 31 These data indicate that GH may play a permissive role in regulating lung cell proliferation in response to mechanical stimuli such as alterations in lung expansion. This may not be restricted to the fetus as GH is thought to play a similar role after birth in regulating compensatory lung growth following hemipneumonectomy which is also thought to be expansion-dependent. 32

The role of retinoids in fetal lung growth and development Much interest in recent years has focused on the role of retinoids (for which vitamin A is a precursor) in fetal lung

development. Retinoids are thought to play a role in cellular proliferation and differentiation as well as airway branching and alveolarization, perhaps by altering the expression of Hox genes. 156 Retinoids exist in a variety of forms, including different isomers of retinoic acid (RA), retinyl esters as well as conjugates with specific binding proteins when circulating and within the cell. 156 Their biological action is mediated via a number of nuclear receptors (RAR) which confer on RA complex and differing responses in the lung. For example, RA administration to newborn animals induces septation and prevents glucocorticoid-induced inhibition of alveolarization in r a t s . 146'157 However, the abolition of RAR[3 by gene knockout increases septation and alveolar number in newborn mice, whereas the administration of a specific RARI3 agonist reduces alveolarization. 158 These data indicate that RAR[3 activation inhibits alveolarization and, therefore, the effect of RA on alveolarization is balanced by an interplay between RA and its receptor subtypes.

C L I N I C A L TREATMENTS FOR INFANTS WITH INAPPROPRIATE LUNG DEVELOPMENT It is clear that fetal lung development is a consequence of a complex interaction between a variety of mechanical and endocrine factors. However, the challenge is to translate this knowledge into therapeutic treatments that will improve the outcome for newborn infants that have inappropriately developed lungs. Inappropriate lung development at birth can arise due to preterm birth, which shortens in utero development time, as well as factors that compromise lung development in utero. With regard to the very preterm infant as a patient, maintaining its respiratory gas requirements is a considerable problem, yet little of what we have learnt from physiological studies in the fetus appears to have been translated into clinical practice. Indeed, many questions need to be addressed, such as the necessity to achieve blood gas tensions similar to those in the mature newborn. If these infants had remained in utero, they would have a PaO 2 of 20-30 mmHg, arterial O 2 saturations of 60-70% and a PaCO 2 of NS0mmHg. Instead, they are often ventilated with high levels of oxygen to maintain adult-like levels of oxygenation, which can damage the lungs and retard lung development (e.g. alveolarization). Although little is known about postnatal lung development in preterm infants, the mechanisms involved are probably similar to those that regulate prenatal lung development. Thus, the basal degree of lung expansion is likely to be an important factor, yet the use of ventilatory practises that specifically facilitate lung development are a minor consideration. For example, the application of a positive end expiratory pressure (PEEP), which increases functional residual capacity, is a common practice in ventilating very preterm infants and is undoubtedly beneficial. However, the primary aim is to improve oxygenation and reduce lung injury, yet appropriate lung development is arguably the key factor that will ultimately

determine the infant's long-term outcome. Thus, do the major objectives of treating preterm infants need re-evaluation in light of what we have learnt from the fetus? The finding that increased fetal lung expansion is a potent stimulus for fetal lung growth and development has prompted the suggestion that this may be used as an in utero therapeutic treatment for fetuses with severely hypoplastic lungs. 15'37'159'160 Indeed, experimental studies have shown that increases in fetal lung expansion, induced by tracheal obstruction, can rapidly reverse fetal lung growth and developmental deficits in Ut8/'O15'161-163 and improve postnatal respiratory function. 164'~65 However, clinical trials in fetal humans with severe pulmonary hypoplasia resulting from a CDH have had mixed success. 42 The principal problems relate to preterm labour, failure to stimulate fetal lung growth in some cases and, in other cases, postnatal respiratory insufficiency inspite of enhanced lung growth. 42 Thus, a significant discrepancy exists between the results obtained from clinical trials and those obtained from animal experiments. Although the reasons for these discrepancies are unknown, they may relate to the stage of lung development at which the treatment was applied, the duration of the tracheal obstruction and the severity of the fetal lung hypoplasia. Most tracheal obstructions in humans have been carried out at

After birth

Fig. 14.6. Instantaneous pulmonary blood flow through the left pulmonary artery before birth in a near-term sheep fetus (left panel) and in the same animal 1 hour after birth (right panel). These flow profiles demonstrate the beat-by-beat changes in pulmonary arterial blood flow. Note the large degree of backflow (- 100 ml/min) during diastole before birth, whereas after birth, this backflow is abolished.

flow through the DA is reversed. Such turbulence would likely elicit the release of endothelial-derived factors (see Chapter 7) that would contribute to the constriction of the DA.

FETAL B R E A T H I N G A N D THE O N S E T OF C O N T I N U O U S BREATHING AT BIRTH Although it is often stated that an infant takes its first breath at birth, it is now clear that respiratory movements begin during fetal life, long before birth. Like postnatal breathing, FBM involve rhythmical contractions of the diaphragm and other inspiratory muscles, such as dilator muscles of the pharynx and larynx, which are driven by brainstem centres that are stimulated by CO2 .64 The major differences between fetal breathing and postnatal breathing are that in the fetus (1) the lungs are liquid-filled and hence tidal volume is very small (see Chapter 9), (2) breathing movements are episodic, (3) breathing movements are inhibited, rather than stimulated, by hypoxia and (4) the breathing movements play no role in gas exchange, but rather result in a net oxygen consumption. Breathing movements in healthy human and ovine fetuses can first be detected during the first half of gestation and continue until the onset of labour. 64 Characteristically, they occur in episodes which become more organised later in gestation as fetal behavioural states become organised. During late gestation, FBM in humans and sheep occur 40-50% of the time 64 and are primarily associated with a state resembling rapid eye movement sleep; during episodes of 'quiet sleep' fetuses are largely apneic. 65 The incidence of FBM is reduced during active labour, but the mechanisms underlying this inhibition are not presently understood. 64 Although it is well established that prostaglandins can inhibit FBM, 66 it is thought that prostaglandins such as PGE 2 are not involved in the labour-related reduction of FBM. 67 Adenosine, which can be released into the fetal circulation from the placenta and fetal liver may play a role, as it is known to inhibit FBM via cerebral adenosine receptors. 68 During parturition and when the umbilical cord is cut at birth, the fetus-neonate may become profoundly hypoxemic, hypercapnic and acidemic, as well as being exposed to a lower environmental temperature with increased heat loss. It will also be exposed to a greatly increased degree of external sensory stimuli; in addition, its behavioural state may change to one of arousal. 69 What triggers continuous breathing at the time of birth is presently the subject of debate, although it is apparent that many factors, such as those listed above, are likely to be involved. A major factor is thought to be increased fetal CO 2 production, perhaps resulting from an increased rate of metabolism at birth, and/or an increased sensitivity to CO 2. Circulating levels of catecholamines are elevated at birth, 4~ which would be expected to stimulate metabolic activity and hence CO 2 production. It has also been shown that a reduction at birth in fetal circulating concentrations of adenosine, which is produced by the placenta, can stimulate thermogenesis TM leading to an increase in CO 2 production. A decrease in the concentrations of one or more circulating factors of placental origin (e.g. PGE 2, adenosine, progesterone metabolites)

that are known to inhibit FBM may facilitate continuous breathing after birth. 66'71'72 Removal of the placenta from the fetal circulation results in an increased ventilatory sensitivity to CO 2 suggesting that an inhibitory factor may act at the level of the fetal brainstem. 69 Central and peripheral chemoreceptors are active in the f e t u s , 73 although it has been suggested that their sensitivity may be tonically suppressed by factors released from the placenta into the fetal circulation such as PGE2, 72 receptors for which have been identified in the fetal brainstem. TM Studies of fetal sheep maintained e x u t e r o by extracorporeal oxygenation with the umbilical cord occluded, thereby eliminating potential effects of placental factors, have shown that continuous breathing is dependent upon blood CO 2 levels, 75 supporting the notion that CO 2 plays a crucial role in the maintenance of continuous breathing after birth. The integrity of the vagus nerves has been shown to be essential for the onset of adequate breathing at birth. 76 Although the critical pathways have not yet been identified, it is likely that volume receptive feedback from the lungs is involved. Studies in unanaesthetized neonatal lambs have shown that the application of negative airway pressures or creation of a tracheostomy, both of which would reduce end-expiratory lung volume (FRC) and the amount of vagal neural traffic from lung volume receptors (pulmonary stretch receptors), result in profound hypoventilation, periodic breathing and active glottic adduction during periods of apnea. 77'78 This indicates that volume receptive vagal feedback at end-expiration, which is normally maintained by an adequate FRC, is essential for continuous breathing in the newborn, and explains, at least in part, the benefits of positive end-expiratory pressure (PEEP) in the treatment of infantile apnea.

CONCLUSIONS During normal gestation, with labour and delivery occurring at term, the fetal lung is well prepared for its critical role of gas exchange after birth. Both endocrine and physical factors play a major role in preparing the lung for extrauterine function. However, if gestation is shortened as a result of preterm birth, maturation of the lung may not have occurred to a sufficient degree, resulting in respiratory compromise; in particular, the lung may not have developed structurally and epithelial type-II cells may not be able to produce sufficient quantities of surfactant, resulting in respiratory distress. If birth occurs in the absence of labour, as a result of caesarian section, this may result in a delay in the clearance of lung liquid after birth, which can result in transient tachypnea and a delay in establishing independent respiratory function. Although much is already known, there remain many unanswered questions relating to physiological and molecular mechanisms underlying the preparation of the lung for birth and its postnatal adaptation to airbreathing.

REFERENCES 1. Liggins GC. Premature delivery of foetal lambs infused with glucocorticoids. J. Endocrinol. 1969; 45:515-23. 2. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972; 50:515-25. 3. Crowley P, Chalmers I, Keirse MJNC. The effects of corticosteroid administration before preterm delivery: an overview of the evidence from controlled trials. Br. J. Obstet. Gynaecol. 1990; 97:11-25. 4. Crowley P. Prophylactic corticosteroids for preterm birth. Cochrane Database Syst. Rev. 2000; CD000065. 5. Jobe AH, Ikegami M. Lung development and function in preterm infants in the surfactant treatment era. Annu. Rev. Physiol. 2000; 62:825-46. 6. Liggins GC. The foetal role in the initiation of parturition in the ewe. In: Wolstenholme GEW, O'Connor M (eds) Foetal Autonomy (Ciba Foundation Symposium). Churchill, London. 1990; 218. 7. Crone RK, Davies P, Liggins GC et al. The effects of hypophysectomy, thyroidectomy, and postoperative infusion of cortisol or adrenocorticotrophin on the structure of the ovine fetal lung.J. Dev. Physiol. 1983; 5:281-8. 8. Kitterman JA, Liggins GC, Campos GA etal. Prepartum maturation of the lung in fetal sheep: relation to cortisol. J. Appl. Physiol. 1981; 51:384-90. 9. Liggins GC, Schellenberg JC, Finberg K etal. The effects of ACTH1-24 or cortisol on pulmonary maturation in the adrenalectomized ovine fetus. J. Dev. Physiol. 1985; 7:105-11. 10. Ballard PL. The glucocorticoid domain in the lung and mechanisms of action. In: Mendelson CR (ed.) Endocrinology of the Lung. Humana Press, Inc., Totowa. 2000; 1-44. 11. Cole TJ, Blendy JA, Monaghan AP et al. Targeted disruption of the glucocorticoid receptor gene blocks adrenergic chromaffin cell development and severely retards lung maturation. Genes Dev. 1995; 9:1608-21. 12. Bland RD. Loss of liquid from the lung lumen in labor: more than a simple "squeeze". Am. J. Physiol. Lung Cell Mol. Physiol. 2001; 280:L602-5. 13. Kitterman JA, Ballard PL, Clements JA et al. Tracheal fluid in fetal lambs: spontaneous decrease prior to birth. J. Appl. Physiol. 1979; 47:985-9. 14. Dickson KA, Maloney JE, Berger PJ. Decline in lung liquid volume before labor in fetal lambs. J. AppL Physiol. 1986; 61:2266-72. 15. Mitzner W, Johnson JWC, Scott R et al. Effect of betamethasone on pressure-volume relationship of fetal rhesus monkey lung.J. Appl. Physiol. 1979; 47:377-82. 16. Harding R, Hooper SB, Dickson KA. A mechanism leading to reduced lung expansion and lung hypoplasia in fetal sheep during oligohydramnios. Am. J. Obstet. Gynecol. 1990; 163:1904-13. 17. Lines A, Hooper SB, Harding R. Lung liquid production rates and volumes do not decrease before labor in healthy fetal sheep.J. Appl. Physiol. 1997; 82:927-32. 18. Dickson KA, Harding R. Restoration of lung liquid volume following its acute alteration in fetal sheep. J. Physiol. 1987; 385:531-43. 19. Hooper SB, Dickson KA, Harding R. Lung liquid secretion, flow and volume in response to moderate asphyxia in fetal sheep.J. Dev. Physiol. 1988; 10:473-85. 20. Olver RE, Ramsden CA, Strang LB et al. The role of amiloride-blockable sodium transport in adrenaline-induced lung liquid reabsorption in the fetal lamb. J. Physiol. 1986; 376:321-40.

21. Hooper SB, Harding R. Fetal lung liquid: a major determinant of the growth and functional development of the fetal lung. Clin. Exp. Pharmacol. Physiol. 1995; 22:235-47. 22. Matalon S, O'Brodovich H. Sodium channels in alveolar epithelial cells: molecular characterization, biophysical properties, and physiological significance. Annu. Rev. Physiol. 1999; 61:627-61. 23. Brown MJ, Olver RE, Ramsden CA et al. Effects of adrenaline and of spontaneous labour on the secretion and absorption of lung liquid in the fetal lamb.J. Physiol. 1983; 344:137-52. 24. Hooper SB, Harding R. Effects of 13-adrenergic blockade on lung liquid secretion during fetal asphyxia. Am. J. Physiol. 1989; 257:R705-10. 25. Wallace MJ, Hooper SB, Harding R. Regulation of lung liquid secretion by arginine vasopressin in fetal sheep. Am. J. Physiol. 1990; 258:R104-11. 26. Barker PM, Brown MJ, Ramsden CA et al. The effect of thyroidectomy in the fetal sheep on lung liquid reabsorption induced by adrenaline or cyclic AMP.J. Physiol. 1988; 407:373-83. 27. Wallace MJ, Hooper SB, Harding R. Role of the adrenal glands in the maturation of lung liquid secretory mechanisms in fetal sheep. Am. J. Physiol. 1996; 270:R1-8. 28. Barker PM, Markiewicz M, Parker KA et al. Synergistic action of triiodothyronine and hydrocortisone on epinephrineinduced reabsorption of fetal lung liquid. Pediatr. Res. 1990; 27:588-91. 29. Wallace MJ, Hooper SB, Harding R. Effects of elevated fetal cortisol concentrations on the volume, secretion and reabsorption of lung liquid.Am. J. Physiol. 1995; 269:R881-7. 30. Kindler PM, Chuang DC, Perks AM. Fluid production by in vitro lungs from near-term fetal guinea pigs: effects of cortisol and aldosterone. Acta Endocrinol. 1993; 129:169-77. 31. Berger S, Bleich M, Schmid W e t al. Mineralocorticoid receptor knockout mice: pathophysiology of Na § metabolism. Proc. Natl. Acad. Sci. USA 1998; 95:9424-9. 32. Dickson KA, Harding R. Compliances of the liquid-filled lungs and chest wall during development in fetal sheep. J. Dev. Physiol. 1991; 16:105-13. 33. Albuquerque CA, Smith KR, Saywers TE etal. Relation between oligohydramnios and spinal flexion in the human fetus. Early Hum. Dev. 2002; 68:119-26. 34. Berger PJ, Kyriakides MA, Smolich JJ et al. Massive decline in lung liquid before vaginal delivery at term in the fetal lamb. Am.J. Obstet. Gynecol. 1998; 178:223-7. 35. Kalache KD, Chaoui R, Marks Bet al. Does fetal tracheal fluid flow during fetal breathing movements change before the onset of labour? Br. J. Obstet. Gynaecol. 2002; 109:514-9. 36. Harding R, Sigger JN, Wickham PJD et al. The regulation of flow of pulmonary fluid in fetal sheep. Respir. Physiol. 1984; 57:47-59. 37. Pfister RE, Ramsden CA, Neil HL et al. Volume and secretion rate of lung liquid in the final days of gestation and labour in the fetal sheep.J. Physiol. 2001; 535:889-99. 38. Waiters DV, Olver RE. The role of catecholamines in lung liquid absorption at birth. Pediatr. Res. 1978; 12:239-42. 39. Harding R, Hooper SB. Regulation of lung expansion and lung growth before birth.J. Appl. Physiol. 1996; 81:209-24. 40. Hagnevik K, Faxelius G, Irestedt Let al. Catecholamine surge and metabolic adaptation in the newborn after vaginal delivery and caesarean section. Acta Paediatr. Scand. 1984; 73:602-9. 41. Avery ME, Cook CD. Volume-pressure relationships of lungs and thorax in fetal, newborn, and adult goats.J. AppL Physiol. 1961; 16:1034-8. 42. Vilos GA, Liggins GC. Intrathoracic pressures in fetal sheep. J. Dev. Physiol. 1982; 4:247-56. 43. Harding R, Bocking AD, Sigger JN. Upper airway resistances in fetal sheep: the influence of breathing activity. J. Appl. Physiol. 1986; 60:160-5.

44. Davey MG, Johns DP, Harding R. Postnatal development of respiratory function in lambs studied serially between birth and 8 weeks. Respir. Physiol. 1998; 113:83-93. 45. Bland RD, Hansen TN, Haberkern CM etal. Lung fluid balance in lambs before and after birth. J. Appl. Physiol. 1982; 53:992-1004. 46. Shannon JM, Jennings SD, Nielsen LD. Modulation of alveolar type II cell differentiated function in vitro. Am. J. Physiol. 1992; 262:L427-36. 47. Danto SI, Shannon JM, Borok Z et al. Reversible transdifferentiation of alveolar epithelial cells. Am. J. Respir. Cell Mol. Biol. 1995; 12:497-502. 48. Gutierrez JA, Gonzalez RF, Dobbs LG. Mechanical distension modulates pulmonary alveolar epithelial phenotypic expression in vitro.Am. J. Physiol. 1998; 274:L196-202. 49. Alcorn D, Adamson TM, Lambert TF et al. Morphological effects of chronic tracheal ligation and drainage in the fetal lamb lung.J. Anat. 1977; 123:649-60. 50. Flecknoe S, Harding R, Maritz G et al. Increased lung expansion alters the proportions of type I and type II alveolar epithelial cells in fetal sheep. Am. J. Physiol. 2000; 278:L1180-5. 51. Flecknoe SJ, Wallace MJ, Harding R et al. Determination of alveolar epithelial cell phenotypes in fetal sheep: evidence for the involvement of basal lung expansion. J. Physiol. 2002; 542:245-53. 52. Flecknoe SJ, Wallace MJ, Harding R et al. Changes in alveolar epithelial cell proportions before and after birth. Am. J. Respir. Crit. Care Med. 2002; 165:A643. 53. Walker AM, Ritchie BC, Adamson TM etal. Effect of changing lung liquid volume on the pulmonary circulation of fetal lambs.J. Appl. Physiol. 1988; 64:61-7. 54. Hooper SB. Role of luminal volume changes in the increase in pulmonary blood flow at birth in sheep. Exp. Physiol. 1998; 83:833-42. 55. Fuhrman BP, Smith-Wright DL, Kulik TJ etal. Effects of static and fluctuating airway pressure on intact pulmonary circulation.J. Appl. Physiol. 1986; 60:114-22. 56. Fuhrman BP, Smith-Wright DL, Venkataraman S etal. Pulmonary vascular resistance after cessation of positive end-expiratory pressure. J. Appl. Physiol. 1989; 66:660-8. 57. Roos A, Thomas LJ, Nagel EL et al. Pulmonary vascular resistance as determined by lung inflation and vascular pressures. J. Appl. Physiol. 1961; 16:77-84. 58. West JB. Pulmonary blood flow and metabolism. In: West JB (ed.) Physiological Basis of Medical Practice, 12th edn. Williams & Wilkins, Baltimore. 1989; 529-536. 59. Heymann MA. Control of the pulmonary circulation in the perinatal period. J. Dev. Physiol. 1984; 6:281-90. 60. Lipsett J, Hunt K, Carati C et al. Changes in the spatial distribution of pulmonary blood flow during the fetal/neonatal transition: an in vivo study in the rabbit. Pediatr. Pulmonol. 1989; 6:213-22.

61. Lipsett J, Gannon B. Regional cycles of perfusion and nonperfusion in the lung of the term fetal rabbit. Pediatr. Pulmonol. 1991; 11:153-60. 62. Teitel DF, Iwamoto HS, Rudolph AM. Changes in the pulmonary circulation during birth-related events. Pediatr. Res. 1990; 27:372-8. 63. Iwamoto HS, Teitel DF, Rudolph AM. Effects of lung distension and spontaneous fetal breathing on hemodynamics in sheep. Pediatr. Res. 1993; 33:639-44. 64. Harding, R. Fetal breathing movements. In: Crystal RG, West JB, Weibel ER etal. (eds) The Lung: Scientific Foundations, 2nd edn. Raven Lippincott, New York. 1997; 1655-63. 65. Dawes GS, Fox HE, Leduc BM et al. Respiratory movements and rapid eye movement sleep in the foetal lamb. J. Physiol. 1972; 220:119-43. 66. Kitterman JA. Arachidonic acid metabolites and control of breathing in the fetus and newborn. Semin. Perinatol. 1987; 11(1):43-52. 67. Wallen LD, Murai DT, Clyman RI et al. Effects of meclofenamate on breathing movements in fetal sheep before delivery. J. Appl. Physiol. 1988; 64(2):759-66. 68. Koos BJ, Maeda T, Jan C. Adenosine A(1) and A(2A) receptors modulate sleep state and breathing in fetal sheep. J. AppL Physiol. 2001; 91:343-50. 69. Adamson SL. Regulation of breathing at birth. J. Dev. Physiol. 1991; 15:45-52. 70. Sawa R, Asakura H, Power GG. Changes in plasma adenosine during simulated birth of fetal sheep. J. Appl. Physiol. 1991; 70(4):1524-8. 71. Crossley KJ, Nicol MB, Hirst JJ etal. Suppression of arousal by progesterone in fetal sheep. Reprod. Fertil. Dev. 1997; 9:767-73. 72. Thorburn GD. The placenta and the control of fetal breathing movements. Reprod. Fertil. Dev. 1995; 7:577-94. 73. Jansen AH, Chernick V. Onset of breathing and control of respiration. Semin. Perinatol. 1988; 12(2):104-12. 74. Tai TC, MacLusky NJ, Adamson SL. Ontogenesis of prostaglandin E2 binding sites in the brainstem of the sheep. Brain Res. 1994; 652:28-39. 75. Kuipers IM, Maertzdorf WJ, De Jong DS et al. Initiation and maintenance of continuous breathing at birth. Pediatr. Res. 1997; 42:163-8. 76. Wong KA, Bano A, Rigaux A et al. Pulmonary vagal innerration is required to establish adequate alveolar ventilation in the newborn lamb. J. Appl. Physiol. 1998; 85:849-59. 77. Johnson P. Physiological aspects of regular, periodic and irregular breathing in adults and in the perinatal period. In: Von Euler C, Lagercrantz H (eds) Central Nervous Control Mechanisms in Breathing. Pergamon Press, Oxford. 1979; 337-51. 78. Harding R. State-related and developmental changes in laryngeal function. Sleep 1980; 3:307-22.

INTRODUCTION

The process of lung aging begins at birth. Development of the respiratory system for a variety of mammalian species has been analyzed both anatomically, as well as physiologically, during early postnatal life. 1-3 However, few studies have quantitatively examined structural changes that occur with aging. Descriptions of age-related alterations in the respiratory system that do exist 4-6 have focused almost exclusively on the gas exchange portions of the lungs. Less information is available for detailed description of cellular and structural changes in the aging process of the tracheobronchial tree. The majority of data related to lung aging is typically confined to brief descriptive observations of control animals through lifetime toxicity studies. 7'8 This chapter will cover aspects of normal aging in a number of mammalian species, focusing primarily on the mouse, rat, and dog. Aging of the human lung will be covered in Chapter 28. Lifespan characteristics for each of these species is different, with the mouse having a median lifespan of 29-30 months, the rat, 30-34 months, the dog, 12-14 years, and the human, 72-78 years. These striking differences in lifespan are likely to have a significant impact on the resultant aging process within the respiratory system of each species. However, there also exists a multitude of morphologic features in aging that cross boundaries for all species. These include postnatal alveolarization of the lungs during early childhood development that is complete in small laboratory species within four to six weeks of birth, in dogs within the first year of life, and in humans within the first eight years of life. Thinning of alveolar septa within the gas exchange portions of the lung is a common characteristic across all species. Pores or fenestrations within alveolar The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

walls are also present in all species. These anatomical structures are thought to serve in cross-collateral ventilation between adjacent alveoli. They may also serve to facilitate the migration of cells such as macrophages within the airspaces, in a short circuit path from one alveolus to another. These pores appear to enlarge during the aging process and may play an important role in the development of an emphysematous condition in the lungs with larger and greater numbers of fenestrations and loss of alveolar septal walls with age. These characteristics have been studied in detail in the mouse, dog, and human. Airspace enlargement is a characteristic feature of aging in all species; however, it is unclear whether such enlargement is always accompanied by the destruction or partial loss of alveolar wall structures. Nevertheless, these general characteristics present in each species described in this chapter can serve to better understand the process of lung aging over the normal lifespan. The most vital function of the lungs is gas exchange. To enable the most efficient exchange of 0 2 and CO 2 air must be brought into close proximity to blood as it passes through the lungs. That portion of the lungs involved in gas exchange represents 80-90% of the total lung volume. Billions of cells in the lungs are arranged to form a delicate, yet sturdy and highly vascular air-tissue interface creating a surface area 25 times greater than that of the external surface of the body. This tissue barrier separates the inspired air from the blood by less than one micrometer. It is easily deformed by the passage of blood through the underlying capillary bed, causing bulging of the walls into the airspace with the outline of red blood cells easily visible through the thin alveolar partitions (Fig. 15.1). This highly efficient surface design for gas exchange is limited to the small space enclosed by the bony thorax and muscular, Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

Fig. 15.1. Alveolar septum from the lungs of a 5-month-old Fischer 344 rat. A fibroblast (F) can be seen at the junction of the septa. A pocket (*) of interstitial matrix is present, but most of the alveolar septum is composed of a thin air-to-blood tissue barrier (bar, 3 l.tm). (Reproduced with permission from Pinkerton KE, Barry BE, O'Neil JJ et al. Morphologic changes in the lung during the life span of Fischer 344 rats. Am. J. Anat. 1982; 164:155-74.)

dome-shaped diaphragm. A highly ordered airway branching beginning at the carina of the trachea gives rise to millions of alveoli lining several generations of alveolar ducts and sacs to bring about this exponential increase in area within a tightly packaged volume. The process of alveolarization during early life leads to the formation of approximately 80% of all alveoli postnatally. However, in later life the number of alveoli may be reduced through a destructive process, with loss of alveolar septal wall surface area and/or the rearrangement of essential extracellular matrix components such as collagen and elastin; these processes allow the alveoli to become stretched and shallow leading to distention of alveolar ducts, referred to as alveolar ductasia. These phenomena are discussed in this chapter as well as in Chapter 28.

AGING, BODY IN MAMMALS

MASS

AND

THE

LUNGS

For the mammalian lung, a strong allometric relationship exists between body mass and the following features: lung volume, lung capillary volume, alveolar surface area, and pulmonary diffusing capacity. 9'1~ Postnatal development is associated with significant increases in body mass and lung volume. For the mouse, the most dramatic increase in body mass occurs during the first month of life. After 2 months, weight increases slowly, but a slight increase occurs up to 19 months of age, usually followed by a decrease after 28 months of age (Table 15.1). For rats, the two most commonly used strains for lifetime studies are SpragueDawley and Fischer 344 rats. The popularity of this species

for lifetime studies is attributed to their small body size, simple housing requirements, as well as a relatively short lifespan of 30-36 months. The Fischer 344 rat has been used extensively because of a slow increase in body mass with age and resistance to pulmonary disease (Fig. 15.2). Dogs cover a vast span in body size and therefore have pulmonary size characteristics that cover a wide range of values compared with mice and rats. Lung characteristics such as airspace volume or alveolar surface area may span at least two orders of magnitude in dogs. However, the lungs of dogs possess many characteristics similar to the human lung, including the presence of respiratory bronchioles, anatomical structures lacking in the lungs of mice and rats. Lung physiology and tissue structure in dogs also change in a similar fashion to humans during the process of lung growth and development. 12 For the human, lung development continues beyond birth, with over 80% of alveoli formed postnatally. This process of alveolarization is thought to be complete by eight years of age. However, it has also been stated that alveolarization may end as early as two years of age. During adolescence, the lungs continue to expand to the dimensions of the thoracic cavity and increase physiological capacity. By the age of 18, lung growth is considered to be complete. The following sections will cover normal aging in the mouse, rat, and dog, with an emphasis on lung anatomy.

THE

MOUSE

Lifespan characteristics of the mouse The white-footed mouse maintained under barrier conditions can survive for up to 8 years, 13'14 the median lifespan being 4.5-6 years. 15 Mouse strains most commonly used in

research, including BALB/c, Swiss-Webster, and C57BL/6, have a shorter median lifespan of 29-30 months. The life span of laboratory mice depends on the strain, nutritional status, and environment in which they are maintained. Special mouse strains have been established, including the senescence-accelerated mouse (SAM), a murine model of rapid aging. 16 Both prone (SAM-P/I) and resistant (SAM-R/I) strains have been developed and have median lifespans of 11.9 and 17.5 months, respectively. 16

The aging mouse lung The majority of studies using mice begin with animals at 1-2 months of age. Studies on the process of pulmonary aging rarely extend beyond 12 months of age. Therefore, the amount of information available on the effects of aging on the structure and function of the lungs in mice, including age-related changes of the airways, vasculature, nerves, lymph vessels, or immune system is limited. Selected changes in the pulmonary parenchyma of the mouse from 1 to 28 months of age have been established. 2'17'1s Studies with SAM have prompted a growing interest in this strain as a potential model of the aging process. Senescence in this mouse strain is marked by behavioral changes, hair lOSS,19 senile amyloidosis, 2~ cataracts, 19 and osteoporosis. 21 Pulmonary studies in SAM have noted accelerated changes in the degree of pulmonary hyperinflation similar to that noted in other mouse strains at older ages. 22'23 Therefore, SAM may provide insights into the aging process in mouse lungs over a shorter period of time, as compared with other conventional mouse strains.

Changes in lung volume of mice during aging Changes in total lung volume with increasing age in mice of differing strains are given in Table 15.2. The volume of the lungs continues to increase in a significant age-dependent

400

MALES

~,

300

E

,~.s ""

o

200

""

FEMALES

. . . . - - - - --

>. e~ 0 m

loo

&.....

A

1 115

5

14

J

26

AGE (months)

Fig. 15.2. Changes in body weight with increasing age in male and female Fischer 344 rats. Each datum point represents the mean +SD of 4 animals. Asterisk on line connecting different age groups denotes significant change in body weight (p< 0.05). (Reproduced with permission from Pinkerton KE, Barry BE, O'Neil JJ et al. Morphologic changes in the lung during the life span of Fischer 344 rats. Am. J. Anat. 1982; 164:155-74.)

manner to age 25-28 months. 17 These age-related increases occur in an approximately linear manner throughout the lifespan of the mouse (Table 15.2). Measurement of lung volume changes is based on a simple technique that utilizes intratracheal instillation of fixative at a standard pressure. For the measurements in Table 15.2, the lungs were removed from the thoracic cavity and degassed before instillation of the fixative at 25-30 cm I-I20; once fixed, lung volume was measured by liquid displacement. 24 Fixed lung volume can serve as a simple reference for measurements of size, volume, and surface area of those structural components present in the lungs by applying morphometric techniques. Intratracheal instillation of fixative at a standard pressure is one of the most accurate and reproducible means of inflating excised lungs. However, in the absence of the chest wall, the increased compliance of pulmonary tissues in excised old mouse lungs could allow distention beyond those volumes permitted within the thoracic limits of the animal. However, for the measurements in Table 15.2, special care was taken to ensure that fixation pressures did not exceed 15 cm H20 after an initial inflating pressure of 25cm H20. Therefore, subsequent measures of parenchymal airspace volume and surface area, as determined using morphometric techniques on these tissues, can be considered as a true reflection of changes due to aging in the lungs of each mouse strain.

in the mouse. Only studies of mice aged 2-3 months have reported the number of epithelial cells per mm of basal lamina in the trachea to be approximately 215. 25 The epithelial cell types in the trachea at this age consist of 10% basal cells, 39% ciliated cells, 49% Clara cells, and 2% unknown cells. The density of epithelial cells does not change in second and third airway generations compared with the trachea. The proportion of epithelial cell types also remains constant. Within the mainstem bronchi of the mouse, basal cells constitute 4% of the total population, ciliated cells 47%, Clara cells 46%, and unknown cells 3%. 25 Epithelial cell density in airway generations 4-6 in the mouse is slightly reduced, with 199 cells per mm of basal lamina. The proportion of epithelial cell types shifts, with basal cells accounting for 1%, ciliated cells 36%, Clara cells 61%, and unknown cells 2%. Epithelial cells of the bronchioles, including the terminal bronchiole, consist of a simple cuboidal to columnar epithelial layer formed by ciliated cells and Clara cells. The relative proportion of Clara cells in the bronchiole is as high as 80% and ciliated cells constitute the remaining cells of the epithelial airway lining. Clara cells within the terminal bronchiole of the mouse have dome-shaped apical surfaces protruding into the lumen of the airway and numerous secretory granules and mitochondria found within the apical cytoplasm. Little is known about the effects of aging on the structure or function of Clara cells in the mouse.

Tracheobronchial airways of aging mice

Lung parenchymal structure in aging mice

In the trachea and proximal bronchi of the mouse, the epithelial cells form a pseudostratified layer, with a staggered arrangement of nuclei located above the basal lamina. The airway epithelium is composed of three primary cell types: basal cells, ciliated cells, and secretory cells. In the mouse, secretory cells present in the trachea form the nonciliated bronchiolar epithelial cells, or Clara cells. In the mouse, the location of these cells within the trachea is unique because Clara cells are typically found only in more distal bronchioles of the airways in most species. Little information is available on epithelial cell populations of the tracheobronchial tree during the aging process

The parenchyma of the lungs is composed of alveoli, alveolar ducts, and alveolar sacs. These structures form - 90% of the total lung volume in the mouse. 17 The alveoli represent the smallest anatomical unit involved in gas exchange and are composed of the airspace bounded by the alveolar wall and its opening into the alveolar duct. The alveolar duct is formed by the airspace shared in common with alveoli opening along a common channel created by the tissue ridges forming the mouth opening of individual alveoli. The branching of alveolar ducts to form discrete alveolar duct generations begins at the bronchiole-alveolar duct junction (BADJ) and ends 3-5 generations away as a blind

alveolar sac formed by several alveoli. The total alveolar surface area formed by these structures in the aging mouse is depicted in Fig. 15.3 for BALB/c mice and two strains of senescent mice (SAM). The proportion of air volume found in ducts and alveoli is given in Table 15.3. Total volume of alveolar air is approximately 0.5 ml at 1 month and steadily increases until 28 months of age. The age-related increase in air volume is more dramatic in young animals compared with older animals. The total volume of air within alveolar ducts and sacs increases in a consistent and significant degree with increasing age in mice (Table 15.3). A 3-fold increase in the air volume of the ducts and sacs occurred from age 1 to 28 months. The total volume of the alveolar wall (Table 15.3) was not significantly changed from 1 to 9 months. In older animals (19 months), alveolar wall volume was significantly greater than in younger animals. An increase in total alveolar surface area (Fig. 15.3) in combination with a continual increase in the total air volumes of the alveoli, ducts, and sacs (Table 15.3) strongly

suggest that the aging process in the mouse lung results in a hyperinflated lung with significant airspace distention. Although there is some evidence for the loss of alveolar surface area in SAM mice with increasing age, there is no loss in alveolar surface area in the BALB/c mouse during the aging process. Interalveolar pores form a communication between adjacent alveoli. Changes in the size and frequency of these pores are best detected using scanning electron microscopy on critical point dried lung tissues. In mice, alveolar pores or wall fenestrations increase in frequency with age. At 1 month, these pores are relatively sparse but increase in size with age. The frequency of interalveolar pores per alveolus varies, depending on location within the lung parenchyma. TM Subpleural and peribronchiolar regions appear to have alveoli that contain higher numbers of interalveolar pores compared with parenchymal tissues in other regions of the lungs. The number of pores per alveolus more than doubles between 1 month and 28 months of age (Fig. 15.4). The total area of interalveolar pores in the aging BALB/cNNia

Fig. 15.3. Alveolar surface area in aging strains of mice. (Reproduced with permission from Pinkerton KE, Cowin LL, Witschi H. Development, growth, and aging of the lungs. In: Mohr U, Dungworth DL, Capen CC et al. (eds) Pathobiology of the Aging Mouse (Volume 1), 1996, pp. 261-72. Washington, DC: ILSI.)

Fig. 15.4. Changes in the number of interalveolar pores per alveolus with aging in the BALB/cNNia mouse. (Reproduced with permission from Pinkerton KE, Cowin LL, Witschi H. Development, growth, and aging of the lungs. In: Mohr U, Dungworth DL, Capen CC et al. (eds) Pathobiology of the Aging Mouse (Volume 1), 1996, pp. 261-72. Washington, DC: ILSI.)

mouse increases throughout life. From I month to 28 months of age, the total area of alveolar pores increases more than 4-fold (Table 15.4). A rapid increase in the number of interalveolar pores early in life has been described in both mice is and dogs. 26 Interalveolar pores are extremely rare during the first 10 days of life in mice, but rapidly increase in number and size after day 14. 27 The increase in the total area of interalveolar pores, as well as the number of interalveolar pores per alveolus, corresponds to the increase in total alveolar surface area during the lifespan of the mouse. Rapid enlargement of the lungs after birth may account in part for the formation of some pores; however, degenerative processes during aging may also be responsible for an increase in the total interalveolar pore surface area as well as the number of interalveolar pores per alveolus. Pore enlargement may occur as a result of the rupture of tissue strands between adjacent pores, especially in the senescent animal. A process of fenestration in the lungs has been described that would allow

these tissues to attenuate and rupture between intervening capillaries. 28 This process could also occur in the aging mouse lungs. The proportion of alveolar wall formed by pores is about 3.5% at 1 month, 5.9% at 2 months, and 8.5-9% in mice older than 9 months. A decrease in pulmonary elasticity with aging has been verified physiologically. The morphological and chemical basis for changes with age in elastic tissue and the organization of the elastic fiber network throughout the lungs and alveoli is not well understood. A number of investigators have examined elastin content in aging mouse lungs by morphology through measurements of total fiber length. 5'17'22 They found aging occurs in the absence of changes in total elastic fiber length, despite an increase in pulmonary volume (Tables 15.2 and 15.3). However, if elastic fiber length is normalized to pulmonary volume, an age-related decrease in elastic fibers was noted in BALB/c mice. 17 This decrease in elastic fibers was associated with an increase in the static compliance of excised lungs of aged mice, due to a progressive loss of elastic recoil pressure. If an increase in pulmonary compliance could occur without the destruction of elastic fibers, an increase in the total elastic fiber length would be expected) However, biochemical analysis of elastic tissue content demonstrated a loss of elastic fibers in aging lungs. During biochemical analysis, special care must be taken to separate pseudoelastin, a form of elastin in human lungs that increases with age, and elastin, or the actual amount of elastin will be overestimated. Pseudoelastin has not been found in mice. Therefore, it has been suggested that this absence of pseudoelastin fibers accounts for the decrease in elastin content in the aging BALB/c mouse lung. 5 Studies of histological changes in lung elastic fibers in SAM mice have found no evidence of destruction of the alveolar wall or elastic fibers during aging. Physiological studies have also demonstrated lung compliance in these mice at age 10 months to be significantly greater than that at 2 months of age. 22 Therefore, age-related changes in lung distension in SAM occur in a manner similar to that noted for other strains of mice. In contrast, for mice of the SAMP/I strain, changes in lung hyperdistention occur in an accelerated manner compared with the resistant strain (SAM-R/l) or other mouse strains. Therefore, SAM-P/I strain of mice could prove useful in the study of senilerelated lung hyperinflation.

Macrophages and lung aging (mouse) Aging is thought to be associated with a decline in immune function. Changes in macrophage function can be an important component in compromised lung defense. Phagocytic cells collected by bronchoalveolar lavage are useful to study the characteristics of these cells. In comparing C57BL/6 mice of differing ages, the number of cells recovered by bronchoalveolar lavage was found to be greater at 26 months than at 1.5 months (Table 15.5). 29 The cell differentials at both ages were similar, with 95-96% of macrophages recovered from the lungs. Neutrophils represented a small fraction (0.1%) in the youngest animals and a slightly higher fraction in older animals (1.8%). Lymphocytes in both young and old animals were approximately 2%. This slight increase in the number of neutrophils and lymphocytes within the lungs of aged animals compared with young animals may reflect a subtle change in the immune system of these animals or the need for greater numbers of cells in old versus young animals to maintain the proper sterility of the lungs. The difference between young and old mice in the numbers of cells recovered by bronchoalveolar lavage may also be a simple reflection of the aging process, with greater numbers of phagocytic cells present in the lungs of old mice. The ability of cells to phagocytize particles has been tested, and has shown that a greater proportion of cells from 26-month-old mice were unable to phagocytize latex spheres than cells from lungs of 1.5-month-old mice. 29

Antioxidant enzyme activity in the aging mouse Antioxidant enzymes are an important means of protecting cells from damage due to gases and particles that have the oxidizing capacity to alter vital cellular components. Superoxide dismutase (SOD), catalase (Cat), and the glutathione (GSH) system all serve to protect against the toxic effects of oxidants. GSH in sufficient concentrations can adequately detoxify oxidants through conjugation. However, if glutathione concentrations become depleted, toxic intermediates can form and injure pulmonary cells. Pulmonary glutathione

concentrations have been determined over the lifespan of male C57BL/6 mice, at ages of 3, 6, 12, 26, and 31 months. GSH concentrations decreased by 30% in the lungs of aged mice, whereas GSSG (oxidized glutathione) cysteine and cystine concentrations remained unchanged. Depletion of pulmonary GSH by injection of acetaminophen demonstrated that young (3--6 months) and mature (12 months) mice recovered hepatic GSH levels more efficiently than senescent mice (31 months), but no differences were noted in the lungs. 3~ These findings suggest that detoxification capacity decreases as age increases in the mouse. However, no information is available regarding cell numbers or cell types responsible for maintaining GSH concentrations in the lungs of mice during the aging process.

Conclusions for lung aging in the mouse Little information is available to adequately describe the effects of aging in mouse lungs. Although changes in air space and tissue volumes and alveolar surface area with aging are known, less is known for cell populations lining the lung airways or forming the alveolar structures of the pulmonary parenchyma over the lifespan of the mouse. A number of studies have implied that lung aging in mice is associated with decreases in specific functional and structural parameters. These include increases in the phagocytic cell populations present in the lung airspaces, but decreased ability to engulf foreign particles. Decreases in antioxidant defense systems have also been noted in the lungs of aging mice. From a structural perspective, hyperinflation of the lungs and increases in interalveolar pore size and number are key features of the lung aging process in mice. Future knowledge of changes in pulmonary cell number, type, distribution, and function with aging would greatly increase our understanding of their impact on pulmonary physiology, metabolism, and immunity in the mouse. Age-related changes could significantly affect the normal function of the lungs as well as greatly increase host susceptibility to injury.

TH E RAT Lifespan characteristics of the rat As previously mentioned, aging has been examined in a number of rat strains, but the two most commonly used for lifetime studies are Sprague-Dawley and Fischer 344 rats. The lifespan characteristics of the Fischer 344 rat have been well-documented. 4'31-38 The majority of rats and mice maintained under barrier conditions show no gross pathological changes of the lungs. The pathology of nasal, laryngeal, tracheal, and respiratory systems that may be associated with aging in the rat has been reviewed. 39

The aging rat lung Development of the respiratory system in the rat has been analyzed morphometrically during early postnatal life 1-3 but few studies have quantitatively examined structural changes related to aging. Descriptions of age-related alterations in

the respiratory system which do exist 4 have focused almost exclusively on the gas exchange portions of the lungs. Less information is available on cellular and structural changes in the tracheobronchial airway tree with age with the exception of the postnatal changes in the nonciliated bronchiolar epithelial (Clara) cell. 2'3 The majority of data related to lung aging in the rat are confined to brief descriptive observations of control animals in lifetime toxicity studies. 7'8

The tracheobronchial tree and epithelium of the aging rat The tracheobronchial epithelium of adult rats varies in terms of the types of cells present and the relative proportions of specific cell types throughout the conducting airway tree. In the trachea, four cell types have been identified: basal cells, serous cells, ciliated cells, and mucous goblet cells. 4~ In contrast, the epithelium of bronchi and bronchioles consists of ciliated cells and nonciliated bronchiolar epithelial (Clara) cells. 4~ The composition of epithelial cells also varies from proximal to distal airway generations with postnatal development. 4~ The transformation of tracheal epithelial cells in the rat from the perinatal period through to adulthood appears to be continual. At birth, the rat trachea contains some mature ciliated cells, obvious secretory cells, and the beginning of basal cell differentiation. 4~ Ciliogenesis begins at about 80% gestation in the rat. Nonciliated secretory cells are obvious in the tracheal epithelium at about 90-95% gestation. We have recently evaluated the general characteristics of the airway epithelium in the aging rat. Epithelial cells were examined in three different airway generations of the left lung of male Fischer 344 rats at 5 and 22 months. 45 Fig. 15.5 shows the 3 airway levels selected (A, B, and C) which are designated as cranial, central, and caudal bronchi, respectively. At each site, ciliated, non-ciliated and basal cells were identified, and the average volume density (cell volume/basal lamina surface area) of each cell type determined (Table 15.6). No significant differences in the

Fig. 15.5. Location of tissue samples taken from the rat lung. (A) Silicone cast of the tracheobronchial airway tree. Lung casts served to standardize sampling. Samples of terminal bronchiole-alveolar duct junctions were taken from the three regions (cranial, central, and caudal) indicated with letters and arrows within the figure. (B) Mediastinal half of a fixed, microdissected rat lung. Note how closely the pathways and sampling regions match the silicone cast shown above. The same level of airway in each figure is designated by A, B and C. (Reproduced with permission from Pinkerton KE, M~nache MG, Plopper CG. Consequences of prolonged inhalation of ozone on F344/N rats: collaborative studies, Part IX. Changes in the tracheobronchial epithelium, pulmonary acinus, and lung antioxidant enzyme activity. Health Effects Institute Research Report No. 65, 1995, pp. 41-98.)

abundance of each cell type were noted between 5 and 22 months; total epithelial cell volume also remained constant. No appreciable differences were noted between each of the three bronchial regions examined. These findings suggest that few changes occur during aging, with a lack of significant shifts in the proportion of epithelial cell types in the tracheobronchial tree of the aging rat. However, far less is known regarding changes in the molecular and biochemical functions of cells with aging.

Lung parenchymal structure in the aging rat Total alveolar airspace volume of the lungs progressively increases over a 2-year period of growth in Fischer 344 rats (Fig. 15.6). Airspace volume was measured in lungs fixed following airway instillation of 2% glutaraldehyde at 20 cm H20; 4'48 prior physiological tests showed that the lungs were fixed at 75% of total lung capacity. 4'49'5~The marked increase in air space volume from I week to 5 months of age

Fig. 15;.6. Total airspace volume (cm3) in glutaraldehyde-fixed lungs of male and female Fischer 344 rats. Each point represents the mean (+ SD) of four animals. Asterisks indicate significant changes between consecutive age groups (p < 0.05). (Reproduced with permission from Pinkerton KE, Vincent R, Plopper CG et al. Normal development, growth, and aging of the lung. In: Mohr U, Dungworth DL, Capen CC (eds) Pathobiology of the Aging Rat (Volume 1), 1992, pp. 97-109. Washington, DC: ILSI.)

was associated with a significant increase in alveolar number. The peak rate of formation of alveoli in the neonatal rat has been estimated to be 1000 per minute. 51 Light microscopic examination of large airways, terminal bronchioles, large blood vessels, alveoli, and alveolar septa demonstrated no detectable differences from 5 to 26 months of age. Rats aged 26 months appeared to have slightly enlarged alveolar ducts compared with younger animals, but otherwise were indistinguishable from the younger animals (Fig. 15.7). The proportion of the lungs forming the lung parenchyma was 0.81 to 0.82 in young and old animals, respectively. 1'4 Epithelial and capillary surface areas of the gas exchange regions of the lungs are given in Table 15.7. Between 1 and 6 weeks of age the surface area formed by epithelial type I cells increased 6-fold, while epithelial type II cell surface area increased 3-fold. The squamous type I cell covered more than 95% of the total alveolar surface, while the cuboidal type II cell covered the remaining 5%. The alveolar type III cell is a rather rare epithelial cell type within the alveoli and contributes very little to the total alveolar surface area; they are found most frequently within alveolar regions near bronchiole-alveolar duct junctions. 42 From 6 weeks to 5 months of age alveolar surface area almost doubled. From 5 to 26 months, total alveolar surface area remained unchanged. Since airspace size increased by 50% from 5 to 26 months in the absence of any significant change in alveolar surface area, alveolar duct enlargement in older animals could explain, in part, how airspace volume increased without a loss of surface area within the lung parenchyma. Enlargement of alveolar ducts was not quantitatively confirmed, but emphysematous changes (alveolar wall destruction) appeared to be absent in the aging Fischer 344 rat. 35'52'53 Changes in capillary surface area with age were proportional to changes in alveolar surface area (Table 15.7). The capillaries within the alveolar septa of the neonatal rat undergo a fascinating transformation from a double capillary system to a single capillary system associated with the growth of secondary alveolar septa to form new alveoli. This reorganization of the pulmonary vasculature is complete by 3 weeks of age. 1'4 From 5 to 26 months of age the

Fig. 15.7. Photomicrographs of the lung parenchyma from Fischer 344 rats aged (A) 5 months and (B) 26 months. A slight increase in alveolar size can be noted in the lungs of 26-month-old rats compared with 5-month-old rats. Scale bar is 100 ~tm.

total surface area of the capillary bed remained unchanged in both male and female rat lungs (Table 15.7).

Alveolar tissue compartments in the aging rat During the first months of life in the rat, alveolar tissue volumes increase dramatically. However, little change is noted in tissue volumes from 5 to 26 months, with the exception of the noncellular component of the interstitium (Table 15.8). The volume of the capillary bed also increases dramatically during the first 5 months of life, but remains relatively unchanged from 5 to 26 months. Alveolar tissue volumes for epithelial, interstitial, and endothelial compartments of the lung parenchyma are presented in Table 15.8. Cell number for each major alveolar cell type is given in Table 15.9 and morphometric characteristics are presented in Table 15.10. The lung parenchyma consists of three tissue compartments, the epithelium, the interstitium, and the endothelium. Alveolar macrophages form a unique tissue compartment of individual cells which freely migrate along the surfaces of alveoli and airways and which are also present within the pulmonary connective tissues as interstitial macrophages. The composition of the alveolar epithelium in the rat changes dramatically during postnatal development (Table 15.8). The volume of type I epithelium increased more than 6-fold from 1 week to 6 weeks, while the type II epithelium displayed a more modest 3-fold increase during the same period. The total surface area of type I and type II cells also increased more than 5-fold from 1 to 6 weeks (Table 15.8). The ratio of type II to type I cells in the lungs at 6 weeks was 1.8, but it decreased to 1.0 by 26 months 4 (Fig. 15.8). The significance of the reduction in type II to type I cell ratio is unknown, but it may contribute to an

altered secretory response in type II cells of old rats since a greater surface area must be served per type II cell to form the surfactant lining layer of the lungs compared with that in younger animals. The type II cell undergoes dramatic changes during the perinatal period. Just prior to birth, the rat type II cell is heavily laden with glycogen and first acquires its characteristic lamellar bodies 48-72 h prior to parturition. From the last day of gestation to a few hours later, the cellular content of glycogen drops from about 10% to zero. On the day of birth the cells have polarized their secretory granules (lamellar bodies) towards the basal pole. The putative immediate precursor of the lamellar body contains eccentrically placed vesicles in addition to slips of phospholipid lamellae and is termed a composite body. These are polarized toward the basal side of type II cells throughout the life span of the animal, but the mature lamellar bodies become randomly distributed between 2 and 6 weeks. 54 In contrast, the Clara cell has its secretory granules polarized toward the apical region like many other secretory cells in the body. Another interesting polarization of type II cell intracellular organelles is that of the light and dark multivesicular bodies. Light multivesicular bodies are not rich in lysosomal enzyme and accumulate endocytosed membrane markers most quickly, like endosomes do in other cells. Dark multivesicular bodies are rich in lysosomal enzymes and are basally polarized like composite bodies. There could be functional differences between the two multivesicular body types, but more direct experimental evidence is needed to define them. Type II cellular composition between birth and adulthood differs mainly by a doubling of the volume density of the lamellar bodies and a 50% increase in mitochondrial

Fig. 15.8. Changes in the ratio of alveolar type II cell number to alveolar type I cell number in the lungs of aging male and female Fischer 344 rats. Each datum point represents the mean + SEM of four animals. Asterisk on the line connecting two age groups denotes a significant change in this ratio (p < 0.05). (Reproduced with permission from Pinkerton KE, Barry BE, O'Neil JJ et al. Morphologic changes in the lung during the life span of Fischer 344 rats. Am. J. Anat. 1982; 164:155-74.)

volume density. Neither the appearance nor the size distribution of lamellar bodies changes during postnatal life, but dramatic changes do occur in mitochondria. They shift from an isolated spheroid form to a highly branched interconnected web in the adult cell (Fig. 15.9). The significance of such a shift is unknown but very similar changes occur in other phyla (such as insect flight muscle) and may accompany cell division. Although the notion is untested, it is easy to see how division of the mitochondria would be simplified by the fetal disconnected form. The adult mitochondria have only a small surface area-volume ratio advantage over the spheroid form and it seems unlikely to be the entire advantage of that shape. It seems possible that the extended shape would allow an entire mitochondrion to respond to very focal cellular changes in high energy phosphates.

Fig. 15;.9. Computer-assisted three-dimensional reconstruction of the mitochondria from an alveolar type II cell within the lungs of a 1-day-old neonate (A) and an adult (B) Sprague-Dawley rat. The globular appearance of individual mitochondria in the neonate is in striking contrast to the filamentous branching mitochondria of the adult. (Reproduced with permission from Pinkerton KE, Vincent R, Plopper CG et al. Normal development, growth, and aging of the lung. In: Mohr U, Dungworth DL, Capen CC (eds) Pathobiology of the Aging Rat (Volume 1), 1992, pp. 97-109. Washington, DC: ILSI.)

One more key change occurs in type II cell morphology during postnatal development. At birth, each type II pneumocyte has as many as 200 cytoplasmic extensions, which perforate the basement membrane, and these are termed foot processes. A fraction of these have close apposition with interstitial lipofibroblasts. Although there is an electron-dense cytoplasmic condensation, there is no intra-gap substructure on tilted sections to suggest a true gap junction. Occasional profiles from serial sections show cytoplasmic bridges between the epithelial type II cell and the mesenchymal lipofibroblast. The function of these three structures is unknown but is widely speculated to be related to epithelial mesenchymal interactions. They can be induced with corticosteroid administration in the perinatal period, and decrease by 20-fold in number with maturation of the lung. In response to injury resulting in interstitial pneumonitis in humans, the foot processes proliferate. Over the lifespan of the Fischer 344 rat, the absolute volume of the cellular interstitium of the lung parenchyma does not change (Table 15.8). Interstitial cell number (Table 15.9) and cell size (Table 15.10) were similar at 1 week and 26 months, although the types and ratios of interstitial cell types forming the parenchyma cell pool were markedly different in neonatal pups compared with adult rats (Fig. 15.10). 55 In contrast, the noncellular components of the interstitium demonstrated dramatic volume changes from 1 week to 5 months of age, increasing approximately 10-fold in total volume. From 5 to 26 months of age interstitial volume continued to change at a lower, but significant rate (Table 15.8). Compared to 5-month-old rats, the interstitial matrix increased 18% by 14 months of age and 39% by 26 months (Fig. 15.11). Females also demonstrated a 36% increase in interstitial matrix volume

Fig. 15.10. Upper micrograph: An alveolar septum from the lungs of a 1-week-old Fischer 344 rat. The alveolar surface is covered by type I (I) and type II (11)epithelial cells. Note the presence of capillaries (C) on both sides of the septum and numerous lipid-containing cells (arrow) in the interstitium (bar is equal to 3 lam). Lower micrograph: An alveolar septum from the lungs of a 6-week-old Fischer 344 rat (bar is equal to 3 ~m). (Reproduced with permission from Pinkerton KE, Barry BE, O'Neil JJ et al. Morphologic changes in the lung during the life span of Fischer 344 rats. Am. J. Anat. 1982; 164:155-74.)

from 81 ml at 5 months; a further 89% increase in matrix volume occurred by 26 months. These changes could be a result of two different conditions: (a) pulmonary edema and (b) an increase in the noncellular matrix components of the lung such as collagen or elastin. Changes in the constituents of the interstitial matrix at the level of the proximal alveolar region (PAR) have been examined in Fischer 344 rats. 56 The PAR consists of alveolar tissue sampled in a perpendicular orientation relative to the axis of the terminal bronchiole, approximately 300-400 ~tm down the alveolar ducts. The overall characteristics of the alveolar septum in this region were similar to those of the more distal parenchyma. Morphometric analysis of the PAR revealed that the increase in interstitial matrix volume in the parenchyma could be attributed almost entirely to a thickening of basement membranes and the deposition of collagen fibers. Basement membrane thickness went from

40-45 nm at 4-6 months to 75-80nm at 20-24 months. Similarly, collagen fiber volume, normalized to the surface of the alveolar epithelium, increased by more than 100% during this same period. Basement membrane volume and collagen fiber volume were similar in the proximal alveolar regions and accounted for 50% of the noncellular matrix at 4-6 months and 80% at 20-24 months. By comparison, no change was noted in the relative volume of elastin and remaining acellular space. The volume ratio of collagen fibers to elastin fibers shifted from 3 to 5 in young adults (4-6 months) to 10 in the older animals (20-24 months). The same collagen to elastin fiber ratio was found in Sprague-Dawley rats as in young Fischer 344 rats. 57 It appears that the volume of the ground substance, relative to epithelial surface, is not significantly modified in the lungs of the aging Fischer 344 rats. Therefore, it is unlikely that edema contributes significantly to the increase

Fig. 15.11. The alveolar septum from the lungs of male (upper micrograph) and female (lower micrograph) Fischer 344 rats 26 months of age. A prominent interstitial matrix space (*) is present in the septum. Some cytoplasmic extensions of interstitial cells are also present. Collagen is present within the matrix. In spite of the increase in the interstitial matrix, a thin air-to-blood tissue barrier is still maintained. Magnification is the same as in Fig. 15;.10. (Reproduced with permission from Pinkerton KE, Barry BE, O'Neil JJ et al. Morphologic changes in the lung during the life span of Fischer 344 rats. Am. J. Anat. 1982; 164:155-74.)

in interstitial matrix volume with age. The absence of detectable changes in the volume of elastin is consistent with its low rate of synthesis and slow turnover in adult animals. 58'59 A progressive thickening of basement membranes may be explained by epithelial and endothelial cell turnover if the cells lay down basement membrane material during each cell cycle, and possibly also during interphase. The increase in the volume of collagen fibers documented morphometrically in adult rats confirms our previous morphologic observations 4 and is in agreement with biochemical studies of collagen in the lungs of Lewis rats 6~ and Fischer 344 rats. 62 Morphologic and biochemical evidence supports the notion that there is a continuous deposition of mature cross-linked extracellular collagen in the lung parenchyma of aging rats which can be interpreted as an age-related excess of collagen (or fibrosis). However, it should be noted that qualitative examination revealed no visible change in the overall gross architecture of the matrix in older animals

compared to younger rats. Although speculative, it is conceivable that the dominant stimulus for collagen fiber deposition, through life, is the stress and strain exerted on fibers and translated to fibroblasts. Following this concept, fibroblasts may add fibrils to the existing network while maintaining the general spatial relationship between collagen fibers and other tissue components, resulting in a net enrichment in collagen. It is not clear how this increase in collagen mass and the shift in the ratio of collagen to elastin with age impact on the micro mechanical behavior of the lungs. A logical assumption would be stiffer lungs, but physiologic measurements suggest otherwise. 4'53 Threedimensional reconstructions in concert with physiologic measurements 57'63 could specifically address this issue in the senescent rat lung. An interesting feature of senescent lungs in the Lewis rat 61 is an abrupt 4-fold decrease in the measured rate of gross collagen synthesis between 15 months and 24 months. More than 80% of newly synthesized collagen is apparently

degraded intracellularly in 15-month-old animals and 60% degraded intracellularly at 24 months. The net result is only a small fraction of product being deposited as extracellular cross-linked collagen. It was suggested that maintenance of a high rate of collagen synthesis provides an adaptive capacity that allows the fibroblast to rapidly redirect the procollagen from intracellular degradation pathways towards secretory pathways in response to injury or pathologic insult. 61 In the Fischer 344 rat there was no decline in fibroblast populations in aging lungs. 4 If these observations 61 can be extended to other strains of rats, then a fall in the rate of collagen synthesis may constitute an intrinsic characteristic of the senescent lung fibroblast in vivo, resulting in a compromised ability to repair collagen fibers. If these assumed dynamics of the lung interstitium prove to be true, there are specific and profound metabolic changes that accompany aging which need to be understood to more fully appreciate the nature of the morphologic and functional alterations seen in normal aging and as a consequence of exposure to toxicants in the aged lung. 64 Acute toxicity studies are seldom performed in senescent animals, while chronic studies are usually not extended through to the last third of the lifespan due to low survival rates. Therefore, our present understanding of age-related interstitial changes is severely limited. The endothelium forms the third major tissue compartment of the alveolar septum. The volume of the endothelium in the lungs of Fischer 344 rats increased more than 3-fold from 1 to 6 weeks of age. Beyond 6 weeks of age, total endothelial cell volume did not increase significantly through to 26 months (Table 15.8). The total surface area of the capillary endothelium, like the epithelial surface, demonstrated nearly a 10-fold increase from 1 week to 6 weeks. From 6 weeks to 5 months, capillary surface area increased by an additional 20%. From 5 months to 26 months, the surface area of the capillary endothelium did not change significantly. Endothelial cell volume and surface area were relatively unchanged in male and female rats from 5 months to 26 months of age (Tables 15.7 and 15.8).

Alveolar macrophages in the aging rat Alveolar macrophages form an important defense against inhaled particulates and pathogens in the lungs. In 1-weekold rats, there were approximately 2 million macrophages, and these increased to 9 million by 6 weeks; numbers increased to 20 million by 5 months of age. Although no further increases in alveolar macrophage number were evident at 26 months (Table 15.10), this population of cells is highly dynamic and numbers can change rapidly by recruitment of monocytes and macrophages from the interstitium and the blood 65 and by the in situ proliferation of cells within the lung airspaces. 66

Conclusions for lung aging in the rat The dynamics of lung growth, development, and aging in the rat is a continuous process that involves every tissue compartment of the lungs. Significant changes in aging

adult rats are primarily within alveolar type II cells and the noncellular portions of the interstitium. Because such changes may influence the response of the lungs to inhaled chemical agents and dusts, the age of the rat should be considered in the evaluation of any experimental study. Different responses of the lungs to inhaled pollutants have been noted in young versus old r a t s . 64'67 Such differences may be due to changes within target cell populations and/or alterations in the functional status of cells through the aging process. Although cellular changes are most prominent during postnatal growth and development, modifications in cells continue up to advanced age. Changes in cell number, size, and function associated with aging are likely to impact on lung physiology, metabolism, and immunity. Such changes could significantly alter the normal functions of the lung and its susceptibility to injury. Therefore, an understanding of the aging process in the rat is essential to the evaluation and interpretation of chronic (lifetime) exposures to a variety of substances that present a potential health risk to all mammalian species.

TH E D O G General characteristics of the lungs in aging dogs Morphological changes in the lungs during aging have been studied most extensively in the dog. As discussed above, a strong allometric relationship exists between body mass and several features of the respiratory tract. Table 15.11 illustrates a number of these morphometric characteristics of the lungs for dogs of differing body mass and age. Changes in alveolar size, enlargement of respiratory bronchioles and alveolar ducts, and the accumulation of anthracotic pigment are all common features of the aging process in the dog lung. In a study of the lungs of 20 beagle dogs, aged < 1 year to 10 years, it was found that the primary lesion in older dogs was large accumulations of pulmonary macrophages containing dust and pigment in the walls of respiratory bronchioles and at the mouth openings of alveoli into alveolar ducts. 68 It was observed that these dustladen macrophages became more prominent with increasing age. Focal pneumonitis was also frequently associated with the accumulation of macrophages. It was further noted that the volume of the lungs occupied by alveolar ducts increased. 68 In the airways, aging was associated with larger submucosal glands and a greater degree of calcification of bronchial cartilage. Due to the typical outdoor habitat of dogs, the accumulation of macrophages loaded with dusts and anthracotic pigment is to be expected with age. Although not studied, impaired clearance of particles may also occur with increasing age, leading to greater retention of pigment with age. The relationship of each of these lung features with age were further analyzed using linear regression to determine which parameters were significantly associated with the aging process. 68 Tissue sections taken from a total of 33 blocks from both lungs of each animal were examined without

knowledge of the dog's identity or age. Scores were based on the degree of change relative to the youngest animals observed. Significant age-related changes were identified as increases in pigment surrounding respiratory bronchioles and alveolar duct mouth openings, enlargement of lumenal size of respiratory bronchioles and alveolar ducts, increased abundance of airway submucosal glands, and bronchial calcification of cartilage with age. Two features not changed with age were subpleural airspace size and the degree of focal pneumonitis. 68

The tracheobronchial tree of the aging dog Little information is available to describe or quantify the aging process of the trachea or conducting airways in the lungs of dogs. It is known that body mass is proportional to the dimensions of the trachea in terms of its length to diameter ratio. In most mammalian species, this ratio is approximately 8:1. Compared to other species, the trachea and central airways of the dog are slightly larger due to their need of dissipating body heat by panting. The relatively large central airways also minimize impedance produced by rapid shallow breathing. Age-related alterations of the bronchi and the non-respiratory bronchiole are unknown with the exception of age-related calcification of cartilaginous plates within the bronchi as well as hypertrophic changes of the submucosal glands. 68 No changes have been described for the non-respiratory bronchioles of dogs with aging. The most peripheral airways in the lungs of dogs are formed by respiratory bronchioles and alveolar ducts. In a study of mucus velocity in the trachea of dogs aged from 1 to 15 years,73 it was found that the velocity increased during the first few years of life, followed by a gradual age-related

reduction beginning around 4 years of age (Fig. 15.12). With extrapolation of equivalent dog years to human years, a close correlation has been found between dogs and humans, with similar patterns of decline in tracheal mucus velocity rates with increasing years during adult life. No information is available to correlate changes in flow rates with cellular and/or functional parameters in epithelial cells lining the trachea. Although extensive studies have examined the branching pattern of the tracheobronchial tree in dogs, 74'75 only one study focused on postnatal growth. TMIn that study, resin casts were made of the bronchial tree of 4 Labrador dogs weighing 0.5, 3.4, 7.5, and 30 kg; TM the smallest dog was one week of age, while the ages of the other dogs were not given. The branching pattern of the lungs was ordered by number, mean diameter, and mean length of branches in each generation; these were plotted by order and expressed according to diameter and length. It was found that the diameter and length of similar airway orders ran in a parallel fashion for dogs of varying size. If these measurements were normalized to body weight, they were found to be identical. This correction factor was equal to the cube root of body weight. These observations provide strong evidence that airway branching morphogenesis of the bronchial tree of the dog is complete at birth and postnatal growth reflects a simple expansion of the length and diameter of each airway generation in the absence of any further airway branching. Lung parenchymal structure in the aging dog To define morphological changes of the aging canine lung, the parenchymal characteristics of the lungs of 14 beagle

Fig. 15.12. Tracheal mucus velocity (mean and SEM) for beagle dogs versus age. The fitted function that describes this relation is V(t)= 11 [1/2 exp (-0.9t)] - 0.6t. The available data for humans are also shown after transforming for age. (Reproduced with permission from Pinkerton KE, Murphy KM, Hyde DM. Morphology and morphometry of the lung. In: Capen CC, Benjamin SA, Hahn FF et al. (eds) Pathobiology of the Aging Dog (Volume 1), 2001; 5:43-56. Iowa State University Press, Ames.)

dogs have been analyzed morphometrically. 6 These dogs, which were also used in another study, 68 ranged from 1 to 10 years of age. The volume and surface densities of tissues and air, the volumes of alveoli, alveolar ducts, and alveolar sacs (i.e. endings of the alveolar duct), as well as the number of alveoli per unit volume of lung parenchyma were measured. It was found that the volume density of alveoli decreased with age, while the volume density of alveolar ducts and sacs increased with age. The number of alveoli per unit volume also decreased with age. There was no correlation between the numerical density of alveolar ducts and alveolar sacs with age. Alveolar tissue density also decreased with age. Multiple regression analysis was also used to determine the relationship between stereological parameters of the lung parenchyma, body weight, age, and diffusion capacity in the dog. 6 Analyses were repeated on males and females, as well as the total population. There were no significant multiple correlations among stereological lung parameters, diffusing capacity, body weight, and age. No sex-related differences in the slope of the regression lines relating to morphometric parenchymal lung values and age were noted. In the earlier study of these dogs 68 a significant agerelated increase in alveolar duct profile area was observed; minimal emphysema was also seen, with only occasional signs of fibrosis associated with focal pneumonitis. These observations were confirmed by the later morphometric measurements, 6 showing an increased volumetric density of alveolar ducts associated with decreases in the volumetric density of alveoli and parenchymal tissue, as well as decreases in the numerical density of alveoli, and the surface density of parenchymal tissues. These observations strongly correlate with a prevalence of lung hyperdistention, or ductasia, in the aging process for dogs. Similar findings have also been noted in the human lung, without a significant decrease in the number of alveoli. In ductasia, those alveoli adjacent to enlarged alveolar ducts in respiratory bronchioles decreased in depth, thus maintaining a constant number of alveoli within the lung. 6 The observed decrease in surface density in the lungs of dogs with age can be explained by an increase in lung volume, a loss of interalveolar septa, a rearrangement of the geometry of the lung by alveolar flattening and duct enlargement, or a combination of any of these anatomical changes. Similar changes have been observed in aged human lungs, with a geometric rearrangement of alveoli in alveolar ducts, resulting in an increase in the average interalveolar septal distance. The total alveolar surface area for the lungs decreased as a result of an increase in the average interalveolar septal distance rather than an increase in lung volume. In these dogs, no significant change in total lung capacity with age was observed. 6s The reduction in volume density of alveolar parenchymal tissue could be interpreted as a loss of interalveolar septa. This type of reduction has also been observed in human lungs. Cross collateral ventilation of the lungs within the parenchyma is accomplished in part through communications formed by alveolar pores, as discussed above. A number of studies have examined the alveolar

Fig. 15.13. Average diameter of alveolar pores plotted against age. Each point represents one dog. Closed circles are males, and open circles are females. (Reproduced with permission from Pinkerton KE, Murphy KM, Hyde DM. Morphology and morphometry of the lung. In: Capen CC, Benjamin SA, Hahn FF et al. (eds) Pathobiology of the Aging Dog (Volume 1), 2001; 5:43-56. Iowa State University Press,Ames.)

pores during aging in dogs. 6'26 These pores appear to be absent at birth, but become prominent within the first year of life. 26 The average number and size of these pores remain constant through the first 10 years of life (Fig. 15.13). Prolonged exposure to a number of environmental air pollutants has been shown to dramatically increase the number and size of these pores 6 with extensive fenestration of alveolar walls.

Conclusions for the aging dog lung Aging of canine lungs has many similarities to that seen in human lungs, but the aging process in dogs occurs over a shorter time frame than in humans. The accumulation of dust-laden macrophages as well as lumenal enlargement of alveolar ducts are hallmarks of this aging process. Although some tissue thinning and loss of alveolar septal tissues may occur, the increase in lumenal size of alveolar ducts is in large measure due to the stretching and shallowing of alveolar outpocketings within these regions of the lung parenchyma. Changes in the relative velocity of mucus flow in the trachea with aging also parallels changes observed in the aging human trachea. Little is known regarding the canine aging process at the cellular level for either the airways or gas exchange portions of the lungs. However, a significant advantage of studying dog lungs is their similarity to the human lung in both composition and structure. These similarities may offer unique opportunities to better understand the aging process of the lungs for both dogs and humans.

OVERALL

CONCLUSIONS

Aging is a natural process in the respiratory system. A number of similarities have been noted in all species

associated with aging. An increase in the total alveolar airspace volume is a natural consequence of aging with enlargement of alveolar ducts immediately beyond the terminal and respiratory bronchioles. This enlargement is typically seen in the form of ductasia, with stretching and shallowing of alveoli in affected regions. Although some destruction of alveolar tissue walls may be evident in the aging lung, such changes appear to be due to an increase in the size and frequency of alveolar pores connecting adjacent alveoli. An increase in the n u m b e r of phagocytic cells in the lung airspace is also a c o m m o n finding, with less efficient phagocytic properties. Increased collagen deposition with focal increases in the interstitium also appears to be a consequence of the aging process. Metabolic functions of cells are also likely to be compromised in the lungs as aging advances. A n u m b e r of environmental factors are likely to accentuate all of these changes during the aging process with the greatest consequences likely to be asthma, emphysema, and chronic obstructive pulmonary disease. These topics are more fully covered in Chapter 28.

REFERENCES 1. Burri PH, Dbaly J, Weibel ER. The postnatal growth of the rat lung: I. Morphometry.Anat. Rec. 1974; 178:711-30. 2. Massaro GD, Davis L, Massaro D. Postnatal development of the bronchiolar Clara cell in rats. Am. J. Physiol. 1984; 247:C197-203. 3. Massaro GD, Massaro D. Development of bronchiolar epithelium in rats. Am. J. Physiol. 1986; 250:R783-8. 4. Pinkerton KE, Barry BE, O'Neil JJ et al. Morphologic changes in the lung during the life span of Fischer 344 rats. Am. J. Anat. 1982; 164:155-74. 5. Ranga V, Kleinerman J, Sorensen J. Age-related changes in elastic fibers and elastin of lung. Am. Rev. Respir. Dis. 1979; 119:369-76. 6. Hyde D, Orthoefer J, Dungworth D et al. Morphometric and morphologic evaluation of pulmonary lesions in beagle dogs chronically exposed to high ambient levels of air pollutants. Lab. Invest. 1978; 4:455-69. 7. Coleman GL, Barthold SW, Osbaldistan GW etal. Pathological changes during aging in barrier-reared Fischer 344 rats.J. Gerontol. 1977; 32:258-78. 8. Goodman DG, Ward JM, Squire RA etal. Neoplastic and non-neoplastic lesions in aging F344 rats. Toxicol. Appl. Pharmacol. 1979; 48:237-48. 9. Pinkerton KE, Gehr P, Crapo JD. Architecture and cellular composition of the air-blood barrier. In: Parent RA (ed.) Treatise on Pulmonary Toxicology, Comparative Biology of the Normal Lung (Volume I), 1992; pp. 121-8. CRC Press, Boca Raton LA. 10. Gehr P, Mwangi DK, Ammann A etal. Design of the mammalian respiratory system. V. Scaling morphometric pulmonary diffusing capacity to body mass: wild and domestic mammals. Respir. Physiol. 1981; 44:61-86. 11. Pinkerton KE, Cowin LL, Witschi H. Development, growth, and aging of the lungs. In: Mohr U, Dungworth DL, Capen CC et al. (eds) Pathobiology of the Aging Mouse (Volume 1), 1996, pp. 261-72. Washington, DC: ILSI. 12. Boyden EA, Thompsett DH. The postnatal growth of the lung in the dog. Acta Anat. 1961; 47:185-215. 13. Sacher GA, Hart RW. Longevity, aging and comparative cellular and molecular biology of the house mouse (Mus musculus)

and the white-footed mouse (Peromyscus leucepus). Birth defects: Original article series 1978; 14:71-96, March of Dimes Foundation, New York. 14. Banfield AWF. The Mammals of Canada, 1974; p. 438. University of Toronto Press, Toronto. 15. Burger J, Gochfield M. Survival and reproduction in Peromyscus leucopus in the laboratory: viable model for aging studies. Growth Dev. Aging 1992; 56:17-22. 16. Takeda T, Hosokawa M, Takeshita S etal. A new murine model of accelerated senescence. Mech. Ageing Dev. 1981; 17:183-94. 17. Kawakami M, Paul JL, Thurlbeck WM. The effect of age on lung structure in male BALB/cNNia inbred mice. Am. J. Anat. 1984; 170-1. 18. Ranga V, Kleinerman J. Interalveolar pores in mouse lungs. Am. Rev. Respir. Dis. 1980; 122:477-81. 19. Hosokawa M, Takeshita S, Higuchi K et al. Cataract and other ophthalmic lesions in senescence accelerated mouse (SAM): morphology and incidence of senescence accelerated ophthalmic changes in mice. Exp. Eye Res. 1984; 38:105-14. 20. Takeshita S, Hosokawa M, Irino M e t a l . Spontaneous ageassociated amyloidosis in senescence accelerated mouse (SAM). Mech. Ageing Dev. 1982; 26:91-102. 21. Matsushita M, Tsuboyama T, Kasai R etal. Age-related changes in bone mass in the senescence accelerated mouse (SAM): SAM-R/3 and SAM-P/6 as new murine models for senile osteoporosis. Am. J. Pathol. 1986; 125:276-83. 22. Kurozumi M, Matsushita T, Hosokawa Metal. Age-related changes in lung structure and function in the senescenceaccelerated mouse (SAM): SAMP/1 as a new murine model of senile hyperinflation of lung. Am. J. Respir. Crit. Care Med. 1994; 149:776-82. 23. Uejima Y, Fukuchi Y, Nagase T et al. A new murine model of aging lung: the senescence accelerated mouse (SAM)-P. Mech. Ageing Dev. 1991; 61:223-36. 24. Scherle WA. A simple method for volumetry of organs in quantitative stereology. Mikroskopie 1970; 26:57-60. 25. Pack RJ, A1-Ugaily LH, Morris G. The cells of the tracheobronchial epithelium of the mouse: quantitative light and electron microscope study. J. Anat. 1981; 132:71. 26. Martin H. The effect of aging on the alveolar pores of Kohn in the dog.Am. Rev. Respir. Dis. 1963; 88:773-8. 27. Amy RWM, Bowes D, Burri PH et al. Postnatal growth of the mouse lung.J. Anat. 1977; 124:131-51. 28. Pump K. Emphysema and its relation to age. Am. Rev. Respir. Dis. 1976; 114:5-13. 29. Higashimoto Y, Fukuchi Y, Shimada Y e t al. The effects of aging on the function of alveolar macrophages in mice. Mech. Aging Dev. 1991; 69:207-17. 30. Chen TS, Richie JP Jr, Lang CA. Life span profiles of glutathione and acetaminophen detoxification. Drug Metab. Dispos. Biol. Fate Chem. 1990; 18:882-7. 31. Boorman GA, Eustis SL. Lung. In: Boorman GA, Eustis SL, Elwell MR etal. (eds) Pathology of the Fischer Rat, 1990; pp. 339-367. Academic Press, New York. 32. Chesky JA, Rockstein M. Life span characteristics in the male Fischer rat. Exp. Aging Res. 1976; 2:399-407. 33. Jacobs BB, Huseby RA. Neoplasms occurring in aged Fischer rats with special reference to testicular, uterine, and thyroid tumors.J. Natl. Cancer Inst. 1967; 39:303-9. 34. Massaro EJ. Mortality and growth characteristics of rat strains commonly used in aging research. Exp. Aging Res. 1980; 6:219-33. 35. Mauderly JL, Likens SA. Relationships of age and sex to function of Fischer 344 rats. Fed. Proc. 1980; 39:10-91. 36. Rockstein M, Chesky JA, Sussman ML. Comparative biology and evolution of aging. In: Finch CE, Hayflick L (eds) Handbook of the Biology of Aging, 1977; pp. 3-34. Van Nostrand Reinhold, New York.

37. Sass B, Rabstein LS, Madison R et al. Incidence of spontaneous neoplasms in F344 rats throughout the natural life-span. J. Natl. Cancer Inst. 1975; 54:1449-56. 38. Snell KC. Spontaneous lesions of the rat. In: Ribelin WE, McCoy JR. (eds) The Pathology of Laboratory Animals, 1965; pp. 211-302. Thomas, Springfield. 39. Boorman GA, Morgan KT, Uriah LC. Nose, larynx and trachea. In: Boorman GA, Eustis SL, Elwell MR et al. (eds) Pathology of the Fischer Rat, 1990, pp. 315-37. Academic Press, New York. 40. Jeffery PK, Reid LM. Ultrastructure of airway epithelium and submucosal glands during development. In: Hodson (ed.) Development of the Lung, 1977, pp. 87-134. Dekker, New York. 41. Plopper CG, Mariassy AT, Wilson DW et al. Comparison of nonciliated tracheal epithelial cells in six mammalian species: Ultrastructure and population densities. Exp. Lung Res. 1983; 5:281-94. 42. Chang L, Mercer RR, Crapo JD. Differential distribution of brush cells in rat lung.Anat. Rec. 1986; 216:49-54. 43. Cireli E. Elektronenmikroskopische anaivse der priiund postnatalen differenzierung des epithels der oberen luftwege der ratte. Z. Mikrosk. Anat. Forsch. 1966; 41:132-78. 44. Kober HJ. Die lumenseitige oberflache der rattentrachea wiihrend der ontogenese. Z. Mikrosk. Anat. Forsch. 1975; 89:399--409. 45. Pinkerton KE, Weller BL, M6nache MG et al. Consequences of prolonged inhalation of ozone on F344/N rats: collaborative studies. Part XIII. A comparison of changes in the tracheobronchial epithelium, and pulmonary acinus in male rats at 3 and 20 months. Health Effects Institute Research Report No. 85, 1998, pp. 1-32. 46. Pinkerton KE, M6nache MG, Plopper CG. Consequences of prolonged inhalation of ozone on F344/N rats: collaborative studies, Part IX. Changes in the tracheobronchial epithelium, pulmonary acinus, and lung antioxidant enzyme activity. Health Effects Institute Research Report No. 65, 1995, pp. 41-98. 47. Pinkerton KE, Vincent R, Plopper CG et al. Normal development, growth, and aging of the lung. In: Mohr U, Dungworth DL, Capen CC (eds) Pathobiology of the Aging Rat (Volume 1), 1992, pp. 97-109. Washington, DC: ILSI. 48. Crapo JD, Peters-Golden M, Marsh-Salin J etal. Pathologic changes in the lungs of oxygen-adapted rats. A morphometric analysis. Lab. Invest. 1978; 39:640-53. 49. Takezawa J, Miller FJ, O'Neil JJ. Single-breath diffusing capacity and lung volumes in small laboratory mammals. J. Appl. Physiol. 1980; 48:1052-9. 50. Hayatdavoudi G, Crapo JD, Miller FJ et al. Factors determining degree of inflation in intratracheally fixed rat lungs. J. Appl. Physiol. 1980; 48:389-93. 51. Randell SH, Mercer RR, Young SL. Postnatal growth of pulmonary acini and alveoli in normal and oxygen-exposed rats studied by serial section reconstructions. Am. J. Anat. 1989; 186:55-68. 52. Liebow AA. Summary: biochemical and structural changes in the aging lung. In: Cander L, Moyer JH (eds)Aging of the lung, 1964; pp. 97-104. Grune and Stratton, New York. 53. Mauderly JL. Effect of age on pulmonary structure and function of immature and adult animals and man. Fed. Proc. 1979; 38:173-7. 54. Young SL, Spain CL, Fram EK et al. Development of type II pneumocytes in the rat lung. Am. J. Physiol. Lung Cell Mol. Physiol. 1991; 260:L113-22.

55. Brody JS, Kaplan NB. Proliferation of alveolar intersitial cells during postnatal lung growth. Evidence for two distinct populations of pulmonary fibroblasts. Am. Rev. Respir. Dis. 1983; 127:763-70. 56. Vincent R, Mercer RR, Chang LY et al. Morphometric study of interstitial matrix in the lungs of the aging rat. FASEB J. 1990; 4:A1915 (Abstract). 57. Mercer RR, Crapo JD. Spatial distribution of collagen and elastin fibers in the lungs.J. Appl. Physiol. 1990; 69:756-65. 58. Rucker RB, Dubick MA. Elastin metabolism and chemistry: potential roles in lung development and structure. Environ. Health Perspect. 1984; 55:179-91. 59. Foster JA, Curtiss SW. The regulation of lung elastin synthesis.Am. J. Physiol. 1990; 259:L13-23. 60. Mays PK, Bishop JE, Laurent GJ. Age-related changes in the proportion of types I and III collagen. Mech. Aging Dev. 1988; 45:203-12. 61. Mays PK, McAnulty RJ, Laurent GJ. Age-related changes in lung collagen metabolism. A role for degradation in regulating lung collagen production. Am. Rev. Respir. Dis. 1989; 140:410-16. 62. Sahebjami H. Lung tissue elasticity during the lifespan of Fischer 344 rats. Exp. Lung Res. 1991; 17:887-902. 63. Mercer RR, Crapo JD. Three-dimensional reconstruction of alveoli in the rat lung for pressure-volume relationships. J. Appl. Physiol. 1987; 62:1480-7. 64. Stiles J, Tyler WS. Age-related morphometric differences in responses of rat lungs to ozone. Toxicol. Appl. Pharmacol. 1988; 92:274-85. 65. Brain JD, Sorokin S, Godieski IJ. Quantification, origin, and fate of pulmonary macrophages. In: Brain JD, Proctor DF, Reid LM (eds) Lung Biology in Health and Disease: Respiratory Defense Mechanisms, part 11 (Volume 5), 1977, p. 849. Dekker, New York. 66. SheUito J, Esparza C, Armstrong C. Maintenance of the normal rat alveolar macrophage cell population. The roles of monocyte influx and alveolar macrophage proliferation in situ. Am. Rev. Respir. Dis. 1987; 135:78-82. 67. Tyler WS, Tyler NK, Last JA et al. Effects of ozone on lung and somatic growth. Pair fed rats after ozone exposure and recovery periods. Toxicology 1987; 46:1-20. 68. Robinson NE, Gillespie JR. Morphologic features of the lungs of aging beagle dogs.Am. Rev. Respir. Dis. 1973; 108:1192-9. 69. Bartlett D Jr, Areson JG. Quantitative lung morphology in newborn mammals. Respir. Physiol. 1977; 2:193-200. 70. Siegwart B, Gehr P, Gil J et al. Morphometric estimation of pulmonary diffusion capacity. IV. The normal dog. Respir. Physiol. 1971; 13:141-59. 71. Crapo JD, Young SL, Fram EK et al. Morphometric characteristics of cells in alveolar region of mammalian lungs. Am. Rev. Respir. Dis. 1983; 128:$42-6. 72. Pinkerton KE, Murphy KM, Hyde DM. Morphology and morphometry of the lung. In: Capen CC, Benjamin SA, Hahn FF et al. (eds) Pathobiology of the Aging Dog (Volume 1), 2001; 5:43-56. Iowa State University Press, Ames. 73. Mauderly JL, Hahn FF. The effects of age on lung function and structure of adult animals. Adv. Vet. Sci. Comp. Med. 1982; 26:35-77. 74. Horsefield K, Cumming G. Morphology of the bronchial tree in the dog. Respir. Physiol. 1976; 26:173-82. 75. Raabe OG, Yeh HC, Schum GM et al. Tracheobronchialgeometry: human, dog, rat, hamster. 1976; Lovelace Foundation, Albuquerque, NM.

Environmental Influences on Lung Development and Aging

ISBN 0 12 324751 9

Part 2

Copyright © 2004 Elsevier

INTRODUCTION Today, infants who are born as early as 22 weeks postconceptional age (term is 40 weeks) may survive if supported by antenatal steroids and postnatal surfactant replacement, mechanical ventilation, supplemental oxygen, antibiotics, and appropriate nutrition. These, and other supports are required because of the sudden event of preterm birth, an abrupt change in environment that profoundly impacts the lung. Two questions are addressed in this chapter. First, what are the causes of preterm birth and, second, how does the abrupt change in environment influence lung development?

Evaluation of birth certificates for most of the country showed an incidence of preterm birth of 9% in 1981, which rose to 11% in 1989. 2 The second source, a multicenter trial, which evaluated preterm birth weight and gestational age also showed a 10% incidence of prematurity. 3 The extensive use of tocolytic agents to arrest preterm labor has not reduced the incidence of preterm birth in the United States or Canada. 4-6 The preterm birth incidence among non-Hispanic African-American women is about twice that of non-Hispanic Caucasian women. 7 Other factors such as maternal smoking, s illicit-drug u s e , 9 previous reproductive history, 1~ incompetence of the cervix, and uterine anomalies are also associated with preterm birth. Mothers who bear their first child before 20 years of age or after 35 years of age are also at high risk, with mothers 35 years or older being at higher risk. 11

CAUSES OF PRETERM B I R T H Factors leading to preterm birth Definition of preterm birth Preterm birth is any delivery, regardless of birth weight, that occurs before 37 completed weeks from the first day of the menstrual cycle. 1 Pregnancies that end before 20-22 completed weeks of gestation are termed abortions. Thus, a reasonable definition of preterm birth is any delivery that occurs between 20 and 37 weeks of gestation.

Incidence of preterm birth In the United States, the overall rate of prematurity is approximately 10%. This estimate derives from two sources. *To whom correspondence should be addressed. The Lung: Development, Aging and the Environment ISBN 0 12 324751 9

Maternal factors dominate the factors that lead to preterm birth. 12 Maternal infection is clinically identified in 30 to 40% of pregnancies complicated by preterm labor. 13 Infection in the maternal cervix or vagina may involve extra-embryonic fetal tissue (i.e. infection of the fetal membranes, called chorioamnionitis) or uterine decidua. Alternatively, low-grade systemic infection in the mother may cross the placenta and inflame the fetus and/or placental villi (called villitis) (Fig. 16.1). On the other hand, at least 35 to 40% of pregnancies ending in preterm labor have no documented infection. TM Because such a large percentage of pregnancies end in preterm labor without clinical evidence of infection, uncertainty persists regarding the role of intrauterine infection Copyright 9 2004 Elsevier All rights of reproduction in any form reserved.

100 -

J Premature

birth Inflammation ,~

J

Initiation of • ~ mechanical IRDSI ventilation /

Pre- or postnatal

"~ Nutrition

infection

Prolonged

~ - " / ~ ' ~

Inflammation

]Recovery]

80-

Supplemental O=

~ Postnatal ~ infection

Ic,DI

mechanical ventilation

Prolonged I~ ~x~supplemental 02

Death

60-

Prior to Discharge

(%)

40-

Nutrition

Fig. 16.1. Factors contributing to the development of respiratory distress syndrome (RDS) and its evolution to chronic lung disease (CLD) following preterm birth.

in initiating preterm labor. Although preterm labor has also been associated with preterm rupture of membranes, pregnancy-induced hypertension, abruption of the placenta, or intrauterine fetal growth retardation, such associations have not been consistently found. 15

Preterm birth and perinatal mortality Deaths of very-low-birth-weight (VLBW) preterm infants (defined as weighing 500-1500gm at birth) have declined since 1988.16 Death before hospital discharge was 26% in 1988 and 20% in 1991. A recent prospective, 14-center survey (the National Institute of Child Health and Human Development Neonatal Research Network) showed that 16% of 4438 VLBW preterm infants died prior to discharge from hospital or long-term care facility in 1995-1996.16 Death prior to discharge, however, was inversely related to birth weight (Fig. 16.2). The incidence of death prior to discharge ranged from 89% for 195 preterm infants who weighed 401-500gm at birth (defined as extremely lowbirth-weight, ELBW) to 3% for 1299 preterm infants who weighed 1251-1500 gm at birth. Perinatal mortality associated with preterm birth is also related to sex of the preterm infant, plurality (twins, triplets, etc.), ethnicity and socioeconomic status, intrauterine growth rate, late-onset sepsis, and neonatal respiratory failure. 16-18 Specific examples include the mortality rate in male preterm infants, which is approximately twice that of female preterm infants when birth occurs before 29 completed weeks of gestation. Twin preterm infants have 3 to 4 times the mortality of singleton preterm infants. Mortality risk in the presence of late-onset sepsis was 18% compared to 7% without infection. 18 Preterm birth and morbidities and adverse outcomes Among the survivors of preterm birth, morbidities such as poor in-hospital growth, patent ductus arteriosus, intracranial hemorrhage, late-onset sepsis, neonatal respiratory failure, necrotizing enterocolitis, and neurological impairments are frequent. Poor in-hospital growth, the most frequent morbidity in a recent study, 16was present in 97% of VLBW infants. Overall morbidity increased from 27% in 1991 to 30% in

20-

0500-750

m

751-1000

1001-1250

1251-1500

Birth Weight Categories (gm)

Fig. 16.2. Histogram summarizing the results of a clinical study that assessed mortality among very-low-birth-weight prematurely born infants, stratified for birth weight. Death prior to discharge from hospital is inversely related to birth weight. Adapted from Lemons and coworkers. 16

1996, primarily because the incidence of chronic lung disease (CLD) of prematurity (also known as bronchopulmonary dysplasia or BPD) in all survivors increased from 19% in 1991 to 23% in 1996.16 The incidence of major morbidity was greatest (63%) among the smallest preterm infants (501-750 gm at birth). Adverse outcomes later in life, such as asthma and neurological impairments, continue to affect a high percentage of prematurely born infants. Because the incidence of morbidities has not changed significantly, the absolute number of surviving preterm infants with morbidities has increased. 16-18 Some VLBW infant survivors of newborn intensive care are now 20 years old. 19 Studies of these infants are providing both good and bad news about adverse outcomes later in life. Encouraging observations are that 51% of the survivors had IQ scores within the normal range, 74% had completed high school, and 41% were continuing their education beyond high school. In addition, alcohol use, illicit-drug use and criminal behavior occurred at a frequency that was not different from peers who were born at normal birth weight (>2500 gm). On the other hand, these infants had more chronic health problems, such as cerebral palsy, blindness and deafness. 19 As a group, they were also shorter, had lower IQ scores and lower scores on academic achievement tests compared to peer young adults who had normal birth weight. The VLBW infants were also less likely to have graduated from high school or enrolled in postsecondary education compared to their young adult peers who had normal birth weight. Whether these differences will diminish as VLBW survivors reach maturity remains an important question.

In summary, preterm birth is an abrupt environmental stress. The incidence of preterm birth is increasing, the size of preterm infants that can be supported is declining, and the incidence of chronic lung disease among the smallest preterm infants remains high. How does the abrupt change in environment influence lung development? This question will be addressed in the next section of this chapter.

PRETERM B I R T H AS AN E N V I R O N M E N T A L INFLUENCE ON LUNG DEVELOPMENT Before term birth, gas exchange is among the functions of the placenta, which is interposed between the fetal and maternal circulations. Because the fetal lung is not required to function as a gas-exchange organ before 37 weeks of gestation, the lung architecture does not develop to facilitate gas exchange, particularly oxygen. In the mature lung, exchange of oxygen and carbon dioxide occurs in relatively large units that are referred to as terminal respiratory units. Terminal respiratory units are all alveolar ducts, together with their accompanying alveoli, that stem from the most proximal respiratory bronchiole. 2~ In the adult human lung, such units contain approximately 100 alveolar ducts and 2000 anatomic alveoli. A terminal respiratory unit is approximately 5 mm in diameter and has a volume of about 0.02ml at functional residual capacity. Together, the two lungs of the adult human contain about 150,000 terminal respiratory units. 21 The function of the terminal respiratory unit is such that diffusion of oxygen and carbon dioxide is so rapid that the partial pressures of each gas are uniform throughout the unit. From each inspired breath, oxygen that reaches the alveolar duct gas diffuses into the alveoli because the incom-

ing air has a higher oxygen concentration than the alveolar gas. Oxygen subsequently diffuses through the air-blood barrier into the red blood cells, where oxygen combines with hemoglobin as the red blood cells flow along the capillaries. Carbon dioxide diffuses in the opposite direction. In the normal adult human lung, the air-blood barrier is exceedingly thin, which facilitates gas diffusion. The average width of this barrier is about 1.5 ktm.21'22 For oxygen, the air-blood barrier's anatomic elements consist, in order, of alveolar epithelium and its subjacent basal lamina, alveolar wall interstitium, basal lamina of the capillary endothelium, the capillary endothelium, plasma, and the membrane of red blood cells. For carbon dioxide, those anatomic elements are traversed in the opposite direction. The terminal respiratory units of the prematurely born infant are incompletely developed so that the air-blood barrier is too thick to allow efficient exchange of oxygen and carbon dioxide. This structural problem is greatest with more immature preterm infants because the development of the terminal respiratory units occurs during the second half of gestation, with the thickness of the gas-exchange barrier inversely related to gestational age. A brief review of the stages of lung development for humans illustrates these features (Table 16.1 and Fig. 16.3). 23 Branching of the bronchial tree is complete by 16 weeks gestation; prior to 17 weeks of gestation, the future bronchi and bronchioles branch into relatively undifferentiated mesenchyme that contains no terminal respiratory units or blood vessels (Fig. 16.3a). This period of lung development is known as the pseudoglandular stage. From weeks 17 to 28 of gestation, the human fetal lung is at the canalicular stage. Lung architecture at this stage is characterized by development of capillaries within the thick partitions of mesenchyme that are delimited by rudimentary air canals. These immature air canals are lined by cuboidal or even columnar epithelial

Fig. 16.3. Normal lung development in humans. Panel a: Although the lung of this infant stillborn at 12-14 weeks of gestation shows considerable autolysis (post-mortem degeneration), a smooth muscle wall (arrow) is visible around airways (AW) that are lined by (largely desquamated) columnar epithelium. There is no development in the intervening mesenchyme (M), which is devoid of blood vessels (BV), except at the periphery of the developing lung tissue. Panel b: By 20 weeks of gestation, smooth-walled respiratory bronchioles and alveolar ducts lined by low cuboidal epithelium (arrow) are visible beyond the larger Y-shaped terminal bronchiole (TB). Numerous capillaries (arrowhead) are visible deep within the developing interstitium (mesenchyme). Panel c: At 30 weeks of gestation, the distal airspaces are subdivided, and many capillaries are present in the thick primary septa (arrow). A terminal bronchiole (TB) and neighboring pulmonary artery (PA) are visible in the center of the panel. Panel d: By 40 weeks of gestation, alveolar development has progressed to a point where the primary septa are thinner (arrow) and secondary septa (crests; arrowhead) protrude into the developing alveoli (A). The alveolar walls are replete with capillaries. Thus, at term, the infant lung clearly resembles its adult form, although considerable growth and development have yet to occur. All four panels show tissue sections that were stained with hematoxylin and eosin, and are the same magnification. (See Color plate 6.)

cells. Although capillary proliferation is robust during this stage of lung development, the developing capillaries are far removed from the epithelium that lines the air canals (Fig. 16.3b). From about week 24 to week 36 of gestation, lung development is at the saccular stage. This stage is characterized structurally by thinning of the mesenchyme, continued proliferation of capillaries, proliferation and expansion of the distal air sacs, and thinning of their lining epithelial ceils (Fig. 16.3c). Combination of these structural transformations creates a thinner connective tissue compartment and decreases the distance between the developing air sacs and the proliferating capillaries. However, intimate structural association between the capillaries and air sac epithelial cells is not reached during this stage. From about 32 weeks of gestation and continuing through the first 18 to

24 months of postnatal life, alveoli are formed by progressive expansion and subdivision (secondary septation, also called secondary crest formation) of the terminal air sacs (Fig. 16.3d). During the alveolarization stage of lung development, the air-blood barrier attains its adult thinness by reduction of the connective tissue compartment in the developing alveolar walls. At the same time, the epithelial cells that line the developing alveoli differentiate into two cell populations. 24 One population, known as alveolar type II epithelial cells, remains cuboidal and develops the cellular organelles that synthesize and secrete surface-active material (surfactant and its apoproteins). Also arising from this population of epithelial cells are alveolar type I (squamous) epithelial cells. These terminally differentiated cells, although fewer in number than their cuboidal counterpart,

cover about 90 to 95% of the alveolar surface area of the peripheral lung. 25 Both structural attributes of alveolar type I epithelial cells provide a large, thin cellular barrier specialized for gas exchange. Before birth, the fetal lungs are filled with liquid that flows from the pulmonary vascular compartment, through the interstitium to the potential airspaces. This liquid then percolates centrally along the conducting airways to the oropharynx, where it is either swallowed or expelled into the amniotic sac. As alveolar type II epithelial cells develop, they secrete surfactant lipids and apoproteins into this fluid. Detection of these secretory products in the amniotic fluid provides a means of assessing lung development prior to birth. 26 Liquid flows from the interstitium into the fetal air spaces because chloride is secreted across the fetal respiratory epithelium into the potential airspaces, 27 causing a concentration gradient that promotes flow of water into the potential airspaces. Near the time of term birth, the secretory activity of the respiratory epithelium switches from a predominantly chloride-secreting membrane to a predominantly sodium-absorbing membrane. 28-3~ The presence of fluid in the potential airspaces, as well as in the mesenchyme surrounding those potential airspaces, is an impediment to gas exchange in the preterm infant. Fluid is an impediment because diffusion is much faster in the gas phase than in water. For example, the solubility of oxygen in water is low (0.03 ml/L x mmHg PaO2) compared to carbon dioxide, the solubility of which is about 20 times more than oxygen (0.7 ml/L • mmHg PaCO2).The advantage for carbon dioxide persists even though the driving pressure for carbon dioxide diffusion is only one tenth that for oxygen entering the blood. Therefore, oxygen diffusion is much slower in the relatively over-hydrated environment of the preterm lung. Balance between adequate production of lung luminal liquid and its drainage is required for normal intrauterine lung growth. 31 When drainage exceeds production, the fetal lung is not exposed to continuous liquid distending pressure and lung growth is inhibited. This occurs in fetuses with prolonged oligohydramnios due to rupture of amniotic membranes. 32 Other conditions that disturb lung growth include diaphragmatic hernia 33 or pulmonary artery occlusion. 34 Structural and functional characteristics of the airways may also be affected by preterm birth. During the canalicular and saccular stages of lung development, the airways have little smooth muscle in their wall, the epithelium is immature, and cell-cell adhesion is weak. Physiological studies of human tracheobronchial segments ex vivo indicate that pressure-volume relations are affected by maturity, 35'36 such that airway compliance was inversely related to maturity. The greater compliance of the airways of preterm infants may contribute to the need for higher airway pressures, and the associated increase in lung volume needed to inflate the collapsed gas exchange regions compared to the airways of term infants. 37 The dearth of smooth muscle and weak

adhesion between cells of the immature airway may make the preterm infant's airways susceptible to injury by the increased airway pressures and lung volumes that are required to effect adequate ventilation. Evidence of injury is manifest as sloughed airway epithelial cells in tracheal aspirates and lung lavage fluid, as well as development of air leaks (interstitial emphysema and pneumothorax). Other structural and functional attributes of the developing lung important to normal lung function are alveolar type II epithelial cells and the surfactant phospholipids and apoproteins that these cells synthesize, secrete and recycle. In the human, type II cells containing lamellar bodies are histologically recognizable at the transition from the canalicular to the saccular stages of lung development (weeks 20 to 24 of gestation). Surfactant synthesis begins later, during the saccular stage (about 30 weeks of gestation). 38 From the saccular stage to the end of term gestation, the concentrations of dipalmitoyl phosphatidylcholine and other surfactant phospholipids increase in lung tissue, lung lavage fluid and amniotic fluid. Surfactant apoproteins (SP) A, B and C, expressed only in lung tissue, are also developmentally regulated but not concordantly. Human SP-A mRNA and protein in lung tissue are not detectable until the saccular stage of lung development (about 30 weeks of gestation). 38-41 Human SP-B and SP-C are detectable in lung tissue at very low levels earlier than SP-A, during the canalicular stage (about 24 weeks of gestation). 42 Detection of SP-B and SP-C proteins, however, is later in development, during the saccular stage (about 30 weeks of gestation). 41 To our knowledge, human SP-D protein has only been detected in amniotic fluid during the saccular stage of lung development (about 24 weeks of gestation). 43'44 Thus, the cellular and biochemical machinery to reduce surface tension at air-liquid interfaces in the future airspaces develops during the third trimester in humans. Many preterm infants are born before the third trimester, and are at risk for respiratory failure because their lungs are deficient in surfactant (Fig. 16.1).

Respiratory distress syndrome When an infant is born prematurely, if its lungs are structurally immature and deficient in surfactant, surface tension forces are high and the alveoli are unstable. 45 Alveolar instability results in alveolar collapse (atelectasis). This atelectasis results in ventilation-perfusion mismatch. Such mismatch results, in turn, in intrapulmonary shunting and thus contributes to poor oxygenation. 46 Opening the collapsed air spaces requires high ventilatory pressure that is transmitted to the immature distal airways and gas exchange regions of the lung. In effect, the extreme effort required to expand the lungs with the first breath must be repeated with each breath because surfactant is not present to prevent collapse of the distal airspaces. Prematurely born infants who have these characteristics usually develop respiratory distress syndrome (RDS; also known as acute lung injury) (Fig. 16.1).

Respiratory distress syndrome affects about 30,000 infants annually in the USA. The incidence is about 50 to 60% of infants who are born before 30 weeks of gestation. 47 The incidence increases with decreasing gestational age.16,18 RDS is more prevalent and severe in male compared to female preterm infants, for reasons that remain unclear. Respiratory distress syndrome is characterized by tachypnea, chest retractions, and cyanosis (hypoxemia) soon after birth. Chest radiographs demonstrate low lung volumes, air bronchograms, and opacities (i.e. reticulargranular infiltrates). Lung function studies have shown increased airway resistance 4s and decreased pulmonary compliance. 49 Pulmonary artery pressure and pulmonary vascular resistance are elevated, particularly in infants with severe RDS. 5~ The pathologic findings in RDS are similar to adults with acute respiratory distress syndrome (ARDS), 51 superimposed on the immature lung. 52 Mechanical ventilation is instituted very early in the course of this disease, and therefore lesions observed reflect both the disease and its treatment. Within the first 3 to 4 h, the lungs may only show uneven ventilation of distal airspaces, resulting from surfactant deficiency, interstitial edema, resulting from incomplete removal of fetal lung liquid or acute injury to microvascular endothelial cells, and congestion of capillaries (Fig. 16.4a). By 12 to 24 h, necrosis of alveolar and bronchiolar epithelial cells develops and the denuded walls become coated by characteristic hyaline membranes, which are brightly eosinophilic transudates of plasma proteins admixed with necrotic epithelial cells and fibrin (Fig. 16.4b). 53 Hyaline membranes accumulate especially at branch points. These pathologic features are similar to the exudative phase of ARDS in older patients, but the alveolar type II epithelial cell hyperplasia that constitutes the proliferative phase of ARDS is not as prominent in autopsy slides of hyaline membrane disease in preterm infants. Mechanical ventilation of preterm infants is frequently associated with ventilator-induced lung injury 54 (Fig. 16.1), even following repeated doses of exogenous surfactant. The injurious effects of mechanical ventilation depend on a number of factors, among which are the magnitude of airway pressure (barotrauma) and lung volume (volutrauma), 55'56 and the concentration of inspired oxygen. Several experimental animal studies have compared indices of ventilator-induced lung injury between conditions that raised airway pressure and increased lung volume. One study distinguished between the effects of pressure and volume by subjecting rats to identical peak airway pressure and either large or small tidal volume ventilation. 57 Small tidal volume ventilation with a high peak airway pressure was accomplished by strapping the thorax and abdomen. Rats subjected to a high airway pressure and large tidal volume ventilation developed increased permeability pulmonary edema and ultrastructural evidence of cellular damage. Strikingly, the rats that were subjected to high peak airway pressure, by strapping the abdomen, and small

tidal volume had no edema or ultrastructural evidence of cellular damage. This and other 58'59 studies, which used normal animals, suggest that large lung volumes, rather than high peak airway pressures per se, are important in the pathogenesis of ventilator-induced lung injury. For this reason, volutrauma is used to describe the injury that is associated with mechanical ventilation. How does volutrauma induce lung injury, especially following preterm birth? Evidence suggests that some of the circumstances that have already been described are exacerbated. For example, regional overinflation increases microvascular permeability, either directly or indirectly (through release of inflammatory mediators from sequestered leukocytes in the lung), which leads to alveolar flooding. Alveolar flooding, in turn, is associated with inactivation of surfactant, which in turn leads to atelectasis and less compliant lungs. If these cycles are not broken, lung injury recurs, necessitating more supplemental oxygen, higher airway pressure and a larger tidal volume. As the extent and duration of mechanical ventilation with supplemental oxygen increases, and lung volutrauma persists or increases, mild RDS may become BPD. Conventional mechanical ventilation, most frequently delivered by a time-cycled, pressure-limited ventilator, is regarded as an important contributing factor in inducing acute lung injury and chronic lung injury (Fig. 16.1). For this reason, other approaches to ventilation continue to be explored, such as high-frequency oscillatory ventilation, high-frequency jet ventilation, ECMO, and liquid perfluorocarbon ventilation. 6~ The aim of these alternative ventilation strategies is to decrease the peak inspiratory pressure and mean airway pressure, thereby resting the lung and reducing ventilator-induced lung injury. Another approach is to avoid mechanical ventilation altogether by use of continuous positive airway pressure delivered through nasal prongs (nasal CPAP). These alternatives offer benefits but each also carries unwanted side effects that limit utility. 61-64 The pathogenesis of increased permeability pulmonary edema during RDS is also associated with accumulation of leukocytes (Fig. 16.1), particularly neutrophils, in the vascular, interstitial and airspace compartments of the lung. 65-67 A recent study that ventilated preterm lambs for 8 h with 100% oxygen showed that the number of circulating neutrophils decreased in the first 30 to 90min of life and that this decrease correlated with an increase in the number of neutrophils that accumulated in the lung's air spaces at the end of the 8 h study. 65 A complimentary, retrospective chart review study in human preterm infants subsequently showed that a low concentration of mature neutrophils in the systemic circulation within 24 h of birth (ascribed to egress of neutrophils from the circulation coupled with an inability of the immature bone marrow to replenish the circulating pool) is associated with more severe respiratory distress during the first postnatal week of life. 68 Such preterm infants were more likely to need mechanical ventilation with supplemental oxygen.

Fig. 16.4. Hyaline membrane disease (HMD; panels a and b) and chronic lung disease of prematurity (CLD; panels c and d) in humans. Panel a: HMD, birth at 25 weeks of gestation followed by 6 h of mechanical ventilation. In the first few hours of life, the preterm, mechanically ventilated lung shows uneven expansion of the distal airspaces, vascular congestion, and interstitial edema. Some patchy hemorrhage also may be present. Panel b: HMD, birth at 29 weeks of gestation followed by 2 days of mechanical ventilation. Hyaline membranes (arrow head) are uniformly present by 12 to 24 h, but may appear within 3 to 4 h of preterm birth. Hyaline membranes are typically most evident at the branch points of distal airways (AW). The distribution of ventilation is uneven, with centriacinar expansion and peripheral collapse. Panels c and d: Infants who progress to CLD show marked simplification of the distal airspaces (compare panels c and d with the expected appearance of the lung at 30 and 40 weeks of gestation in Fig. 16.3, panels c and d, respectively; all four panels are at the same magnification). Panel c: CLD, birth at 22 weeks of gestation followed by 44 days of mechanical ventilation (28-29 weeks post-conceptual age). The distal airspaces appear as distended circles that are filled, in this specimen, with cellular debris (*). Thick, cellular mesenchyme (M) separates the adjacent airspaces. Panel d: CLD, birth at 24 weeks of gestation followed by 151 days of mechanical ventilation (45 weeks post-conceptual age). Simplification and overdistension of the distal airspaces (DAS) is clearly evident. The primary septa (arrow) are thick and cellular. Secondary septa (crests; arrowhead) are infrequently visible; those that are visible are short, thick and devoid of capillaries near their tip. Panels a and b show tissue sections that were stained with hematoxylin and eosin. Panels c and d show tissue sections that were stained with Masson's trichrome. All four panels are the same magnification. (See Color plate 7.)

Because lung immaturity is a consistent feature of RDS, strategies have been developed to accelerate lung development before preterm delivery. One strategy is administration of hormones, such as glucocorticoids. 69'7~The seminal studies of Liggins demonstrated that glucocorticoids accelerated lung maturation of fetal sheep and decreased the incidence of RDS in human infants after antenatal corticosteroid therapy. 71'72 The maturational effect of corticosteroid treatment on lung structure was first described by Kikkawa and colleagues. 73 Among the structural effects, corticosteroids (betamethasone or dexamethasone) accelerate the maturation of alveolar type II epithelial cells, which are the source of

pulmonary surfactant. For these reasons, mothers who are in preterm labor, and at risk of giving birth to a preterm infant between 24 and 35 weeks, of gestation, are treated antenatally with glucocorticoids. 74 Evidence has been demonstrated that maternal treatment with thyroid-releasing hormone enhances the beneficial effects of antenatal glucocorticoids by further promoting lung maturation from structural, biochemical and functional standpoints, including surfactant production. 75'76 For example, combination treatment increased the content of saturated phosphatidylcholine, stability of alveoli, and clearance of lung liquid. 77-82 However, a recent follow-up

study of children two years after receiving antenatal combination therapy of thyroid-releasing hormone and corticosteroids showed delayed mental development as well as more respiratory problems, ventilation days, and chronic lung disease. 83 These results, in combination with meta-analysis,84 have led to the recommendation that antenatal thyroidreleasing hormone treatment should not be given to women at risk of preterm birth. A treatment strategy to reduce lung stiffness (i.e. increase lung compliance) after preterm birth is to instill surfactant into the airways. 85 Surfactant replacement therapy is beneficial because once it becomes widely and thinly distributed in the lung, the exogenous surfactant reduces surface tension at air-liquid interfaces, thereby stabilizing alveoli when they are deflated. After surfactant replacement, oxygenation improves swiftly, followed for several hours by progressive improvement in gas exchange and lung mechanics.86'87 Other improvements, at least in preterm lambs, are reduced vascular injury and edema. 88 Thus, surfactant replacement therapy reduces the incidence and severity of RDS following preterm birth.

100 Yipp and Tan, 1991 Kraybill et al., 1987

80

Preterm Infants Who Developed BPD

(%)

60m

40-

20-

I I i

0 500-750

751-1000

1001-1250

1251-1500

Birth Weight C a t e g o r i e s ( g m )

Fig. 16.5. Histogram summarizing the results from two clinical studies that assessed the incidence of bronchopulmonary dysplasia (BPD) among birth weight categories for prematurely born infants. The incidence of BPD is inversely related to birth weight. Adapted from Yipp and Tan16Sand Kraybill and colleagues. 166

Chronic lung disease of prematurity Survival of prematurely born infants with RDS has significantly increased since the introduction of antenatal steroids and postnatal surfactant replacement therapies, and gentler ventilation strategies. 16'6~176 As described earlier, 84% of preterm infants weighing 500-1500gin at birth now survive. Unfortunately, about 8000-10,000 preterm infants who develop RDS in the United States annually, and who receive antenatal glucocorticoids and postnatal surfactant replacement, go on to develop CLD of prematurity (also called BPD) (Fig. 16.1). The reported incidence of CLD ranges between 15 and 60% of preterm infants. 91-93 As expected, smaller infants were at greater risk for CLD than larger preterm infants (Fig. 16.5). Other risk factors for CLD include male gender, white race, greater severity of RDS, and a low Apgar score 1 rain after birth. After adjusting for these risk factors, development of CLD is associated most with a need for high inspired oxygen 96 h after birth, high peak inspiratory pressure, and high fluid intake. The original description of BPD was made by Northway and colleagues (none of whom were neonatologistsI). 94 These authors defined BPD based upon clinical, radiological and pathological criteria. The principal clinical criteria were the requirement for supplemental oxygen for at least 28 days after birth (a duration selected because that is a month; W. Northway, personal communication). The major radiological criteria were cyst formation and hyperexpansion mixed with atelectasis. The characteristic pathological findings at autopsy depended upon the duration of disease prior to death. Classic hyaline membrane disease was seen in the first 2-3 days. A regenerative phase was seen over the next week that consisted of necrosis and regeneration ofbronchiolar and alveolar epithelium, in addition to hyaline membranes.

Over the next 10 days, hyperplasia of bronchiolar and arteriolar smooth muscle, fibrosis, and dilated or collapsed distal air sacs ensued. In the final stage, these alternating areas of overdistention and atelectasis became fixed. The term 'bronchopulmonary' was originally chosen because it identified airway and blood vessel involvement; "dysplasia" was chosen because it described the failure of the distal air sacs to develop into normal anatomic alveoli (W. Northway; personal communication). Today, CLD is usually clinically defined as either oxygen dependency at 28 days of life with appropriate radiological findings or oxygen dependency at 36 weeks' postconceptional age. 95'96 As originally described by Northway etal., 94 BPD was diagnosed in preterm infants who did not recover from RDS. However, treatment of respiratory failure from other causes, including meconium aspiration pneumonia, 97 neonatal pneumonia, 98 congestive heart failure, 99 or congenital diaphragmatic hernia, 1~176 may be associated with CLD. Prematurely born infants who develop CLD today, compared to those who developed BPD over 30 years ago, are much younger (23-26 weeks of gestation compared to 31-34 weeks of gestation) and smaller (