Advances in Botanical Research, Volume 41

Advances in BOTANICAL RESEARCH Incorporating Advances in Plant Pathology Editor-in-Chief J. A. CALLOW School of Biosc...

Author: J. A. Callow

69 downloads 780 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Advances in

BOTANICAL RESEARCH Incorporating Advances in Plant Pathology

Editor-in-Chief J. A. CALLOW

School of Biosciences, The University of Birmingham, UK

Editorial Board A. R. HARDHAM J. S HESLOP-HARRISON M. KREIS R. A. LEIGH E. LORD D. G. MANN P. R. SHEWRY D. SOLTIS

Australian National University, Canberra, Australia University of Leicester, UK Universite de Paris-Sud, Orsay, France University of Cambridge, Cambridge, UK University of California, Riverside, USA Royal Botanic Garden, Edinburgh, UK IACR-Long Ashton Research Station, UK University of Florida at Gainesville, USA

CONTRIBUTORS TO VOLUME 41

ARNAUD COMPLAINVILLE Institut des Sciences du Ve´ge´tal, Centre National de la Recherche Scientifique, 91198 Gif sur Yvette, France JAMES E. COOPER Department of Applied Plant Science, Queen’s University Belfast, Belfast BT9 5PX, United Kingdom MARTIN CRESPI Institut des Sciences du Ve´ge´tal, Centre National de la Recherche Scientifique, 91198 Gif sur Yvette, France CLAIRE HALPIN University of Dundee, Plant Science Research Group, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom CELIA HANSEN Department of Biology, University of Leicester, Leicester LE1 7RH, United Kingdom J. S. HESLOP-HARRISON Department of Biology, University of Leicester, Leicester LE1 7RH, United Kingdom HAMLYN G. JONES University of Dundee, Plant Science Research Group, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom

CONTENTS OF VOLUMES 30–40

Contents of Volume 30 Nitrate and Ammonium Nutrition of Plants: Physiological and Molecular Perspectives G. FORDE and D. T. CLARKSON Secondary Metabolites in Plant–Insect Interactions: Dynamic Systems of Induced and Adaptive Responses J. A. PICKETT, D. W. M. SMILEY and C. M. WOODCOCK Biosynthesis and Metabolism of Caffeine and Related Purine Alkaloids in Plants H. ASHIHARA and A. CROZIER Arabinogalactan-Proteins in the Multiple Domains of the Plant Cell Surface M. D. SERPE and E. A. NOTHNAGEL Plant Disease Resistance: Progress in Basic Understanding and Practical Application N. T. KEEN

Contents of Volume 31 PLANT TRICHOMES Edited by D. L. Hallahan and J. C. Gray Trichome Diversity and Development E. WERKER

x


Structure and Function of Secretory Cells A. FAHN Monoterpenoid Biosynthesis in Glandular Trichomes of Labiate Plants D. L. HALLAHAN Current and Potential Exploitation of Plant Glandular Trichome Productivity S. O. DUKE, C. CANEL, A. M. RIMANDO, M. R. TELLEZ, M. V. DUKE and R. N. PAUL Chemotaxonomy Based on Metabolites from Glandular Trichomes O. SPRING Anacardic Acids in Trichomes of Pelagonium: Biosynthesis, Molecular Biology and Ecological Effects D. J. SCHULTZ, J. I. MEDFORD, D. COX-FOSTER, R. A. GRAZZINI, R. CRAIG and R. O. MUMMA Specification of Epidermal Cell Morphology B. J. GLOVER and C. MARTIN Trichome Initiation in Arabidopsis A. R. WALKER and M. D. MARKS Trichome Differentiation and Morphogenesis in Arabidopsis ¨ LSKAMP and V. KIRIK M. HU Trichome Plasmodesmata: A Model System for Cell-to-Cell Movement F. WAIGMANN and P. ZAMBRYSKI


Contents of Volume 32 PLANT PROTEIN KINASES Edited by M. Kreis and J. C. Walker Plant Protein-Serine/Threonine Kinases: Classification into Subfamilies and Overview of Function D. G. HARDIE Bioinformatics: Using Phylogenetics and Databases to Investigate Plant Protein Phosphorylation E. R. INGHAM, T. P. HOLTSFORD and J. C. WALKER Protein Phosphatases: Structure, Regulation and Function S. LUAN Histidine Kinases and the Role of Two-component Systems in Plants G. E. SCHALLER Light and Protein Kinases J. C. WATSON Calcium-dependent Protein Kinases and their Relatives E. M. HRABAK Receptor-like Kinases in Plant Development K. U. TORII and S. E. CLARK A Receptor Kinase and the Self-incompatibility Response in Brassica J. M. COCK Plant Mitogen-activated Protein Kinase Signalling Pathways in the Limelight S. JOUANNIC, A.-S. LEPRINCE, A. HAMAL, A. PICAUD, M. KREIS and Y. HENRY

xi

xii


Plant Phosphorylation and Dephosphorylation in Environmental Stress Responses in Plants K. ICHIMURA, T. MIZOGUCHI, R. YOSHIDA, T. YUASA and K. SHINOZAKI Protein Kinases in the Plant Defence Response G. SESSA and G. B. MARTIN SNF1-Related Protein Kinases (SnRKs) – Regulators at the Heart of the Control of Carbon Metabolism and Partitioning N. G. HALFORD, J.-P. BOULY and M. THOMAS Carbon and Nitrogen Metabolism and Reversible Protein Phosphorylation D. TOROSER and S. C. HUBER Protein Phosphorylation and Ion Transport: A Case Study in Guard Cells J. LI and S. M. ASSMANN

Contents of Volume 33 Foliar Endophytes and Their Interactions with Host Plants, with Specific Reference to the Gymnospermae W.-M. KRIEL, W. J. SWART and P. W. CROUS Plants in Search of Sunlight D. KOLLER The Mechanics of Root Anchorage A. R. ENNOS


xiii

Molecular Genetics of Sulphate Assimilation M. J. HAWKESFORD and J. L. WRAY Pathogenicity, Host-specificity, and Population Biology of Tapesia spp., Causal Agents of Eyespot Disease of Cereals J. A. LUCAS, P. S. DYER and T. D. MURRAY

Contents of Volume 34 BIOTECHNOLOGY OF CEREALS Edited by Peter Shewry Cereal Genomics K. J. EDWARDS and D. STEVENSON Exploiting Cereal Genetic Resources R. J. HENRY Transformation and Gene Expression P. BARCELO, S. RASCO-GAUNT, C. THORPE and P. A. LAZZERI Opportunities for the Manipulation of Development of Temperate Cereals J. R. LENTON Manipulating Cereal Endosperm Structure, Development and Composition to Improve End Use Properties P. R. SHEWRY and M. MORELL Resistance to Abiotic Freezing Stress in Cereals M. A. DUNN, G. O’BRIEN, A. P. C. BROWN, S. VURAL and M. A. HUGHES

xiv


Genetics and Genomics of the Rice Blast Fungus Magnaporthe grisea: Developing an Experimental Model for Understanding Fungal Diseases of Cereals N. J. TALBOT and A. J. FOSTER Impact of Biotechnology on the Production of Improved Cereal Varieties R. G. SOLOMON and R. APPELS Overview and Prospects P. R. SHEWRY, P. A. LAZZERI and K. J. EDWARDS

Contents of Volume 35 Recent Advances in the Cell Biology of Chlorophyll Catabolism ¨ RTENSTEINER H. THOMAS, H. OUGHAM and S. HO The Microspore: A Haploid Multipurpose Cell A. TOURAEV, M. PFOSSER and E. HEBERLE-BORS The Seed Oleosins: Structure Properties and Biological Role J. NAPIER, F. BEAUDOIN, A. TATHAM and P. SHEWRY Compartmentation of Proteins in the Protein Storage Vacuole: A Compound Organelle in Plant Cells L. JIANG and J. ROGERS Intraspecific Variation in Seaweeds: The Application of New Tools and Approaches C. MAGGS and R. WATTIER Glucosinolates and their Degradation Products R. F. MITHEN


Contents of Volume 36 PLANT VIRUS VECTOR INTERACTIONS Edited by R. Plumb Aphids: Non-persistent Transmission T. P. PIRONE and K. L. PERRY Persistent Transmission of Luteoviruses by Aphids B. REAVY and M. A. MAYO Fungi M. J. ADAMS Whitefly Transmission of Plant Viruses J. K. BROWN and H. CZOSNEK Beetles R. C. GERGERICH Thrips As Vectors of Tospoviruses D. E. ULLMAN, R. MEIDEROS, L. R. CAMPBELL, A. E. WHITFIELD, J. L. SHERWOOD and T. L. GERMAN Virus Transmission by Leafhoppers, Planthoppers and Treehoppers (Auchenorrhyncha, Homoptera) E. AMMAR and L. R. NAULT Nematodes S. A. MacFARLANE, R. NEILSON and D. J. F. BROWN Other Vectors R. T. PLUMB

xv

xvi


Contents of Volume 37 ANTHOCYANINS IN LEAVES Edited by K. S. Gould and D. W. Lee Anthocyanins in Leaves and Other Vegetative Organs: an Introduction D. W. LEE and K. S. GOULD Le Rouge et le Noir: Are Anthocyanins Plant Melanins? G. S. TIMMINS, N. M. HOLBROOK and T. S. FEILD Anthocyanins in Leaves: History, Phylogeny and Development D. W. LEE The Final Steps in Anthocyanin Formation: a Story of Modification and Sequestration C. S. WINEFIELD Molecular Genetics and Control of Anthocyanin Expression B. WINKEL-SHIRLEY Differential Expression and Functional Significance of Anthocyanins in Relation to Phasic Development in Hedera helix L. W. P. HACKETT Do Anthocyanins Function as Osmoregulators in Leaf Tissues? L. CHALKER-SCOTT The Role of Anthocyanins for Photosynthesis of Alaskan Arctic Evergreens During Snowmelt S. F. OBERBAUER and G. STARR Anthocyanins in Autumn Leaf Senescence D. W. LEE


xvii

A Unified Explanation for Anthocyanins in Leaves? K. S. GOULD, S. O. NEILL and T. C. VOGELMANN

Contents of Volume 38 An Epidemiological Framework For Disease Management C. A. GILLIGAN Golgi-independent Trafficking of Macromolecules to the Plant Vacuole D. C. BASSHAM Phosphoenolpyruvate Carboxykinase: Structure, Function and Regulation R. P. WALKER and Z.-H. CHEN Developmental Genetics of the Angiosperm Leaf C. A. KIDNER, M. C. P. TIMMERMANS, M. E. BYRNE and R. A. MARTIENSSEN A Model for the Evolution and Genesis of the Pseudotetraploid Arabidopsis thaliana Genome Y. HENRY, A. CHAMPION, I. GY, A. PICAUD, A. LECHARNY and M. KREIS

Contents of Volume 39 Cumulative Subject Index Volumes 1–38

xviii


Contents of Volume 40 Starch Synthesis in Cereal Grains K. TOMLINSON and K. DENYER The Hyperaccumulation of Metals by Plants M. R. MACNAIR Plant Chromatin — Learning from Similarities and Differences J. BRZESKI, J. DYCZKOWSKI, S. KACZANOWSKI, P. ZIELENKIEWICZ and A. JERZMANOWSKI The Interface Between the Cell Cycle and Programmed Cell Death in Higher Plants: from Division unto Death D. FRANCIS The Importance of Extracellular Carbohydrate Production by Marine Epipelic Diatoms G. J. C. UNDERWOOD and D. M. PATERSON Fungal Pathogens of Insects: Cuticle Degrading Enzymes and Toxins A. K. CHARNLEY

Multiple Responses of Rhizobia to Flavonoids During Legume Root Infection

JAMES E. COOPER

Department of Applied Plant Science, Queen’s University Belfast, Belfast BT9 5PX, United Kingdom

I. II. III. IV. V.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Overview of Early Legume-Rhizobia Interactions . . . . . . . . . . . . . . . . . . . . . Synthesis and Release of Flavonoids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identification of Flavonoids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diversity and Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Chemoattractants and Growth Stimulators . . . . . . . . . . . . . . . . . . . . . . . . B. Inducers of Nodulation Genes Required for Nod Factor Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Induction of a Type III Secretion System and a Type I Secreted Protein. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Further Flavonoid-Dependent Gene Expression in Rhizobia . . . . . . . VI. Reception of Flavonoids by Rhizobia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Interactions with Regulatory NodD Proteins . . . . . . . . . . . . . . . . . . . . . . B. Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 3 7 12 15 16 17 19 26 31 31 35 41 42 42

ABSTRACT In the formation of legume-rhizobia symbioses flavonoids released from roots and seeds are best known as inducers of the bacterial nodulation genes that control the synthesis of reciprocal chitolipooligosaccharide signals (Nod factors) to the prospective host plant. However, successful symbiotic development requires the transmission Advances in Botanical Research, Vol. 41 Incorporating Advances in Plant Pathology 0065-2296/04 $35.00

Copyright 2004, Elsevier Ltd. All rights reserved.

2

J. E. COOPER

of a variety of other signals from rhizobia to plant roots and flavonoids initiate or mediate the production of most of them. This review considers the wide range of rhizobial responses to these secondary plant metabolites in the early phases of symbiotic interaction, including chemotaxis, growth stimulation, degradation, Nod factor synthesis, protein secretion by type I and III systems, surface polysaccharide production and expression of many new genes and proteins whose functions are only beginning to be analysed. Attention is also drawn to aspects of flavonoid-rhizobia interaction, such as release of compounds from inoculated roots and the reception of nodulation gene inducers by regulatory NodD proteins, that will need to be revisited by researchers if a complete understanding of the molecular dialogue between the partners is to be achieved.

I. INTRODUCTION The ability of legumes to capture (fix) atmospheric nitrogen endows them with special significance among agricultural plants: their productivity is theoretically independent of soil nitrogen status and extraneous fertilizer applications and they provide important grain and forage crops, both in temperate and in tropical zones. Nitrogen fixation can only occur when these plants are in the symbiotic state and the agents of fixation are soil bacteria from the genera Rhizobium, Bradyrhizobium, Sinorhizobium, Mesorhizobium and Azorhizobium—collectively known as rhizobia—that invade the root or stem cortex. Successful infections result in the formation of nodules into which atmospheric nitrogen diVuses to be reduced to ammonia by the nitrogenase enzyme of rhizobial bacteroids. Rhizobia exhibit varying degrees of specificity toward their hosts; with one known exception, Parasponia, infections are confined to the Leguminosae, and among rhizobial species and biovars particular specificity is often further displayed toward individual or small groups of legume genera, especially those originating in temperate or cool regions (Table I). Symbiotic promiscuity is a feature of some rhizobia and may be more widespread, especially among tropical strains, than hitherto appreciated (Perret et al., 2000). Strains with very broad legume host ranges do exist, an example being Rhizobium sp. NGR234, which can nodulate at least 112 legume genera (Pueppke and Broughton, 1999), and certain legumes (e.g., Phaseolus and Vigna) are considered to be nonselective hosts for rhizobia (Lewin et al., 1987; Michiels et al., 1998). The progression to the symbiotic state by two initially independent, freeliving partners is governed by reciprocal signal generation and perception, which has been described as a ‘‘molecular dialogue’’ (De´ narie´ et al., 1993). A class of plant secondary metabolites, the flavonoids, is responsible not only for initiating the formation of a symbiosis but also for influencing many of the subsequent events needed for successful root infections. This chapter reviews the full range of rhizobial responses to legume flavonoids,

RESPONSES OF RHIZOBIA TO FLAVONOIDS

3

TABLE I Some Species of Rhizobia and Their Legume Hosts Speciesa,b

Hosts nodulated

Rhizobium leguminosarum bv. phaseoli bv. trifolii bv. viciae Rhizobium etli Rhizobium galegae Rhizobium lupini Rhizobium tropici Sinorhizobium fredii c Sinorhizobium meliloti Mesorhizobium loti Bradyrhizobium japonicum Azorhizobium caulinodansd Rhizobium spp.e Bradyrhizobium spp.

Phaseolus Trifolium Pisum, Lens, Vicia Phaseolus Galega Lupinus Phaseolus, Leucaena Glycine, etc. Medicago, Melilotus, Trigonella Lotus, Astragalus Glycine, Macroptilium, Vigna Sesbania Vigna, Arachis, Desmodium, Lotus, etc. Sarothamnus, Ulex, etc.

a The taxonomy and nomenclature of the rhizobia are the subjects of much debate and controversy. See Broughton (2003); Farrand et al. (2003); Sawada et al. (2003); Young et al. (2001, 2003). b Rhizobium and Sinorhizobium species are relatively fast growing in laboratory culture media. Bradyrhizobium species grow more slowly and Mesorhizobium species display an intermediate growth rate. c Includes strain USDA257, which has a very broad host range. d Stem-nodulating and exceptional among rhizobia in fixing nitrogen in the free-living state. e Includes strain NGR234, which can nodulate at least 112 legume genera. Sometimes referred to as Sinorhizobium sp. strain NGR234.

including chemotaxis, growth stimulation, degradation, protein secretion, and expression of nodulation (nod) and other symbiotically active genes. Particular attention is paid to the recently characterized type III protein secretion systems in rhizobia, the release of flavonoids from inoculated roots and their reception by regulatory NodD proteins. The last two aspects of symbiotic interaction, despite having received much research attention for more than one-and-a-half decades, will need to be revisited if their precise biochemical and molecular mechanisms are to be clarified.

II. OVERVIEW OF EARLY LEGUME-RHIZOBIA INTERACTIONS The currently known components of the early infection phase are identified in Fig. 1. In virtually all examples studied to date symbiotic interaction is initiated by micromolar or nanomolar concentrations of flavonoids or isoflavonoids

4

J. E. COOPER

Fig. 1. Early signalling events in legume-rhizobia symbioses. Pathways dependent on or mediated by flavonoids are shown in gray. Some pathways are not operative in all symbioses. AHL, N-acyl homoserine lactone; EPS, extracellular polysaccharides; KPS, capsular polysaccharides; LPS, lipopolysaccharides.

in legume root or seed exudates. These compounds may initially assist rhizosphere colonization by acting as chemoattractants or, less likely, as growth enhancers for rhizobia. Other factors may play a more prominent role than flavonoids in stimulating rhizobial growth in the legume rhizosphere. For example, Sinorhizobium meliloti responds to very low concentrations of


5

external biotin through a regulatory locus, bioS (Streit and Phillips, 1997), resulting in increased growth rate and root colonization of Medicago sativa (Streit et al., 1996). A crucial contribution to the infection process is made when flavonoids interact with the constitutively expressed internal proteins of rhizobial regulatory nodD genes to form a transcriptional activator of other nod genes whose protein products are responsible for the synthesis of reciprocal signal molecules to the host plant root—the chitolipooligosaccharide Nod factors. This interaction constitutes the first of many elements that influence host specificity in legume-rhizobia symbioses. Flavonoids and isoflavonoids are not, however, inert compounds in this context, because rhizobia are capable of metabolizing them to yield a plethora of polycyclic and monocyclic phenolic products, some of which themselves possess nod gene-inducing or gene-inhibiting properties. Nonflavonoid nod gene inducers are also secreted by some legumes, in the form of betaines or aldonic acids, but compared to flavonoids these compounds are active only at higher concentrations. The term nod is used at this point and hereafter in the manner adopted by Downie (1998): as a generic designation for nodulation genes (e.g., nod, nol, and noe), except when referring to specific examples (e.g., nodA). Chitolipooligosaccharide Nod factors are essential signals for rhizobial entry into legume roots (Relic´ et al., 1994), and the success or otherwise of the infection process is in large part determined by their structural features. Application of nanomolar or femtomolar concentrations of purified rhizobial Nod factor to the roots of an appropriate legume host elicits the following responses, which can be detected by biochemical, molecular, and microscopical analysis: (1) deformation of root hairs (Lerouge et al., 1990) accompanied by root hair plasma membrane depolarization (Ehrhardt et al., 1992; Felle et al., 1995); (2) rapid increases then oscillations in intracellularfree calcium in root hairs, often referred to as calcium spiking (Ehrhardt et al., 1996; Gehring et al., 1997; Wais et al., 2000, 2002; Walker et al., 2000); (3) changes in the root hair cytoskeleton (Ca´ rdenas et al., 1998; Timmers et al., 1998); (4) preinfection thread formation in deformed root hairs (van Brussel et al., 1992); and (5) localized cortical cell division at the sites of root nodule primordia (Lo´ pez-Lara et al., 1995; Spaink, 1992; Spaink et al., 1993). Inhibition by Nod factor of the reactive oxygen-generating system in Medicago truncatula roots, indicating a plant defense suppression function, has recently been reported (Shaw and Long, 2003). Nod factors alone can induce some of the plant genes (nodulins) that are expressed in the preinfection, infection, nodule development, and nodule function phases of symbiotic interaction, some examples of the more rapidly expressed genes being enod12 (Scheres et al., 1990), enod40 (Kouchi and Hata, 1993), rip1

6

J. E. COOPER

(Cook et al., 1995), and dd23b (Crockard et al., 2002). Auxin flow in roots at the earliest stages of nodule formation is perturbed by Nod factors in conjunction with endogenous root flavonoids acting as auxin transport inhibitors (Boot et al., 1999; Mathesius et al., 1998). The contribution of flavonoids in this case appears to involve regulation of auxin breakdown by peroxidase (Mathesius, 2001). These findings confirm the earlier conclusion of Jacobs and Rubery (1988) that flavonoids regulate auxin transport in plant roots. Nod factors also control the number of nodules formed on a root system by inducing an autoregulation response in the host plant (van Brussel et al., 2002). The nature of signal-transduction pathways leading from the perception of Nod factors to symbiosis-related gene activation is currently the subject of intensive research (Cullimore et al., 2001; Goedhart et al., 2003; Limpens and Bisseling, 2003; Oldroyd, 2001). A breakthrough was achieved with the discovery of a symbosis receptor-like kinase (SYMRK) gene in Lotus (Stracke et al., 2002) and a nodulation receptor Kinase (NORK) in Medicago (Endre et al., 2002) that is required for early signal transduction in both rhizobial and mycorrhizal symbioses. More recently two genes that encode LysM receptor-like kinases that function upstream of SYMRK and could be direct receptors for rhizobial Nod factors were discovered in Lotus japonicus (Madsen et al., 2003; Radutoiu et al., 2003). Likewise, in Medicago truncatula two receptor-like kinase genes have been recognized as encoders of potential Nod factor receptors (Limpens et al., 2003). Other genes that are involved in the transduction of rhizobial Nod factor signals but are not required for mycorrhizal infection have also been identified in this legume (Ben Amor et al., 2003; Oldroyd and Long, 2003). In addition to Nod factors, flavonoids induce the synthesis and release by rhizobia of proteins that fulfill a variety of functions during plant infection. They include several that are associated with a type III secretion system and another, NodO, which is secreted by a type I system. Transcriptional and proteomics analyses have identified many other rhizobial genes and proteins whose expression is flavonoid-dependent but whose functions have yet to be defined. Flavonoids, either directly or via the Nod factors and secreted proteins whose synthesis depends on them, are therefore of prime significance as signal molecules and mediators of host specificity in legume-rhizobia symbioses. Other compounds are also required for successful symbiotic development, and their biosynthesis and structural features are influenced by flavonoids in many cases. Included in this category are the various surface polysaccharides of rhizobia that fulfill host recognition or defense avoidance/suppression functions: extracellular polysaccharides (EPS), lipopolysaccharides (LPS), K-antigen or capsular polysaccharides (KPS), and cyclic glucans (for reviews,


7

see Becker and Pu¨ hler, 1998; Becker et al., 2000; Carlson et al., 1999; Fraysse et al., 2003; Kannenberg and Brewin, 1994; Mitho¨ fer, 2002; Noel and Duelli, 2000; Price, 1999; Spaink, 2000). Quorum-sensing N-acyl homoserine lactone (AHL) signals, used by rhizobia to coordinate the behaviour of individual cells in a population, act as autoinducers of rhizosphere-expressed (rhi) genes in rhizobia that influence host nodulation (Wisniewski-Dye and Downie, 2002) and are essential for the expression of certain exopolysaccharide synthesis genes in S. meliloti (Marketon et al., 2003). AHL signals also elicit responses from prospective host plants in the form of changes in the accumulation of proteins (Mathesius et al., 2003). In the legume partner carbohydrate-binding proteins on root surfaces, the lectins, have been regarded as important determinants of host recognition following the pioneering studies of Bohlool and Schmidt (1974), Dazzo and Hubbell (1975), and Hamblin and Kent (1973). Despite the eVorts of various research groups in the intervening period, many details of lectin function remain unresolved, although it appears likely that they mediate host specificity through selective interactions with Nod factors and/or rhizobial surface polysaccharides (Bhattacharya et al., 2002; Etzler et al., 1999; Hirsch, 1999; Kalsi and Etzler, 2000; van Rhijn et al., 1998, 2001).

III. SYNTHESIS AND RELEASE OF FLAVONOIDS Flavonoids are secondary metabolic products of the central phenylpropanoid pathway and the acetate-malonate pathway of plants. Thus all flavonoids are derivatives of phenylalanine from the shikimic acid pathway and malonyl CoA from the acetyl CoA carboxylase reaction. Condensation of 4-coumaroyl CoA from the phenylpropanoid pathway with three molecules of malonyl CoA by chalcone synthase (CHS) creates the central chalcone precursor from which all other flavonoid structures are ultimately derived (Fig. 2). The main flavonoid subclasses (e.g., chalcones, flavones, flavanones, flavonols, flavan 3-ols, proanthocyanidins, isoflavones, isoflavans, pterocarpans) contain numerous compounds involved in many plant functions (Shirley, 1996; Woo et al., 2002), including pigmentation, protection against ultraviolet (UV) light, pollen fertility, regulation of auxin transport, and hydrogen peroxide scavenging, as well as interactions with symbiotic microorganisms or defense against pathogens. Detailed accounts of their structures and biosynthesis are available (Aoki et al., 2000; Dixon, 1999; Forkmann and Heller, 1999), and their significance for metabolic engineering in plants has been emphasized by Dixon and Steele (1999).

8

J. E. COOPER

Fig. 2. Flavonoid biosynthesis in legumes. CA4H, cinnamic acid 4-hydroxylase; CHI, chalcone isomerase; CHR, chalcone reductase; CHS, chalcone synthase; CL, coumaroyl-CoA ligase; FLS, flavonol synthase; FNS, flavone synthase; F3,4R (FDR), flavan 3,4-diol reductase; F3H, flavanone 3-hydrolase; HFR (DFR), 3-OHflavanone 4-reductase (dihydroflavonol 4-reductase); IFS, isoflavone synthase; IFR, isoflavone reductase; I20 H, isoflavone 20 -hydroxylase; PAL, phenylalanine ammonia lyase; PAS, proanthocyanidin synthase; PTS, pterocarpan synthase; VER, vestitone reductase. (From Jain and Nainawatee, 2002; StaVord, 1997.) Reprinted by permission from The Botanical Review, 63(1) ß 1997. The New York Botanical Garden Press.


9

Flavonoids acting as primary signals to rhizobia have been found in legume seed coat and root exudates. When deposited on seed coats, flavonoids are simply released into the surrounding aqueous environment during imbibition without involvement of any metabolic regulation (Hartwig and Phillips, 1991). Storage of flavonoids in roots and their release from epidermal tissues are, however, subject to internal metabolic controls, and strong evidence exists for a process of concurrent synthesis and release. For example, U-14C incorporated into phenylalanine was found in root exudate flavonoids (Maxwell and Phillips, 1990). Other data linking synthesis to release have been reviewed by Jain and Nainawatee (2002), including inhibition of phenylalanine ammonia lyase (PAL) by 2-aminooxy-3-phenylpropionic acid (AOPP), which decreased the synthesis of 7,40 -dihydroxyflavanone by 90–95% and its exudation by 50% (Amrhein and Godeke, 1977). Exudation of both 7,40 -dihydroxyflavone and 4,40 -dihydroxy-20 -methoxychalcone was also tightly linked to their concurrent synthesis. A relatively high proportion of unlabelled to labelled 7,40 -dihydroxyflavanone in the root exudate of AOPP-treated plants indicated that this compound could also be released from a presynthesized pool within the root. Indirect evidence for a linkage between synthesis and release of flavonoid signals to rhizobia comes from experiments in which inhibition of PAL by 2-aminoindan-2-phosphonic acid (AIP) was accompanied by decreased nodulation of Medicago roots (Zon and Amrhein, 1992). Flavonoids may be released as aglycones or glycosidic conjugates (Maxwell and Phillips, 1990). The latter are inherently more soluble and may therefore have a greater potential for diVusion from the root surface before being hydrolyzed to the aglycone form by rhizobia, other soil microorganisms, or plant exoenzymes (Hartwig and Phillips, 1991). Rhizobia themselves may be able to alter the hydrophobicity of flavonoid aglycones: complexation of luteolin with cyclosophoraoses produced by S. meliloti markedly enhances the solubility of this nod gene inducer (Lee et al., 2003). The presence of rhizobia in the legume rhizosphere also influences the quantity and perhaps the types of flavonoids released from roots. Increases in overall flavonoid-dependent nod gene-inducing activity of root extracts or exudates following inoculation with homologous rhizobia (the so-called Ini response—increase in nod gene-inducing flavonoids) have been reported for white clover (Rolfe et al., 1988), vetch (van Brussel et al., 1990), soybean (Cho and Harper, 1991), and alfalfa (Dakora et al., 1993a). Inoculation with heterologous rhizobia can also generate an Ini eVect in alfalfa that is lower than that produced by homologous S. meliloti (Dakora et al., 1993a) and in white clover, in which inoculation with Rhizobium leguminosarum bv. viciae gave a fourfold increase in nod gene-inducing activity of root extracts

10

J. E. COOPER

compared to inoculation with homologous R. leguminosarum bv. trifolii (Rolfe et al., 1988). In terms of specific compounds, root exudates of Phaseolus vulgaris inoculated with R. leguminosarum bv. phaseoli contained greater quantities of the phytoalexin coumestrol (3,9-dihydroxycoumestan) and its isoflavonoid precursor daidzein (40 ,7-dihydroxyisoflavone) than did exudates of sterile plants (Dakora et al., 1993b). Bolanos-Vasquez and Werner (1997) also found increased quantities of daidzein, naringenin (40 ,5,7-trihydroxyflavanone), liquiritigenin (40 ,7-dihydroxyflavanone), and isoliquiritigenin (20 ,40 ,4-trihydroxychalcone) in root exudates of this legume following inoculation with homologous rhizobia. In soybean root exudate elevated concentrations of daidzein, genistein (40 ,5,7-trihydroxyisoflavone), and coumestrol were detected after inoculation with wild-type Bradyrhizobium japonicum. Purified Nod factors from B. japonicum and Rhizobium sp. NGR234 produced the same eVect, whereas a B. japonicum mutant lacking the ability to synthesize Nod factors had no influence on the release of these compounds (Schmidt et al., 1994). Root exudates of Medicago sativa inoculated with S. meliloti were qualitatively diVerent with respect to flavonoid content compared with exudates from sterile plants (Dakora et al., 1993a); the former contained three extra compounds in the form of an aglycone and a glycoside of medicarpin and a nod gene inducer identified as formononetin-7-O-600 -O-malonylglucoside, a conjugate of the medicarpin precursor formononetin (7-hydroxy-40 methoxyisoflavone). This conjugate may have been identified incorrectly on account of the unavailability of an authentic standard for comparison of nuclear magnetic resonance (1H-NMR) and fast atom bombardment mass spectrometry (FAB-MS) (Schlaman et al., 1998); when formononetin-7-O600 -O-malonylglucoside was later purified from alfalfa roots, it failed to induce nod genes in S. meliloti (Coronado et al., 1995). Recourt et al. (1991) reported six new flavanones and two new chalcones in root exudates of Vicia sativa inoculated with R. leguminosarum bv. viciae, although it is not clear whether some of the flavanones were among those previously isolated, but not identified to the level of individual compounds, from sterile root exudate of the same legume by Zaat et al. (1989). In Trifolium subterraneum an additional nod gene-inducing compound, thought to be 40 ,7-dihydroxyflavone, was found in root exudates 3 days after inoculation with R. leguminosarum bv. trifolii and a diVerent, uncharacterized inducer was detected 2 days later (Lawson et al., 1996). Whether or not new flavonoids found in root exudates of plants after inoculation with rhizobia are the products of altered internal plant biosynthetic pathways, changes in the patterns of release of preformed compounds from internal pools, or biotransformations by rhizobia in the


11

rhizosphere, is often diYcult to ascertain. Rhizobia have been shown to degrade legume flavonoids to yield a great variety of flavonoid and other phenolic metabolites. Degradative activity was displayed toward flavonoids presented as authentic compounds at 100 mM (Rao et al., 1991), 10 mM (Rao and Cooper, 1994, 1995), and an even lower, nod gene-inducing concentration of 2 nM (Rao et al., 1996), as well as toward naturally occurring flavonoids in unconcentrated root exudates (Steele et al., 1999). Incubations of exudates from sterile plants with wild-type rhizobia have yielded conflicting results: the medicarpin found in inoculated root exudates of Medicago sativa by Dakora et al. (1993a) could not be identified in exudate from uninoculated plants after its incubation with S. meliloti, but Steele et al. (1999) detected new flavonoids in root exudate from uninoculated Lotus pedunculatus after incubation with Mesorhizobium loti. No studies of this type have included mutant rhizobia lacking degradative activity toward flavonoids. Challenging legume roots with Nod factor alone would eliminate the possibility of rhizobial degradation and permit the unequivocal identification of any new flavonoids released into root exudate. Nod factor inoculation does produce a general Ini response in Vicia sativa (van Brussel et al., 1990). However, in the only study to combine a Nod factor treatment with detection of individual flavonoids, analysis was restricted to three compounds already present in sterile soybean root exudate, all of which were released in greater quantities from Nod factor treated roots (Schmidt et al., 1994). If new flavonoids were released from roots after Nod factor treatment, in this study they would have remained unidentified. There is much evidence to show that the presence of rhizobia or their Nod factors elicits changes in enzyme activity and gene expression in the plant phenylpropanoid biosynthetic pathway. Increases in PAL and, especially, CHS expression have been reported in several legumes, but the timing of the response varies from several hours to several days postinoculation. PAL is encoded by a multigene family, and at least seven diVerent CHS isoforms have been characterized in Medicago sativa (McKhann and Hirsch, 1994) and six in Phaseolus vulgaris (Ryder et al., 1987). Inoculation of Glycine max with B. japonicum induces expression of subsets of the PAL and CHS gene families, but this was considered to be a postinfection event (Estabrook and SenguptaGopalan, 1991). CHS transcript accumulation in Vicia sativa reached its maximum approximately 2 days after inoculation with R. leguminosarum bv. viciae (Recourt et al., 1992), the CHS5 gene in Trifolium subterraneum roots was up-regulated within 6 hours of inoculation (Lawson et al., 1994), and in Medicago sativa root hairs CHS6-4 and CHS4-1 were up-regulated several days after inoculation with S. meliloti (McKhann et al., 1997). Nod factor stimulated CHS1 expression in root hairs of Vigna unguiculata 1 day

12

J. E. COOPER

after treatment (Krause et al., 1997) and in Medicago microcallus suspensions isoflavone reductase (IFR) gene expression was induced by relatively high concentrations of Nod factor from the cognate S. meliloti (Savoure´ et al., 1994). Enhanced expression of flavonoid biosynthesis genes has sometimes been accompanied by detection of new flavonoids in roots or root exudates (Lawson et al., 1996; McKhann et al., 1997; Recourt et al., 1991). However, it is often diYcult to distinguish changes connected with internal processes, such as auxin transport inhibition (Mathesius et al., 2000), from those that may aVect nod gene induction. Linking increases in gene transcription to the appearance of specific flavonoids is another problem; it is worth noting that the only certain consequence of increased CHS expression is the production of more chalcones. Radio-carbon tracing studies together with precise histochemical localizations of individual flavonoids in roots, as advocated by StaVord (1997), would help to clarify this aspect of plant response to rhizobial inoculation.

IV. IDENTIFICATION OF FLAVONOIDS For many years the preferred technique for separating and identifying flavonoids in plant tissues and their exudates has been high-performance liquid chromatography (HPLC), usually employing a variable wavelength UV detector. HPLC protocols have been developed specifically for the analysis of flavonoids in legumes (Graham, 1991a,b). UV spectroscopy is often supplemented by mass spectrometry (MS) or nuclear magnetic resonance (NMR) for confirmation of compound structures. The coupling of instruments for separation with those providing structural data has had a profound eVect in the field of phytochemical analysis, opening the way for more sophisticated approaches that are particularly suitable for screening tissue extracts or exudates for their full complement of flavonoid compounds (Kite et al., 2003). Hostettmann et al. (1996) recommended the application of hyphenated techniques such as liquid-chromatography-mass spectrometry (LC-MS) and HPLC-UV (LC-UV) as screening systems for flavonoids and related compounds in plants on the grounds that they oVer a complete and rapid analysis of small amounts of material. Because flavonoids exhibit characteristic UV spectra, photodiode array detection and postcolumn derivatization can provide much information on these compounds and their substitution patterns. A schematic diagram of the instrumentation used by Hostettmann et al. to analyze plant extracts for phenolic compounds of potential medicinal value is shown in Fig. 3. A single injection of a plant extract or exudate suYces to provide all MS and UV data for its polyphenolic constituents (excluding UV spectral shift measurements).


13

Fig. 3. Scheme for LC-UV-MS analysis of plant tissue extracts or exudates for flavonoids, incorporating shift reagents to provide extra UV spectral information. (Modified from Hostettmann et al., 1996.)

This system was used successfully to identify flavonol glycosides in nonderivatized methanolic extracts of Epilobium leaves (Ducrey et al., 1995), and further information on hydroxylation patterns and sugar positions was obtained by means of postcolumn derivatization with up to five diVerent UV shift reagents prior to diode array detection. A variant of this arrangement, with separate LC-UV and gas chromatography-mass spectrometry (GC-MS) analyses, was used by Bolanos-Vasquez and Werner (1997) to identify isoflavonoids and flavonoids in methanolic extracts of Phaseolus vulgaris root exudates. Another analytical system, again based on multiple separation-identification techniques, enabled Steele et al. (1999) to identify flavonoids in seed and root exudates of Lotus pedunculatus. In this case a preseparation step, which involved high-performance thin-layer chromatography with densitometry (HPTLC-UV) to provide UV absorption spectra of individual spots in the range 200–500 nm (Scheidemann and Wetzel, 1997), was inserted before two further hyphenated techniques operating in parallel: capillary zone electrophoresis coupled to a diode array detector (CZE-UV) and GC-MS (Fig. 4). Postrun analysis software permitted comparisons between unknown sample compound UV spectra and a reference library of spectra from authentic flavonoids and related compounds. Examples of results for two flavonoids—quercetin (3,30 ,40 ,5,7-pentahydroxyflavonol) and

14

J. E. COOPER

Fig. 4. Scheme for separation and identification of flavonoids in legume seed and root exudates using a combination of high-performance thin-layer chromatographydensitometry (HPTLC-UV), capillary zone electrophoresis coupled to a diode array detector (CZE-UV), and gas chromatography-mass spectrometry (GC-MS). (From Steele et al., 1999.)

catechin (3,30 ,40 ,5,7-pentahydroxyflavan)—detected in seed exudates of Lotus pedunculatus by CZE-UV and GC-MS are shown in Fig. 5. Capillary electrophoresis has been applied to the profiling of isoflavonoids and their glycosidic conjugates in legume root extracts (Baggett et al., 2002), with results from micellar electrokinetic capillary chromatography (MEKC) showing good correlation with results from HPLC. Capillary electrophoresis coupled to mass spectrometry (CE-MS) can also be an eVective alternative to LC-MS for flavonoid identification in plant extracts (Huck et al., 2002).


15

Fig. 5. Identification of quercetin (a) and catechin (b) in seed exudates of Lotus pedunculatus from UV and mass spectral comparisons with authentic standard compounds. (From Steele et al., 1999.)

V. DIVERSITY AND FUNCTIONS In the earliest phases of their interaction with legumes, rhizobia display a variety of responses to the presence of flavonoids in the rhizosphere. Some compounds are chemoattractants for rhizobia and may stimulate their growth. The same, or other, flavonoids act either as inducers or anti-inducers for the transcription of rhizobial nod genes. Compounds that induce nod

16

J. E. COOPER

genes can also induce genes that encode the synthesis and release of proteins from some rhizobia via type I or type III secretion systems. Surface polysaccharide structures may be modified in the presence of flavonoids, and the expression of other genes, whose functions in many cases have not yet been determined, is also known to be flavonoid-dependent. A. CHEMOATTRACTANTS AND GROWTH STIMULATORS

Rhizobia exhibit positive chemotaxis toward unfractionated legume epidermal exudates (Gaworzewska and Carlile, 1982) and to individual compounds found therein, including a number of flavonoids. Luteolin, 4,40 -dihydroxy-20 -methoxychalcone, 7,40 -dihydroxyflavone, and 7,40 -dihydroxyflavanone from alfalfa all induce positive chemotaxis in S. meliloti (Caetano-Anolle´ s et al., 1988; Dharmatilake and Bauer, 1992), and in the case of luteolin chemotaxis is a nodDdependent process (Caetano-Anolle´ s et al., 1988). For R. leguminosarum bv. phaseoli, apigenin (40 ,5,7-trihydroxyflavone), luteolin, umbelliferone, and acetosyringone all act as chemoattractants (Aguilar et al., 1988), whereas naringenin, kaempferol (3,40 ,5,7-tetrahydroxyflavonol)and apigenin are chemoattractants for R. leguminosarum bv. viciae (Armitage et al., 1988). In the case of B. japonicum, Kape et al. (1991) found no chemotaxis to isoflavonoids from its soybean host; however, hydroxycinnamic acids were strong chemoattractants. Similarly, Barbour et al. (1991) concluded that isoflavones were not the principal attractants of B. japonicum in soybean seed and root exudate. In at least one legume, Medicago sativa, rhizobia are attracted to that region of the root from which nod gene-regulating flavonoids are exuded (Gulash et al., 1984; Peters and Long, 1988). Rhizobia are positively chemotactic to many other compounds and sometimes more strongly so than toward flavonoids. Examples include sugars (Bowra and Dilworth, 1981), common amino acids (Barbour et al., 1991; Go¨tz et al., 1982), dicarboxylic acids (Barbour et al., 1991), a glycoprotein (Currier and Strobel, 1977), as well as aromatic acids, hydroxyaromatic acids, and simple phenolic compounds (Aguilar et al., 1988; Parke et al., 1985). Chemotaxis to flavonoids or any other compounds does not, however, appear to be an essential component of the infection process: rhizobial mutants lacking flagellae, motility, or chemotactic behaviour produce as many nodules as wild-type strains (Ames et al., 1980). Depending on concentration, flavonoids are potentially toxic to bacteria, and inhibitory eVects on rhizobial growth have been reported. Medicarpin and kievitone from soybean roots were strong inhibitors of growth for B. japonicum, R. lupini, and two fast-growing Lotus rhizobia, whereas phaseolin and maakiain were slightly less inhibitory. These compounds had no eVect on the growth of clover, pea, and alfalfa rhizobia (Pankhurst and


17

Biggs, 1980). Soybean rhizobia are normally sensitive to the phytoalexin glyceollin (Parniske et al., 1991). In contrast, Lameta and Jay (1987) reported a stimulatory eVect of low concentrations of daidzein on the growth of B. japonicum. At concentrations of 1–10 mM, quercetin, quercetin-3-Ogalactoside, luteolin, and luteolin-7-O-glucoside increased the growth rate of S. meliloti in a defined minimal medium (Hartwig et al., 1991). The mechanism underlying growth stimulation in S. meliloti by luteolin appears to be independent of nod gene induction, because mutants lacking regulatory nodD genes also exhibited enhanced growth rates (Hartwig et al., 1991). A positive eVect of quercetin on the growth of S. meliloti was also recorded by Jain and Nainawatee (1999). In the same study elevated levels of exopolysaccharide production and increased activity of citric acid cycle and 6-phosphoglucanate pathway enzymes was observed after culture medium supplementation with naringenin; however, there was no eVect on growth and protein content. Genistein, naringenin, chrysin, and apigenin all promoted the growth of S. fredii USDA257 in late log phase (Lin et al., 1999). The degradative activity of rhizobia toward flavonoids is considered in detail in the section on metabolism, appearing later in this chapter. B. INDUCERS OF NODULATION GENES REQUIRED FOR NOD FACTOR SYNTHESIS

The availability of lacZ reporter fusions to some of the genes required for Nod factor synthesis (e.g., nodABC) permitted the discovery that transcription of these genes required a factor or factors from the host plant (Innes et al., 1985; Mulligan and Long, 1985; Rossen et al., 1985; Zaat et al., 1987). The first nod gene-inducing flavonoids to be discovered in this way were the flavones luteolin (5,7,30 ,40 -tetrahydroxyflavone) (Peters et al., 1986) and 7,40 -dihydroxyflavone (Redmond et al., 1986), the former having been isolated from the seed coat of Medicago sativa and the latter from roots of Trifolium repens. They are nod gene inducers for S. meliloti and R. leguminosarum bv. trifolii, respectively. Shortly after these important initial discoveries, apigenin-7-O-glucoside (5,7,40 -trihydroxyflavone-7-O-glucoside) and eriodictyol (5,7,30 ,40 -tetrahydroxyflavanone) in Pisum sativum (pea) were identified as inducers for R. leguminosarum bv. viciae (Firmin et al., 1986). Inducers from these and other flavonoid subclasses such as chalcones, flavonols, anthocyanidins, and isoflavonoids have, subsequently, been isolated from a variety of legumes, including Glycine max (soybean) (Banfalvi et al., 1988; D’Arcy-Lameta, 1986; Kape et al., 1992; Kosslak et al., 1987; Smit et al., 1992); Phaseolus vulgaris (common bean) (Bolanos-Vasquez and Werner, 1997; Hungria et al., 1991a,b, 1992); Medicago sativa (alfalfa) (Hartwig et al., 1990a; Maxwell et al.,

18

J. E. COOPER

1989; Phillips et al., 1994); Vicia sativa (vetch) (Zaat et al., 1989); Sesbania (Messens et al., 1991); Vigna subterranea (African Bambara groundnut); Vigna unguiculata (cowpea) (Dakora, 2000), and Galega orientalis (Suominen et al., 2003). Combinations of inducers can be more eVective than single flavonoid compounds. Synergistic inducing eVects of eriodictyol and naringenin with genistein, and liquiritigenin or isoliquiritigenin with daidzein, were observed for R. leguminosarum bv. phaseoli (Bolanos-Vasquez and Werner, 1997; Hungria et al., 1992). Hesperetin (30 ,5,7-trihydroxy-40 -methoxyflavanone) and naringenin were found by Begum et al. (2001) to generate higher levels of nod gene expression in pea rhizobia than either compound alone. Some flavonoids act as antagonists (anti-inducers) of nod gene induction by other flavonoids (Djordjevic et al., 1987; Firmin et al., 1986; Kosslak et al., 1990; Peters and Long, 1988; Zuanazzi et al., 1998), an eVect that may be based on competitive inhibition because it can be overcome by increasing inducer concentration (Peters and Long, 1988). The fact that inducers and anti-inducers are often present in the exudates of a single legume species has prompted the suggestion that in vivo levels of nod gene induction are the net outcome of positive and negative flavonoid eVects on the process (Jain and Nainawatee, 2002; Rolfe et al., 1988; Zuanazzi et al., 1998). Compounds that are inducers for certain rhizobia can be anti-inducers for others: the isoflavones genistein and daidzein are inducers of nod gene expression in B. japonicum and Rhizobium sp. NGR234, but they are anti-inducers for R. leguminosarum bv. trifolii and viciae. The ecological significance of antiinducers is unclear; some workers have used anti-inducing compounds that are not known to be released by the host of the microsymbiont under study, and Schlaman et al. (1998) pointed out that levels of nod gene-inducing activity recoverable from alfalfa rhizosphere soil (Leo´ n-Barrios et al., 1993) were so low that no plant benefit could be envisaged from decreasing them further. On the other hand, flavonoid-independent repression of nod gene expression by proteins such as NolR and NolA does occur (see the section, ‘‘Interactions with Regulatory NodD Proteins,’’ later in the chapter) and, in S. meliloti, mutants displaying significantly reduced expression of nodF (involved in providing key structural components to the nonreducing terminus of a Nod factor; see Table III and Fig. 7) are fully capable of nodulating alfalfa (Wells and Long, 2003). The fact that strains within B. japonicum respond diVerentially to flavonoid inducers and anti-inducers (Cunningham et al., 1991; Kosslak et al., 1990) casts doubt on the idea that plants release certain flavonoids simply to inhibit nod gene induction by other flavonoids. Some root exudate flavonoids have a dual function in symbiotic communication: daidzein, genistein, and isoliquiritigenin are


19

inducers both of nod genes and of resistance to the soybean phytoalexin glyceollin in B. japonicum (Kape et al., 1992; Parniske et al., 1991). Certain nonflavonoid compounds also act as nod gene inducers but only at much higher concentrations (in the micromolar or millimolar range compared with nanomolar concentrations for most flavonoids). Examples are the betaines stachydrine and trigonelline from Medicago species (Phillips et al., 1992, 1995) and the aldonic acids erythronic and tetronic acid from Lupinus albus (Gagnon and Ibrahim, 1998). As noted by Aoki et al. (2000), these aldonic acids are inducers for Lotus-nodulating rhizobia (as measured by Nod factor production in M. loti rather than the usual nod gene expression assay using reporter gene fusions) as well as lupin rhizobia (R. lupini), but their presence in Lotus tissues or epidermal exudates has not been established. The natural nod gene inducers for M. loti remain unidentified despite the detection of many flavonoids in Lotus roots, root nodules, and root exudates (Cooper and Rao, 1992; Morris and Robbins, 1992; Pankhurst and Jones, 1979; Robbins et al., 1995; Steele et al., 1999; Wagner et al., 1996) and attempts to determine the inducing properties of most of them (Lo´pez-Lara et al., 1995). An example of nonflavonoid induction of Rhizobium sp. NGR234 nod genes by the simple phenolic compounds vanillin and isovanillin in seedling extracts of a nonlegume (i.e., wheat) has also been reported (Le Strange et al., 1990). A list of nod gene inducers of legume origin is presented in Table II, and the structures of inducing compounds from four flavonoid subclasses are shown in Fig. 6. C. INDUCTION OF A TYPE III SECRETION SYSTEM AND A TYPE I SECRETED PROTEIN

It has recently been established that flavonoid nod gene inducers are also required for the transcription of type III secretion system (TTSS) genes that are found in some rhizobia. TTSS genes occur in several gram-negative plant and animal pathogens and are characterized by secretion of proteins into the extracellular environment or directly into eukaryotic cytoplasm when contact is made with host cells (Cornelis, 2000; Cornelis and Van Gijsegem, 2000; He, 1998; Hueck, 1998). The functions of such proteins are subversion of the mammalian immune system in the case of animal pathogens such as Shigella and Yersinia (Cornelis and Wolf-Watz, 1997) and elicitation of a hypersensitive response in resistant plants or disease in susceptible ones by plant pathogenic bacteria (He, 1998). Probably the first indication that TTSS components occurred in rhizobia was obtained by Sadowsky et al. (1988), who discovered two daidzein- and genistein-inducible genes in S. fredii. These genes showed no significant

20

TABLE II Rhizobial nod Gene Inducers Isolated from Legumes Under Sterile Conditions Host legume Medicago sativa

Vicia sativa Trifolium repens Glycine max

Luteolin (5,7,30 40 -tetrahydroxyflavone) Chrysoeriol (30 -methoxy-5,7,40 -trihydroxyflavone) Liquiritigenin (7,40 -dihydroxyflavanone) 7,40 -Dihydroxyflavone Methoxychalcone (4,40 -dihydroxy-20 -methoxychalcone) Stachydrine (betaine) Trigonelline (betaine) Apigenin-7-O-glucoside (5,7,40 -trihydroxyflavone-7-O-glucoside) Eriodictyol (5,7,30 ,40 -tetrahydroxyflavanone) 3,5,7,30 -Tetrahydroxy-40 -methoxyflavanone 7,30 -Dihydroxy-40 -methoxyflavanone Four more partially characterized flavanones 7,40 -Dihydroxyflavone Geraldone (7,40 -dihydroxy-30 -methoxyflavone) 40 -Hydroxy-7-methoxyflavone Daidzein (7,40 -dihydroxyisoflavone) Genistein (5,7,40 -trihydroxyisoflavone) Coumestrol (3,9-dihydroxycoumestan) Isoliquiritigenin (4,20 ,40 -trihydroxychalcone) Genistein-7-O-glucoside Genistein-7-O-(600 -O-malonylglucoside) Daidzein-7-O-(600 -O-malonylglucoside)

Reference Peters et al. (1986) Hartwig et al. (1990a) Maxwell et al. (1989) Maxwell et al. (1989) Maxwell et al. (1989) Phillips et al. (1992) Phillips et al. (1992) Firmin et al. (1986) Firmin et al. (1986) Zaat et al. (1989) Zaat et al. (1989) Zaat et al. (1989) Redmond et al. (1986) Redmond et al. (1986) Redmond et al. (1986) Kosslak et al. (1987) Kosslak et al. (1987) Kosslak et al. (1987) Kape et al. (1992) Smit et al. (1992) Smit et al. (1992) Smit et al. (1992)

J. E. COOPER

Pisum sativum

Compound

Phaseolus vulgaris

Sesbania rostrata Lupinus albus Galega orientalis

Hungria et al. (1991a) Hungria et al. (1991a) Hungria et al. (1991a) Hungria et al. (1991a) Hungria et al. (1991a) Hungria et al. (1991a) Hungria et al. (1991b) Hungria et al. (1991b) Hungria et al. (1991b) Bolanos-Vasquez and Werner (1997) Bolanos-Vasquez and Werner (1997) Bolanos-Vasquez and Werner (1997) Bolanos-Vasquez and Werner (1997) Dakora (2000) Dakora (2000) Dakora (2000) Messens et al. (1991) Gagnon and Ibrahim (1998) Gagnon and Ibrahim (1998) Suominen et al. (2003)


Cowpea

Delphinidin (3,5,7,30 ,40 ,50 -hexahydroxyflavylium) Kaempferol (3,5,7,,40 -tetrahydroxyflavonol) Malvidin (3,5,7,40 -pentahydroxy-30 ,50 -dimethoxyflavylium) Myricetin (3,5,7,30 ,40 ,50 -hexahydroxyflavone) Petunidin (3,5,7,40 ,50 -pentahydroxy-30 -methoxyflavylium) Quercetin (3,5,7,30 ,40 -pentahydroxyflavonol) Eriodictyol Genistein Naringenin (5,7,40 -trihydroxyflavanone) Daidzein Liquiritigenin Isoliquiritigenin Coumestrol Daidzein Genistein Coumestrol Liquiritigenin Erythronic acid (aldonic acid) Tetronic acid (aldonic acid) Uncharacterized chalcone

21

22

J. E. COOPER

Fig. 6. Rhizobial nod gene inducers from four flavonoid subclasses. The carbon numbering convention for chalcones diVers from that in flavones, flavanones, and isoflavones. B-ring numbering in both the flavone and isoflavone is shown for comparison.

homologies to any proteins in the databases available at that time and lacked the highly conserved nod box sequence in the 50 region that is required for the expression of many nodulation genes. One of these genes (ORF1) is now known to share a high degree of homology with an ORF (y4yP) that has been only recently identified in the TTSS of S. fredii USDA257 and Rhizobium sp. NGR234 (Krishnan et al., 2003). Subsequently, the other gene (ORF2) was shown to need both a flavonoid and a functional nodD1 gene for induction and has been named nolJ (Boundy-Mills et al., 1994). It has not been identified as a rhizobial TTSS component. Another clue to the presence of such a system in rhizobia came from work by Krishnan and Pueppke (1993), which demonstrated that S. fredii USDA257 exported new proteins, designated signal responsive (SR), subsequent to nod gene induction by flavonoids. Protein production and secretion were dependent on the presence of isoflavonoid or flavonoid nod gene inducers (e.g., genistein, luteolin, naringenin) and export occurred without N-terminal processing— a characteristic of protein secretion by TTSSs. A plasmid-borne locus in USDA257 involved in the regulation of soybean cultivar specificity,


23

nolXWBTUV, contains flavonoid-inducible genes whose protein products share sequence homologies with components of TTSSs (Meinhardt et al., 1993). The finding that disruption of the locus was accompanied by absence of SR protein secretion strengthened the possibility that some products of the nolXWBTUV locus are responsible for synthesis of TTSS components (Krishnan et al., 1995). This study also proved that, as with nod gene induction in rhizobia, flavonoid-induced protein secretion requires the presence of regulatory NodD proteins. In addition, other intermediary regulators such as y4xI (a homologue of the Xanthomonas campestris hrp regulator HrpG), which is under NodD1 control and has recently been renamed TtsI (Marie et al., 2003), appear to be required for transcriptional activation of TTSS genes (Viprey et al., 1998). The presence of TTSSs in several rhizobia has now been established by nucleotide sequencing, including Rhizobium sp. NGR234 (Freiberg et al., 1997); M. loti MAFF303099 (Kaneko et al., 2000); B. japonicum USDA110 (Go¨ ttfert et al., 2001); R. etli CFN42 (National Center for Biotechnology Information [NCBI] database, accession number U80928); and S. fredii USDA257 (Krishnan et al., 2003). In the case of B. japonicum, Krause et al. (2002) have proposed a model for a regulatory cascade (initiated by genistein reception by NodD1 and NodV and involving the transcriptional activator protein NodW) controlling expression of genes in the TTSS cluster. Protein secretion by TTSS has been confirmed in Rhizobium sp. NGR234 (Marie et al., 2003; Viprey et al., 1998) and S. fredii (Krishnan et al., 1995). A model for the process involving two cytoplasmic proteins, five inner membrane proteins, two outer membrane proteins, and one lipoproteinassociated outer membrane protein was first proposed by Viprey et al. (1998). This has since been modified by Marie et al. (2003) to allow allocation of putative transfer or eVector functions to secreted proteins, with the latter type determining the plant response to the rhizobial TTSS. The genetic organization of TTSSs in NGR234, which cannot form N-fixing nodules on soybean rhizobia, and in USDA257, which can, is 98% identical (Krishnan et al., 2003). In contrast, the TTSSs of two soybean-nodulating rhizobia— S. fredii USDA257 and B. japonicum USDA110—share only limited sequence homology and markedly diVerent genetic organization. Rhizobial genes encoding TTSSs appear to be clustered within regions of 35–47 kb (Marie et al., 2001), and the clusters contain genes that encode the secretion apparatus as well as the secreted proteins themselves. Fully sequenced TTSS clusters in rhizobia contain ORFs that are homologous to TTSS genes in the animal pathogen Yersinia (ysc genes) and plant bacterial pathogens (hrc genes) and are designated rhc (Rhizobium conserved) (Viprey et al., 1998). In animal and plant pathogens, as well as the rhizobial symbiont NGR234, nine

24

J. E. COOPER

such conserved genes in the TTSS clusters are thought to encode the core protein secretion apparatus (Bogdanove et al., 1996; Freiberg et al., 1997). Other genes that play a role in protein secretion are termed tts (Marie et al., 2003). Homologues of rhc genes have been found via genomic hybridizations in Bradyrhizobium elkanii USDA76, B. japonicum CB756, and by mutagenesis in S. fredii HH103 but not in S. meliloti 2011 (Viprey et al., 1998). Another strain of S. meliloti, 1021, also lacks homologues of genes encoding a TTSS (Galibert et al., 2001), as does R. leguminosarum bv. viciae 3841 (J.A. Downie, personal communication). Marie et al. (2001) proposed that proteins secreted by rhizobial TTSSs be termed Nops (Nodulation outer proteins, encoded by nop genes) to reflect the accepted nomenclature for Yersinia outer proteins (Yops). Five flavonoid-, NodD1-, TtsI-, and Rhcdependent secreted Nops have been identified in NGR234, and wild-type USDA257 appeared also to secrete the same five proteins plus one other of approximate size 36 kilodaltons (kDa), after induction with apigenin (Marie et al., 2003). One 7-kDa secreted protein from NGR234, NopA, shared 99% identity with a homologue from USDA257, and database searches established that NopA homologues were present in all other TTSS-containing rhizobia but not in those lacking a TTSS. NopA is the NGR234 homologue of Nop7 from USDA257 (H. B. Krishnan, personal communication), which was recently shown to be associated with pili structures (see next paragraph and Krishnan et al., 2003) and may fulfill a transfer function. Two other secreted proteins from NGR234, NopX and NopL, were found to be absent from the TTSSs of some other rhizobia and were considered to possess transfer and eVector functions, respectively (Marie et al., 2003). Homologues of NGR234 NopX and NopL exist in USDA257; the first of these, and probably also the second (designated Nop38), has also been found in association with USDA257 pili (next paragraph and Krishnan et al., 2003). Plant and animal pathogens with TTSSs are thought to deliver secreted proteins into the plasma membrane and cytosol of eukaryotic host cells (Brown et al., 2001; Casper-Lindley et al., 2002; Gala´ n and Collmer, 1999; Lee, 1997) via structures, termed needle complexes (Kubori et al., 1998), or other surface appendages including pili (Roine et al., 1997). Recently, pili, whose production depends on a functional nodD1 gene and the presence of a flavonoid or isoflavonoid nod gene inducer, have been discovered in S. fredii USDA257 (Krishnan et al., 2003). The S. fredii nod gene inducers genistein, daidzein, apigenin, and luteolin all stimulated pili production, as did soybean seed exudate, but no pili were produced in the absence of genistein or the presence of the noninducers biochanin A and umbelliferone. Mutations in some USDA257 rhc genes also negated pili formation in genistein-induced cells containing a functional nodD1 gene. Furthermore, biochemical analysis


25

showed that at least three Nops—Nop7 (now designated NopA), Nop38 (thought to be NopL, but confirmation by Western blotting with NopL antibodies is required; H.B. Krishnan, personal communication), and NopX—were associated with purified pili. It is not known whether eVector Nops of symbiotic bacteria are delivered directly into legume host cell cytoplasm, but NopL from Rhizobium sp. NGR234 has recently been shown to serve as a substrate for plant protein kinases, and its phosphorylation was inhibited by a mitogen-activated protein (MAP) kinase inhibitor, pointing to a role in the modulation of MAP kinase pathways (Bartsev et al., 2003). More work is needed to determine the mechanisms governing the delivery and reception of Nops, but in terms of function they do appear to make an important contribution to the formation of successful symbioses. TTSS mutants exhibit diVerent and inconsistent symbiotic phenotypes compared to wild-type strains, ranging from no eVect, to increases or decreases in nodule number and changes in nodule N fixation capacity, to altered host specificity. To cite a few examples: wild-type Rhizobium sp. NGR234 and S. fredii HH103 form ineVective (nonfixing) nodule-like structures on the roots of Crotalaria juncea and Erythrinia variegata, respectively, but TTSS defective mutants form eVective nodules (Marie et al., 2001). Further studies with C. juncea showed that Nops from wild-type NGR234 were impairing nodule development in this host (Marie et al., 2003). An even more drastic alteration of phenotype is displayed by USDA257 after TTSS disruption. In this case rhc mutants acquire the ability to form eVective root nodules on a soybean cultivar, McCall, which cannot be nodulated at all by the wild-type parent strain (Krishnan et al., 2003). Interestingly, the same rhc mutants form significantly fewer nodules than wild-type USDA257 on another soybean cultivar, Peking (Krishnan et al., 2003). Clearly, TTSSs in rhizobia can influence nodulation in positive or negative ways, depending on the type of host plant involved, and various explanations have been invoked to account for their eVects. These include diVerences in flavonoid inducer exudation among host plants, varying levels of recognition of Nops among prospective hosts, the absence or modification of a Nops receptor, and the elicitation or avoidance of a defense response (Marie et al., 2001). Another flavonoid-inducible, NodD-dependent rhizobial gene encoding a secreted protein, NodO, has been found, but only in very few rhizobia: R. leguminosarum bv. viciae (de Maagd et al., 1989a, b) and a broad host range strain, Rhizobium sp. BR816, isolated from Leucaena leucocephala (van Rhijn et al., 1996). This protein is released by a diVerent, type I, secretion system that, like type III, is also found in gram-negative bacteria, where it controls the release of compounds such as the a-haemolysin toxin of Escherichia coli into the extracellular space. Unlike TTSSs, the genes required for the type

26

J. E. COOPER

I secretion apparatus (the protein secretion genes prsDE) are not linked to the gene encoding the secreted protein, and their expression is not flavonoiddependent (Scheu et al., 1992). NodO is a Ca2þ-binding protein with partial homology to E. coli haemolysin (Economou et al., 1990; Sutton et al., 1994). In R. leguminosarum bv. viciae, nodO can partially complement the nodulation defect of a nodEFL deletion mutant (Downie and Surin, 1990), and in R. leguminosarum bv. trifolii transfer of nodO to a nodE-deficient mutant extends the host range to include Vicia hirsuta (Economou et al., 1994). Host ranges of other rhizobia are also extended on receipt of nodO even when they possess a functional nodE gene (van Rhijn et al., 1996; Vlassak et al., 1998). The nodE gene encodes a -ketoacyl synthase responsible for synthesis of an unsaturated fatty acid, located at the nonreducing terminus of Nod factors (Table III and Fig. 7), which influences host specificity (Demont et al., 1993; Pacios Bras et al., 2000; Spaink et al., 1991). In R. leguminosarum bv. viciae and trifolii both NodE and NodF are involved in fatty acid synthesis, for which a model combining elongation of the unsaturated acyl chain by NodE and the provision of acyl groups by NodF has been proposed (Geiger et al., 1998). Although NodO is not involved in synthesizing the fatty acid component of Nod factors (or in any other facet of Nod factor synthesis), it can suppress nodulation defects brought about by the absence of this or another Nod factor substituent, a carbamoyl group, in several rhizobial species (Vlassak et al., 1998). Original proposals for the mode of action of NodO invoked a capacity to form ion channels that permit cation movement across and concomitant depolarization of the plasma membrane of plant cells (Economou et al., 1994; Sutton et al., 1994). Such changes are among the first to be observed when roots are challenged with Nod factors (Ehrhardt et al., 1992, 1996; Felle et al., 1996; Gehring et al., 1997). More recently nodO has been identified as a gene that promotes infection thread development in root hairs (Walker and Downie, 2000). D. FURTHER FLAVONOID-DEPENDENT GENE EXPRESSION IN RHIZOBIA

In addition to the examples given in the preceding sections, it is apparent from both transcriptional (Ampe et al., 2003; Perret et al., 1994, 1999) and proteomics (Chen et al., 2000; Guerreiro et al., 1997, 1999) studies that the expression of other rhizobial genes, located on symbiotic plasmids or in the chromosome, is flavonoid dependent. Perret et al. (1994), using a combination of competitive RNA hybridization, subtractive DNA hybridization, and shotgun sequencing, found several flavonoid-inducible transcripts on the large symbiotic plasmid of Rhizobium sp. NGR234 (pNGR234a) that shared no homologies with known nodulation genes but strong homologies


27

TABLE III Nodulation Gene Products Required for Synthesis and Release of Nod Factors Protein

Function

Biosynthesis of glucosamine (chitin) oligosaccharide backbone NodM Glucosamine synthase NodCa N-acetyl-glucosamine transferase NodBa Deacetylase, acting at the nonreducing end of glucosamine oligosaccharide Biosynthesis and transfer of fatty acid moiety at nonreducing terminus NodF Acyl carrier protein NodE -Ketoacyl synthase NodAa Acyl transferase involved in N-acylation of deacetylated nonreducing terminus of glucosamine oligosaccharide Modification of nonreducing terminus NodS Methyl transferase NodU Carbamoyl transferase NolO Carbamoyl transferase NodL O-acetyl transferase, O-acetylates at 6-C position Modification of reducing terminus NodP,Q ATP sulphurylase and APS kinase, provide activated sulphur for sulphated Nod factors NodH Sulphotransferase NoeE Sulphotransferase involved in sulphation of fucose NolK GDP fucose synthesis NodZ Fucosyl transferase NolL O-acetyltransferase; involved in acetyl-fucose formation NodX O-acetyltransferase, specifically O-acetylates the 6-C of the terminal nonreducing sugar of the penta-N-acetylglucosamine of R. leguminosarum TOM from Afghanistan pea NoeI 2-O-methyltransferase involved in 2-O-methylation of fucose Secretion of Nod factors NodIa ABC transporter component carrying an ATPase domain NodJa ABC transporter sub-unit APS, adenosine-50 -phosphosulphate kinase; ATP, adenosine triphosphate; GDP, guanosine diphosphate. Sources: Downie (1998); Pacios Bras et al. (2000); Vance (2002). a

Present in all rhizobia.

to a number of other prokaryotic genes and proteins. One symbiotically active ORF was highly homologous to the leucine responsive regulatory protein (Lrp) of E. coli; it was present in Rhizobium sp. NGR234 but not in the closely related S. fredii USDA257. Further, detailed studies showed that daidzein enhanced the transcription of 147 previously silent ORFs on pNGR234a and that genes involved in Nod factor biosynthesis were more rapidly induced than some others whose products are required at a later stage of interaction with a host plant (Kobayashi et al., 2004; Perret et al., 1999).

Fig. 7. Composite Nod factor structure showing the range of possible substitutions on the oligochitin backbone. Nod proteins responsible for structural modifications are indicated where known. Ac, acetyl; Ara, arabinosyl; Cb, carbamoyl; Fuc, fucosyl; Gro, glycerol; Man, mannosyl; Me, methyl; S, sulphate. (From Bladergroen and Spaink, 1998; Pacios Bras et al., 2000.)


29

Proteomics analyses have identified new proteins whose expression levels are influenced by the presence of nod gene-inducing flavonoids in the bacterial growth medium. For example, two proteins that did not show sequence matches with any known nod gene products were induced in R. leguminosarum bv. trifolii by 7,40 dihydroxyflavone (Guerreiro et al., 1997). The expression of several proteins that appeared to be encoded by pSyma of S. meliloti was positively regulated by luteolin and none of them matched the products of any previously identified luteolin-regulated gene (Chen et al., 2000). Other proteins were down-regulated in the presence of luteolin, or expressed only in the absence of pSyma, or accumulated in maximum amounts when pSyma was either present or absent. At the level of protein expression it is clear therefore that luteolin exerts both positive and negative regulatory eVects on plasmid and chromosomal genes in S. meliloti. Two proteins with homologies to a molecular chaperone, GroEL, that is thought to assist partially folded proteins in attaining a correctly folded configuration (Ogawa and Long, 1995) were up-regulated by luteolin. It was suggested that this fulfilled the need for specific folding requirements of other luteolininduced proteins and that another up-regulated, 30S ribosomal protein was indicative of a luteolin influence on the cell’s translational machinery (Chen et al., 2000). Proteomics analyses are likely to underestimate the extent of flavonoid-dependent gene expression in rhizobia for several reasons, such as culture conditions that aVect transcriptional activity (Girard et al., 1996) and, consequently, the numbers and amounts of proteins expressed, and the limited resolving capability of two-dimensional gel electrophoresis which may limit their discrimination. Also, some nod gene products known to be induced by flavonoids (e.g., NodABC) are not always detectable on gels (Chen et al., 2000; Guerreiro et al., 1999). The three main types of surface polysaccharide in rhizobia (EPS, KPS, and LPS) all contribute to symbiotic development, and flavonoids have been shown to influence their structures either during or after biosynthesis. EPS produced by S. fredii USDA193 were of lower average molecular mass and had a reduced uronic acid content when genistein was present as a nod gene inducer in the growth medium. Genistein-mediated changes in EPS structure were also Sym plasmid-dependent (Dunn et al., 1992). In Rhizobium sp. S-2, isolated from the pigeon pea plant, EPS production was both naringeninand Sym plasmid-dependent and EPS determinants were found on the large symbiotic plasmid of this strain (Pandya and Desai, 1998). Fraysse et al. (2003) proposed that EPS, even when encoded by chromosomal genes, could be subject to modification by symbiotic plasmid genes whose expression was directly or indirectly induced by plant flavonoids. In the case of KPS, production of the secondary 2-O-MeMan-containing K-antigen of S. fredii

30

J. E. COOPER

USDA205 was up-regulated by the nod gene inducer apigenin (Forsberg and Reuhs, 1997; Reuhs et al., 1993, 1994,1995). A reciprocal influence of K-antigens from S. meliloti on (iso)flavonoid biosynthesis in alfalfa was noted by Becquart de Kozak et al. (1997). The O-antigen polysaccharide (OPS) component of LPS in S. fredii USDA205 is altered in both carbohydrate composition and mass range when apigenin is introduced into the growth medium (Reuhs et al., 1994). When cultured in the presence of nod gene-inducing flavonoids, Rhizobium sp. NGR234 produces large quantities of a new high molecular weight, rhamnose-rich LPS, but in the absence of inducers only fast-migrating lower molecular weight LPSs are synthesized (Fraysse et al., 2002). Furthermore, LPS biosynthesis genes are located in nod regions of the NGR234 large symbiotic plasmid (Freiberg et al., 1997), an arrangement that could permit structural adaptation of LPSs before bacterial contact with a prospective host (Fraysse et al., 2003). Interestingly, LPS structural changes in some rhizobia appear to be brought about directly by flavonoids without a requirement for nodD or any other gene on the symbiotic plasmid (Noel et al., 1996; Tao et al., 1992). Such is the case with R. etli, in which LPS alterations induced by low pH and anthocyanins (Duelli and Noel, 1997) appear to involve methylation of particular OPS residues (Noel and Duelli, 2000). Some of the flavonoid-dependent eVects on LPS that were described previously can be understood in the context of a highly integrated glycolipid chemistry in rhizobia (Cedergren et al., 1995; Price, 1999) that can provide a common origin for structural elements, such as sulfation, methylation, and fatty acyl composition, that are shared between chitolipooligosaccharides (Nod factors) and lipopolysaccharides. This applies also to membrane phospholipids, which can contain flavonoid-/NodD-dependent, nodFE-derived fatty acids found in Nod factors produced by the same organism (Cedergren et al., 1995; Geiger et al., 1994), as well as other fatty acids that are syntheszsed by the nodFE gene products but not incorporated into Nod factors (Geiger et al., 1998). One membrane phospholipid in S. meliloti, phosphatidylcholine, has been shown to be essential for nodulation of Medicago (Lo´ pez-Lara et al., 2003; Sohlenkamp et al., 2003). It can be synthesized using choline exuded from the host plant (de Rudder et al., 1999), but it is not known whether flavonoid-/NodD-dependent, nodFEderived fatty acyl residues are components of its structure (O. Geiger, personal communication). Turning from synthesis to catabolism, a recent study (Baumberger et al., 2003) has identified two genistein-inducible polysaccharide degradation genes in B. japonicum that would be of great significance for infection should they be needed for localized degradation of root hair cell walls. These genes


31

are gunA2 and pgl, encoding, respectively, an endoglucanase (cellulase) and a polygalacturonase. Finally in this section, flavonoid repression of a group of three rhi genes, rhiABC, on the symbiotic plasmid of R. leguminosarum bv. viciae, has been reported. The rhiABC operon requires AHLs for induction (Gray et al., 1996; Rodelas et al., 1999) and is regulated by a fourth rhi gene product, RhiR, whose expression is repressed by flavonoids (e.g., hesperetin) that are nod gene inducers for this organism. Thus transcription of rhiA is decreased in the presence of flavonoids and the eVect is NodD-dependent (Cubo et al., 1992; Economou et al., 1989). Mutations in rhi genes do not appear to aVect nodulation of Vicia hirsuta adversely unless the closely linked nodFEL genes are missing (Cubo et al., 1992).

VI. RECEPTION OF FLAVONOIDS BY RHIZOBIA A. INTERACTIONS WITH REGULATORY NodD PROTEINS

Activation of nodulation genes whose protein products are required for Nod factor synthesis is mediated by regulatory NodD proteins, the constitutively expressed products of the nodD genes. NodD belongs to the LysR family of transcriptional regulators (Schell, 1993), and it binds in the absence of inducers to conserved nucleic acid sequences, termed nod boxes (Fisher and Long, 1993; Rostas et al., 1986), in the promoter regions of inducible nodulation genes (Goethals et al., 1992). In the presence of appropriate plant flavonoids and NodD these genes are transcribed; their various protein products (see Table III) act collectively to synthesize the reciprocal Nod factor signal (see Fig. 7) to the plant root that is a determinant of host specificity. In B. japonicum and R. etli the nodD1 gene itself is preceded by a nod box sequence and transcription is enhanced in the presence of its own product (NodD1) and certain isoflavonoid glycosides (Smit et al., 1992). The regulation of nodulation genes, their functions, and the structures and properties of the Nod factors they ultimately encode have been extensively and exhaustively reviewed (Broughton et al., 2000; D’Haeze and Holsters, 2002; Downie, 1998; Pacios Bras et al., 2000; Perret et al., 2000; Schlaman et al., 1998; Schultze and Kondorosi, 1998; Spaink, 2000). Consequently, this section is principally concerned with the reception of flavonoids by rhizobia and the uncertainties surrounding the nature of their interactions with NodD proteins. All rhizobia possess nodD genes, but the number of homologues varies from one in R. leguminosarum bv. trifolii and R. leguminosarum bv. viciae to

32

J. E. COOPER

between two and five in B. japonicum, Rhizobium sp. NGR234, S. meliloti, R. etli, and R. tropici. Mutation of the single nodD gene in R. leguminosarum bv. trifolii results in loss of nodulating ability on clovers, and inactivation of nodD1 in NGR234 similarly produces a Nod phenotype on its many host genera (Broughton et al., 2000). In some other rhizobia such as S. meliloti and B. japonicum, nod gene induction appears to involve more complicated regulatory mechanisms. The three NodD proteins in S. meliloti do not all act as flavonoid receptors; NodD1 interacts with flavonoids and NodD2 with betaines and 4,40 -dihydroxy-20 -methoxychalcone (Hartwig et al., 1990b; Phillips et al., 1992), but, exceptionally, NodD3 regulates expression of nod genes in the absence of any compound in plant exudates. Mutations in all three nodD genes are required to eliminate host nodulation by S. meliloti. In this organism another symbiotic regulatory gene, syrM, (which is also flavonoid independent) acts in conjunction with nodD3 to provide selfamplifying positive regulation of nod genes in developing root nodules (Swanson et al., 1993). Recent work with mutants of S. meliloti that lack a-isopropyl-malate synthase, the first enzyme in the leucine biosynthetic pathway, has shown that a leucine-related metabolic intermediate may also be required as an inducer, in addition to luteolin, for activation of nodulation genes by NodD1 (Sanjua´ n-Pinilla et al., 2002). A further example of diVerential responses to flavonoids can be found in B. japonicum: Glycine max releases genistein and daidzein that induce the B. japonicum nodYABCSUIJ operon and certain isoflavone glycosides that do not, except in the presence of a suboptimal genistein concentration (Smit et al., 1992). B. japonicum possesses two genes, nodVW, which are distinct from and supplementary to nodD and are involved in the regulation of Nod factor synthesis via isoflavonoid inducers. This two-component system relies on NodV, a sensory kinase, to recognize flavonoids that do not normally interact with NodD, whereas NodW activates gene transcription (Go¨ ttfert et al., 1990, 1992; Loh et al., 1997; Sanjua´ n et al., 1994). The system is responsible for extending the host range of this organism to legumes such as Macroptilium atropurpureum, Vigna radiata and Vigna unguiculata. The previously mentioned isoflavonoid-inducible cellulase (gunA2) and polygalacturonase (pgl) genes in B. japonicum (Baumberger et al., 2003) require both NodD and NodW for their expression, even though neither of these two putative targets contains a promoter with a nod box consensus. It was suggested that in this case a hierarchical regulatory cascade may operate, in which NodD or NodW (or both) control an unidentified (so far) regulatory gene whose product in turn controls gunA and pgl expression. Yet another regulatory system is present in S. fredii and involves the nolJ, nolBTUV, and nolX transcriptional units (Bellato et al., 1996; Meinhardt et al., 1993). These


33

genes also lack nod box sequences in their promoters, but they too are isoflavonoid-inducible and NodD dependent. In contrast to these examples of flavonoid selectivity by diVerent regulatory Nod proteins in a single bacterium, all three separate NodDs of R. etli appeared to react in the same way to a range of flavonoid inducers from common bean (Hungria et al., 1992). Regulation of nod gene expression is also subject to negative control by repressor proteins such as NolR and NolA whose production is flavonoid independent (Schlaman et al., 1998). An excess of Nod factors in the rhizosphere is apparently detrimental to eYcient nodulation and can aVect the spectrum of hosts that are nodulated (Fellay et al., 1998; Gillette and Elkan, 1996; Knight et al., 1986; Sadowsky et al., 1991). It may also trigger unwanted host defense reactions (Savoure´ et al., 1997). In R. leguminosarum bv. viciae the single nodD gene is negatively autoregulated by its own product, NodD (Rossen et al., 1985). Relationships between flavonoid structure and nodulation gene-inducing or inhibiting properties are apparent. A hydroxyl group at the 7-carbon position in the flavonoid skeleton, regardless of other OH substitutions, is a feature of inducing compounds for Rhizobium sp. NGR234. Hydroxylation at the 7- and 40 -carbon positions is a common feature of inducers from many legumes (see Table II), and testing of more than 1000 flavonoids led to the conclusion that hydroxylation at the position equivalent to the 7-carbon was a feature of compounds that had inducing or anti-inducing activity in many diVerent rhizobia (Cunningham et al., 1991). Rhizobia with narrower host ranges appear to require a more specific pattern of substitutions in the basic flavonoid structure to ensure interaction with NodD. Thus the nod genes of R. leguminosarum bv. viciae are induced by flavonoids with hydroxyl groups at the 5-, 7-, and 40 -carbon positions. Is narrow host range in rhizobia attributable to a high degree of specificity between flavonoids and NodD? Certainly an organism such as Rhizobium sp. NGR234 can interact with many flavonoid inducers and nodulates legumes in many genera. However, other examples do not provide such a correlation; R. leguminosarum bv. viciae responds to a wide variety of flavonoids but has a narrow host range. In general it can be concluded that there is no universal flavonoid inducer for rhizobial nodulation genes; rhizobia associated with a particular host tend to be more responsive to inducers released from that plant and variation among NodDs of diVerent rhizobia with regard to flavonoid reception is not of itself suYcient to define host specificity. How do NodD proteins interact with coinducing flavonoids to activate nodulation gene transcription? Despite the presumption, often reiterated in the literature over many years (a notable exception being the review of

34

J. E. COOPER

rhizobial nod gene regulation by Schlaman et al., 1998), that flavonoids form a complex with NodD and in so doing eVect conformational changes at the nod box binding sites that activate transcription, there is no direct evidence for physical interaction between the two molecules. Specific binding of NodD itself to DNA, in the presence or absence of flavonoids, has been confirmed by gel electrophoresis retardation assays (Fisher and Long, 1989, 1993; Goethals et al., 1992; Hong et al., 1987; Kondorosi et al., 1989; Schlaman et al., 1992). NodD binding to nodulation gene promoters protects a 75- to 20-base pair (bp) region from the transcriptional initiation sites, including nearly the entire nod box sequence (Fisher and Long, 1989; Kondorosi et al., 1989; Machado et al., 1998). Working with R. leguminosarum bv. viciae, Okker et al. (2001) confirmed that NodD binds to DNA in the absence of inducer compounds, and by means of single base pair substitutions along the nodF nod box, they were able to show that mutations in the nod box LysR motif abolished flavonoid-dependent promoter activity even though binding of NodD was unaVected. This finding implies that flavonoids are required as coinducers when NodD interacts with the entire nod box or that the LysR motif is not the only sequence involved in NodD binding (Schlaman et al., 1998). NodD interacts with two binding sites in the nod box (Fisher and Long, 1993), and a recent study (Feng et al., 2003) provided evidence that it binds to target DNA through anchoring the two half-sites of the nod box as a tetramer. An imperfect inverted repeat, (AT-N10-GAT) in each half-site is critical for NodD binding, and mutation of the inverted repeat at the distal half-site allowed NodD to activate nodA transcription in vivo in the absence of a flavonoid coinducer. Yeh et al. (2002) showed that the chaperonin GroESL is required for recombinant S. meliloti NodD1 to respond to its inducer, luteolin, in vitro. This interaction results in increased binding of NodD1 to target DNA at luteolin concentrations up to 100 mM and decreased binding at even higher concentrations. A decrease in NodD binding to nod box DNA at naringenin concentrations higher than 10 mM was noted in R. leguminosarum by Hu et al. (2000), who argued that inducer concentrations of this order could accumulate in the cytoplasmic membranes of rhizobia in vivo. Speculation on the mode of transcriptional repression of nod genes by NodD in the absence of flavonoids takes account of the fact that this protein can induce a bend in DNA at the site of nod gene promoters (Feng et al., 2003; Fisher and Long, 1993; Schlaman et al., 1998). The flavonoid coinducer may interact with NodD in an as-yet undefined manner and location to produce a change in bend that enables RNA polymerase to initiate transcription of the nod gene promoters (Hu et al., 2000). As noted by Yeh et al. (2002), even if a general mechanism for flavonoid-NodD interaction can be established, the relationships among


35

coinducer structural specificity, NodD activity, and legume host range will still require explanation. B. METABOLISM

Many microorganisms can degrade phenolic compounds of plant origin, including condensed tannins and flavonoids (Arunachalam et al., 2003; Bhat et al., 1998; Ibrahim and Abul-Hajj, 1990). Degradation may occur under aerobic or anaerobic conditions, and metabolites (e.g., acetate, butyrate, -ketoadipate) from various pathways are channelled into the citric acid cycle (Bhat et al., 1998). Barz (1970) and Barz et al. (1970) demonstrated that bacteria in legume rhizospheres could degrade flavonoids such as formononetin and daidzein, and rhizobia themselves are known to be capable of catabolizing many aromatic substances (Latha and Mahadevan, 1997). An indication that bacteria could participate in the aerobic degradation of flavonoids came from a study of the utilization of the flavan-3-ol catechin by B. japonicum, in which protocatechuic acid was detected among the metabolites (Muthukumar et al., 1982). More detailed analyses of degradation products resulted in the detection of phloroglucinol carboxylic acid, phloroglucinol, resorcinol, hydroxyquinol, maleyl acetate, and -carboxy-cis,cismuconate, in addition to protocatechuic acid (Hopper and Mahadevan, 1997). Another rhizobial isolate from root nodules of Leucaena leucocephala was shown to use catechin as a sole carbon source and to yield protocatechuic acid, phloroglucinol carboxylic acid, phloroglucinol, resorcinol, catechol, and hydroxyquinol as by-products (Gajendiran and Mahadevan, 1988). These and many other simple phenolics can be used by rhizobia as sole carbon and energy sources (Gajendiran and Mahadevan, 1990; Parke and Ornston, 1984; Vela et al., 2002). Indirect evidence for degradation of the flavone chrysin and the flavanone naringenin by B. japonicum was obtained from HPLC analyses of supernatants from spent flavonoid-supplemented cultures in which no trace of either compound could be detected (Kosslak et al., 1990). Rhizobial cleavage of the flavone nucleus of a flavonoid was first demonstrated by Rao et al. (1991); incubation of the pentahydroxyflavonol quercetin, supplied at a concentration of 100 mM in an arabinose-based, defined growth medium, with M. loti or a Lotus-nodulating Bradyrhizobium strain, yielded protocatechuic acid and phloroglucinol among the metabolites after 3 days (Fig. 8). A pattern of multiple C-ring fissions with the potential for transient chalcone formation was proposed to account for the production of these conserved A- and B-ring products. Among bacteria, reports of flavone type ring cleavage have been largely confined to anaerobes such as Clostridium and Eubacterium in the mammalian intestinal tract (Hur and

36

J. E. COOPER

Fig. 8. Proposed mechanism for fission of the flavone nucleus (C-ring) in the pentahydroxyflavonol quercetin and principal degradation products, phloroglucinol, and protocatechuic acid, detected in extracted culture supernatants by GC-MS. (From Rao et al., 1991.)

Rafii, 2000; Hur et al., 2000, 2002; Krumholz and Bryant, 1986; Schoefer et al., 2002; Winter et al., 1989). However, Pillai and Swarup (2002) found that a plant-growth-promoting rhizobacterial strain of Pseudomonas putida could utilize a variety of flavonoids as sole carbon sources and was capable of aerobic catabolism of quercetin by a similar, if not identical, mechanism to that proposed by Rao et al. (1991) for Lotus-nodulating rhizobia. In the case of P. putida the first step is dehydroxylation to naringenin, followed by hydrolysis and C-ring cleavage to yield phloroglucinol and protocatechuate among the products, a process that can be viewed as a reversal of quercetin biosynthesis in plants, which involves the hydroxylation of naringenin by flavanone-3-hydroxylase. Before this finding, flavonoid degradation by Pseudomonas spp. was thought to occur primarily via A-ring cleavage mechanisms (JeVrey et al., 1972a, b; Rao and Cooper, 1994; Rao et al., 1991; Schultz et al., 1974). Quercetin is not a nod gene inducer for M. loti (Lo´ pez-Lara et al., 1995), but other rhizobia can degrade their own presumptive nod gene inducers by mechanisms that originate with C-ring fission (Rao and Cooper, 1994, 1995). Other general features of the process are apparent, such as the formation of C-ring modification compounds in addition to conserved A- and B-ring monocyclic hydroaromatics. Also, flavonoids with OH substitutions at the 5- and 7-carbon positions yield phloroglucinol as the main conserved A-ring product, whereas those with a single OH substitution at the 7-carbon position yield resorcinol. Conserved B-ring products are more varied and include p-coumaric acid (4-hydroxycinnamic acid), p-hydroxybenzoic acid (4-hydroxybenzoic acid), protocatechuic, and phenylacetic and caVeic acids (Rao and Cooper, 1994). When incubated with its principal nod gene inducer, luteolin, S. meliloti yielded new closed C-ring metabolites (tetrahydroxyflavanone and apigenin) in addition to a suite of monocyclic phenolic degradation products (Cooper and Rao, 1995).


37

The same general pattern of degradation applies to the isoflavonoid nod gene inducers daidzein and genistein during incubation with B. japonicum USDA110spc4, and a detailed analysis of products revealed that significant amounts of isoflavonoid were transformed (Rao and Cooper, 1995). Proposed degradation pathways for both compounds, based on products detected in culture supernatants by GC-MS and HPLC-UV, are shown in Fig. 9. As previously noted for the catabolism of quercetin by P. putida the rhizobial degradation of these isoflavonoids can be regarded as a reversal of their plant biosynthetic pathways. A notable feature of this scheme is the requirement for aryl (B-ring) migration from the 3-carbon to the 2-carbon position on the C-ring to account for the presence of chalcones among the metabolites. The pathways allow for the possibility of multiple sources for some products; for example, umbelliferone could be derived from any or all of the following compounds in the daidzein scheme: resorcinol, coumestrol, or p-coumaric acid. The formation of several new metabolites from single isoflavonoid precursors could have implications for the root infection process. For example, as already noted, isoliquiritigenin is a strong nod gene inducer for B. japonicum (Kape et al., 1992), whereas umbelliferone and coumestrol can act as moderately eVective inducers or inhibitors, depending on the strain of B. japonicum (Kosslak et al., 1990). Two degradation products from daidzein and genistein, namely umbelliferone and phenylacetic acid, reduced the nod gene-inducing activity of genistein in B. japonicum USDA110spc4 by significant amounts (Rao and Cooper, 1995). Current models of NodD-flavonoid interaction take no account of the degradation of nod gene inducers by the receiving organism and the formation of new compounds that may influence gene induction. Findings from studies of flavonoid degradation by rhizobia point to a more complicated version of flavonoid-induced nod gene expression than the one proposed by Hubac et al. (1993, 1994), which involved enhanced retention of a single, inert inducer compound (luteolin) in the outer membrane of S. meliloti. One study has revealed a positive correlation between inducer catabolism and nodulation: in S. meliloti, genes involved in demethylation of the betaine nod gene inducer stachydrine are required for eYcient host nodulation and are grouped on the symbiotic plasmid in a region that contains the nodulation genes (Goldmann et al., 1994). The reports of flavonoid and isoflavonoid nod gene-inducer degradation, as previously described, were based on experiments in which authentic compounds were added to defined growth media at a concentration of 10 mM. If one discounts studies of rhizobial influence on flavonoid release from roots, in which the occurrence of some compounds may be attributable to flavonoid modification or degradation in the rhizosphere (see the section on the synthesis

Fig. 9. Proposed pathways for degradation of the isoflavones genistein and daidzein by Bradyrhizobium japonicum, based on products detected by GC-MS and HPLC-UV analyses of extracted culture supernatants. Isomerization of chalcones and flavanones is indicated by reversible arrows. (From Rao and Cooper, 1995.)


39

and release of flavonoids, earlier in this chapter), he or she should consider that very few experiments providing information on rhizobial biotransformations of flavonoids in legume epidermal exudates have been undertaken. Steele et al. (1999) detected two unidentified new flavonoids as well as elevated concentrations of protocatechuic acid and phloroglucinol in root exudate from Lotus pedunculatus following incubation with M. loti. Incubation of the same exudate with R. leguminosarum bv. trifolii also yielded two new phenolic compounds that were diVerent than those formed by M. loti. No nod gene induction assays were performed with the new compounds. The monocyclic phenolics 4-hydroxybenzoate and protocatechuate, which are often found among the products of rhizobial flavonoid metabolism (Rao and Cooper, 1994, 1995; Rao et al., 1991), are known to serve as growth substrates for rhizobia (Chen et al., 1984; Hussein et al., 1974; Lorite et al., 1998; Muthukumar et al., 1982; Parke and Ornston, 1984; Wong et al., 1991). Both compounds, together with another flavonoid degradation product, p-coumaric acid (4-coumaric acid), can be metabolized via the protocatechuate branch of the -ketoadipate pathway (Fig. 10) to enter the citric acid cycle via succinyl CoA and acetyl CoA (Parke, 1997; Parke et al., 1991). Some rhizobia can also metabolize 4-hydroxybenzoate via catechol or salicylic acid and gentisic acid prior to ring cleavage (Muthukumar et al., 1982). Microbial degradation of protochatechuate to -ketoadipate may also proceed via the catechol branch of the pathway, whereas phloroglucinol carboxylic acid and phloroglucinol may reach the same endpoint via intermediates such as resorcinol, hydroxyhydroquinone, and maleyl acetate (Bhat et al., 1998). Distribution of the -ketoadipate pathway is widespread among the rhizobia (Parke, 1997), and it appears that enzymes of the pathway are inducible in Rhizobium species but constitutive in Bradyrhizobium (Parke and Ornston, 1986). Catabolism of 4-hydroxybenzoate and protocatechuate via the protocatechuate branch of the pathway in Agrobacterium tumefaciens is mediated by a regulatory gene, pcaQ, which encodes an activator that responds to -carboxy-cis,cis-muconate and controls the expression of five genes in the pathway: pcaDCHGB (Parke, 1993, 1995). Like NodD, the activator protein PcaQ is a member of the LysR family of transcriptional regulators (Parke, 1996a), but in a phylogenetic tree of LysR-type proteins it falls outside the group that contains NodD (Schlaman et al., 1992). Homologues of pcaQ have been found in rhizobia (Parke, 1996b), and the pattern of induction of pca genes in A. tumefaciens is similar to that in R. leguminosarum bv. trifolii (Parke, 1995). The genetic organization and regulation of the -ketoadipate pathways in various bacteria, including rhizobia, have been reviewed by Harwood and Parales (1996). No definitive evidence has yet been obtained to show that flavonoid degradation

40

J. E. COOPER

Fig. 10. The protocatechuate branch of the -ketoadipate (3-oxoadipate) pathway in rhizobia with gene designations shown for each step.

by rhizobia is accompanied by channelling of some monocyclic phenolic metabolites into the protocatechuate branch of the -ketoadipate pathway. However, the detection of -carboxy-cis,cis-muconate during incubation of catechin with B. japonicum (Hopper and Mahadevan, 1997) and elevated


41

levels of pcaH expression in S. meliloti growing in luteolin-supplemented culture medium (T. P. McCorry and J. E. Cooper, unpublished results) suggest that this is the case. Finally, an intriguing pointer to the fate of certain carbon atoms in a nod gene-inducing flavonoid during assimilation and metabolism by R. leguminosarum bv. viciae was obtained from a radio-carbon tracing study using A-ring labelled 14C-naringenin (Rao et al., 1996). Two hours after presenting this compound to the bacterium at a nod gene-inducing concentration of 2 nM (12.5 Kilobecquerels [kBq]), radio activity (2189 disintegrations per minute [dpm]) was detected in a supernatant fraction corresponding to one of the main biologically active Nod factors of this organism, NodRlv-IV (Ac, C18:4) (Spaink et al., 1991). Further analysis revealed that 95% of this signal was located in the fatty acid side chain at the nonreducing terminus that is a determinant of host specificity for this Nod factor. Incorporation of carbon atoms that were originally in naringenin, into the fatty acyl moiety occurred despite the fact that the flavonoid was supplied at a concentration that was 250 times lower than cold acetate in the incubation medium. This is the sole available example of the contribution of structural elements (carbon atoms) from a nod gene-inducing flavonoid to a Nod factor and it is not known whether such an intimate linkage between primary and reciprocal signal molecules is a general feature of rhizobial responses to inducing compounds. The contribution of various signalling events to the specificity of legumerhizobia symbioses has been referred to throughout this review, but, with the qualification that flavonoid reception determines the production or modifies the nature of many other rhizobial signal molecules, there appears to be no single step that ensures a successful outcome in the form of a nitrogen-fixing root nodule. Certainly, flavonoid interaction with NodD and/or Nod factor eVects on root hair curling and cortical cell division can be highly specific in some partnerships but, as argued by Fellay et al. (1995), symbiotic control can be spread over a number of steps, none of which need be specific. The infection process can therefore be viewed as a series of hurdles carrying varying probabilities of successful negotiation, which has the capacity to account for all known combinations of macrosymbiont and microsymbiont.

VII. CONCLUSION Since the identification of luteolin and 7, 40 -dihydroxyflavone as activators of rhizobial nod gene expression in 1986, it has been accepted that flavonoids are key participants in the molecular dialogue between plant and bacterium that eventually leads to a functioning root nodule. Appreciation of their

42

J. E. COOPER

contribution to this process was enhanced by the discovery that the reciprocal Nod factor signals, whose synthesis they induce, are the agents of numerous changes in the host root at the physical, biochemical, and molecular genetic levels, which are all associated with infection and nodule development. However, as this chapter has attempted to demonstrate, Nod factors represent only one of many types of signals transmitted from rhizobia to the host plant and a close analysis of the literature reveals that flavonoids initiate, regulate, or mediate the production of most of them. As examples, flavonoid inducers are required for the synthesis of type III proteins, and their secretion machinery, the type I secreted protein NodO, polysaccharide-degrading enzymes, and numerous other proteins whose functions have yet to be determined. The three main classes of rhizobial surface polysaccharides—EPS, KPS, and LPS—are all subject to flavonoid-mediated qualitative or quantititative changes either during or after biosynthesis. Although gene induction or up-regulation are the most common functions attributable to flavonoids, they can also act as repressors, as in the case of rhi genes in R. leguminosarum bv. viciae. Evidently, as initiators of the rhizobial response to the presence of a legume that involves multiple reciprocal signals whose synthesis is directly or indirectly dependent upon them, flavonoids cannot simply be regarded as one type of signal among several with approximately equal significance. On the contrary, they appear to be a sine qua non for infection of virtually all legumes studied to date. New flavonoid-inducible genes and their products continue to be discovered and their roles during host infection elucidated. Future transcriptional, proteomics, and mutational analyses, together with the mining of data from rhizobial genome sequencing projects, will undoubtedly uncover more examples to further expand the scope of flavonoid influence on symbiotic development.

ACKNOWLEDGMENTS I wish to thank Allan Downie, Otto Geiger, Hari Krishnan, and Wolfgang Streit for providing information on their recent research, J. R. Rao for many helpful discussions, and Oonagh McMeel for the preparation of figures.

REFERENCES Aguilar, J. M. M., Ashby, A. M., Richards, A. J. M., Loake, G. J., Watson, M. D. and Shaw, C. H. (1988). Chemotaxis of Rhizobium leguminosarum biovar phaseoli towards flavonoid inducers of the symbiotic nodulation genes. Journal of General Microbiology 134, 2741–2746.


43

Ames, P., Schluederberg, S. A. and Bergman, K. (1980). Behavioural mutants of Rhizobium meliloti. Journal of Bacteriology 141, 722–727. Ampe, F., Kiss, E., Sabourdy, F. and Batut, J. (2003). Transcriptome analysis of Sinorhizobium meliloti during symbiosis. Genome Biology 4, R15. Amrhein, N. and Godeke, K. H. (1977). a-Aminooxy- -phenylpropionic acid, a potent inhibitor of L-phenylalanine ammonia lyase in vitro and in vivo. Plant Science Letters 8, 313–317. Aoki, T., Akashi, T. and Ayabe, S. (2000). Flavonoids of leguminous plants: Structure, biological activity, and biosynthesis. Journal of Plant Research 113, 475–488. Armitage, J. P., Gallagher, A. and Johnston, A. W. B. (1988). Comparison of the chemotactic behaviour of Rhizobium leguminosarum with and without the nodulation plasmid. Molecular Microbiology 2, 743–748. Arunachalam, M., Mohan, N., Sugadev, R., Chellappan, P. and Mahadevan, A. (2003). Degradation of (+)-catechin by Acinetobacter calcoaceticus MTC127. Biochimica et Biophysica Acta 1621, 261–265. Baggett, B. R., Cooper, J. D., Hogan, E. T., Carper, J., Paiva, N. L. and Smith, J. T. (2002). Profiling isoflavonoids found in legume root extracts using capillary electrophoresis. Electrophoresis 23, 1642–1651. Banfalvi, Z., Nieuwkoop, A., Schell, M., Besl, L. and Stacey, G. (1988). Regulation of nod gene expression in Bradyrhizobium japonicum. Molecular and General Genetics 214, 420–424. Barbour, W. M., Hattermann, D. R. and Stacey, G. (1991). Chemotaxis of Bradyrhizobium japonicum to soybean exudates. Applied and Environmental Microbiology 57, 2635–2639. Bartsev, A. V., Boukli, N. M., Deakin, W. J., Staehelin, C. and Broughton, W. J. (2003). Purification and phosphorylation of the eVector protein NopL from Rhizobium sp. NGR234. FEBS Letters 554, 271–274. Barz, W. (1970). Isolation of rhizosphere bacterium capable of degrading flavonoids. Phytochemistry 9, 1745–1749. Barz, W., Adamek, C. and Berlin, J. (1970). The degradation of formononetin and daidzein in Cicer arietinum and Phaseolus aureus. Phytochemistry 9, 1735–1744. Baumberger, I. C., Fraefel, N., Go¨ ttfert, M. and Hennecke, H. (2003). New NodWor NifA-regulated Bradyrhizobium japonicum genes. Molecular PlantMicrobe Interactions 16, 342–351. Becker, A. and Pu¨ hler, A. (1998). Production of exopolysaccharides. In ‘‘The Rhizobiaceae’’ (H. P. Spaink, A. Kondorosi and J. P. P. Hooykaas, eds.), pp. 97–118. Kluwer, Dordrecht, The Netherlands. Becker, A., Niehaus, K. and Pu¨ hler, A. (2000). The role of rhizobial extracellular polysaccharides (EPS) in the Sinorhizobium meliloti-alfalfa symbiosis. In ‘‘Prokaryotic Nitrogen Fixation’’ (E. W. Triplett, ed.), pp. 433–447. Horizon Scientific Press, Wymondham, United Kingdom. Becquart de Kozak, I., Reuhs, B. L., BuVard, D., Breda, C., Kim, J. S., Esnault, R. and Kondorosi, A. (1997). Role of the K-Antigen subgroup of capsular polysaccharides in the early recognition process between Rhizobium meliloti and alfalfa leaves. Molecular Plant-Microbe Interactions 10, 114–123. Begum, A. A., Leibovitch, S., Migner, P. and Zhang, F. (2001). Specific flavonoids induced nod gene expression and pre-activated nod genes of Rhizobium leguminosarum increased pea (Pisum sativum L.) and lentil (Lens culinaris L.) nodulation in controlled growth chamber environments. Journal of Experimental Botany 52, 1537–1543. Bellato, C. M., Balatti, P. A., Pueppke, S. G. and Krishnan, H. B. (1996). Proteins from cells of Rhizobium fredii bind to DNA sequences preceding nolX, a

44

J. E. COOPER

flavonoid-inducible nod gene that is not associated with a nod box. Molecular Plant-Microbe Interactions 9, 457–463. Ben Amor, B., Shaw, S. L., Oldroyd, G. E. D., Maillet, F., Penmetsa, R. V., Cook, D., Long, S. R., De´ narie´ , J. and Gough, C. (2003). The NFP locus of Medicago truncatula controls an early step of Nod factor signal transduction upstream of a rapid calcium flux and root hair deformation. Plant Journal 34, 495–506. Bhat, T. K., Singh, B. and Sharma, O. P. (1998). Microbial degradation of tannins: A current perspective. Biodegradation 9, 343–357. Bhattacharya, I., Gautam, H. and Das, H. R. (2002). Characterization of LPS mutants of peanut specific Bradyrhizobium (GN17). Glycoconjugate Journal 19, 395–402. Bladergroen, M. R. and Spaink, H. P. (1998). Genes and signal molecules involved in the rhizobia-Leguminosae symbiosis. Current Opinion in Plant Biology 1, 353–359. Bogdanove, A. J., Bauer, D. W. and Beer, S. V. (1996). Erwinia amylovora secretes harpin via a type III pathway and contains a homolog of yopN of Yersinia spp. Journal of Bacteriology 178, 1720–1730. Bohlool, B. B. and Schmidt, E. L. (1974). Lectins: A possible basis for specificity in the Rhizobium-legume root nodule symbiosis. Science 185, 269–271. Bolanos-Vasquez, M. C. and Werner, D. (1997). EVects of Rhizobium tropici, R. etli, and R. leguminosarum bv. phaseoli on nod gene-inducing flavonoids in root exudates of Phaseolus vulgaris. Molecular Plant-Microbe Interactions 10, 339–346. Boot, K. J. M., van Brussel, A. A. N., Tak, T., Spaink, H. P. and Kijne, J. W. (1999). Lipochitin oligosaccharides from Rhizobium leguminosarum bv. viciae reduce auxin transport capacity in Vicia sativa subsp. nigra roots. Molecular Plant-Microbe Interactions 12, 839–844. Boundy-Mills, K. L., Kosslak, R. M., Tully, R. E., Pueppke, S. G., Lohrke, S. and Sadowsky, M. J. (1994). Induction of the Rhizobium fredii nod box-independent nodulation gene nolJ requires a functional nodD1 gene. Molecular Plant-Microbe Interactions 7, 305–308. Bowra, B. J. and Dilworth, M. J. (1981). Motility and chemotaxis towards sugars in Rhizobium leguminosarum. Journal of General Microbiology 126, 231–235. Broughton, W. J. (2003). Roses by other names: Taxonomy of the Rhizobiaceae. Journal of Bacteriology 185, 2975–2979. Broughton, W. J., Jabbouri, S. and Perret, X. (2000). Keys to symbiotic harmony. Journal of Bacteriology 182, 5641–5652. Brown, I. R., Mansfield, J. W., Taira, S., Roine, E. and Romantschuk, M. (2001). Immunocytochemical localization of HrpA and HrpZ supports a role for the Hrp pilus in the transfer of eVector proteins from Pseudomonas syringae pv. tomato across the host plant cell wall. Molecular Plant-Microbe Interactions 14, 394–404. Caetano-Anolle´ s, G., Crist-Estes, D. K. and Bauer, W. D. (1988). Chemotaxis of Rhizobium meliloti to the plant flavone luteolin requires functional nodulation genes. Journal of Bacteriology 170, 3164–3169. Ca´ rdenas, L., Vidali, L., Dominguez, J., Perez, H., Sanchez, F., Hepler, P. K. and Quinto, C. (1998). Rearrangement of actin microfilaments in plant root hairs responding to Rhizobium etli nodulation signals. Plant Physiology 116, 871–877. Carlson, R. W., Reuhs, B. L., Forsberg, L. S. and Kannenberg, E. L. (1999). Rhizobial cell surface carbohydrates: Their structures, biosynthesis and


45

functions. In ‘‘Genetics of Bacterial Polysaccharides’’ (J. B. Goldberg, ed.), pp. 53–90. CRC Press, Boca Raton, Fla. Casper-Lindley, C., Dahlbeck, D., Clark, E. T. and Staskawicz, B. J. (2002). Direct biochemical evidence for type III secretion-dependent translocation of the AvrBs2 eVector protein into plant cells. Proceedings of the National Academy of Sciences of the United States of America 99, 8336–8341. Cedergren, R. A., Lee, J. G., Ross, K. L. and Hollingsworth, R. I. (1995). Common links in the structure and cellular-localization of Rhizobium chitolipooligosaccharides and general Rhizobium membrane phospholipid and glycolipid components. Biochemistry 34, 4467–4477. Chen, H. C., Higgins, J., Oresnik, I. J., Hynes, M. F., Natera, S., Djordjevic, M. A., Weinman, J. J. and Rolfe, B. G. (2000). Proteome analysis demonstrates complex replicon and luteolin interactions in pSyma-cured derivatives of Sinorhizobium meliloti strain 2011. Electrophoresis 21, 3833–3842. Chen, Y. P., Glenn, A. R. and Dilworth, M. J. (1984). Uptake and oxidation of aromatic substrates by Rhizobium leguminosarum MNF-3841 and Rhizobium trifolii TA1. FEMS Microbiology Letters 21, 201–205. Cho, M. J. and Harper, J. E. (1991). EVect of inoculation and nitrogen on isoflavonoid concentration in wild-type and nodulation-mutant soybean roots. Plant Physiology 95, 435–442. Cook, D., Dreyer, D., Bonnet, D., Howell, M., Nony, E. and Van den Bosch, K. (1995). Transient induction of a peroxidase gene in Medicago truncatula precedes infection by Rhizobium meliloti. Plant Cell 7, 43–55. Cooper, J. E. and Rao, J. R. (1992). Localized changes in flavonoid biosynthesis in roots of Lotus pedunculatus after infection by Rhizobium loti. Plant Physiology 100, 444–450. Cooper, J. E. and Rao, J. R. (1995). Flavonoid metabolism by rhizobia: Mechanisms and products. Symbiosis 19, 91–98. Cornelis, G. R. (2000). Type III secretion: A bacterial device for close combat with cells of their eukaryotic host. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 355, 681–693. Cornelis, G. R. and Van Gijsegem, F. (2000). Assembly and function of type III secretory systems. Annual Review of Microbiology 54, 735–774. Cornelis, G. R. and Wolf-Watz, H. (1997). The Yersinia Yop virulon: A bacterial system for subverting eukaryotic cells. Molecular Microbiology 23, 861–867. Coronado, C., Zuanazzi, J. A. S., Sallaud, C., Quirion, J. C., Esnault, R., Husson, H. P., Kondorosi, A. and Ratet, P. (1995). Alfalfa root flavonoid production is nitrogen regulated. Plant Physiology 108, 533–542. Crockard, M. A., Bjourson, A. J., Dazzo, F. B. and Cooper, J. E. (2002). A white clover nodulin gene, dd23b, encoding a cysteine cluster protein, is expressed in roots during the very early stages of interaction with Rhizobium leguminosarum biovar trifolii and after treatment with chitolipooligosaccharide Nod factors. Journal of Plant Research 115, 439–447. Cubo, M. T., Economou, A., Murphy, G., Johnston, A. W. B. and Downie, J. A. (1992). Molecular characterization and regulation of the rhizosphereexpressed genes rhiABCR that can influence nodulation by Rhizobium leguminosarum biovar viciae. Journal of Bacteriology 174, 4026–4035. Cullimore, J. V., Ranjeva, R. and Bono, J. J. (2001). Perception of lipo-chitooligosaccharidic Nod factors in legumes. Trends in Plant Science 6, 24–30. Cunningham, S., Kollmeyer, W. D. and Stacey, G. (1991). Chemical control of interstrain competition for soybean nodulation by Bradyrhizobium japonicum. Applied and Environmental Microbiology 57, 1886–1892.

46

J. E. COOPER

Currier, W. W. and Strobel, G. A. (1977). Chemotaxis of Rhizobium spp. to a glycoprotein produced by birdsfoot trefoil roots. Science 196, 434–436. Dakora, F. D. (2000). Commonality of root nodulation signals and nitrogen assimilation in tropical grain legumes belonging to the tribe Phaseoleae. Australian Journal of Plant Physiology 27, 885–892. Dakora, F. D., Joseph, C. M. and Phillips, D. A. (1993a). Alfalfa (Medicago sativa L.) root exudates contain isoflavonoids in the presence of Rhizobium meliloti. Plant Physiology 101, 819–824. Dakora, F. D., Joseph, C. M. and Phillips, D. A. (1993b). Common bean root exudates contain elevated levels of daidzein and coumestrol in response to Rhizobium inoculation. Molecular Plant-Microbe Interactions 6, 665–668. D’Arcy-Lameta, A. (1986). Study of soybean and lentil root exudates. II. Identification of some polyphenolic compounds, relation with plantlet physiology. Plant and Soil 92, 113–123. Dazzo, F. B. and Hubbell, H. D. (1975). Cross-reactive antigens and lectins as determinants of symbiotic specificity in the Rhizobium-clover association. Applied Microbiology 36, 1017–1033. De Maagd, R. A., Spaink, H. P., Pees, E., Mulders, I. H. M., Wijfjes, A., WijVelman, C. A., Okker, R. J. H. and Lugtenberg, B. J. J. (1989a). Localization and symbiotic function of a region on the Rhizobium leguminosarum Sym plasmid Prl1ji responsible for a secreted, flavonoid-inducible 50 kilodalton protein. Journal of Bacteriology 171, 1151–1157. De Maagd, R. A., Wijfjes, A. H. M., Spaink, H. P., Ruiz-Sainz, J. E., WijVelman, C. A., Okker, R. J. H. and Lugtenberg, B. J. J. (1989b). nodO, a new nod gene of the Rhizobium leguminosarum biovar viciae Sym plasmid Prl1ji, encodes a secreted protein. Journal of Bacteriology 171, 6764–6770. Demont, N., Debelle´ , F., Aurelle, H., De´ narie´ , J. and Prome´ , J.-C. (1993). Role of the Rhizobium meliloti nodF and nodE genes in the biosynthesis of lipooligosaccharidic nodulation factors. Journal of Biological Chemistry 268, 20134–20142. De´ narie´ , J., Debelle´ , F., Truchet, G. and Prome´ , J. C. (1993). Rhizobium and legume nodulation: A molecular dialogue. In ‘‘New Horizons in Nitrogen Fixation’’ (R. Palacios, J. Moira and W. E. Newton, eds.), pp. 19–30. Kluwer, Dordrecht, The Netherlands. de Rudder, K. E. E., Sohlenkamp, C. and Geiger, O. (1999). Plant-exuded choline is used for rhizobial membrane lipid biosynthesis by phosphatidylcholine synthase. Journal of Biological Chemistry 274, 20011–20016. D’Haeze, W. and Holsters, M. (2002). Nod factor structures, responses and perception during initiation of nodule development. Glycobiology 12, 79R–105R. Dharmatilake, A. J. and Bauer, W. D. (1992). Chemotaxis of Rhizobium meliloti towards nodulation gene-inducing compounds from alfalfa roots. Applied and Environmental Microbiology 58, 1153–1158. Dixon, R. A. (1999). Isoflavonoids: Biochemistry, molecular biology and biological functions. In ‘‘Comprehensive Natural Products Chemistry, Vol. I: Polyketides and Other Secondary Metabolites Including Fatty Acids and Their Derivatives’’ (U. Sankawa, ed.), pp. 773–823. Elsevier, Oxford, England. Dixon, R. A. and Steele, C. L. (1999). Flavonoids and isoflavonoids: A gold mine for metabolic engineering. Trends in Plant Science 4, 394–400. Djordjevic, M. A., Redmond, J. W., Batley, M. and Rolfe, B. G. (1987). Clovers secrete specific phenolic compounds which either stimulate or repress nod gene expression in Rhizobium trifolii. EMBO Journal 6, 1173–1179.


47

Downie, J. A. (1998). Functions of rhizobial nodulation genes. In ‘‘The Rhizobiaceae’’ (H. P. Spaink, A. Kondorosi and J. P. P. Hooykaas, eds.), pp. 387–402. Kluwer, Dordrecht, The Netherlands. Downie, J. A. and Surin, B. P. (1990). Either of two nod gene loci can complement the nodulation defect of a nod deletion mutant of Rhizobium leguminosarum bv. viciae. Molecular and General Genetics 222, 81–86. Ducrey, B., Wolfender, J.-L., Marston, A. and Hostettmann, K. (1995). Analysis of flavonol glycosides of thirteen Epilobium species (Onagraceae) by LC-UV and thermospray LC-MS. Phytochemistry 38, 129–137. Duelli, D. M. and Noel, K. D. (1997). Compounds exuded by Phaseolus vulgaris that induce a modification of Rhizobium etli lipopolysaccharide. Molecular Plant-Microbe Interactions 10, 903–910. Dunn, M. F., Pueppke, S. G. and Krishnan, H. B. (1992). The nod gene inducer genistein alters the composition and molecular mass distribution of extracellular polysaccharides produced by Rhizobium fredii USDA193. FEMS Microbiology Letters 97, 107–112. Economou, A., Hawkins, F. K. L., Downie, J. A. and Johnston, A. W. B. (1989). Transcription of rhiA, a gene on a Rhizobium leguminosarum bv viciae Sym plasmid, requires RhiR and is repressed by flavonoids that induce nod genes. Molecular Microbiology 3, 87–93. Economou, A., Hamilton, W. D. O., Johnston, A. W. B. and Downie, J. A. (1990). The Rhizobium nodulation gene nodO encodes a Ca2+-binding protein that is exported without N-terminal cleavage and is homologous to hemolysin and related proteins. EMBO Journal 9, 349–354. Economou, A., Davies, A. E., Johnston, A. W. B. and Downie, J. A. (1994). The Rhizobium leguminosarum biovar viciae nodO gene can enable a nodE mutant of Rhizobium leguminosarum biovar trifolii to nodulate vetch. Microbiology-UK 140, 2341–2347. Ehrhardt, D. W., Atkinson, E. M. and Long, S. R. (1992). Depolarization of alfalfa root hair membrane potential by Rhizobium meliloti Nod factors. Science 256, 998–1000. Ehrhardt, D., Wais, R. and Long, S. R. (1996). Calcium spiking in plant root hairs responding to Rhizobium nodulation signals. Cell 85, 673–681. Endre, G., Kereszt, A., Kevei, Z., Mihacea, S., Kalo, P. and Kiss, G. B. (2002). A receptor kinase gene regulating symbiotic nodule development. Nature 417, 962–966. Estabrook, E. M. and Sengupta-Gopalan, C. (1991). DiVerential expression of phenylalanine ammonia lyase and chalcone synthase during soybean nodule development. Plant Cell 3, 299–308. Etzler, M. E., Kalsi, G., Ewing, N. N., Roberts, N. J., Day, R. B. and Murphy, J. B. (1999). A Nod factor-binding lectin with apyrase activity from legume roots. Proceedings of the National Academy of Sciences of the United States of America 96, 5856–5861. Farrand, S. K., van Berkum, P. B. and Oger, P. (2003). Agrobacterium is a definable genus of the family Rhizobiaceae. International Journal of Systematic and Evolutionary Microbiology 53, 1681–1687. Fellay, R., Rochepeau, P., Relic´ , B. and Broughton, W. J. (1995). Signals to and emanating from Rhizobium largely control symbiotic specificity. In ‘‘Pathogenesis and Host Specificity in Plant Diseases. Histopathological, Biochemical, Genetic and Molecular Bases, Vol. I: Prokaryotes’’ (U. S. Singh, R. P. Singh and K. Kohmoto, eds.), pp. 199–220. Pergamon/Elsevier, Oxford, England.

48

J. E. COOPER

Fellay, R., Hanin, M., Montorzi, G., Frey, J., Freiberg, C., Golinowski, W., Staehelin, C., Broughton, W. J. and Jabbouri, S. (1998). nodD2 of Rhizobium sp. NGR234 is involved in the repression of the nodABC operon. Molecular Microbiology 27, 1039–1050. Felle, H. H., Kondorosi, E., Kondorosi, A. and Schultze, M. (1995). Nod signalinduced plasma membrane potential changes in alfalfa root hairs are diVerentially sensitive to structural modifications of the lipochitooligosaccharide. Plant Journal 7, 939–947. Felle, H. H., Kondorosi, E., Kondorosi, A. and Schultze, M. (1996). Rapid alkalinization in alfalfa root hairs in response to rhizobial lipochitooligosaccharide signals. Plant Journal 10, 295–301. Feng, J., Li, Q., Hu, H. L., Chen, X. C. and Hong, G. F. (2003). Inactivation of the nod box distal half-site allows tetrameric NodD to activate nodA transcription in an inducer-independent manner. Nucleic Acids Research 31, 3143–3156. Firmin, J. L., Wilson, K. E., Rossen, L. and Johnston, A. W. B. (1986). Flavonoid activation of nodulation genes in Rhizobium reversed by other compounds present in plants. Nature 324, 90–92. Fisher, R. F. and Long, S. R. (1989). DNA footprint analysis of the transcriptional activator proteins NodD1 and NodD3 on inducible nod gene promoters. Journal of Bacteriology 171, 5492–5502. Fisher, R. F. and Long, S. R. (1993). Interactions of NodD at the nod box. NodD binds to two distinct sites on the same face of the helix and induces a bend in the DNA. Journal of Molecular Biology 233, 336–348. Forkmann, G. and Heller, W. (1999). Biosynthesis of flavonoids. In ‘‘Comprehensive Natural Products Biochemistry, Vol. I: Polyketides and Other Secondary Metabolites Including Fatty Acids and Their Derivatives’’ (U. Sankawa, ed.), pp. 713–748. Elsevier, Oxford, England. Forsberg, L. S. and Reuhs, B. L. (1997). Structural characterization of the K antigens from Rhizobium fredii USDA257: Evidence for a common structural motif, with strain-specific variation, in the capsular polysaccharides of Rhizobium spp. Journal of Bacteriology 179, 5366–5371. Fraysse, N., Jabbouri, S., Treilhou, M., Couderc, F. and Poinsot, V. (2002). Symbiotic conditions induce structural modifications of Sinorhizobium sp. NGR234 surface polysaccharides. Glycobiology 12, 741–748. Fraysse, N., Couderc, F. and Poinsot, V. (2003). Surface polysaccharide involvement in establishing the Rhizobium-legume symbiosis. European Journal of Biochemistry 270, 1365–1380. Freiberg, C., Fellay, R., Bairoch, A., Broughton, W. J., Rosenthal, A. and Perret, X. (1997). Molecular basis of symbiosis between Rhizobium and legumes. Nature 387, 394–401. Gagnon, H. and Ibrahim, R. K. (1998). Aldonic acids: A novel family of nod gene inducers of Mesorhizobium loti, Rhizobium lupini, and Sinorhizobium meliloti. Molecular Plant-Microbe Interactions 11, 988–998. Gajendiran, N. and Mahadevan, A. (1988). Utilization of catechin by Rhizobium sp. Plant and Soil 108, 263–266. Gajendiran, N. and Mahadevan, A. (1990). Utilization of phenolic substances by Rhizobium sp. Indian Journal of Experimental Biology 28, 1136–1140. Gala´ n, J. E. and Collmer, A. (1999). Type III secretion machines: Bacterial devices for protein delivery into host cells. Science 284, 1322–1328. Galibert, F., Finan, T. M., Long, S. R., Pu¨ hler, A., Abola, P., Ampe, F., BarloyHubler, F., Barnett, M. J., Becker, A., Boistard, P., Bothe, G., Boutry, M., Bowser, L., Buhrmester, J., Cadieu, E., Capela, D., Chain, P., Cowie, A.,


49

Davis, R. W., Dre´ ano, S., Federspiel, N. A., Fisher, R. F., Gloux, S., Godrie, T., GoVeau, A., Golding, B., Gouzy, J., Gurjal, M., HernandezLucas, I., Hong, A., Huizar, L., Hyman, R. W., Jones, T., Kahn, D., Kahn, M. L., Kalman, S., Keating, D. H., Kiss, E., Komp, C., Lalaure, V., Masuy, D., Palm, C., Peck, M. C., Pohl, T. M., Portetelle, D., Purnelle, B., Ramsperger, U., Surzycki, R., The´ bault, P., Vandenbol, M., Vorho¨ lter, F. J., Weidner, S., Wells, D. H., Wong, K., Yeh, K. C. and Batut, J. (2001). The composite genome of the legume symbiont Sinorhizobium meliloti. Science 293, 668–672. Gaworzewska, E. T. and Carlile, M. J. (1982). Positive chemotaxis of Rhizobium leguminosarum and other bacteria towards root exudates from legumes and other plants. Journal of General Microbiology 128, 1179–1188. Gehring, C. A., Irving, H. R., Kabbara, A. A., Parish, R. W., Boukli, N. M. and Broughton, W. J. (1997). Rapid, plateau-like increases in intracellular free calcium are associated with Nod factor-induced root hair deformation. Molecular Plant-Microbe Interactions 10, 791–802. Geiger, O., Thomas-Oates, J. E., Glushka, J., Spaink, H. P. and Lugtenberg, B. J. J. (1994). Phospholipids of Rhizobium contain nodE-determined highly unsaturated fatty acid moieties. Journal of Biological Chemistry 269, 11090–11097. Geiger, O., Glushka, J., Lugtenberg, B. J. J., Spaink, H. P. and Thomas-Oates, J. E. (1998). NodFE-dependent fatty acids that lack an alpha-beta unsaturation are subject to diVerential transfer, leading to novel phospholipids. Molecular Plant-Microbe Interactions 11, 33–44. Gillette, W. K. and Elkan, G. H. (1996). Bradyrhizobium (Arachis) sp. strain NC92 contains two nodD genes involved in the repression of nodA and a nolA gene required for the eYcient nodulation of host plants. Journal of Bacteriology 178, 2757–2766. Girard, L., Valderrama, B., Palacios, R., Romero, D. and Davila, G. (1996). Transcriptional activity of the symbiotic plasmid of Rhizobium etli is aVected by diVerent environmental conditions. Microbiology-UK 142, 2847–2856. Goedhart, J., Bono, J.-J., Bisseling, T. and Gadella, T. W. J. (2003). Identical accumulation and immobilization of sulfated and nonsulfated Nod factors in host and nonhost root hair cell walls. Molecular Plant-Microbe Interactions 16, 884–892. Goethals, K., van Montagu, M. and Holsters, M. (1992). Conserved motifs in a divergent nod box of Azorhizobium caulinodans ORS571 reveal a common structure in promoters regulated by LysR-type proteins. Proceedings of the National Academy of Sciences of the United States of America 89, 1646–1650. Goldmann, L., Lecouer, L., Message, B., Dela Rue, M., Schoonejans, E. and Tepfer, D. (1994). Symbiotic plasmid genes essential to the catabolism of proline betaine, or stachydrine, are also required for eYcient nodulation by Rhizobium meliloti. FEMS Microbiology Letters 115, 305–311. Go¨ ttfert, M., Grob, P. and Hennecke, H. (1990). Proposed regulatory pathway encoded by the nodV and nodW genes, determinants of host specificity in Bradyrhizobium japonicum. Proceedings of the National Academy of Sciences of the United States of America 87, 2680–2684. Go¨ ttfert, M., Holzhauser, D., Bani, D. and Hennecke, H. (1992). Structural and functional analysis of two diVerent nodD genes in Bradyrhizobium japonicum USDA110. Molecular Plant-Microbe Interactions 5, 257–265.

50

J. E. COOPER

Go¨ ttfert, M., Rothlisberger, S., Kundig, C., Beck, C., Marty, R. and Hennecke, H. (2001). Potential symbiosis-specific genes uncovered by sequencing a 410 kilobase DNA region of the Bradyrhizobium japonicum chromosome. Journal of Bacteriology 183, 1405–1412. Go¨ tz, R., Limmer, N., Ober, K. and Schmitt, R. (1982). Motility and chemotaxis in two strains of Rhizobium with complex flagella. Journal of General Microbiology 128, 789–798. Graham, T. L. (1991a). A rapid, high-resolution high-performance liquid chromatography profiling procedure for plant and microbial aromatic secondary metabolites. Plant Physiology 95, 584–593. Graham, T. L. (1991b). Flavonoid and isoflavonoid distribution in developing soybean seedling tissues and in seed and root exudates. Plant Physiology 95, 594–603. Gray, K. M., Pearson, J. P., Downie, J. A., Boboye, B. E. A. and Greenberg, E. P. (1996). Cell-to-cell signaling in the symbiotic nitrogen-fixing bacterium Rhizobium leguminosarum: Autoinduction of a stationary phase and rhizosphere-expressed genes. Journal of Bacteriology 178, 372–376. Guerreiro, N., Redmond, J. W., Rolfe, B. G. and Djordjevic, M. A. (1997). New Rhizobium leguminosarum flavonoid-induced proteins revealed by proteome analysis of diVerentially displayed proteins. Molecular Plant-Microbe Interactions 10, 506–516. Guerreiro, N., Djordjevic, M. A. and Rolfe, B. G. (1999). Proteome analysis of the model microsymbiont Sinorhizobium meliloti: Isolation and characterisation of novel proteins. Electrophoresis 20, 818–825. Gulash, M., Ames, P., Larosiliere, R. C. and Bergman, K. (1984). Rhizobia are attracted to localized sites on legume roots. Applied and Environmental Microbiology 48, 149–152. Hamblin, J. and Kent, S. P. (1973). Possible role of phytohemagglutinin in Phaseolus vulgaris L. Nature New Biology 245, 28–29. Hartwig, U. A. and Phillips, D. A. (1991). Release and modification of nod geneinducing flavonoids from alfalfa seeds. Plant Physiology 95, 804–807. Hartwig, U. A., Maxwell, C. A., Joseph, C. M. and Phillips, D. A. (1990a). Chrysoeriol and luteolin released from alfalfa seeds induce nod genes in Rhizobium meliloti. Plant Physiology 92, 116–122. Hartwig, U. A., Maxwell, C. A., Joseph, C. M. and Phillips, D. A. (1990b). EVects of alfalfa nod gene-inducing flavonoids on nodABC transcription in Rhizobium meliloti strains containing diVerent nodD genes. Journal of Bacteriology 172, 2769–2773. Hartwig, U. A., Joseph, C. M. and Phillips, D. A. (1991). Flavonoids released naturally from alfalfa seeds enhance growth rate of Rhizobium meliloti. Plant Physiology 95, 797–803. Harwood, C. S. and Parales, R. E. (1996). The beta-ketoadipate pathway and the biology of self-identity. Annual Review of Microbiology 50, 553–590. He, S. Y. (1998). Type III protein secretion systems in plant and animal pathogenic bacteria. Annual Review of Phytopathology 36, 363–392. Hirsch, A. M. (1999). Role of lectins (and rhizobial exopolysaccharides) in legume nodulation. Current Opinion in Plant Biology 2, 320–326. Hong, G. F., Burn, J. E. and Johnston, A. W. B. (1987). Evidence that DNA involved in the expression of nodulation (nod ) genes in Rhizobium binds to the product of the regulatory gene nodD. Nucleic Acids Research 15, 9677–9690. Hopper, W. and Mahadevan, A. (1997). Degradation of catechin by Bradyrhizobium japonicum. Biodegradation 8, 159–165.


51

Hostettmann, K., Marston, A., Wolfender, J.-L. and Maillard, M. (1996). Screening for flavonoids and related compounds in medicinal plants by LC-UV-MS and subsequent isolation of bioactive compounds. In ‘‘Flavonoids and Bioflavonoids’’ (S. Antus, M. Gabor and K. Vetschera, eds.), pp. 35–51. Akademiai Kiado, Budapest. Hu, H. L., Liu, S. T., Yang, Y., Chang, W. Z. and Hong, G. F. (2000). In Rhizobium leguminosarum, NodD represses its own transcription by competing with RNA polymerase for binding sites. Nucleic Acids Research 28, 2784–2793. Hubac, C., Ferran, J., Guerrier, D., Tremolieres, A. and Kondorosi, A. (1993). Luteolin absorption in Rhizobium meliloti wild-type and mutant strains. Journal of General Microbiology 139, 1571–1578. Hubac, C., Ferran, J., Tremolieres, A. and Kondorosi, A. (1994). Luteolin uptake by Rhizobium meliloti: Evidence for several steps including an active extrusion process. Microbiology-UK 140, 2769–2774. Huck, C. W., Stecher, G., Ahrer, W., Stoggl, W. M., Buchberger, W. and Bonn, G. K. (2002). Analysis of three flavonoids by CE-UV and CE-ESI-MS. Determination of naringenin from a phytomedicine. Journal of Separation Science 25, 904–908. Hueck, C. J. (1998). Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiology and Molecular Biology Reviews 62, 379–433. Hungria, M., Joseph, C. M. and Phillips, D. A. (1991a). Anthocyanidins and flavonols, major nod gene inducers from seeds of a black-seeded common bean (Phaseolus vulgaris L.). Plant Physiology 97, 751–758. Hungria, M., Joseph, C. M. and Phillips, D. A. (1991b). Rhizobium nod gene inducers exuded naturally from roots of common bean (Phaseolus vulgaris L.). Plant Physiology 97, 759–764. Hungria, M., Johnston, A. W. B. and Phillips, D. A. (1992). EVects of flavonoids released naturally from bean (Phaseolus vulgaris) on NodD-regulated gene transcription in Rhizobium leguminosarum bv. phaseoli. Molecular PlantMicrobe Interactions 5, 199–203. Hur, H. G. and Rafii, F. (2000). Biotransformation of the isoflavonoids biochanin A, formononetin, and glycitein by Eubacterium limosum. FEMS Microbiology Letters 192, 21–25. Hur, H. G., Lay, J. O., Beger, R. D., Freeman, J. P. and Rafii, F. (2000). Isolation of human intestinal bacteria metabolizing the natural isoflavone glycosides daidzin and genistin. Archives of Microbiology 174, 422–428. Hur, H. G., Beger, R. D., Heinze, T. M., Lay, J. O., Freeman, J. P., Dore, J. and Rafii, F. (2002). Isolation of an anaerobic intestinal bacterium capable of cleaving the C-ring of the isoflavonoid daidzein. Archives of Microbiology 178, 8–12. Hussein, Y. A., Tewfik, M. S. and Hamdi, Y. A. (1974). Degradation of certain aromatic compounds by rhizobia. Soil Biology and Biochemistry 6, 377–381. Ibrahim, A. R. S. and Abul-Hajj, Y. J. (1990). Microbiological transformation of flavone and isoflavone. Xenobiotica 20, 363–373. Innes, R. W., Kuempel, P. L., Plazinski, J., Canter-Cremers, H., Rolfe, B. G. and Djordjevic, M. A. (1985). Plant factors induce expression of nodulation and host-range genes in Rhizobium trifolii. Molecular and General Genetics 201, 426–432. Jacobs, M. and Rubery, P. H. (1988). Naturally occurring auxin transport regulators. Science 241, 346–349. Jain, V. and Nainawatee, H. S. (1999). Flavonoids influence growth and saccharide metabolism of Rhizobium meliloti. Folia Microbiologica 44, 311–316.

52

J. E. COOPER

Jain, V. and Nainawatee, H. S. (2002). Plant flavonoids: Signals to legume nodulation and soil microorganisms. Journal of Plant Biochemistry and Biotechnology 11, 1–10. JeVrey, A. M., Jerina, D. M., Self, R. and Evans, W. C. (1972a). The bacterial degradation of flavonoids: Oxidative fission of the A-ring of dihydrogossypetin by a Pseudomonas sp. Biochemical Journal 130, 383–390. JeVrey, A. M., Knight, M. and Evans, W. C. (1972b). The bacterial degradation of flavonoids: Hydroxylation of the A-ring of taxifolin by a soil pseudomonad. Biochemical Journal 130, 373–380. Kalsi, G. and Etzler, M. E. (2000). Localization of a Nod factor-binding protein in legume roots and factors influencing its distribution and expression. Plant Physiology 124, 1039–1048. Kaneko, T., Nakamura, Y., Sato, S., Asamizu, E., Kato, T., Sasamoto, S., Watanabe, A., Idesawa, K., Ishikawa, A., Kawashima, K., Kimura, T., Kishida, Y., Kiyokawa, C., Kohara, M., Matsumoto, M., Matsuno, A., Mochizuki, Y., Nakayama, S., Nakazaki, N., Shimpo, S., Sugimoto, M., Takeuchi, C., Yamada, M. and Tabata, S. (2000). Complete genome structure of the nitrogen-fixing symbiotic bacterium Mesorhizobium loti. DNA Research 7, 331–338. Kannenberg, E. L. and Brewin, N. J. (1994). Host-plant invasion by Rhizobium: The role of cell surface components. Trends in Microbiology 2, 277–283. Kape, R., Parniske, M. and Werner, D. (1991). Chemotaxis and nod gene activity of Bradyrhizobium japonicum in response to hydroxycinnamic acids and isoflavonoids. Applied and Environmental Microbiology 57, 316–319. Kape, R., Parniske, M., Brandt, S. and Werner, D. (1992). Isoliquiritigenin, a strong nod gene-inducing and glyceollin resistance-inducing flavonoid from soybean root exudate. Applied and Environmental Microbiology 58, 1705–1710. Kite, G. C., Veitch, N. C., Grayer, R. J. and Simmonds, M. S. J. (2003). The use of hyphenated techniques in comparative phytochemical studies of legumes. Biochemical Systematics and Ecology. 31, 813–843. Knight, C. D., Rossen, L., Robertson, J. G., Wells, B. and Downie, J. A. (1986). Nodulation inhibition by Rhizobium leguminosarum multicopy nodABC genes and analysis of early stages of plant infection. Journal of Bacteriology 166, 552–558. Kobayashi, H., Naciri-Graven, Y., Broughton, W. J. and Perret, X. (2004). Flavonoids induce temporal shifts in gene-expression of nod-box controlled loci in Rhizobium sp. NGR234. Molecular Microbiology 51, 335–347. Kondorosi, E., Gyuris, J., Schmidt, J., John, M., Duda, E., HoVmann, B., Schell, J. and Kondorosi, A. (1989). Positive and negative control of nod gene expression in Rhizobium meliloti is required for optimal nodulation. EMBO Journal 8, 1331–1340. Kosslak, R. M., Bookland, R., Barkei, J., Paaren, H. E. and Appelbaum, E. R. (1987). Induction of Bradyrhizobium japonicum common nod genes by isoflavones isolated from Glycine max. Proceedings of the National Academy of Sciences of the United States of America 84, 7428–7432. Kosslak, R. M., Joshi, R. S., Bowen, B. A., Paaren, H. E. and Appelbaum, E. R. (1990). Strain-specific inhibition of nod gene induction in Bradyrhizobium japonicum by flavonoid compounds. Applied and Environmental Microbiology 56, 1333–1341. Kouchi, H. and Hata, S. (1993). Isolation and characterization of novel nodulin cDNAs representing genes expressed at early stages of soybean nodule development. Molecular and General Genetics 238, 106–119.


53

Krause, A., Lan, V. T. T. and Broughton, W. J. (1997). Induction of chalcone synthase expression by rhizobia and Nod factors in root hairs and roots. Molecular Plant-Microbe Interactions 10, 388–393. Krause, A., Doerfel, A. and Go¨ ttfert, M. (2002). Mutational and transcriptional analysis of the type III secretion system of Bradyrhizobium japonicum. Molecular Plant-Microbe Interactions 15, 1228–1235. Krishnan, H. B. and Pueppke, S. G. (1993). Flavonoid inducers of nodulation genes stimulate Rhizobium fredii USDA257 to export proteins into the environment. Molecular Plant-Microbe Interactions 6, 107–113. Krishnan, H. B., Kuo, C. I. and Pueppke, S. G. (1995). Elaboration of flavonoidinduced proteins by the nitrogen-fixing soybean symbiont Rhizobium fredii is regulated by both nodD1 and nodD2, and is dependent on the cultivarspecificity locus, nolXWBTUV. Microbiology-UK 141, 2245–2251. Krishnan, H. B., Lorio, J., Kim, W. S., Jiang, G. Q., Kim, K. Y., De Boer, M. and Pueppke, S. G. (2003). Extracellular proteins involved in soybean cultivarspecific nodulation are associated with pilus-like surface appendages and exported by a type III protein secretion system in Sinorhizobium fredii USDA257. Molecular Plant-Microbe Interactions 16, 617–625. Krumholz, L. R. and Bryant, M. P. (1986). Eubacterium oxidoreducens sp. Nov. requiring H2 or formate to degrade gallate, pyrogallol, phloroglucinol and quercetin. Archives of Microbiology 144, 8–14. Kubori, T., Matsushima, Y., Nakamura, D., Uralil, J., Lara-Tejero, M., Sukhan, A., Gala´ n, J. E. and Aizawa, S. (1998). Supramolecular structure of the Salmonella typhimurium type III protein secretion system. Science 280, 602–605. Lameta, A. D. and Jay, M. (1987). Study of soybean and lentil root exudates. III. Influence of soybean isoflavonoids on the growth of rhizobia and some rhizospheric microorganisms. Plant and Soil 101, 267–272. Latha, S. and Mahadevan, A. (1997). Role of rhizobia in the degradation of aromatic substances. World Journal of Microbiology and Biotechnology 13, 601–607. Lawson, C. G. R., Djordjevic, M. A., Weinman, J. J. and Rolfe, B. G. (1994). Rhizobium inoculation and physical wounding result in the rapid induction of the same chalcone synthase copy in Trifolium subterraneum. Molecular Plant-Microbe Interactions 7, 498–507. Lawson, C. G. R., Rolfe, B. G. and Djordjevic, M. A. (1996). Rhizobium inoculation induces condition-dependent changes in the flavonoid composition of root exudates from Trifolium subterraneum. Australian Journal of Plant Physiology 23, 93–101. Lee, C. A. (1997). Type III secretion systems: Machines to deliver bacterial proteins into eukaryotic cells? Trends in Microbiology 5, 148–156. Lee, S. H., Seo, D. H., Park, H. L., Choi, Y. J. and Jung, S. H. (2003). Solubility enhancement of a hydrophobic flavonoid, luteolin by the complexation with cyclosophoraoses isolated from Rhizobium meliloti. Antonie van Leeuwenhoek 84, 201–207. Leo´ n-Barrios, M., Dakora, F. D., Joseph, C. M. and Phillips, D. A. (1993). Isolation of Rhizobium meliloti nod gene inducers from alfalfa rhizosphere soil. Applied and Environmental Microbiology 59, 636–639. Lerouge, P., Roche, P., Faucher, C., Maillet, F., Truchet, G., Prome´ , J.-C. and De´ narie´ , J. (1990). Symbiotic host-specificity of Rhizobium meliloti is determined by a sulfated and acylated glucosamine oligosaccharide signal. Nature 344, 781–784. Le Strange, K. K., Bender, G. L., Djordjevic, M. A., Rolfe, B. G. and Redmond, J. W. (1990). The Rhizobium strain NGR234 nodD1 gene product responds to activation by the simple phenolic compounds vanillin and isovanillin

54

J. E. COOPER

present in wheat seedling extracts. Molecular Plant-Microbe Interactions 3, 214–220. Lewin, A., Rosenberg, C., Meyer, H., Wong, C. H., Nelson, L., Manen, J. F., Stanley, J., Dowling, D. N., De´ narie´ , J. and Broughton, W. J. (1987). Multiple host-specificity loci of the broad host range Rhizobium sp. NGR234 selected using the widely compatible legume Vigna unguiculata. Plant Molecular Biology 8, 447–459. Limpens, E. and Bisseling, T. (2003). Signaling in symbiosis. Current Opinion in Plant Biology 6, 343–350. Limpens, E., Franken, C., Smit, P., Willemse, J., Bisseling, T. and Geurts, R. (2003). LysM domain receptor kinases regulating rhizobial Nod factor-induced infection. Science 302, 630–633. Lin, C. C., Chen, Y. C., Song, S. C. and Lin, L. P. (1999). Flavonoids as inducers of extracellular proteins and exopolysaccharides of Sinorhizobium fredii. Biology and Fertility of Soils. 30, 83–89. Loh, J., Garcia, M. and Stacey, G. (1997). NodV and NodW, a second flavonoid recognition system regulating nod gene expression in Bradyrhizobium japonicum. Journal of Bacteriology 179, 3013–3020. Lo´ pez-Lara, I. M., van den Berg, J. D. J., Thomas-Oates, J. E., Glushka, J., Lugtenberg, B. J. J. and Spaink, H. P. (1995). Structural identification of the lipo-chitin oligosaccharide nodulation signals of Rhizobium loti. Molecular Microbiology 15, 627–638. Lo´ pez-Lara, I. M., Sohlenkamp, C. and Geiger, O. (2003). Membrane lipids in plantassociated bacteria: Their biosynthesis and possible functions. Molecular Plant-Microbe Interactions 16, 567–579. Lorite, M. J., Sanjua´ n, J., Velasco, L., Olivares, J. and Bedmar, E. J. (1998). Characterization of Bradyrhizobium japonicum pcaBDC genes involved in 4-hydroxybenzoate degradation. Biochimica et Biophysica Acta 1397, 257–261. Machado, D., Pueppke, S. G., Vinardel, J. M., Ruiz-Sainz, J. E. and Krishnan, H. B. (1998). Expression of nodD1 and nodD2 in Sinorhizobium fredii, a nitrogenfixing symbiont of soybean and other legumes. Molecular Plant-Microbe Interactions 11, 375–382. Madsen, E. B., Madsen, L. H., Radutoiu, S., Olbryt, M., Rakwalska, M., Szczyglowski, K., Sato, S., Kaneko, T., Tabata, S., Sandal, N. and Stougaard, J. (2003). A receptor kinase gene of the LysM type is involved in legume perception of rhizobial signals. Nature 425, 637–640. Marie, C., Broughton, W. J. and Deakin, W. J. (2001). Rhizobium type III secretion systems: Legume charmers or alarmers? Current Opinion in Plant Biology 4, 336–342. Marie, C., Deakin, W. J., Viprey, V., Kopcinska, J., Golinowski, W., Krishnan, H. B., Perret, X. and Broughton, W. J. (2003). Characterization of Nops, nodulation outer proteins, secreted via the type III secretion system of NGR234. Molecular Plant-Microbe Interactions 16, 743–751. Marketon, M. M., Glenn, S. A., Eberhard, A. and Gonzalez, J. E. (2003). Quorum sensing controls exopolysaccharide production in Sinorhizobium meliloti. Journal of Bacteriology 185, 325–331. Mathesius, U. (2001). Flavonoids induced in cells undergoing nodule organogenesis in white clover are regulators of auxin breakdown by peroxidase. Journal of Experimental Botany 52, 419–426. Mathesius, U., Schlaman, H. R. M., Spaink, H. P., Sautter, C., Rolfe, B. G. and Djordjevic, M. A. (1998). Auxin transport inhibition precedes root nodule


55

formation in white clover roots and is regulated by flavonoids and derivatives of chitin oligosaccharides. Plant Journal 14, 23–34. Mathesius, U., Charon, C., Rolfe, B. G., Kondorosi, A. and Crespi, M. (2000). Temporal and spatial order of events during the induction of cortical cell divisions in white clover by Rhizobium leguminosarum bv. trifolii inoculation or localized cytokinin addition. Molecular Plant-Microbe Interactions 13, 617–628. Mathesius, U., Mulders, S., Gao, M. S., Teplitski, M., Caetano-Ano´ lles, G., Rolfe, B. G. and Bauer, W. D. (2003). Extensive and specific responses of a eukaryote to bacterial quorum-sensing signals. Proceedings of the National Academy of Sciences of the United States of America 100, 1444–1449. Maxwell, C. A. and Phillips, D. A. (1990). Concurrent synthesis and release of nod gene-inducing flavonoids from alfalfa roots. Plant Physiology 93, 1552–1558. Maxwell, C. A., Hartwig, U. A., Joseph, C. M. and Phillips, D. A. (1989). A chalcone and two related flavonoids released from alfalfa roots induce nod genes of Rhizobium meliloti. Plant Physiology 91, 842–847. McKhann, H. I. and Hirsch, A. M. (1994). Isolation of chalcone synthase and chalcone isomerase cDNAs from alfalfa (Medicago sativa): Highest transcript levels occur in young roots and root tips. Plant Molecular Biology 24, 767–777. McKhann, H. I., Paiva, N. L., Dixon, R. A. and Hirsch, A. M. (1997). Chalcone synthase transcripts are detected in alfalfa root hairs following inoculation with wild-type Rhizobium meliloti. Molecular Plant-Microbe Interactions 10, 50–58. Meinhardt, L. W., Krishnan, H. B., Balatti, P. A. and Pueppke, S. G. (1993). Molecular cloning and characterization of a Sym plasmid locus that regulates cultivar-specific nodulation of soybean by Rhizobium fredii USDA257. Molecular Microbiology 9, 17–29. Messens, E., Geelen, D., van Montagu, M. and Holsters, M. (1991). 7,40 -Dihydroxyflavanone is the major Azorhizobium nod gene-inducing factor present in Sesbania rostrata seedling exudate. Molecular Plant-Microbe Interactions 4, 262–267. Michiels, J., Dombrecht, B., Vermeiren, N., Xi, C.-W., Luyten, E. and Vanderleyden, J. (1998). Phaseolus vulgaris is a non-selective host for nodulation. FEMS Microbiology Ecology 26, 193–205. Mitho¨ fer, A. (2002). Suppression of plant defence in rhizobia-legume symbiosis. Trends in Plant Science 7, 440–444. Morris, P. and Robbins, M. P. (1992). Condensed tannin formation by Agrobacterium rhizogenes transformed root and shoot organ cultures of Lotus corniculatus. Journal of Experimental Botany 43, 221–231. Mulligan, J. T. and Long, S. R. (1985). Induction of Rhizobium meliloti nodC expression by plant exudate requires NodD. Proceedings of the National Academy of Sciences of the United States of America 82, 6609–6613. Muthukumar, G., Arunakumari, A. and Mahadevan, A. (1982). Degradation of aromatic compounds by Rhizobium spp. Plant and Soil 69, 163–169. Noel, K. D. and Duelli, D. M. (2000). Rhizobium lipopolysaccharide and its role in symbiosis. In ‘‘Prokaryotic Nitrogen Fixation’’ (E. W. Triplett, ed.), pp. 415–431. Horizon Scientific Press, Wymondham, United Kingdom. Noel, K. D., Duelli, D. M., Tao, H. and Brewin, N. J. (1996). Antigenic change in the lipopolysaccharide of Rhizobium etli CFN42 induced by exudates of Phaseolus vulgaris. Molecular Plant-Microbe Interactions 9, 180–186.

56

J. E. COOPER

Ogawa, J. and Long, S. R. (1995). The Rhizobium meliloti groELc locus is required for regulation of early nod genes by the transcription activator NodD. Genes and Development 9, 714–729. Okker, R. J. H., Spaink, H. P., Lugtenberg, B. J. J. and Schlaman, H. R. M. (2001). Mutants in the nodFEL promoter of Rhizobium leguminosarum bv viciae reveal a role of individual nucleotides in transcriptional activation and protein binding. Archives of Microbiology 175, 152–160. Oldroyd, G. E. D. (2001). Dissecting symbiosis: Developments in Nod factor signal transduction. Annals of Botany 87, 709–718. Oldroyd, G. E. D. and Long, S. R. (2003). Identification and characterization of nodulation-signaling pathway 2, a gene of Medicago truncatula involved in Nod factor signaling. Plant Physiology 131, 1027–1032. Pacios Bras, C., Spaink, H. P. and Stuurman, N. (2000). Structure and function of Nod factors. In ‘‘Prokaryotic Nitrogen Fixation’’ (E. W. Triplett, ed.), pp. 365–383. Horizon Scientific Press, Wymondham, United Kingdom. Pandya, S. and Desai, A. (1998). Localization of Nod, Nif, and acidic exopolysaccharide determinants on a large plasmid in slow-growing cowpea Rhizobium sp. S2. Current Microbiology 36, 36–40. Pankhurst, C. E. and Biggs, D. R. (1980). Sensitivity of Rhizobium to selected isoflavonoids. Canadian Journal of Microbiology 26, 542–545. Pankhurst, C. E. and Jones, W. T. (1979). EVectiveness of Lotus root nodules. III. EVect of combined nitrogen on nodule eVectiveness and flavolan biosynthesis in plant roots. Journal of Experimental Botany 30, 1109–1118. Parke, D. (1993). Positive regulation of phenolic catabolism in Agrobacterium tumefaciens by the pcaQ gene in response to beta-carboxy-cis,cis-muconate. Journal of Bacteriology 175, 3529–3535. Parke, D. (1995). Supraoperonic clustering of pca genes for catabolism of the phenolic compound protocatechuate in Agrobacterium tumefaciens. Journal of Bacteriology 177, 3808–3817. Parke, D. (1996a). Characterization of PcaQ, a LysR-type transcriptional activator required for catabolism of phenolic compounds, from Agrobacterium tumefaciens. Journal of Bacteriology 178, 266–272. Parke, D. (1996b). Conservation of PcaQ, a transcriptional activator of pca genes for catabolism of phenolic compounds, in Agrobacterium tumefaciens and Rhizobium species. Journal of Bacteriology 178, 3671–3675. Parke, D. (1997). Acquisition, reorganization, and merger of genes: Novel management of the beta-ketoadipate pathway in Agrobacterium tumefaciens. FEMS Microbiology Letters 146, 3–12. Parke, D. and Ornston, L. N. (1984). Nutritional diversity of Rhizobiaceae revealed by auxanography. Journal of General Microbiology 130, 1743–1750. Parke, D. and Ornston, L. N. (1986). Enzymes of the beta-ketoadipate pathway are inducible in Rhizobium and Agrobacterium spp. and constitutive in Bradyrhizobium spp. Journal of Bacteriology 165, 288–292. Parke, D., Rivelli, M. and Ornston, L. N. (1985). Chemotaxis to aromatic and hydroaromatic acids: Comparison of Bradyrhizobium japonicum and Rhizobium trifolii. Journal of Bacteriology 163, 417–422. Parke, D., Rynne, F. and Glenn, A. (1991). Regulation of phenolic catabolism in Rhizobium leguminosarum biovar trifolii. Journal of Bacteriology 173, 5546–5550. Parniske, M., Ahlborn, B. and Werner, D. (1991). Isoflavonoid-inducible resistance to the phytoalexin glyceollin in soybean rhizobia. Journal of Bacteriology 173, 3432–3439.


57

Perret, X., Fellay, R., Bjourson, A. J., Cooper, J. E., Brenner, S. and Broughton, W. J. (1994). Subtraction hybridization and shot gun sequencing: A new approach to identify symbiotic loci. Nucleic Acids Research 22, 1335–1341. Perret, X., Freiberg, C., Rosenthal, A., Broughton, W. J. and Fellay, R. (1999). High resolution transcriptional analysis of the symbiotic plasmid of Rhizobium sp. NGR234. Molecular Microbiology 32, 415–425. Perret, X., Staehelin, C. and Broughton, W. J. (2000). Molecular basis of symbiotic promiscuity. Microbiology and Molecular Biology Reviews 64, 180–201. Peters, N. K. and Long, S. R. (1988). Alfalfa root exudates and compounds which promote or inhibit induction of Rhizobium meliloti nodulation genes. Plant Physiology 88, 396–400. Peters, N. K., Frost, J. W. and Long, S. R. (1986). A plant flavone, luteolin, induces expression of Rhizobium meliloti nodulation genes. Science 233, 977–980. Phillips, D. A., Joseph, C. M. and Maxwell, C. A. (1992). Trigonelline and stachydrine released from alfalfa seeds activate NodD2 protein in Rhizobium meliloti. Plant Physiology 99, 1526–1531. Phillips, D. A., Dakora, F. D., Sande, E., Joseph, C. M. and Zon, J. (1994). Synthesis, release, and transmission of alfalfa signals to rhizobial symbionts. Plant and Soil 161, 69–80. Phillips, D. A., Wery, J., Joseph, C. M., Jones, A. D. and Teuber, L. R. (1995). Release of flavonoids and betaines from seeds of seven Medicago species. Crop Science 35, 805–808. Pillai, B. V. S. and Swarup, S. (2002). Elucidation of the flavonoid catabolism pathway in Pseudomonas putida PML2 by comparative metabolic profiling. Applied and Environmental Microbiology 68, 143–151. Price, N. P. J. (1999). Carbohydrate determinants of Rhizobium-legume symbioses. Carbohydrate Research 317, 1–9. Pueppke, S. G. and Broughton, W. J. (1999). Rhizobium sp. strain NGR234 and R. fredii USDA257 share exceptionally broad, nested host ranges. Molecular Plant-Microbe Interactions 12, 293–318. Radutoiu, S., Madsen, L. H., Madsen, E. B., Felle, H. H., Umehara, Y., Grønlund, M., Sato, S., Nakamura, Y., Tabata, S., Sandal, N. and Stougaard, J. (2003). Plant recognition of symbiotic bacteria requires two LysM receptor-like kinases. Nature 425, 585–592. Rao, J. R. and Cooper, J. E. (1994). Rhizobia catabolize nod gene-inducing flavonoids via C-ring fission mechanisms. Journal of Bacteriology 176, 5409–5413. Rao, J. R. and Cooper, J. E. (1995). Soybean nodulating rhizobia modify nod gene inducers daidzein and genistein to yield aromatic products that can influence gene-inducing activity. Molecular Plant-Microbe Interactions 8, 855–862. Rao, J. R., Sharma, N. D., Hamilton, J. T. G., Boyd, D. R. and Cooper, J. E. (1991). Biotransformation of the pentahydroxyflavone quercetin by Rhizobium loti and Bradyrhizobium strains (Lotus). Applied and Environmental Microbiology 57, 1563–1565. Rao, J. R., Cooper, J. E., Everaert, E. S. W. and De Cooman, L. (1996). Assimilation of nod gene inducer 14C-naringenin and the incorporation of labelled carbon atoms into the acyl side chain of a host-specific Nod factor produced by Rhizobium leguminosarum bv. viciae. Plant and Soil 186, 63–67. Recourt, K., Schripsema, J., Kijne, J. W., van Brussel, A. A. N. and Lugtenberg, B. J. J. (1991). Inoculation of Vicia sativa subsp. nigra roots with Rhizobium

58

J. E. COOPER

leguminosarum biovar viciae results in release of nod gene-activating flavanones and chalcones. Plant Molecular Biology 16, 841–852. Recourt, K., van Tunen, A. J., Mur, L. A., van Brussel, A. A. N., Lugtenberg, B. J. J. and Kijne, J. W. (1992). Activation of flavonoid biosynthesis in roots of Vicia sativa subsp. nigra plants by inoculation with Rhizobium leguminosarum biovar viciae. Plant Molecular Biology 19, 411–420. Redmond, J. W., Batley, M., Djordjevic, M. A., Innes, R. W., Kuempel, P. L. and Rolfe, B. G. (1986). Flavones induce expression of nodulation genes in Rhizobium. Nature 323, 632–635. Relic´ , B., Perret, X., Estrada-Garcia, M. T., Kopcinska, J., Golinowski, W., Krishnan, H. B., Pueppke, S. G. and Broughton, W. J. (1994). Nod factors of Rhizobium are a key to the legume door. Molecular Microbiology 13, 171–178. Reuhs, B. L., Carlson, R. W. and Kim, J. S. (1993). Rhizobium fredii and Rhizobium meliloti produce 3-deoxy-D-manno-2-octulosonic acid-containing polysaccharides that are structurally analogous to group II K antigens (capsular polysaccharides) found in Escherichia coli. Journal of Bacteriology 175, 3570–3580. Reuhs, B. L., Kim, J. S., Badgett, A. and Carlson, R. W. (1994). Production of cellassociated polysaccharides of Rhizobium fredii USDA205 is modulated by apigenin and host root extract. Molecular Plant-Microbe Interactions 7, 240–247. Reuhs, B. L., Williams, M. N. V., Kim, J. S., Carlson, R. W. and Cote, F. (1995). Suppression of the Fix phenotype of Rhizobium meliloti exoB mutants by lpsZ is correlated to a modified expression of the K-polysaccharide. Journal of Bacteriology 177, 4289–4296. Robbins, M. P., Thomas, B. and Morris, P. (1995). Phenylpropanoid defense responses in transgenic Lotus corniculatus. II. Modeling plant defense responses in transgenic root cultures using thiol and carbohydrate elicitors. Journal of Experimental Botany 46, 513–524. Rodelas, B., Lithgow, J. K., Wisniewski-Dye, F., Hardman, A., Wilkinson, A., Economou, A., Williams, P. and Downie, J. A. (1999). Analysis of quorum-sensing-dependent control of rhizosphere-expressed (rhi) genes in Rhizobium leguminosarum bv. viciae. Journal of Bacteriology 181, 3816–3823. Roine, E., Wei, W. S., Yuan, J., Nurmiaho-Lassila, E. L., Kalkkinen, N., Romantschuk, M. and He, S. Y. (1997). Hrp pilus: An hrp-dependent bacterial surface appendage produced by Pseudomonas syringae pv. tomato DC3000. Proceedings of the National Academy of Sciences of the United States of America 94, 3459–3464. Rolfe, B. G., Batley, M., Redmond, J. W., Richardson, A. E., Simpson, R. J., Bassam, B. J., Sargent, C. L., Weinman, J. J., Djordjevic, M. A. and Dazzo, F. B. (1988). Phenolic compounds secreted by legumes. In ‘‘Nitrogen Fixation: Hundred Years After’’ (H. Bothe, F. J. de Bruijn and W. E. Newton, eds.), pp. 405–409. Gustav Fischer, Stuttgart, Germany. Rossen, L., Shearman, C. A., Johnston, A. W. B. and Downie, J. A. (1985). The nodD gene of Rhizobium leguminosarum is autoregulatory and in the presence of plant exudate induces the nodABC genes. EMBO Journal 4, 3369–3373. Rostas, K., Kondorosi, E., Horvath, B., Simoncsits, A. and Kondorosi, A. (1986). Conservation of extended promoter regions of nodulation genes in Rhizobium. Proceedings of the National Academy of Sciences of the United States of America 83, 1757–1761.


59

Ryder, T. B., Hedrick, S. A., Bell, J. N., Liang, X. W., Clouse, S. D. and Lamb, C. J. (1987). Organization and diVerential activation of a gene family encoding the plant defense enzyme chalcone synthase in Phaseolus vulgaris. Molecular and General Genetics 210, 219–233. Sadowsky, M. J., Olson, E. R., Foster, V. E., Kosslak, R. M. and Verma, D. P. S. (1988). Two host-inducible genes of Rhizobium fredii and characterization of the inducing compound. Journal of Bacteriology 170, 171–178. Sadowsky, M. J., Cregan, P. B., Go¨ ttfert, M., Sharma, A., Gerhold, D., RodriguezQuinones, F., Keyser, H. H., Hennecke, H. and Stacey, G. (1991). The Bradyrhizobium japonicum nolA gene and its involvement in the genotypespecific nodulation of soybeans. Proceedings of the National Academy of Sciences of the United States of America 88, 637–641. Sanjua´ n, J., Grob, P., Go¨ ttfert, M., Hennecke, H. and Stacey, G. (1994). NodW is essential for full expression of the common nodulation genes in Bradyrhizobium japonicum. Molecular Plant-Microbe Interactions 7, 364–369. Sanjua´ n-Pinilla, J. M., Munoz, S., Nogales, J., Olivares, J. and Sanjua´ n, J. (2002). Involvement of the Sinorhizobium meliloti leuA gene in activation of nodulation genes by NodD1 and luteolin. Archives of Microbiology 178, 36–44. Savoure´ , A., Magyar, Z., Pierre, M., Brown, S., Schultze, M., Dudits, D., Kondorosi, A. and Kondorosi, E. (1994). Activation of the cell cycle machinery and the isoflavonoid biosynthesis pathway by active Rhizobium meliloti Nod signal molecules in Medicago microcallus suspensions. EMBO Journal 13, 1093–1102. Savoure´ , A., Sallaud, C., El Turk, J., Zuanazzi, J., Ratet, P., Schultze, M., Kondorosi, A., Esnault, R. and Kondorosi, E. (1997). Distinct response of Medicago suspension cultures and roots to Nod factors and chitin oligomers in the elicitation of defense-related responses. Plant Journal 11, 277–287. Sawada, H., Kuykendall, L. D. and Young, J. M. (2003). Changing concepts in the systematics of bacterial nitrogen-fixing legume symbionts. Journal of General and Applied Microbiology 49, 155–179. Scheidemann, P. and Wetzel, A. (1997). Identification and characterization of flavonoids in the root exudate of Robinia pseudoacacia. Trees 11, 316–321. Schell, M. A. (1993). Molecular biology of the LysR family of transcriptional regulators. Annual Review of Microbiology 47, 597–626. Scheres, B., Vandewiel, C., Zalensky, A., Horvath, B., Spaink, H., Van Eck, H., Zwartkruis, F., Wolters, A. M., Gloudemans, T., van Kammen, A. and Bisseling, T. (1990). The Enod12 gene product is involved in the infection process during the pea-Rhizobium interaction. Cell 60, 281–294. Scheu, A. K., Economou, A., Hong, G. F., Ghelani, S., Johnston, A. W. B. and Downie, J. A. (1992). Secretion of the Rhizobium leguminosarum nodulation protein NodO by hemolysin-type systems. Molecular Microbiology 6, 231–238. Schlaman, H. R. M., Okker, R. J. H. and Lugtenberg, B. J. J. (1992). Regulation of nodulation gene expression by NodD in rhizobia. Journal of Bacteriology 174, 5177–5182. Schlaman, H. R. M., Phillips, D. A. and Kondorosi, E. (1998). Genetic organisation and transcriptional regulation of rhizobial nodulation genes. In ‘‘The Rhizobiaceae’’ (H. P. Spaink, A. Kondorosi and P. J. J. Hooykaas, eds.), pp. 361–386. Kluwer, Dordrecht, The Netherlands. Schmidt, P. E., Broughton, W. J. and Werner, D. (1994). Nod factors of Bradyrhizobium japonicum and Rhizobium sp. NGR234 induce flavonoid accumula-

60

J. E. COOPER

tion in soybean root exudate. Molecular Plant-Microbe Interactions 7, 384–390. Schoefer, L., Mohan, R., Braune, A., Birringer, M. and Blaut, M. (2002). Anaerobic C-ring cleavage of genistein and daidzein by Eubacterium ramulus. FEMS Microbiology Letters 208, 197–202. Schultz, E., Engle, F. E. and Wood, J. M. (1974). New oxygenases in the degradation of flavones and flavanones by Pseudomonas putida. Biochemistry 13, 1768–1776. Schultze, M. and Kondorosi, A. (1998). Regulation of symbiotic root nodule development. Annual Review of Genetics 32, 33–57. Shaw, S. L. and Long, S. R. (2003). Nod factor inhibition of reactive oxygen eZux in a host legume. Plant Physiology 132, 2196–2204. Shirley, B. W. (1996). Flavonoid biosynthesis: ‘New’ functions for an ‘old’ pathway. Trends in Plant Science 1, 377–382. Smit, G., Puvanesarajah, V., Carlson, R. W., Barbour, W. M. and Stacey, G. (1992). Bradyrhizobium japonicum nodD1 can be specifically induced by soybean flavonoids that do not induce the nodYABCSUIJ operon. Journal of Biological Chemistry 267, 310–318. Sohlenkamp, C., Lo´ pez-Lara, I. M. and Geiger, O. (2003). Biosynthesis of phosphatidylcholine in bacteria. Progress in Lipid Research 42, 115–162. Spaink, H. P. (1992). Rhizobial lipo-oligosaccharides: Answers and questions. Plant Molecular Biology 20, 977–986. Spaink, H. P. (2000). Root nodulation and infection factors produced by rhizobial bacteria. Annual Review of Microbiology 54, 257–288. Spaink, H. P., Sheeley, D. M., van Brussel, A. A. N., Glushka, J., York, W. S., Tak, T., Geiger, O., Kennedy, E. P., Reinhold, V. N. and Lugtenberg, B. J. J. (1991). A novel highly unsaturated fatty acid moiety of lipo-oligosaccharide signals determines host specificity of Rhizobium. Nature 354, 125–130. Spaink, H. P., Wijfjes, A. H. M., van Vliet, T. B., Kijne, J. W. and Lugtenberg, B. J. J. (1993). Rhizobial lipo-oligosaccharide signals and their role in plant morphogenesis: Are analogous lipophilic chitin derivatives produced by the plant? Australian Journal of Plant Physiology 20, 381–392. StaVord, H. A. (1997). Roles of flavonoids in symbiotic and defense functions in legume roots. Botanical Review 63, 27–39. Steele, H. L., Werner, D. and Cooper, J. E. (1999). Flavonoids in seed and root exudates of Lotus pedunculatus and their biotransformation by Mesorhizobium loti. Physiologia Plantarum 107, 251–258. Stracke, S., Kistner, C., Yoshida, S., Mulder, L., Sato, S., Kaneko, T., Tabata, S., Sandal, N., Stougaard, J., Szczyglowski, K. and Parniske, M. (2002). A plant receptor-like kinase required for both bacterial and fungal symbiosis. Nature 417, 959–962. Streit, W. R. and Phillips, D. A. (1997). A biotin-regulated locus, bioS, in a possible survival operon of Rhizobium meliloti. Molecular Plant-Microbe Interactions 10, 933–937. Streit, W. R., Joseph, C. M. and Phillips, D. A. (1996). Biotin and other watersoluble vitamins are key growth factors for alfalfa root colonization by Rhizobium meliloti 1021. Molecular Plant-Microbe Interactions 9, 330–338. Suominen, L., Luukkainen, R., Roos, C. and Lindstrom, K. (2003). Activation of the nodA promoter by the nodD genes of Rhizobium galegae induced by synthetic flavonoids or Galega orientalis root exudate. FEMS Microbiology Letters 219, 225–232. Sutton, J. M., Lea, E. J. A. and Downie, J. A. (1994). The nodulation signaling protein NodO from Rhizobium leguminosarum biovar viciae forms ion


61

channels in membranes. Proceedings of the National Academy of Sciences of the United States of America 91, 9990–9994. Swanson, J. A., Mulligan, J. T. and Long, S. R. (1993). Regulation of syrM and nodD3 in Rhizobium meliloti. Genetics 134, 435–444. Tao, H., Brewin, N. J. and Noel, K. D. (1992). Rhizobium leguminosarum CFN42 lipopolysaccharide antigenic changes induced by environmental conditions. Journal of Bacteriology 174, 2222–2229. Timmers, A. C. J., Auriac, M. C., de Billy, F. and Truchet, G. (1998). Nod factor internalization and microtubular cytoskeleton changes occur concomitantly during nodule diVerentiation in alfalfa. Development 125, 339–349. van Brussel, A. A. N., Recourt, K., Pees, E., Spaink, H. P., Tak, T., WijVelman, C. A., Kijne, J. W. and Lugtenberg, B. J. J. (1990). A biovar-specific signal of Rhizobium leguminosarum bv. viciae induces increased nodulation geneinducing activity in root exudate of Vicia sativa subsp. nigra. Journal of Bacteriology 172, 5394–5401. van Brussel, A. A. N., Bakhuizen, R., Van Spronsen, P. C., Spaink, H. P., Tak, T., Lugtenberg, B. J. J. and Kijne, J. W. (1992). Induction of preinfection thread structures in the leguminous host plant by mitogenic lipooligosaccharides of Rhizobium. Science 257, 70–72. van Brussel, A. A. N., Tak, T., Boot, K. J. M. and Kijne, J. W. (2002). Autoregulation of root nodule formation: Signals of both symbiotic partners studied in a split-root system of Vicia sativa subsp. nigra. Molecular Plant-Microbe Interactions 15, 341–349. Vance, C. P. (2002). Root-bacteria interactions: Symbiotic N2 fixation. In ‘‘Plant Roots’’ (Y. Waisal, A. Eshel and U. Kafkafi, eds.), pp. 839–868. Marcel Dekker, New York. van Rhijn, P., Luyten, E., Vlassak, K. and Vanderleyden, J. (1996). Isolation and characterization of a pSym locus of Rhizobium sp. BR816 that extends nodulation ability of narrow host range Phaseolus vulgaris symbionts to Leucaena leucocephala. Molecular Plant-Microbe Interactions 9, 74–77. van Rhijn, P., Goldberg, R. B. and Hirsch, A. M. (1998). Lotus nodulation specificity is changed by the presence of a soybean lectin gene. Plant Cell 10, 1233–1249. van Rhijn, P., Fujishige, N. A., Lim, P. O. and Hirsch, A. M. (2001). Sugar-binding activity of pea lectins enhances heterologous infection of transgenic alfalfa plants by Rhizobium leguminosarum bv. viciae. Plant Physiology 126, 133–144. Vela, S., Haggblom, M. M. and Young, L. Y. (2002). Biodegradation of aromatic and aliphatic compounds by rhizobial species. Soil Science 167, 802–810. Viprey, V., Del Greco, A., Golinowski, W., Broughton, W. J. and Perret, X. (1998). Symbiotic implications of type III protein secretion machinery in Rhizobium. Molecular Microbiology 28, 1381–1389. Vlassak, K. M., Luyten, E., Verreth, C., van Rhijn, P., Bisseling, T. and Vanderleyden, J. (1998). The Rhizobium sp. BR816 nodO gene can function as a determinant for nodulation of Leucaena leucocephala, Phaseolus vulgaris, and Trifolium repens by a diversity of Rhizobium spp. Molecular Plant-Microbe Interactions 11, 383–392. Wagner, M. L., Strittmatter, C. D., Kade, M., Rivero, M., Ricco, R. A. and Gurni, A. A. (1996). Phytochemical survey of Lotus tenuis Waldst-and-Kit. Phyton International Journal of Experimental Botany 58, 141–146. Wais, R. J., Galera, C., Oldroyd, G., Catoira, R., Penmesta, R. V., Cook, D., Gough, C., De´ narie´ , J. and Long, S. R. (2000). Genetic analysis of calcium spiking responses in nodulation mutants of Medicago truncatula. Proceedings of the

62

J. E. COOPER

National Academy of Sciences of the United States of America 97, 13407–13412. Wais, R. J., Keating, D. H. and Long, S. R. (2002). Structure-function analysis of Nod factor-induced root hair calcium spiking in Rhizobium-legume symbiosis. Plant Physiology 129, 211–224. Walker, S. A. and Downie, J. A. (2000). Entry of Rhizobium leguminosarum bv. viciae into root hairs requires minimal Nod factor specificity, but subsequent infection thread growth requires nodO or nodE. Molecular Plant-Microbe Interactions 13, 754–762. Walker, S. A., Viprey, V. and Downie, J. A. (2000). Dissection of nodulation signaling using pea mutants defective for calcium spiking induced by Nod factors and chitin oligomers. Proceedings of the National Academy of Sciences of the United States of America 97, 13413–13418. Wells, D. H. and Long, S. R. (2003). Mutations in rpoBC suppress the defects of a Sinorhizobium meliloti relA mutant. Journal of Bacteriology 185, 5602–5610. Winter, J., Moore, L. H., Dowell, V. R. and Bokkenheuser, V. D. (1989). C-ring cleavage of flavonoids by human intestinal bacteria. Applied and Environmental Microbiology 55, 1203–1208. Wisniewski-Dye, F. and Downie, J. A. (2002). Quorum-sensing in Rhizobium. Antonie van Leeuwenhoek 81, 397–407. Wong, C. M., Dilworth, M. J. and Glenn, A. R. (1991). Evidence for two uptake systems in Rhizobium leguminosarum for hydroxyaromatic compounds metabolized by the 3-oxoadipate pathway. Archives of Microbiology 156, 385–391. Woo, H.-H., Kuleck, G., Hirsch, A. M. and Hawes, M. C. (2002). Flavonoids: Signal molecules in plant development. In ‘‘Flavonoids in Cell Function’’ (B. Buslig and J. Manthey, eds.), pp. 51–60. Kluwer Academic/Plenum, New York. Yeh, K. C., Peck, M. C. and Long, S. R. (2002). Luteolin and GroESL modulate in vitro activity of NodD. Journal of Bacteriology 184, 525–530. Young, J. M., Kuykendall, L. D., Martinez-Romero, E., Kerr, A. and Sawada, H. (2001). A revision of Rhizobium Frank 1889, with an emended description of the genus, and the inclusion of all species of Agrobacterium Conn 1942 and Allorhizobium undicola de Lajudie et al. 1998 as new combinations: Rhizobium radiobacter, R. rhizogenes, R. rubi, R. undicola and R. vitis. International Journal of Systematic and Evolutionary Microbiology 51, 89–103. Young, J. M., Kuykendall, L. D., Martinez-Romero, E., Kerr, A. and Sawada, H. (2003). Classification and nomenclature of Agrobacterium and Rhizobium: A reply to Farrand et al. International Journal of Systematic and Evolutionary Microbiology 53, 1689–1695. Zaat, S. A. J., WijVelman, C. A., Spaink, H. P., van Brussel, A. A. N., Okker, R. J. H. and Lugtenberg, B. J. J. (1987). Induction of the nodA promoter of Rhizobium leguminosarum Sym plasmid Prl1ji by plant flavanones and flavones. Journal of Bacteriology 169, 198–204. Zaat, S. A. J., Schripsema, J., WijVelman, C. A., van Brussel, A. A. N. and Lugtenberg, B. J. J. (1989). Analysis of the major inducers of the Rhizobium nodA promoter from Vicia sativa root exudate and their activity with diVerent nodD genes. Plant Molecular Biology 13, 175–188. Zon, J. and Amrhein, N. (1992). Inhibitors of phenylalanine ammonia lyase-2aminoindan-2-phosphonic acid and related compounds. Liebigs Annalen Der Chemie 6, 625–628. Zuanazzi, J. A. S., Clergeot, P. H., Quirion, J. C., Husson, H. P., Kondorosi, A. and Ratet, P. (1998). Production of Sinorhizobium meliloti nod gene activator and repressor flavonoids from Medicago sativa roots. Molecular Plant-Microbe Interactions 11, 784–794.

Investigating and Manipulating Lignin Biosynthesis in the Postgenomic Era

CLAIRE HALPIN

University of Dundee, Plant Science Research Group, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom

I. II. III. IV.

V. VI.

VII. VIII. IX. X. XI.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lignification in Plant Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Structure of the Lignin Polymer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Monolignol Biosynthetic Pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. The Basic Pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Recent Revisions to the Pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. DiVerent Pathway Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monolignol Polymerization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cell Biology of Lignin Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Immunolocalization of Lignin Biosynthetic Enzymes . . . . . . . . . . . . . . B. Do Enzyme Complexes Promote Metabolite Channelling? . . . . . . . . . C. Export of the Monolignols to the Cell Wall . . . . . . . . . . . . . . . . . . . . . . . D. Alternative Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transcriptional Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Transcription Factors and Regulatory Elements . . . . . . . . . . . . . . . . . . . B. Transcript Profiling Using Microarrays . . . . . . . . . . . . . . . . . . . . . . . . . . . Metabolic Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis of Lignin Content, Structure, and Composition . . . . . . . . . . . . . . Genes Involved in Lignin Biosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Plants with Modified Expression of Lignin Biosynthetic Genes . . . . . . . . A. Phenylalanine Ammonia Lyase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Cinnamic Acid 4-Hydroxylase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. 4-Coumarate:CoA Ligase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Advances in Botanical Research, Vol. 41 Incorporating Advances in Plant Pathology 0065-2296/04 $35.00

64 65 67 68 68 68 69 70 71 71 72 73 74 76 76 77 78 79 81 81 82 83 83


64

C. HALPIN

D. Hydroxycinnamoyl-CoA:Shikimate/Quinate Hydroxycinnamoyl Transferase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. P-Coumarate 3-Hydroxylase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. CaVeoyl-CoA O-Methyltransferase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. CaVeic Acid O-Methyltransferase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Ferulate 5-Hydroxylase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Cinnamoyl-CoA Reductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. Cinnamyl Alchohol Dehydrogenase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . K. Manipulation of Multiple Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XII. Commercial Applications of Modified Lignin Plants. . . . . . . . . . . . . . . . . . . XIII. Conclusions and Perspectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

84 84 85 85 87 87 89 90 92 94 95

ABSTRACT Enormous progress has been made over the last decade in understanding and manipulating the pathway for lignin biosynthesis in plants. The roles of most of the genes on the pathway have now investigated by reverse genetic approaches in a variety of species, including trees, and field trails underpinning commercial exploitation have taken place. Despite this, many basic aspects of the lignin pathway are still very poorly understood. Little is known about the cell biology of the process, for example, and much has still to be learned about how the pathway is controlled and regulated at transcriptional and biochemical levels. The advent of post-genomic technologies such as transcript and metabolite profiling offer new opportunities for probing the lignin pathway and its inter-relationship with other pathways and developmental processes in plants. This review describes recent advances in understanding lignification while highlighting the areas where significant further work is needed.

I. INTRODUCTION Lignification is a fundamental developmental process unique to higher land plants. Over the past decade the biochemistry and molecular biology of lignification have been extensively studied, an eVort driven both by scientific and by commercial interests. Lignin is the second most abundant biopolymer on earth, constituting 20% of total organic carbon, and is thought to have played a critical role in the evolution of land plants by conferring structural rigidity to strengthening tissues (sclerenchyma fibers), and by waterproofing xylem vessels. Lignin has major influence on the ease of wood pulping during papermaking, on the digestibility of forage crops, and on the postharvest quality of certain vegetables. It is also critical for normal plant health, development, and disease resistance. Most of the genes involved in lignin biosynthesis have been cloned, and a significant amount of research eVort has been committed to determining the function of individual genes by reverse genetics. Genetic engineering of the

INVESTIGATING AND MANIPULATING LIGNIN BIOSYNTHESIS

65

pathway has been achieved in a variety of species including trees, and field trials to underpin commercial exploitation have already taken place. Despite, or perhaps because of, this headlong rush toward biotechnological applications, some basic aspects of lignification have largely been ignored. The spatial organization of the pathway, how it is regulated, and its role in the myriad complex interactions that result in normal plant development are all areas that have received relatively little research attention to date. Although the focus on the biotechnological applications of lignin manipulation in part explains the lack of progress in elucidating these more fundamental scientific questions, progress has also been hampered by the fact that such issues were diYcult to address with pregenomic technologies. The recent development of novel tools, particularly some of the technologies of functional genomics, is beginning to impact these areas of research. The application of techniques such as microarray analyses, metabolite profiling, and T-DNA mutagenesis oVer enormous potential for illuminating the full interactions of the lignin pathway with other plant metabolic processes and for elucidating the full consequences of manipulating lignin biosynthesis in transgenic crops. This chapter describes much of the recent work aimed at understanding lignification and highlights areas where a significant amount of further work is needed to give an overview of our current state of knowledge and ignorance of this important developmental process.

II. LIGNIFICATION IN PLANT DEVELOPMENT Lignin is deposited in the secondary wall of certain diVerentiating cell types. Although predominantly found in sclerenchyma fibers and in the tracheids and vessels of vascular tissue, lignin is also critical to the functioning of other, less obvious cell types. For example, loss of lignification in a subset of valve margin cells in fruits of the Arabidopsis shp1 shp2 mutant is implicated in the failure of the fruits to dehisce (Liljegren et al., 2000). Lignin can also be deposited in response to infection or certain abiotic stimuli, even in cell types that do not normally contain it. Nevertheless, lignin deposition during normal plant development is cell specific and tightly controlled both temporally and spatially. Several Arabidopsis mutants illustrate how alteration in this temporal and spatial control may result in premature lignification (Mele et al., 2003) and/or ‘‘inappropriate’’ lignification of pith parenchyma cells (Newman et al., 2004; Zhong et al., 2000a), endodermal cells (Cano-Delgado et al., 2000), or epidermal cells (Mele et al., 2003). The study of such ‘‘ectopic lignification’’ mutants is beginning to reveal the complex, interconnected networks that can influence and control lignin

66

C. HALPIN

deposition. The eli1 (ectopic lignification 1) mutants, for example, have shown reduced cellulose synthesis as a result of mutations in the cellulose synthase gene CESA3 (Cano-Delgado et al., 2003). This cellulose deficiency is proposed to activate lignin synthesis and defense responses through jasmonate and ethylene and other signalling pathways. The induction of lignification in tissues with reduced cellulose synthesis suggests that plant cells may have mechanisms to monitor, and perhaps to attempt to maintain, cell wall integrity. Ethylene signalling was also altered in elp1 (ectopic deposition of lignin in pith 1), a mutant in a chitinase-like gene, AtCTL1. Ethylene production in the mutant was significantly greater than in the wild-type, and certain mutant phenotypes, including exaggerated hook curvature and increased root hair formation, could be rescued by ethylene inhibitors, whereas other phenotypes such as reduced root and hypocotyl length could be partially rescued (Zhong et al., 2002). The ectopic lignification was, however, unaVected by ethylene inhibitors, indicating that although these studies reveal aspects of the signalling cascades that may regulate lignification, no simple conclusions can yet be drawn. Further complexity is added by a very recent report indicating that misexpression of a specific transcription factor, AtMYB61, can cause ectopic lignification in Arabidopsis (Newman et al., 2004). In the det3 (de-etiolated 3) mutant the spatial control of AtMYB61 expression is lost and this apparently induces both the ectopic lignification and dark photo-morphogenic phenotypes of this mutant. It is still unclear how these two pleomorphic phenotypes are connected and how AtMYB61 expression is influenced by the det3 locus, which encodes the C subunit of a vacuolar ATPase (adenosine triphosphatase). Given that many cellular processes may be disrupted in this mutant, it is possible that ethylene and jasmonic acid signalling could be altered, as in the case of eli1. Further investigation is needed to determine whether common or diVerent mechanisms and pathways are responsible for the ectopic lignification phenotypes of these various, apparently unrelated, mutants. The importance of lignin to maintaining the integrity of certain cell walls is graphically illustrated by the collapsed xylem vessels seen in mutant or transgenic plants with reduced expression of certain specific lignin biosynthetic genes (Franke et al., 2002a; Jones et al., 2001; Piquemal et al., 1998). Many of these plants have radically reduced lignin content and show a range of developmental abnormalities including stunted growth and, often, altered leaf morphology, highlighting the critical role lignin plays in normal plant development. However, the link between reduced lignin content and adverse phenotypes is not absolute (Chabannes et al., 2001a; O’Connell et al., 2002) and nonpolymeric products of the lignin pathway may also have as yet illdefined roles in modifying development. When phenolic metabolism and


67

lignin biosynthesis were repressed in tobacco by heterologous expression of a Myb gene from the Antirrhinum plant, growth was inhibited and leaves showed signs of premature cell death (Tamagnone et al., 1998a). A deficiency in dehydrodiconiferyl alcohol glucosides (DCGs; synthesized from the monolignol coniferyl alcohol) was implicated in causing the leaf phenotypes. DCGs are cytokinin-like growth promoters that have been implicated in the regulation of cell division and expansion (Binns et al., 1987; Tamagnone et al., 1998b; Teutonico et al., 1991). Levels of DCGs were greatly reduced in the Myboverexpressing plants, and suspension cultured cells from these plants regained a normal phenotype when fed the glucoside (Tamagnone et al., 1998b). Further work is needed therefore to resolve the relative roles of deficiencies in lignin and in soluble phenolics in causing the various phenotypes that indicate altered development in certain lignin-modified plants.

III. STRUCTURE OF THE LIGNIN POLYMER Lignin content and composition varies among plant taxa, species, cell types, and even from one part of the cell wall to another. Lignin is a polymer predominantly composed of para-coumaryl ( p-coumaryl) alcohol, coniferyl alcohol, and sinapyl alcohol, the three ‘‘monolignols’’ (Fig. 1). These units diVer only in their degree of methoxylation and give rise to three diVerent types of lignin; that is, p-hydroxyphenyl (H), guaiacyl (G), and syringyl (S) lignin, respectively. The monolignol composition of the polymer varies considerably among taxa, tissues, and cell types. In gymnosperms, lignin is composed almost entirely of guaiacyl units, with small amounts of p-hydroxyphenyl units. The proportion of H lignin is greatly increased in compression wood, which forms in regions under compressive stress, such as the underside of branches. Lignin in most dicot angiosperms is a combination of guaiacyl and syringyl monomers, present in varying proportions according to species. In the monocot

Fig. 1. alcohol.

The monolignols para-coumaryl alcohol, coniferyl alcohol, and sinapyl

68

C. HALPIN

grasses, lignin additionally contains a significant proportion of H units. A variety of chemical linkages including noncondensed -O-4 ether bonds and condensed carbon-carbon bonds (e.g., - , -1, -5, and 5-5) can connect the monomers in lignin. Gymnosperm lignin has relatively more condensed carbon-carbon bonds than angiosperm lignin because of the predominance of G units where the aromatic C5 position is free to make ring-to-ring linkages. These carbon-carbon bonds are resistant to chemical degradation, which explains why gymnosperm woods are harder to pulp than angiosperm woods using chemical pulping processes.

IV. THE MONOLIGNOL BIOSYNTHETIC PATHWAY A. THE BASIC PATHWAY

Many aspects of the lignin biosynthetic pathway have been comprehensively discussed in recent reviews (see Anterola and Lewis, 2002; Baucher et al., 2003; Boerjan et al., 2003; Humphreys and Chapple, 2002), so only a brief description of the pathway and related unresolved issues will be given in this chapter. Lignin is a product of the phenylpropanoid pathway. Early reactions on the pathway are catalyzed by phenylalanine ammonia lyase (PAL), cinnamate 4-hydroxylase (C4H) and 4-coumarate:CoA ligase (4CL). Subsequent hydroxylation and methylation reactions add one or two methoxyl groups to the phenyl ring. Four enzymes—cinnamate 3-hydroxylase (C3H), caVeoyl-CoA O-methyltransferase (CCoAOMT), ferulate 5-hydroxylase (F5H), and caVeate O-methyltransferase (COMT)—are involved in these reactions. Two successive reduction reactions catalyzed by cinnamoyl-CoA reductase (CCR) and cinnamyl alcohol dehydrogenase (CAD) are involved in the final steps of monolignol synthesis. All nine of the enzymes referred to so far have been cloned, and their involvement in lignin biosynthesis has been clearly demonstrated in vivo because transgenic or mutant plants deficient in any of these activities have significant changes to either lignin content or lignin monomeric composition. B. RECENT REVISIONS TO THE PATHWAY

The monolignol biosynthetic pathway has undergone many recent revisions and it is likely that the substrate(s) and position(s) in the pathway of some of the enzymes have still not been definitively determined. A major revision concerns the position in the angiosperm pathway where guaiacyl precursors can be converted, by the addition of a methoxyl group, into syringyl lignin precursors. This occurs in two reactions catalyzed by the enzymes F5H and COMT. According to the traditional lignin pathway, these reactions were


69

believed to occur at the level of the acids, converting ferulic acid into sinapic acid. However, recent work indicates these reactions really occur at the position of the aldehydes and possibly the alcohols (Chen et al., 1999; Humphreys et al., 1999; Osakabe et al., 1999). Although these pathway revisions were originally proposed on the basis of biochemical evidence, such as the substrate specificity of the enzymes assayed in vitro, they are entirely consistent with results previously obtained from plants manipulated or mutated to reduce COMT or F5H expression. Recent discoveries suggest the possible involvement of at least two additional enzymes in the production of the monolignols. The cloning of C3H and the identification of mutants defective in it (Franke et al., 2002a,b; Schoch et al., 2001) have indicated that p-coumaroyl shikimate and p-coumaroyl quinate are likely to be important intermediates in lignin biosynthesis. These two compounds and not p-coumarate as previously thought, are the preferred substrates for C3H (Schoch et al., 2001), which in turn suggests that a novel hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyltransferase (HCT), capable of conjugating the shikimate and quinate groups onto p-coumarate, must also be involved in lignin biosynthesis. Recently the gene encoding HCT has been cloned (HoVmann et al., 2003), and an important role for it in plant development and lignin biosynthesis has been verified using reverse genetic approaches (HoVmann et al., 2004). Similarly, a second novel gene, encoding a sinapyl alcohol dehydrogenase (SAD), has recently been discovered in poplar (Li et al., 2001a). This enzyme apparently preferentially converts sinapaldehyde into sinapyl alcohol, indicating a potential role in the lignin pathway. SAD expression also correlates with S lignin biosynthesis both spatially and temporally. On the basis of these data a new pathway for lignin biosynthesis has been proposed where CAD exclusively catalyzes the conversion of coniferaldehyde to coniferyl alcohol while SAD converts sinapaldehyde to sinapyl alcohol (Li et al., 2001a). The data indicating a potential role for SAD in lignin biosynthesis are convincing but nonetheless circumstantial. Further work, particularly the production of SAD mutants or SADsuppressed transgenics, is needed to validate and further investigate the role of SAD in vivo. C. DIFFERENT PATHWAY MODELS

The current uncertainty about the exact sequence of reactions that contribute to lignin biosynthesis is reflected in the recent literature, in which numerous apparently diVerent models of the pathway can be found. Many still view the pathway as the traditional ‘‘metabolic grid’’ or network where

70

C. HALPIN

certain routes may nevertheless be favored over others in diVerent tissues or cell types (see Anterola and Lewis, 2002; Boerjan et al., 2003). However, certain research groups favor models in which the main lignin pathway is a streamlined linear sequence of reactions leading to G lignin synthesis with a single branch point allowing for S lignin biosynthesis (Li et al., 2001a, 2003). Still others have suggested a metabolic channeling model that allows for essentially independent pathways for G and S lignin synthesis (Dixon et al., 2001). Although the distinction between these diVerent models is sometimes more apparent than real (most pathways allow for the potential existence of other minor routes), this situation must be very confusing for the nonexpert reader. Further research is needed to clarify this most basic aspect of lignin biosynthesis (i.e., to outline the full sequence of reactions on the main biosynthetic pathway in normal wild-type plants). The extent to which the pathway is the same in diVerent dicot angiosperms is also an area for further clarification because substrate specificities of some enzymes, notably 4CL, diVer among species.

V. MONOLIGNOL POLYMERIZATION Several diVerent possible mechanisms have been proposed for the polymerization of monolignols within the cell wall (reviewed by Boerjan et al., 2003; Lewis, 1999). Traditionally it has been believed that monolignols are converted into phenoxy radicals by cell wall oxygenases, although the identity of the oxygenases involved is still an issue of some debate. Peroxidases, laccases, and other phenoloxidases have all been proposed to be involved (Dean and Eriksson, 1994; McDougall et al., 1994; Richardson et al., 1997; Savidge and Udagama-Randeniya, 1992), but the multiplicity of such enzymes that exist in plants and the possibility of functional redundancy make the exact role of specific enzymes diYcult to investigate. Considerable controversy also exists as to whether the subsequent polymerization of the monolignols is a random, spontaneous phenomenon (Hatfield and Vermerris, 2001), or whether it is highly ordered (Gang et al., 1999) and perhaps mediated by dirigent proteins such as the one isolated from Forsythia, which is capable of catalyzing the stereoselective coupling of two coniferyl alcohol radicals into the lignan pinoresinol (Davin et al., 1997). This debate is not likely to be drawn to a rapid conclusion, particularly because good evidence from transgenic or mutant plants supporting the roles of particular cell wall enzymes is still lacking. New ideas also need to be incorporated into this discussion, such as the recent proposal that redox shuttle-mediated oxidation may be ¨ nnerud et al., 2002). involved (O


71

VI. CELL BIOLOGY OF LIGNIN BIOSYNTHESIS A. IMMUNOLOCALIZATION OF LIGNIN BIOSYNTHETIC ENZYMES

Considering the enormous amount of research expended on lignin biosynthesis in recent years, it is surprising how little is definitely known about the cell biology of the process. It is generally accepted that the monolignols are synthesized inside the cell, then exported to the cell wall where they are polymerized into lignin. However, a degree of uncertainty still surrounds the subcellular localization of the lignin biosynthetic enzymes. It was originally believed that the enzymes might reside within the endomembrane system because this made it easy to envisage monolignol export to the cell wall in membrane-bound vesicles. However, this assumption was brought into question once lignin biosynthetic genes began to be cloned. Examination of the sequences of many lignin biosynthetic genes (including PAL, CCoAOMT, COMT, 4CL, CCR, and SAD) reveals that most of these genes do not contain signal sequences or other recognizable subcellular targeting domains. The absence of such targeting signals strongly suggests that many lignin enzymes may be cytosolic. However, the issue of the subcellular location of these enzymes is given further complexity by repeated claims of enzyme:enzyme complexes, perhaps membrane-bound, that might act to promote metabolic channelling of phenylpropanoid intermediates (Dixon et al., 2001; Guo et al., 2002; Rasmussen and Dixon, 1999; Winkel-Shirley et al., 1999). One might expect that many uncertainties about the cell biology of lignification might be simply resolved by immunolocalization of the individual lignin biosynthetic enzymes. Although attempts have been made to immunolocalize certain enzymes in xylem cells, results from diVerent groups have been somewhat contradictory. PAL, controlling the entry point to the phenylpropanoid pathway, is the most extensively studied enzyme involved in lignin biosynthesis. A small number of diverse studies have looked at its subcellular localization, and the results suggest that it may be present in multiple locations. In lignifying bean xylem, PAL was found by immunocytochemistry to be mostly cytosolic but sometimes membrane associated and sometimes in vacuoles (Smith et al., 1994). In diVerentiating Zinnia tracheary elements, PAL was found to be dispersed between cytosol, the endomembrane system, and cell walls (Nakashima et al., 1997). Similar confusion surrounds the localization of CAD, the final enzyme of monolignol synthesis. Two independent immunolocalization experiments contradict each other. Nakashima et al. (1997) report that in Zinnia, CAD is found in the cytosol, in Golgi-derived vesicles, and in secondary cell walls (an identical localization to that was reported for PAL in the same paper). Samaj et al. (1998)

72

C. HALPIN

report that in poplar, CAD is found predominantly in the cytoplasm and is ‘‘randomly’’ associated with endoplasmic reticulum (ER) and Golgi vesicles but is not present in the cell wall. It is possible that some of these results are anomalous because of artifacts that arise because of the need to fix and section materials before visualization. For example, localization of CAD and PAL in cell walls could be a result of the spread of epitopes from the cytosol during the physical disruption of preparation and sectioning. By contrast, Takabe et al. (2001) report finding little evidence for PAL or CAD localization on membranes or in the cell wall in poplar, and concluding that these enzymes are cytosolic. A recent study immunolocalizing CAD, COMT, and CCR in maize and sugarcane also concludes that all three enzymes are mainly cytosolic (Ruelland et al., 2003). Similarly, other researchers have indicated a cytosolic location for CCoAOMT and/or COMT in alfalfa, poplar, and eucalyptus (Kersey et al., 1999; Takeuchi et al., 2001). Despite the uncertainly about the subcellular localization of many lignin biosynthetic enzymes, at least three are clearly localized on membranes. C4H, like other plant Cyt P450-dependent monoxygenases (P450s), is a membrane protein with its catalytic site in the cytoplasm. It has been purified from microsomes (Gabriac et al., 1991) and was shown by immunolocalization to be abundant in Golgi in French bean (Smith et al., 1994). However, more recent work expressing a C4H:GFP fusion in Arabidopsis suggests that C4H is exclusively localized to the ER (Ro et al., 2001). Although the other two P450 enzymes, C3H and F5H, that act downstream of C4H on the lignin pathway, are certain to be membrane proteins, their exact location within cells has not been formally determined. F5H activity has been demonstrated in microsome membranes (Grand et al., 1984). B. DO ENZYME COMPLEXES PROMOTE METABOLITE CHANNELLING?

Although many lignin biosynthetic enzymes are probably cytosolic, a peripheral association with membranes, mediated by interaction with membrane proteins, is possible. Assembly of complexes of lignin enzymes on membranes is an attractive idea, because this could facilitate the metabolic channelling that some research groups have claimed for the pathway. It is also easier to envisage how the monolignols might get sequestered into export vesicles if the enzymes producing those monolignols are membrane associated. A possible association of PAL and C4H on microsome membranes has been suggested to explain metabolic channelling from Phe to p-coumarate (Czichi and Kindl, 1977; Rasmussen and Dixon, 1999; Wagner and Hrazdina, 1984). However, no data directly supports interactions between lignin enzymes, and many of the enzymes, including CAD, can


73

be purified to homogeneity from aqueous cell extracts without the use of detergents or reagents that disrupt protein:protein interactions. Channelling between F5H and COMT has also been repeatedly proposed. Although a small proportion of total COMT activity could be found together with F5H activity in crude microsomal preparations from alfalfa, no improved catalytic eYciency in the production of sinapaldehyde could be demonstrated (Guo et al., 2002). The authors concluded that their results provided only limited support for the concept of metabolic channelling in the biosynthesis of S monolignol precursors. C. EXPORT OF THE MONOLIGNOLS TO THE CELL WALL

It is assumed by many that the monolignols are exported as glucosides, although no direct evidence supports this. The cambial sap of conifers contains high concentrations of the monolignol glucoside coniferin (see Whetten and SederoV, 1995), which has been assumed to be a stored source of monomers for subsequent lignification of developing tracheids. Some angiosperms also store monolignol glucosides, but most do not. A large multigene family of uridine diphospho (UDP)-glucosyltransferases (UDPGs) has been identified in Arabidopsis (Li et al., 2001b), but so far only circumstantial evidence based on substrate specificities of the recombinant enzymes suggests which gene(s) could be involved in monolignol modification (Lim et al., 2001). Two of the recombinant UDPGs were able to 4-Oglucosylate sinapyl alcohol into syringin, and one of these could also glucosylate coniferyl alcohol to coniferin in vitro (Lim et al., 2001). However, a direct role for such reactions during lignin biosynthesis remains to be proven. Traditionally, it has been assumed that the monolignols are exported in membrane-bound vesicles, because vesicle-like structures filled with ultraviolet (UV)-fluorescent material are abundant in lignifying cells. However, cell wall phenolics such as ferulic acid are also highly UV fluorescent and no direct evidence confirms that these vesicles actually transport monolignols. More recently, high-resolution transmission electron microscopy of pine xylem has confirmed that developing tracheids actively making secondary walls have a highly developed trans-Golgi network with unusual structures and large associated vesicles (Samuels et al., 2002). These vesicles were apparently involved in exporting cell wall polysaccharides such as mannans to the cell wall, but they were also stained with osmium, indicating that they might also transport phenolics. A rarely considered alternative to the hypothesis of vesicular transport of monolignols is the possibility that the monomers might be able to cross the

74

C. HALPIN

plasma membrane directly, either as glucosides or aglycones. Either way, it is not known whether the monolignols can cross the relevant membrane (either vesicle or plasma membrane) by simple diVusion or whether the action of a specific transporter is needed. If monolignols are indeed transported as glucosides, the activity of a cell wall glucosidase is needed to remove the sugar residue before polymerization. A coniferin -glucosidase capable of deglucosylating one of the monolignols has been cloned from pine (Dharmawardhana et al., 1999). Immunolocalization of this enzyme indicates that it is extracellular, residing in the secondary cell walls of developing xylem, a location compatible with its proposed role in deglycosylating lignin precursors before polymerization (Samuels et al., 2002). Reverse genetic approaches now need to be applied to more directly investigate whether lignin precursors are transported as glucosides and to what extent coniferin -glucosidase is needed for lignin biosynthesis. Given that such approaches would be technically diYcult to implement in pine, the identification of coniferin -glucosidase orthologues in more easily manipulated species such as Arabidopsis or tobacco should be a priority, but no information on this yet exists in the literature. D. ALTERNATIVE MODELS

A number of diVerent potential models can therefore be proposed to explain the cell biology of lignin biosynthesis. The simplest model is based purely on the fact that no direct evidence supports a more complicated scheme. According to this model, the lignin pathway operates by free diVusion of substrates between enzymes located in the cytoplasm and the P450 enzymes located on a membrane. Monolignol products either diVuse through the plasma membrane or are transported across via a specific transporter, either as free alcohols or as glucosides (Fig. 2A). A more complicated model takes into account the possibility of metabolite channelling on the pathway. According to this model, lignin biosynthetic enzymes located in the cytosol could be sequestered onto membranes by the membraneintegrated P450 enzymes (C4H, C3H, and F5H). The membrane involved might be the ER, Golgi, or a post-Golgi compartment. After synthesis, the monolignols would be glucosylated by a UDPG interacting with the membrane-associated complex of lignin enzymes (Plant UDPGs are assumed to be cytosolic enzymes.). The monolignol glucosides would then be transported in vesicles to the plasma membrane for export to the cell wall (Fig. 2B). The currently available evidence does not allow us to distinguish even between these two extreme models, and it is likely that in reality the


75

Fig. 2. Two possible models of lignin biosynthesis. (A) Simple model—the lignin pathway operates by free diVusion of substrates between enzymes located in the cytoplasm, and the P450 enzymes (C4H, C3H, and F5H) located on a membrane and monolignol products diVuse through the plasma membrane. (B) More complicated model—lignin biosynthetic enzymes located in the cytosol are sequestered onto membranes by the membrane-integrated P450 enzymes. Monolignols are glucosylated by a UDP-glucosyltransferase, cross into vesicles (mediated by a specific transporter D), and are transported to the plasma membrane for export to the cell wall. Current knowledge does not allow us to distinguish between these two extreme models and it is likely that, in reality, elements of both models are involved.

route(s) of synthesis and transport of monolignols may involve elements of both models. To date, no conclusive model of the spatial organization of the lignin pathway exists and many aspects of the cell biology of the process have been largely neglected. The use of sophisticated modern techniques, including live-imaging microscopy of fluorescently labelled lignin biosynthetic enzymes, should enhance this area of research, but only one study so far has taken this approach (Ro et al., 2001). It is remarkable that more work has not been performed to reveal the cell biology underpinning this major export pathway of many plant cells. Knowledge of the spatial organization of the lignin pathway at the subcellular level will also ultimately underpin biotechnological advances and may lead to new ideas on how the processes of lignification and secondary cell wall deposition can be manipulated to useful ends.

76

C. HALPIN

VII. TRANSCRIPTIONAL REGULATION A. TRANSCRIPTION FACTORS AND REGULATORY ELEMENTS

Few regulatory or transcription factor genes have been identified that can promote or inhibit expression of lignin biosynthetic enzymes. Results from the Arabidopsis genome initiative suggest that plants have a particularly large complement of transcription factor genes and only a small fraction have yet been characterized. Now that information on the full number of Arabidopsis transcription factors is available, rapid progress can be expected in this area. Quite a bit is already known about transcriptional regulation of the general phenylpropanoid pathway, which represents the early steps in lignin biosynthesis (Leyva et al., 1992; Sablowski et al., 1994), but little is understood about the regulation of the later steps more dedicated to monolignol production. A growing body of evidence suggests that Myb proteins are involved (Jin et al., 2000; Martin and Paz-Ares, 1997; Tamagnone et al., 1998a). Consistent with this premise, conserved motifs (AC elements or PAL-boxes), which are commonly recognition sites for Myb binding, are found in the promoters of some lignin biosynthesis genes, including PAL (Lois et al., 1989), C4H (Bell-Lelong et al., 1997), 4CL (HauVe et al., 1993), CCoAOMT (Chen et al., 2000), CCR (Lacombe et al., 2000), and CAD (Lauvergeat et al., 2002), although not all have been verified by experimental functional analyses. In some cases, AC elements have been shown to reside within promoter regions that are essential for expression in vascular tissues, particularly xylem, suggesting that these elements, in consort with other promoter sequences, may be important in directing tissue-specific expression (Hatton et al., 1995; Lacombe et al., 2000; Lauvergeat et al., 2002). This idea is supported by a recent genome-wide study of lignification genes in Arabidopsis, which described the presence of AC elements in the promoters of all but one (C4H) of the genes needed for the biosynthesis of G lignin during vascular development (Raes et al., 2003). Other conserved elements such as H-boxes or G-boxes are found in certain lignin biosynthetic genes and are thought to be associated with stress and defense responsiveness. Some lignin biosynthesis genes are induced by wounding, and the cis-elements responsible for wounding responsiveness have been localized within their promoters (see Lauvergeat et al., 2002). Certain lignin biosynthetic genes have also been shown to be circadianclock controlled in Arabidopsis (Harmer et al., 2000), and their expression can be induced by a whole range of other triggers, but the regulatory mechanisms underlying these responses remain largely undetermined. Some progress is beginning to be made to identify specific transcription factors or other regulatory proteins involved in modulating the


77

expression of lignin biosynthesis genes. A Myb protein from Pinus taeda, PtMYB4, has recently been shown to bind to AC elements and to induce expression of certain lignin biosynthesis genes when overexpressed in tobacco (PatzlaV et al., 2003). Overexpression of another Myb protein, Arabidopsis PAP1, can activate all areas of phenylpropanoid metabolism, including lignin biosynthesis (Borevitz et al., 2000). A PAL-box binding protein, Ntlim1, with a zinc finger motif and similarity to members of the LIM protein family, has been isolated from tobacco (Kawaoka and Ebinuma, 2001). Transgenic tobacco containing an antisense Ntlim1 gene showed a 20% decrease in lignin content and reduced activity of PAL, 4CL, and CAD (Kawaoka and Ebinuma, 2001). Arabidopsis mutants in the BREVIPEDICELLUS (BP) gene, a KNOX gene primarily involved in internode patterning, showed increased and aberrant lignin deposition and the BP gene was shown to bind to promoters of some lignin pathway genes (Mele et al., 2003). Significant future work is necessary to delineate and characterize the diVerent, probably overlapping, regulatory elements and the corresponding proteins that bind to them, which are required for the control of lignin gene expression spatially, temporally, and in response to biotic and abiotic stresses.

B. TRANSCRIPT PROFILING USING MICROARRAYS

Transcript profiling of lignifying cells or tissues is beginning to add a new dimension to this area of research. Global gene expression analysis has the potential to reveal these networks of coordinated regulation and to suggest, perhaps, unexpected interactions among metabolic pathways and physiological processes. Similarly, comparison of coregulated lignin genes may reveal new common regulatory elements, an approach that has already proved useful in other systems. Transcript profiling in plants is just beginning, but even microarrays that cover only a small portion of a genome can yield useful data. A hybrid aspen xylem microarray containing 2995 unique expressed sequence tags (ESTs) has recently been produced and used to analyze tissue-specific transcript profiles from distinct development zones within the wood-forming tissues of wild-type (Hertzberg et al., 2001) and transgenic (Israelsson et al., 2003) plants. This analysis revealed that the genes for lignin biosynthetic enzymes and transcription factors and other potential regulators of xylogenesis are under strict transcriptional regulation, being expressed only in tissues at a particular developmental stage within the xylem (Hertzberg et al., 2001). Among other results this work indicated that a homologue for the dirigent protein gene was induced

78

C. HALPIN

coincident with lignification, and that two peroxidases (not previously linked to lignification) and two laccases were expressed in lignifying cells, providing supportive evidence for a role for these genes in lignification. Transcript profiling of all known genes involved in monolignol production in conifers performed on Pinus taeda cell suspension cells indicated that PAL, C4H, C3H, 4CL, CCoAOMT, CCR, and CAD were coordinately up-regulated on transfer of cells to a medium containing 8% sucrose and 20 mM potassium iodide, whereas a putative acid/ester O-methyltransferase that had been suggested to be involved in lignin biosynthesis was not up-regulated, which suggests that it is not, in fact, involved in monolignol production (Anterola et al., 2002).

VIII. METABOLIC REGULATION In the aforementioned Pinus taeda study the eVects of increasing Phe supply on lignin gene transcript levels was also addressed. By increasing exogenously supplied Phe to saturating levels, transcript levels of PAL, 4CL, CCoAOMT, CCR, and CAD increased, whereas those for C4H and C3H were only slightly up-regulated. These data, along with metabolic profiling data, were interpreted in the light of preexisting literature on genetically manipulated plants to suggest that carbon allocation to the monolignol pathway and its distribution toward the synthesis of the diVerent monolignols in conifers is controlled by Phe supply and by diVerential modulation of C4H and C3H (Anterola et al., 2002). This work highlights the complex interrelationships that must exist both within and between plant metabolic pathways at the biochemical level. Metabolites are not just the products of these pathways but often also act as sensitive sensors and regulators so that changes in metabolite pattern can sometimes lead to complex changes in partitioning of flux within metabolic networks (Li et al., 2000). Although research into this area is just beginning, it is clear that metabolic regulation plays important roles on the phenylpropanoid pathway at various levels. Certain phenylpropanoid intermediates are known to have feedforward (Loake et al., 1992) or feedback (Blount et al., 2000) regulatory properties acting at the level of enzyme activity (Bolwell et al., 1986) or gene transcription (Loake et al., 1992; Mavandad et al., 1990). For example, PAL activity is reduced in transgenic tobacco deficient in C4H, presumably due to feedback modulation (Blount et al., 2000), because cinnamic acid has been shown to inhibit PAL expression at the transcriptional level (Blount et al., 2000; Mavandad et al., 1990). Phenylpropanoid intermediates can also modulate pathway reactions at the biochemical level. For


79

example, 5-hydroxyconiferaldehyde, the preferred substrate for COMT, has been shown to inhibit the possible competing reactions of caVeate and 5-hydroxyferulate methylation (Li et al., 2000). Similarly, in mixed reactions in vitro containing both coniferaldehyde and ferulate, coniferaldehyde acted as a noncompetitive inhibitor of 5FH, suggesting that, in vivo, in the presence of coniferaldehyde, syringyl monolignols are only synthesized from coniferaldehyde (Osakabe et al., 1999). Although plant metabolite profiling is in its infancy, some researchers have already developed metabolite profiling protocols that can quantify lignin pathway intermediates and plant phenolic glucosides from xylem extracts of wild-type and transgenic plants (Meyermans et al., 2000). Analysis of xylem from CCoAOMT–down-regulated poplars indicated that while lignin content decreased, the amount of certain soluble phenolics, notably glucosides of phenolic acids, increased. In particular, O4- -D-glucopyranosyl-sinapic acid, a storage or detoxification product of sinapic acid, accumulated to 10% of soluble phenolics, allowing Meyermans et al. (2000) to infer that endogenously produced sinapic acid is not a major precursor in syringyl lignin biosynthesis in poplar. By contrast, Chen et al. (2003) were unable to detect diVerences in monolignol pathway intermediates in the soluble extracts of stems or leaves from COMT– or CCoAOMT–down-regulated alfalfa, although caVeoyl glucoside apparently accumulated in CCoAOMT– down-regulated stems (Chen et al., 2003; Guo et al., 2001a). Further development and application of metabolic profiling techniques will undoubtedly provide a productive area of research into the monlignol biosynthetic pathway in the future, potentially clarifying the accuracy or otherwise of the diVerent current models of the pathway, and illuminating the ways that the monolignol pathway interacts with other branches of phenylpropanoid metabolism.

IX. ANALYSIS OF LIGNIN CONTENT, STRUCTURE, AND COMPOSITION Determining the content, structure, and composition of lignin in plant materials is a challenging problem because of the complexity and heterogeneity of the polymer and its cross-links to other wall components. Indeed, it is not currently possible to fully elucidate lignin structure and composition in any species. Most techniques for lignin characterization rely on initial solubilization of a lignin fraction under relatively harsh chemical conditions. It is important to realize that the resulting soluble fraction does not represent native lignin either quantitatively or qualitatively. In particular, condensed

80

C. HALPIN

lignin fractions resistant to degradation are absent from such preparations and solubilized fractions are often modified (by condensation, oxidation, addition, or substitution) during extraction. Lignin analysis techniques also vary considerably in sensitivity, and minor lignin components that can be detected by one technique may appear absent using a diVerent method. These problems mean that no single technique is completely reliable or unbiased and that it is usually necessary to use several methods to get representative data. This is particularly important when dealing with ‘‘novel’’ lignins from genetically manipulated plants in which apparent discrepancies in the literature can often be attributed to the distinct techniques used by diVerent laboratories. A detailed description of the various methods available for lignin determination is outside the scope of this review and therefore only a brief summary will be given. A more complete explanation of the methods and their limitations can be found elsewhere (see Anterola and Lewis, 2002; Baucher et al., 2003). Lignin is most frequently quantified by gravimetric techniques after extracting the other, more soluble, wall components (Klason lignin determination) or, alternatively, by extracting the lignin component itself (e.g., with thioglycolic acid or acetyl bromide). Lignin structure and composition can be determined by analysis of its degradation products derived by pyrolysis gas chromatography-mass spectrometry (pyrolysis GC-MS), alkaline nitrobenzene oxidation, thioacidolysis, or derivatization followed by reductive cleavage (DFRC). The targets of these techniques are predominantly the -O-4 linkages, the most abundant linkages in ‘‘normal’’ lignin. However, the proportion of -O–4 linked units may be dramatically reduced in certain genetically modified plants, where the degree of condensation of lignin tends to increase. Two-dimensional nuclear magnetic resonance (NMR) spectroscopy of isolated lignin fractions has allowed an improved picture of lignin structure to be obtained from particular genetically modified plants (Ralph et al., 1998, 2001). Methods that allow the whole of the polymer to be analyzed in situ, such as solid-state NMR or FT-IR (Fouriertransform infrared spectroscopy), although generally not as sensitive as ‘‘wet’’ chemical methods, can also be useful when dealing with novel lignins that are hard to extract. Recently a method has been described that is claimed to fully dissolve finely ground cell wall material in such a way that lignin structures remain intact, allowing the whole lignin fraction to be analyzed by high-resolution solution-state NMR methods for the first time (Lu and Ralph, 2003). This method oVers great potential for more complete determination of lignin structures and deserves further testing on a range of transgenic and mutant plants to determine the extent of its versatility and usefulness.


81

X. GENES INVOLVED IN LIGNIN BIOSYNTHESIS Most of the genes involved in monolignol production have been cloned in a variety of species. In addition, the recent whole genome sequencing of the model plant Arabidopsis has enabled the assembly of complete inventories of the genes potentially involved in the monolignol biosynthetic pathway in this species (Costa et al., 2003; Goujon et al., 2003b; Raes et al., 2003). Comparison of 59,797 ESTs from wood-forming tissues of loblolly pine with the genome sequence of Arabidopsis suggests a substantial conservation of gene sequences between this woody gymnosperm and the herbaceous angiosperm (Kirst et al., 2003). For contigs of 1 kb or more of high-quality sequence, more than 90% had an apparent Arabidopsis homologue. These results suggest that a common set of genes for woodiness may exist in all seed plants, supporting the use of Arabidopsis for comparative genomics of other angiosperms and even gymnosperms. Analysis of sequences for putative monolignol biosynthetic genes in Arabidopsis show that most are encoded by multigene families, although evidence of a real involvement in lignification is available for only one or a few genes in each family. In contrast to this general redundancy of sequences, a couple of genes involved in lignin synthesis (e.g., C4H, HCT) are unique in Arabidopsis or have only one homologue (e.g., F5H) (Raes et al., 2003). There is also evidence that some lignin biosynthetic genes (e.g., PAL, 4CL, CAD) have been duplicated during Arabidopsis evolution (Goujon et al., 2003b).

XI. PLANTS WITH MODIFIED EXPRESSION OF LIGNIN BIOSYNTHETIC GENES A significant amount of research eVort has gone into determining the exact roles of diVerent genes in lignin biosynthesis by modifying their expression in transgenic plants. A variety of reverse genetic approaches have been used to either suppress gene activity with antisense RNA/cosuppressing transgenes, or to overexpress genes. This work has concentrated on herbaceous and woody species that are easy to transform, such as tobacco and poplar. Mutant plants defective in particular lignin biosynthetic enzymes have also been identified in maize, sorghum, loblolly pine, and Arabidopsis. With the advent of modern techniques allowing for saturation mutagenesis in Arabidopsis, it is to be expected that mutants for all lignin biosynthesis genes may soon be identified in this species. Such mutants often have the advantage of being ‘‘knockouts,’’ in which the function of a gene is completely disrupted. This is in contrast to the situation in transgenics made by gene suppression technologies, in which a certain amount of expression is still possible from

82

C. HALPIN

the target gene. On the other hand, the significant possibility for functional redundancy within the multigene families that exist for many lignin genes in Arabidopsis means that knockout mutants in single genes may have no phenotype. Gene suppression technologies have the advantage here in potentially being able to target several genes within a gene family provided they have suYciently high homology to each other. Both knockout mutants and transgenics are therefore likely to be useful in elucidating the potential roles of the less well-studied genes likely to be involved in lignin synthesis (i.e., glucosyltransferases, -glucosidases, dirigent proteins) in the same way that they have been fundamental in promoting our current understanding of the roles of the main enzymes involved in monolignol production. The following section summarizes work to date for each enzyme. A. PHENYLALANINE AMMONIA LYASE

Phenylalanine ammonia lyase catalyzes the first step of the general phenylpropanoid pathway, a step that is common to the production of many metabolites including flavonoids, coumarins, and phytoalexins, not just lignin. PAL expression has been significantly suppressed in tobacco and results in a range of phenotypes including reduced growth, altered leaf shape, reduced pollen viability (Elkind et al., 1990), and increased susceptibility to the fungal pathogen Cercospora nicotianae (Maher et al., 1994). Plants with low PAL activity have thinner cell walls in the secondary xylem (Bate et al., 1994; Elkind et al., 1990) and reduced lignin content. In particular, the incorporation of G units into the noncondensed fraction of lignin is reduced and, consequently, S:G increases (Korth et al., 2001; Sewalt et al., 1997a). PAL overexpression results in a small increase in Klason lignin and a decrease in the amount of S units, yielding a twofold reduction in the S:G ratio when lignin was analyzed by thioacidolysis (Korth et al., 2001). The level of chlorogenic acid (3-caVeoylquinic acid) has been correlated with PAL activity in leaves and stems of both PAL-silenced and PAL-overexpressing tobacco (Bate et al., 1994; Blount et al., 2000; Elkind et al., 1990; Howles et al., 1996; Korth et al., 2001; Maher et al., 1994). PAL suppression decreases the chlorogenic acid content of tobacco leaves (Bate et al., 1994; Blount et al., 2000), whereas PAL overexpression increases it (Bate et al., 1994; Howles et al., 1996). No Arabidopsis PAL mutants have yet been described in the literature. Arabidopsis has four PAL genes and all are expressed in the inflorescence stem, a tissue with a high proportion of lignifying cells. However, the presence of an AC element in the promoters of the PAL1 and PAL2 genes suggests that these genes are the most likely candidates to be involved in monolignol synthesis in lignifying vascular cells (Raes et al., 2003). It has been suggested that PAL1 and PAL2 may be functionally redundant and


83

that a mutation in just one of these genes may not be enough to induce a phenotype (Goujon et al., 2003b). B. CINNAMIC ACID 4-HYDROXYLASE

Like PAL, the enzyme before it on the general phenylpropanoid pathway, C4H, is necessary to the production of many plant metabolites including lignin, and its suppression in transgenic tobacco significantly reduces lignin content. However, in contrast to PAL, C4H suppression particularly aVects the incorporation of S units, promoting a corresponding decrease in the S:G ratio (Sewalt et al., 1997a). Thus C4H and PAL deficiencies cause opposing changes to lignin monomeric composition, a finding that is hard to reconcile with the sequential positioning of the two enzymes on current lignin biosynthesis pathways. A variety of explanations are possible but suggest either that C4H has a diVerent role and position on the pathway than that currently assumed, or that metabolic channelling, mediated by C4H and other enzymes in complex, can operate on the lignin pathway to direct precursors toward S lignin biosynthesis (Dixon et al., 2001). Arabidopsis contains a single C4H gene and ref3 plants (reduced epidermal fluorescence; Ruegger et al., 1999) have a mutation in this gene (Franke et al., 2002b). It is not clear whether these plants are null mutants, and it is possible that they possess some residual C4H activity. Reduction of C4H activity in this mutant has similar eVects to C4H suppression in tobacco (i.e., the plants have decreased lignin content with altered composition), being particularly deficient in S units. The mutant also has developmental abnormalities and is stunted with increased branching and is male sterile. C. 4-COUMARATE:COA LIGASE

Transgenic plants with reduced 4CL activity have been produced in tobacco (Kajita et al., 1996, 1997), Arabidopsis (Lee et al., 1997), and aspen (Hu et al., 1999; Li et al., 2003). In all cases, plants with significant deficiencies in 4CL had lignin content reduced by 25–50%. By combining data from all of these transgenics, Anterola and Lewis (2002) estimate that 4CL activity has to be suppressed by more than 60% before the deficiency has a significant impact on lignin content. Despite consistent results in terms of lignin content reduction, these diVerent transgenics appear to display inconsistent changes to lignin composition, with S units apparently predominantly reduced in tobacco, only G units reduced in Arabidopsis, and no change to the relative proportions of S and G units in aspen (Hu et al., 1999; Kajita et al., 1996, 1997; Lee et al., 1997). These diVerential eVects may result from suppression of distinct 4CL isoforms by the diVerent transgenes used in each experiment or from the diVerent substrate specificities of certain 4CL isoforms in diVerent species. Equally,

84

C. HALPIN

these discrepancies could also simply reflect the diYculties in directly comparing results obtained from diVerent plants grown under varying conditions that were analyzed using diVerent lignin analytical techniques, each with their own general inadequacies and limitations (for discussion, see Anterola and Lewis, 2002). In tobacco and poplar, 4CL deficiency was associated with a higher amount of cell-wall–bound hydroxycinnamic acids ( p-coumaric, ferulic, and sinapic acids). Tobacco plants with the greatest lignin reductions were stunted and had collapsed xylem vessels (Kajita et al., 1997). However, in aspen, 4CL suppression has been reported to enhance growth (Hu et al., 1999), although the growth eVects were apparently not reproduced in a subsequent experiment in which the transgene was expressed from a diVerent promoter (Li et al., 2003). Increased cellulose contents were also reported for 4CL-suppressed aspen in both studies. These results may suggest that reduced carbon flow toward lignin synthesis increases the availability of carbon for cellulose synthesis, although it has been suggested that much of the apparent cellulose increase may be accounted for by the proportional increase expected in the cellulose component of whole tissues when the lignin component is reduced (see Anterola and Lewis, 2002 for discussion). No 4CL mutants have been identified in Arabidopsis, probably because of functional redundancy between the 4CL1 and 4CL2 genes (Goujon et al., 2003b). Of the four 4CL genes in Arabidopsis, these two (i.e., 4CL1 and 4CL2) are thought to be the best candidates for monolignol synthesis (Ehlting et al., 1999), and AC elements are present in their promoters (Raes et al., 2003). D. HYDROXYCINNAMOYL-COA:SHIKIMATE/QUINATE HYDROXYCINNAMOYL TRANSFERASE

HCT has only recently been cloned from tobacco (HoVmann et al., 2003) and a single report indicates the consequences of HCT deficiency in plants. HoVmann et al. (2004) describe HCT-silenced Nicotiana benthamiana which have decreased lignin syringyl units and increased p-hydrophenyl units, confirming the function of the acyltransferase in phenylpropanoid biosynthesis. E. P-COUMARATE 3-HYDROXYLASE

Because of its very recent isolation, no data exists on the eVects of C3H suppression in tobacco or woody plants. However, by screening Arabidopsis plants for reduced epidermal fluorescence, the ref 8 mutant was identified and subsequently shown to be defective in the CYP98A3 gene, the single Arabidopsis C3H gene (Franke et al., 2002a,b). Lignin content in this mutant is reduced by 60–80% and is almost entirely composed of p-coumaryl alcohol (H) units with large amounts of esterified p-coumaric acid. The mutant


85

accumulates a range of soluble p-coumarate esters instead of sinapoyl malate. Ref8 plants are severely dwarfed and have collapsed xylem vessels, increased cell wall degradability, and a higher susceptibility to fungal colonization (Franke et al., 2002a). These data indicate that C3H is a major control point in the production of C3- and C5-substituted phenylpropanoid lignin precursors. F. CAFFEOYL-COA O-METHYLTRANSFERASE

Suppression of CCoAOMT reduces lignin content by 12–50% in transgenic tobacco (Pinc,on et al., 2001a; Zhong et al., 1998), alfalfa (Guo et al., 2001a; Marita et al., 2003a), and poplar (Meyermans et al., 2000; Zhong et al., 2000b). In all studies the amount of G units was reduced, consistent with the proposed role for CCoAOMT in G lignin synthesis. In some studies, S units were also reduced, although small increases in the S:G ratio reflected a predominant influence on G units (Meyermans et al., 2000; Zhong et al., 1998). In alfalfa, and in one tobacco study, the amount of S units was not reduced (Guo et al., 2001a; Pinc,on et al., 2001a). In poplar the lignin produced by CCoAOMT-suppressed plants has been shown to be less cross-linked than normal (Zhong et al., 2000b). Vessel cell walls also showed enhanced fluorescence, possibly a result of the increased levels of free and bound p-hydroxybenzoic acid that were detected. Similarly, increased amounts of methanol-extractable phenolics including the O-b-D-glucosides of caVeic acid, sinapic acid, and vanillic acid, were detected in the wood of the transgenic poplars (Meyermans et al., 2000), whereas soluble caVeoyl glucoside accumulated in stem extracts of transgenic alfalfa (Guo et al., 2001a). These glucosides may result from a detoxification of accumulating hydroxycinnamic acids, as indicated by feeding experiments with caVeic and sinapic acids (Meyermans et al., 2000). The accumulation of O-b-Dglucopyranosyl–sinapic acid in plants with reduced S lignin supports the hypothesis that sinapic acid is not the main precursor for S units in vivo. Of all the CCoAOMT plants produced, only tobacco plants from one experiment showed an obvious phenotype with altered growth and flower development (Pinc,on et al., 2001a). Arabidopsis mutants in CCoAOMT genes have not yet been identified. G. CAFFEIC ACID O-METHYLTRANSFERASE

COMT is one of the most well-studied lignin genes and data on the results of its suppression exist for tobacco, poplar, alfalfa, and maize transgenics (Atanassova et al., 1995; Dwivedi et al., 1994; Guo et al., 2001a; Jouanin et al., 2000; Ni et al., 1994; Tsai et al., 1998; Van Doorsselaere et al., 1995)

86

C. HALPIN

and in maize, sorghum, and Arabidopsis mutants (Bout and Vermerris, 2003; Goujon et al., 2003c; Suzuki et al., 1997; Vignols et al., 1995). The predominant and consistent eVect of COMT suppression in all these species is a drastic reduction in the lignin S:G ratio resulting from a reduction in S unit amounts. An unusual monomer, the 5-hydroxyguaiacyl (5OHG) unit is also present in the polymer in both transgenic (Atanassova et al., 1995; Guo et al., 2001a; Jouanin et al., 2000; Lapierre et al., 1999; Marita et al., 2003a; Tsai et al., 1998; Van Doorsselaere et al., 1995) and mutant plants (Chabbert et al., 1994; Goujon et al., 2003c; Suzuki et al., 1997). In some plants the level of 5OHG units even exceeded that of S units (Jouanin et al., 2000), whereas in the Arabidopsis knockout mutant, no S units could be detected and the plants had increased levels of 5OHG and G units. These data are therefore consistent with a predominant and essential role for COMT in the production of S lignin units. Similarly, in COMT-suppressed alfalfa the reduction in both G and S units (Guo et al., 2001a; Marita et al., 2003a) is consistent with the work of Parvathi et al. (2001), who showed that in this species COMT is also involved in the methylation of caVeyl aldehyde. Severe suppression of COMT may sometimes moderately reduce total lignin content in most species. Reports on this are not consistent, however, and sometimes, similar plants are reported to have (Jouanin et al., 2000; Ni et al., 1994), or to have no (Atanassova et al., 1995; Dwivedi et al., 1994; Van Doorsselaere et al., 1995), reduced lignin, or the apparent reduction in lignin amount is dependent on the technique used to analyze the polymer (Guo et al., 2001a; Marita et al., 2003a). These data likely highlight the caution that has to be exercised when comparing lignin data from plants grown for diVerent lengths of time under varying environmental conditions and analyzed using diVerent techniques (Baucher et al., 2003). The changes in composition in lignin of COMT-suppressed plants have consequent eVects on lignin structure, which is also greatly altered. In poplar the proportion of -O-4 linkages appear to be reduced, while condensed C-C linkages are increased. Biphenyl (5–5) and phenylcoumaran ( -5) linkages are particularly more abundant (Guo et al., 2001a; Jouanin et al., 2000; Lapierre et al., 1999), making the lignin more similar to softwood lignin. These changes, and the reduction in free phenolic groups on -O-4-linked G units (Lapierre et al., 1999) probably contribute to the fact that this poplar wood is more diYcult to pulp (Pilate et al., 2002). In COMT-suppressed alfalfa, - , -1, and -5 linkages involving S units are absent (Guo et al., 2001a), whereas in COMT–down-regulated poplar, free phenolic groups in -O-4-linked G units were less abundant (Lapierre et al., 1999). In general, however, COMT-suppressed plants have no obvious external phenotype, although in poplar, when the bark is removed, the wood of the


87

COMT-suppressed lines has a rose color (Van Doorsselaere et al., 1995) or a reddish-brown color (Jouanin et al., 2000; Tsai et al., 1998) compared to the whitish, wild-type wood; this attribute has been ascribed to an increased amount of coniferaldehyde (Tsai et al., 1998). H. FERULATE 5-HYDROXYLASE

Because of the recent appreciation that 5-hydroxylation of lignin precursors occurs preferentially at the cinnamaldehyde level, this enzyme is sometimes referred to as coniferaldehyde 5-hydroxylase or Cald5H, although many researchers still use the original name. F5H expression has recently been manipulated in tobacco, Arabidopsis, and poplar/aspen (Franke et al., 2000; Li et al., 2003; Meyer et al., 1998; Sibout et al., 2002), but its role in lignin biosynthesis has been appreciated for more than a decade because of the early identification of an Arabidopsis F5H mutant, fah1 (Chapple et al., 1992; Meyer et al., 1996). This mutant produces a lignin deficient in S units with enhanced proportions of phenylcoumaran ( -5) and dibenzodioxocin (biphenyl; 5-5) linkages (Marita et al., 1999). F5H overexpression in Arabidopsis, tobacco, or poplar results in lignin almost entirely composed of S units, containing no phenylcoumaran or dibenzodioxocin structures (Franke et al., 2000; Marita et al., 1999; Meyer et al., 1998). These studies confirm that F5H plays a crucial role in the production of S lignin units and in determining lignin monomer composition. I. CINNAMOYL-COA REDUCTASE

Reports describing the eVects of CCR deficiency in transgenic or mutant plants give a consistent indication of the important role the enzyme plays in controlling lignin content. CCR-suppressed transgenic tobacco (O’Connell et al., 2002; Piquemal et al., 1998; Ralph et al., 1998) and Arabidopsis (Goujon et al., 2003a) have approximately 50% less Klason lignin than wild-type plants, as does an Arabidopsis ccr mutant called irregular xylem (irx4) (Jones et al., 2001). Many of these CCR-deficient plants display similar aberrant phenotypes (Fig. 3) including stunted growth, altered leaf morphology, and collapsed or irregular xylem vessels. The mechanical weakness of the vessel walls appears to be the result of disorganization and loosening of the secondary walls of fibers and vessels, where there is a particular deficiency in noncondensed lignin units (Chabannes et al., 2001a,b; Goujon et al., 2003a; Pinc,on et al., 2001b). Thioacidolysis of lignin from both CCR-deficient Arabidopsis and tobacco has confirmed an overall reduction in noncondensed lignin, particularly in -O-4-linked guaiacyl

88

C. HALPIN

Fig. 3. Phenotype of CCR-suppressed plants. Tobacco plants suppressed in CCR activity (left) frequently show altered leaf shape compared to wild-type plants (right).

units (Goujon et al., 2003a; O’Connell et al., 2002). The more pronounced decrease in noncondensed G units compared to S units likely accounts for the increased S:G ratio of lignin from CCR-deficient tobacco (O’Connell et al., 2002; Piquemal et al., 1998). A similar increase in thioacidolysis S:G units has also been seen in some CCR-deficient Arabidopsis lines, depending on the growth conditions (Goujon et al., 2003a). Increased amounts of ferulic acid and sinapic acid are incorporated into the cell walls of CCRdeficient plants (Goujon et al., 2003a; O’Connell et al., 2002; Piquemal et al., 1998) and, in tobacco, feruloyl-tyramine is also detected (Ralph et al., 1998), possibly representing an alternative sink for feruloyl-CoA. The presence of unusual phenolics such as ferulic acid and sinapic acid may account for the unusual orange-brown color of xylem cell walls in CCR-suppressed tobacco, because semi-in vivo incorporation of these two hydroxycinnamic acids into stem sections resulted in a comparable phenotype (Piquemal et al., 1998). The altered structure of lignin in CCR-suppressed tobacco is also indicated by the higher amount of alkali-labile material that can be released from the extractives-free polymer and by the fact that there is an increased proportion of free phenolic groups in the noncondensed fraction of lignin (O’Connell et al., 2002).


89

J. CINNAMYL ALCOHOL DEHYDROGENASE

The roles of CAD in lignin biosynthesis have been extensively studied in a wide variety of species. Transgenic plants with reduced CAD activity have been produced in tobacco (Halpin et al., 1994; Hibino et al., 1995; Stewart et al., 1997; Yahiaoui et al., 1998), poplar (Baucher et al., 1996), and alfalfa (Baucher et al., 1999), whereas cad mutants exist in pine (MacKay et al., 1997), maize (Halpin et al., 1998), Arabidopsis (Sibout et al., 2003), and probably sorghum (Pillonel et al., 1991). CAD catalyzes the last step in the production of the monolignols, the reduction of cinnamaldehydes into cinnamyl alcohols. Despite its important role in monolignol production, lignin content is not reduced, or is only slightly reduced in most CAD-deficient plants. It appears that other phenolics such as the aldehyde substrates of CAD can be incorporated into lignin and to some extent compensate for the reduced availability of monolignols. The CAD-deficient bm1 mutant of maize has been reported to have significant reductions in lignin in certain genetic backgrounds (Colenbrander et al., 1973; Kuc and Nelson, 1964), but no change in lignin content was detected when the mutation was introduced into other maize varieties (Marita et al., 2003b), highlighting the way that genetic and environmental factors can potentially influence the results of diVerent experiments. The changes to lignin composition identified in CAD-deficient plants are for the most part consistent with the role traditionally envisaged for CAD in catalyzing the reduction of all three lignin cinnamaldehydes ( p-coumaryl-, coniferyl- and sinapyl-aldehyde). Increased levels of cinnamaldehydes have been detected in the lignin of CAD-antisense tobacco (Halpin et al., 1994; Ralph et al., 1998, 2001) and poplar (Kim et al., 2002), as well as in the pine (Ralph et al., 1997), Arabidopsis (Sibout et al., 2003), and maize (Marita et al., 2003b) cad mutants. In tobacco, poplar, and Arabidopsis, increases in both coniferaldehyde and sinapaldehyde were identified (Halpin et al., 1994; Kim et al., 2002; Ralph et al., 2001; Sibout et al., 2003), suggesting that CAD is indeed involved in the reduction of both of these cinnamaldehydes. Similarly, in CAD-antisense poplar and in the maize bm1 mutant (Halpin et al., 1998; Lapierre et al., 1999) the proportion of S and G units in lignin thioacidolysis products is not altered, whereas in CAD-antisense tobacco and alfalfa (Baucher et al., 1999; Halpin et al., 1994) the S:G ratio is reduced. This suggests that in CAD-deficient plants the deposition of S lignin units is equally or more aVected than that of G lignin units. All of this data is at variance with a recent claim that a specific sinapyl alcohol dehydrogenase, and not CAD, is responsible for sinapaldehyde reduction in angiosperms (Li et al., 2001a).

90

C. HALPIN

Fig. 4. Wood color in CAD antisense tobacco. Tobacco plants with very low levels of CAD activity (right) have a bright red color in the woody xylem compared to the cream color of wild-type plants (left).

A striking characteristic of transgenic plants down-regulated for CAD is the red or brownish color of xylem tissues (Fig. 4), initially observed in the maize ‘‘brown-midrib’’ (bm) mutants. This color has been attributed to the incorporation of cinnamaldehydes in the polymer because synthetic dhydrogenation polymers (DHPs) of coniferyl alcohol and coniferaldehyde also form a red polymer (Higuchi et al., 1994). The cross-coupling of cinnamaldehydes into the lignin polymer may result in a more extended conjugated system, which causes the red color (Higuchi et al., 1994). The unusual monomer, dihydroconiferyl alcohol, is apparently also incorporated into the lignin of the pine cad mutant, where it accounts for 30% of the polymer compared to only 3% in wild-type plants (Ralph et al., 1997). Altered structure in the lignin polymer of CAD-suppressed plants is also indicated by its increased extractability in alkali (Baucher et al., 1996; Bernard-Vailhe et al., 1996; Halpin et al., 1994; MacKay et al., 1999; Yahiaoui et al., 1998). In transgenic poplar and tobacco, the lignin was enriched in free phenolic groups in both S and G units (Lapierre et al., 1999; O’Connell et al., 2002). Similarly, the proportion of G units with free phenolic groups was increased in the Arabidopsis Atcad-D mutant (Sibout et al., 2003). This increase in free phenolic groups may be important in altering the solubility of lignin, which in turn has implications for the ease with which wood from these plants can be pulped (Lapierre et al., 1999; O’Connell et al., 2002). K. MANIPULATION OF MULTIPLE GENES

Although the vast majority of work aimed at manipulating lignin biosynthesis in transgenic plants has focused on modifying the expression of single genes, some researchers are beginning to explore the possibilities of


91

producing novel lignins by simultaneous manipulation of multiple genes. Stacking transgenes or traits in this way may potentially provide great opportunities for crop and tree improvement in the future (Halpin and Boerjan, 2003; Halpin et al., 2001). Both COMT and CCoAOMT have simultaneously been suppressed in tobacco (Pinc,on et al., 2001a; Zhong et al., 1998) and alfalfa (Guo et al., 2001a). In tobacco, a greater reduction in Klason lignin content was achieved in comparison with the respective single transformants (Pinc,on et al., 2001a), but this was not the case in alfalfa (Guo et al., 2001a). In both species, lignin S:G ratio was reduced. However, in alfalfa, only S units were decreased (Guo et al., 2001a), whereas in tobacco both G and S units were aVected (Zhong et al., 1998). It has been proposed that additional enzymes may be involved in the methylation of G unit precursors in alfalfa, to explain the relative preservation of G units in this species (Guo et al., 2001a). Simultaneous suppressions in COMT and CCR expression have been achieved in tobacco by crossing plants down-regulated in the single genes (Pinc,on et al., 2001b). Progeny of this cross had reduced activity of both enzymes and had intermediate phenotypes between those of the two parents. In terms of lignin modification, however, the eVects of CCR suppression appeared to predominate and plants had reduced lignin content and increased S:G ratio, whereas characteristics typical of COMT suppression were not detected (Pinc,on et al., 2001b). This may reflect insuYcient levels of COMT down-regulation to promote lignin modification. Homozygous transgenics suppressed in either CAD or CCR were similarly crossed to produce tobacco plants down-regulated in both genes (Chabannes et al., 2001a). Combinatorial suppression of both genes appeared to have a synergistic eVect in reducing lignin quantity, which was decreased by approximately 50% compared to 32% and 12% reductions in plants hemizygous for CCR- and CAD-suppressing transgenes, respectively. Nevertheless, NMR spectra of isolated lignin fractions showed that the structure of the polymer was closer to that of wild-type plants than to the CCR- or CAD-deficient parents. Similarly, the phenotype of the CAD/CCR suppressed plants was normal with only slight alterations in the vessel shape, indicating that under certain circumstances, plants can tolerate important reductions in lignin content without developing adverse phenotypes. Tobacco plants doubly suppressed in CCR/COMT, in CAD/CCR, or in CAD/COMT have been produced using chimeric constructs consisting of partial sense sequences for both target genes (Abbott et al., 2002). Plants suppressed in CAD and COMT were also produced by crossing transgenics suppressed in either enzyme so that the two diVerent strategies could be

92

C. HALPIN

compared. Greater levels of enzyme suppression and greater changes to lignin were achieved using the single chimeric construct for CAD/COMT suppression, compared to the crossed transgenics, indicating the greater eYciency of the chimeric construct strategy. The same approach was used successfully to suppress all three genes (CAD, COMT, and CCR) together. The resulting plants were severely stunted and had characteristics of COMT, CCR, and CAD suppression in lignin (Abbott et al., 2002), further illustrating the usefulness and eVectiveness of this strategy. Suppression of 4CL has been combined with overexpression of F5H by cotransformation of two transgenes in aspen (Li et al., 2003). Additive eVects were observed compared to transformants with the individual transgenes. The trees had a higher S:G ratio along with a 52% decrease in lignin and proportional increases in cellulose. Thus multigene manipulation can be used to great eVect to concurrently improve several valuable wood quality traits.

XII. COMMERCIAL APPLICATIONS OF MODIFIED LIGNIN PLANTS Lignin has a major influence on a number of plant characteristics that aVect the agronomic quality or commercial value of crops or plantation trees. The digestibility of forage crops is improved in mutants where lignin content is reduced (see Cherney et al., 1991 for a review). Some of these increased digestibility ‘‘brown-midrib’’ mutants have been shown to be deficient in enzymes of monolignol biosynthesis, notably COMT and CAD (Halpin et al., 1998; Vignols et al., 1995), suggesting a route to digestibility improvement via genetic engineering of lignin biosynthetic genes. Consistent with this idea, transgenic tobacco and alfalfa plants suppressed in CAD have been shown to have slightly improved in situ cell wall degradability when fed to sheep (Baucher et al., 1999; Bernard-Vailhe et al., 1998), whereas alfalfa, tobacco, and Stylosanthes plants deficient in COMT also had improved digestibility (Barriere et al., 2003; Bernard-Vailhe et al., 1996; Guo et al., 2001b; Rae et al., 2001; Sewalt et al., 1997b). Cell walls of the Arabidopsis C3H mutant (ref8) have increased susceptibility to polysaccharide hydrolases (Franke et al., 2002a), whereas enzymatic digestibility of transgenic tobacco suppressed in PAL or transgenic Arabidopsis down-regulated for CCR indicated a similar digestibility improvement (Goujon et al., 2003a; Sewalt et al., 1997b). Obviously, to be useful, genetic manipulation of lignin content must be accomplished without adverse eVects on plant phenotype, disease resistance, or agronomic characters such as lodging resistance. In practice


93

these restrictions are likely to restrict the choice of genes for manipulation to those later on the monolignol branch pathway, such as COMT and CAD. Lignin modification could also be used as a means to manipulate the rate with which plant residues decompose in soils. Lignin is inherently recalcitrant to decay and also protects the cell wall carbohydrates that are covalently linked to it from microbial attack. Reducing the amount of lignin in potentially disease-harboring plant residues could possibly improve decomposition rates, perhaps, reducing rotation times for certain crops. Although this area has received little attention to date, one study already illustrates the potential for altering decomposition rates by manipulating lignin (Hopkins et al., 2001). Lignin limits the ease with which wood can be pulped because removal of lignin from cellulose requires the use of harsh mechanical or toxic chemical processes. To make pulping easier and more environmentally benign, lignin contents of wood could be reduced or lignin extractability could be improved by modifying the polymer structure. Several pulping studies have been performed on transgenic lignin-modified plants and results illustrate the potential of genetic engineering for improving pulping properties of woods. Greenhouse-grown transgenic tobacco and poplar with down-regulated CAD activity have been subjected to chemical pulping analyses (Baucher et al., 1996; Jouanin et al., 2000; Lapierre et al., 1999; O’Connell et al., 2002). In both species the changes to lignin structure in CAD-suppressed plants resulted in a greater ease of pulping by the chemical Kraft process, and the kappa number, a measure of residual lignin in the pulp after cooking, was reduced compared to that of wild-type plants. Subsequent bleaching of the pulps was also easier and there were no detrimental changes to other pulp properties. Most importantly, these pulping improvements were maintained when the transgenic poplars were grown for 4 years in the field at two diVerent sites in France and the United Kingdom (Pilate et al., 2002). Wood from the pine cad mutant has also been subjected to Kraft processes, but in this case no enhanced delignification was evident (MacKay et al., 1999), probably reflecting the diVerences in lignin structure between angiosperms and gymnosperms. Soda pulping of the mutant, however, produced pulp with a lower kappa number than wild-type plants, providing a potential opportunity for lowering consumption of sulfide during pulping. Pulping of wood from field-grown COMT–down-regulated poplars indicates that the modified lignin is more diYcult to extract and kappa numbers are consequently higher, whereas pulp brightness after bleaching is lower than that of wild-type trees (Pilate et al., 2002). These data suggest that, contrary to original expectations, depleting COMT activity reduces wood quality for Kraft pulping.

94

C. HALPIN

Transgenic tobacco with reduced levels of CCR has also been subjected to Kraft pulping. As in the case of CAD-deficient tobacco, the low-CCR plants produced pulp with a lower kappa number than that of wild-type plants (O’Connell et al., 2002). After bleaching, however, pulp from the CCRsuppressed plants had reduced brightness, which was apparently caused by a higher content of unextracted chlorophyll (O’Connell et al., 2002). Similarly, tobacco suppressed in 4CL has been shown to be improved for Kraft pulping and subsequent bleaching, exhibiting a higher eYciency of delignification and higher pulp yield than corresponding control material (Kajita et al., 2002). Analysis of greenhouse-grown poplar overexpressing F5H has demonstrated particularly significant improvements of Kraft pulping eYciency. Pulps had lower kappa numbers and increased brightness compared to wild-type (Huntley et al., 2003). The authors estimate that this genetic improvement could increase pulp throughput by 60% while concomitantly decreasing the consumption of pulping chemicals. Taken together the data currently available suggest that suppression of CAD, CCR, or 4CL and overexpression of F5H provide the best currently tested options for improving Kraft pulping eYciency. In particular, CAD suppression and F5H overexpression present the most promising opportunities, because manipulation of these genes in a variety of species has shown no adverse phenotypes to be associated with the genetic changes and clear pulping benefits have been demonstrated.

XIII. CONCLUSIONS AND PERSPECTIVES A decade of intensive research into the lignin biosynthesis has greatly improved our understanding of the genes and reactions involved and how to manipulate them toward useful ends, although some clarity is still needed on the accuracy of diVerent proposed pathway models. Progress in these areas is, however, balanced by a complete lack of consensus, based on conflicting pieces of incomplete information, on the cell biology of the process. Similarly, the transcriptional and metabolic regulation of the pathway and of its interactions with other cellular processes is only beginning to be investigated. Postgenomic technologies such as transcript and metabolite profiling are already being applied to address these questions and oVer the prospect of rapidly increasing our understanding and appreciation of the lignin pathway and its wider role in plant metabolism and development by comparing these processes in wild-type and transgenic plants modified in cell wall biosynthesis. Such analyses should provide a comprehensive and holistic view on cell wall assembly at a level that was not possible just a few years ago. In addition to the identification of novel


95

genes involved in wood formation, whose function can be further studied by reverse genetics, applications of these techniques will shed light onto the interrelations between the biochemical pathways leading to the biosynthesis of the diVerent cell wall macromolecules and onto their relationship to plant growth and development.

REFERENCES Abbott, J. C., Barakate, A., Pinc,on, G., Legrand, M., Lapierre, C., Mila, I., Schuch, W. and Halpin, C. (2002). Simultaneous suppression of multiple genes by single transgenes. Downregulation of three unrelated lignin biosynthetic genes in tobacco. Plant Physiology 128, 844–853. Anterola, A. M. and Lewis, N. G. (2002). Trends in lignin modification: A comprehensive analysis of the eVects of genetic manipulations/mutations on lignification and vascular integrity. Phytochemistry 61, 221–294. Anterola, A. M., Jeon, J. H., Davin, L. B. and Lewis, N. G. (2002). Transcriptional control of monolignol biosynthesis in Pinus taeda: Factors aVecting monolignol ratios and carbon allocation in phenylpropanoid metabolism. Journal of Biological Chemistry 277, 18272–18280. Atanassova, R., Favet, N., Martz, F., Chabbert, B., Tollier, M. T., Monties, B., Fritig, B. and Legrand, M. (1995). Altered lignin composition in transgenic tobacco expressing O-methyltransferase sequences in sense and antisense orientation. The Plant Journal 8, 465–477. Barriere, Y., Guillet, C., GoVner, D. and Pichon, M. (2003). Genetic variation and breeding strategies for improved cell wall digestibility in annual forage crops. A review. Animal Research 52, 193–228. Bate, N. J., Orr, J., Ni, W. T., Meromi, A., Nadlerhassar, T., Doerner, P. W., Dixon, R. A., Lamb, C. J. and Elkind, Y. (1994). Quantitative relationship between phenylalanine ammonia-lyase levels and phenylpropanoid accumulation in transgenic tobacco identifies a rate-determining step in natural product synthesis. Proceedings of the National Academy of Sciences of the United States of America 91, 7608–7612. Baucher, M., Chabbert, B., Pilate, G., Van Doorsselaere, J., Tollier, M. T., PetitConil, M., Cornu, D., Monties, B., Van Montagu, M., Inze, D., Jouanin, L. and Boerjan, W. (1996). Red xylem and higher lignin extractability by down-regulating a cinnamyl alcohol dehydrogenase in poplar. Plant Physiology 112, 1479–1490. Baucher, M., Bernard-Vailhe, M. A., Chabbert, B., Besle, J. M., Opsomer, C., Van Montagu, M. and Botterman, J. (1999). Down-regulation of cinnamyl alcohol dehydrogenase in transgenic alfalfa (Medicago sativa L.) and the eVect on lignin composition and digestibility. Plant Molecular Biology 39, 437–447. Baucher, M., Halpin, C., Petit-Conil, M. and Boerjan, W. (2003). Lignin: Genetic engineering and impact on pulping. Critical Reviews in Biochemistry and Molecular Biology 38, 305–350. Bell-Lelong, D. A., Cusumano, J. C., Meyer, K. and Chapple, C. (1997). Cinnamate4-hydroxylase expression in Arabidopsis. Regulation in response to development and the environment. Plant Physiology 113, 729–738. Bernard-Vailhe, M. A., Migne, C., Cornu, A., Maillot, M. P., Grenet, E., Besle, J. M., Atanassova, R., Martz, F. and Legrand, M. (1996). EVect of modification

96

C. HALPIN

of the O-methyltransferase activity on cell wall composition, ultrastructure and degradability of transgenic tobacco. Journal of the Science of Food and Agriculture 72, 385–391. Bernard-Vailhe, M. A., Besle, J. M., Maillot, M. P., Cornu, A., Halpin, C. and Knight, M. (1998). EVect of down-regulation of cinnamyl alcohol dehydrogenase on cell wall composition and on degradability of tobacco stems. Journal of the Science of Food and Agriculture 76, 505–514. Binns, A. N., Chen, R. H., Wood, H. N. and Lynn, D. G. (1987). Cell-division promoting activity of naturally-occurring dehydrodiconiferyl glucosides: Do cell-wall components control cell-division? Proceedings of the National Academy of Sciences of the United States of America 84, 980–984. Blount, J. W., Korth, K. L., Masoud, S. A., Rasmussen, S., Lamb, C. and Dixon, R. A. (2000). Altering expression of cinnamic acid 4-hydroxylase in transgenic plants provides evidence for a feedback loop at the entry point into the phenylpropanoid pathway. Plant Physiology 122, 107–116. Boerjan, W., Ralph, J. and Baucher, M. (2003). Lignin biosynthesis. Annual Review of Plant Biology 54, 519–546. Bolwell, G. P., Cramer, C. L., Lamb, C. J., Schuch, W. and Dixon, R. A. (1986). L-Phenylalanine ammonia-lyase from Phaseolus –vulgaris. Modulation of the levels of active enzyme by trans-cinnamic acid. Planta 169, 97–107. Borevitz, J. O., Xia, Y. J., Blount, J., Dixon, R. A. and Lamb, C. (2000). Activation tagging identifies a conserved MYB regulator of phenylpropanoid biosynthesis. Plant Cell 12, 2383–2393. Bout, S. and Vermerris, W. (2003). A candidate-gene approach to clone the sorghum Brown midrib gene encoding caVeic acid O-methyltransferase. Molecular Genetics and Genomics 269, 205–214. Cano-Delgado, A. I., MetzlaV, K. and Bevan, M. W. (2000). The eli1 mutation reveals a link between cell expansion and secondary cell wall formation in Arabidopsis thaliana. Development 127, 3395–3405. Cano-Delgado, A., Penfield, S., Smith, C., Catley, M. and Bevan, M. (2003). Reduced cellulose synthesis invokes lignification and defense responses in Arabidopsis thaliana. The Plant Journal 34, 351–362. Chabannes, M., Barakate, A., Lapierre, C., Marita, J. M., Ralph, J., Pean, M., Danoun, S., Halpin, C., Grima-Pettenati, J. and Boudet, A. M. (2001a). Strong decrease in lignin content without significant alteration of plant development is induced by simultaneous down-regulation of cinnamoyl CoA reductase (CCR) and cinnamyl alcohol dehydrogenase (CAD) in tobacco plants. The Plant Journal 28, 257–270. Chabannes, M., Ruel, K., Yoshinaga, A., Chabbert, B., Jauneau, A., Joseleau, J. P. and Boudet, A. M. (2001b). In situ analysis of lignins in transgenic tobacco reveals a diVerential impact of individual transformations on the spatial patterns of lignin deposition at the cellular and subcellular levels. The Plant Journal 28, 271–282. Chabbert, B., Tollier, M. T., Monties, B., Barriere, Y. and Argillier, O. (1994). Biological variability in lignification of maize: Expression of the brown midrib bm3 mutation in 3 maize cultivars. Journal of the Science of Food and Agriculture 64, 349–355. Chapple, C. C. S., Vogt, T., Ellis, B. E. and Somerville, C. R. (1992). An arabidopsis mutant defective in the general phenylpropanoid pathway. Plant Cell 4, 1413–1424. Chen, C. Y., Meyermans, H., Burggraeve, B., De Rycke, R. M., Inoue, K., De Vleesschauwer, V., Steenackers, M., Van Montagu, M. C., Engler, G. J. and


97

Boerjan, W. A. (2000). Cell-specific and conditional expression of caVeoylcoenzyme A-3-O-methyltransferase in poplar. Plant Physiology 123, 853–867. Chen, F., Yasuda, S. and Fukushima, K. (1999). Evidence for a novel biosynthetic pathway that regulates the ratio of syringyl to guaiacyl residues in lignin in the diVerentiating xylem of Magnolia kobus DC. Planta 207, 597–603. Chen, F., Duran, A. L., Blount, J. W., Sumner, L. W. and Dixon, R. A. (2003). Profiling phenolic metabolites in transgenic alfalfa modified in lignin biosynthesis. Phytochemistry 64, 1013–1021. Cherney, J. H., Cherney, D. J. R., Akin, D. E. and Axtell, J. D. (1991). Potential of brown-midrib, low-lignin mutants for improving forage quality. Advances in Agronomy 46, 157–198. Colenbrander, V. L., Lechtenberg, L. F. and Bauman, L. F. (1973). Digestibility and feeding value of brown midrib corn stover silage. Journal of Animal Science 37, 294–295. Costa, M. A., Collins, R. E., Anterola, A. M., Cochrane, F. C., Davin, L. B. and Lewis, N. G. (2003). An in silico assessment of gene function and organization of the phenylpropanoid pathway metabolic networks in Arabidopsis thaliana and limitations thereof. Phytochemistry 64, 1097–1112. Czichi, U. and Kindl, H. (1977). Phenylalanine ammonia-lyase and cinnamic acid hydroxylase as assembled consecutive enzymes on microsomal membranes of cucumber cotyledons: Cooperation and subcellular distribution. Planta 134, 133–143. Davin, L. B., Wang, H. B., Crowell, A. L., Bedgar, D. L., Martin, D. M., Sarkanen, S. and Lewis, N. G. (1997). Stereoselective bimolecular phenoxy radical coupling by an auxiliary (dirigent) protein without an active center. Science 275, 362–366. Dean, J. F. D. and Eriksson, K. E. L. (1994). Laccase and the deposition of lignin in vascular plants. Holzforschung 48, 21–33. Dharmawardhana, D. P., Ellis, B. E. and Carlson, J. E. (1999). cDNA cloning and heterologous expression of coniferin beta-glucosidase. Plant Molecular Biology 40, 365–372. Dixon, R. A., Chen, F., Guo, D. J. and Parvathi, K. (2001). The biosynthesis of monolignols: A metabolic grid, or independent pathways to guaiacyl and syringyl units? Phytochemistry 57, 1069–1084. Dwivedi, U. N., Campbell, W. H., Yu, J., Datla, R. S. S., Bugos, R. C., Chiang, V. L. and Podila, G. K. (1994). Modification of lignin biosynthesis in transgenic nicotiana through expression of an antisense o-methyltransferase gene from populus. Plant Molecular Biology 26, 61–71. Ehlting, J., Buttner, D., Wang, Q., Douglas, C. J., Somssich, I. E. and Kombrink, E. (1999). Three 4-coumarate: Coenzyme A ligases in Arabidopsis thaliana represent two evolutionarily divergent classes in angiosperms. The Plant Journal 19, 9–20. Elkind, Y., Edwards, R., Mavandad, M., Hedrick, S. A., Ribak, O., Dixon, R. A. and Lamb, C. J. (1990). Abnormal-plant development and down-regulation of phenylpropanoid biosynthesis in transgenic tobacco containing a heterologous phenylalanine ammonia-lyase gene. Proceedings of the National Academy of Sciences of the United States of America 87, 9057–9061. Franke, R., McMichael, C. M., Meyer, K., Shirley, A. M., Cusumano, J. C. and Chapple, C. (2000). Modified lignin in tobacco and poplar plants overexpressing the Arabidopsis gene encoding ferulate 5-hydroxylase. The Plant Journal 22, 223–234. Franke, R., Hemm, M. R., Denault, J. W., Ruegger, M. O., Humphreys, J. M. and Chapple, C. (2002a). Changes in secondary metabolism and deposition of

98

C. HALPIN

an unusual lignin in the ref8 mutant of Arabidopsis. The Plant Journal 30, 47–59. Franke, R., Humphreys, J. M., Hemm, M. R., Denault, J. W., Ruegger, M. O., Cusumano, J. C. and Chapple, C. (2002b). The Arabidopsis REF8 gene encodes the 3-hydroxylase of phenylpropanoid metabolism. The Plant Journal 30, 33–45. Gabriac, B., Werckreichhart, D., Teutsch, H. and Durst, F. (1991). Purification and immunocharacterization of a plant cytochrome P450: The cinnamic acid 4-hydroxylase. Archives of Biochemistry and Biophysics 288, 302–309. Gang, D. R., Costa, M. A., Fujita, M., Dinkova-Kostova, A. T., Wang, H. B., Burlat, V., Martin, W., Sarkanen, S., Davin, L. B. and Lewis, N. G. (1999). Regiochemical control of monolignol radical coupling: A new paradigm for lignin and lignan biosynthesis. Chemistry and Biology 6, 143–151. Goujon, T., Ferret, V., Mila, I., Pollet, B., Ruel, K., Burlat, V., Joseleau, J. P., Barriere, Y., Lapierre, C. and Jouanin, L. (2003a). Down-regulation of the AtCCR1 gene in Arabidopsis thaliana: EVects on phenotype, lignins and cell wall degradability. Planta 217, 218–228. Goujon, T., Sibout, R., Eudes, A., MacKay, J. and Joulanin, L. (2003b). Genes involved in the biosynthesis of lignin precursors in Arabidopsis thaliana. Plant Physiology and Biochemistry 41, 677–687. Goujon, T., Sibout, R., Pollet, B., Maba, B., Nussaume, L., Bechtold, N., Lu, F. C., Ralph, J., Mila, I., Barriere, Y., Lapierre, C. and Jouanin, L. (2003c). A new Arabidopsis thaliana mutant deficient in the expression of O-methyltransferase impacts lignins and sinapoyl esters. Plant Molecular Biology 51, 973–989. Grand, C. (1984). Ferulic acid 5-hydroxylase: A new cytochrome p-450-dependent enzyme from higher-plant microsomes involved in lignin synthesis. FEBS Letters 169, 7–11. Guo, D., Chen, F., Inoue, K., Blount, J. W. and Dixon, R. A. (2001a). Downregulation of caVeic acid 3-O-methyltransferase and caVeoyl CoA 3-Omethyltransferase in transgenic alfalfa: Impacts on lignin structure and implications for the biosynthesis of G and S lignin. Plant Cell 13, 73–88. Guo, D., Chen, F., Wheeler, J., Winder, J., Selman, S., Peterson, M. and Dixon, R. A. (2001b). Improvement of in-rumen digestibility of alfalfa forage by genetic manipulation of lignin O-methyltransferases. Transgenic Research 10, 457–464. Guo, D., Chen, F. and Dixon, R. A. (2002). Monolignol biosynthesis in microsomal preparations from lignifying stems of alfalfa (Medicago sativa L.). Phytochemistry 61, 657–667. Halpin, C. and Boerjan, W. (2003). Stacking transgenes in forest trees. Trends in Plant Sciences 8, 363–365. Halpin, C., Knight, M. E., Foxon, G. A., Campbell, M. M., Boudet, A. M., Boon, J. J., Chabbert, B., Tollier, M.-T. and Schuch, W. (1994). Manipulation of lignin quality by down-regulation of cinnamyl alcohol dehydrogenase. The Plant Journal 6, 339–350. Halpin, C., Holt, K., Chojecki, J., Oliver, D., Chabbert, B., Monties, B., Edwards, K. and Foxon, G. A. (1998). Maize brown-midrib (bm1)—a mutation aVecting the cinnamyl alcohol dehydrogenase gene. The Plant Journal 14, 545–553. Halpin, C., Barakate, A., Askari, B. M., Abbott, J. C. and Ryan, M. D. (2001). Enabling technologies for manipulating multiple genes on complex pathways. Plant Molecular Biology 47, 295–310.


99

Harmer, S. L., Hogenesch, L. B., Straume, M., Chang, H. S., Han, B., Zhu, T., Wang, X., Kreps, J. A. and Kay, S. A. (2000). Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science 290, 2110–2113. Hatfield, R. and Vermerris, W. (2001). Lignin formation in plants. The dilemma of linkage specificity. Plant Physiology 126, 1351–1357. Hatton, D., Sablowski, R., Yung, M. H., Smith, C., Schuch, W. and Bevan, M. (1995). Two classes of Cis sequences contribute to tissue-specific expression of a Pal2 promoter in transgenic tobacco. Plant Journal 7, 859–876. HauVe, K. D., Lee, S. P., Subramaniam, R. and Douglas, C. J. (1993). Combinatorial interactions between positive and negative cis-acting elements control spatial patterns of 4Cl-1 expression in transgenic tobacco. Plant Journal 4, 235–253. Hertzberg, M., Aspeborg, H., Schrader, J., Andersson, A., Erlandsson, R., Blomqvist, K., Bhalerao, R., Uhlen, M., Teeri, T. T., Lundeberg, J., Sundberg, B., Nilsson, P. and Sandberg, G. (2001). A transcriptional roadmap to wood formation. Proceedings of the National Academy of Sciences of the United States of America 98, 14732–14737. Hibino, T., Takabe, K., Kawazu, T., Shibata, D. and Higuchi, T. (1995). Increase of cinnamaldehyde groups in lignin of transgenic tobacco plants carrying an antisense gene for cinnamyl alcohol-dehydrogenase. Bioscience Biotechnology and Biochemistry 59, 929–931. Higuchi, T., Ito, T., Umezawa, T., Hibino, T. and Shibata, D. (1994). Red-brown color of lignified tissues of transgenic plants with antisense cad gene—winered lignin from coniferyl aldehyde. Journal of Biotechnology 37, 151–158. HoVmann, L., Maury, S., Martz, F., GeoVroy, P. and Legrand, M. (2003). Purification, cloning, and properties of an acyltransferase controlling shikimate and quinate ester intermediates in phenylpropanoid metabolism. Journal of Biological Chemistry 278, 95–103. HoVmann, L., Besseau, S., GeoVroy, P., Ritzenthaler, C., Meyer, D., Lapierre, C., Pollet, B. and Legrand, M. (2004). Silencing of hydroxycinnamoy-coenzme A shikimate/quinate hydroxycinnamoyltransferase affects phenylpropanoid biosynthesis. Plant Cell 16, 1446–1465. Hopkins, D. W., Webster, E. A., Chudek, J. A. and Halpin, C. (2001). Decomposition of stems from tobacco plants with genetic modifications to lignin biosynthesis. Soil Biology and Biochemistry 33, 1455–1462. Howles, P. A., Sewalt, V. J. H., Paiva, N. L., Elkind, Y., Bate, N. J., Lamb, C. and Dixon, R. A. (1996). Overexpression of L-phenylalanine ammonia-lyase in transgenic tobacco plants reveals control points for flux into phenylpropanoid biosynthesis. Plant Physiology 112, 1617–1624. Hu, W. J., Harding, S. A., Lung, J., Popko, J. L., Ralph, J., Stokke, D. D., Tsai, C. J. and Chiang, V. L. (1999). Repression of lignin biosynthesis promotes cellulose accumulation and growth in transgenic trees. Nature Biotechnology 17, 808–812. Humphreys, J. M., Hemm, M. R. and Chapple, C. (1999). New routes for lignin biosynthesis defined by biochemical characterization of recombinant ferulate 5-hydroxylase, a multifunctional cytochrome P450-dependent monooxygenase. Proceedings of the National Academy of Sciences of the United States of America 96, 10045–10050. Humphreys, J. M. and Chapple, C. (2002). Rewriting the lignin roadmap. Current Opinion in Plant Biology 5, 224–229. Huntley, S. K., Ellis, D., Gilbert, M., Chapple, C. and Mansfield, S. D. (2003). Significant increases in pulping eYciency in C4H-F5H-transformed

100

C. HALPIN

poplars: Improved chemical savings and reduced environmental toxins. Journal of Agricultural and Food Chemistry 51, 6178–6183. Israelsson, M., Eriksson, M. E., Hertzberg, M., Aspeborg, H., Nilsson, P. and Moritz, T. (2003). Changes in gene expression in the wood-forming tissue of transgenic hybrid aspen with increased secondary growth. Plant Molecular Biology 52, 893–903. Jin, H. L., Cominelli, E., Bailey, P., Parr, A., Mehrtens, F., Jones, J., Tonelli, C., Weisshaar, B. and Martin, C. (2000). Transcriptional repression by AtMYB4 controls production of UV-protecting sunscreens in Arabidopsis. EMBO Journal 19, 6150–6161. Jones, L., Ennos, A. R. and Turner, S. R. (2001). Cloning and characterization of irregular xylem4 (irx4): A severely lignin-deficient mutant of Arabidopsis. The Plant Journal 26, 205–216. Jouanin, L., Goujon, T., de Nadai, V., Martin, M. T., Mila, I., Vallet, C., Pollet, B., Yoshinaga, A., Chabbert, B., Petit-Conil, M. and Lapierre, C. (2000). Lignification in transgenic poplars with extremely reduced caVeic acid O-methyltransferase activity. Plant Physiology 123, 1363–1373. Kajita, S., Katayama, Y. and Omori, S. (1996). Alterations in the biosynthesis of lignin in transgenic plants with chimeric genes for 4-coumarate: Coenzyme A ligase. Plant and Cell Physiology 37, 957–965. Kajita, S., Hishiyama, S., Tomimura, Y., Katayama, Y. and Omori, S. (1997). Structural characterization of modified lignin in transgenic tobacco plants in which the activity of 4-coumarate:coenzyme A ligase is depressed. Plant Physiology 114, 871–879. Kajita, S., Ishifuji, M., Ougiya, H., Hara, S., Kawabata, H., Morohoshi, N. and Katayama, Y. (2002). Improvement in pulping and bleaching properties of xylem from transgenic tobacco plants. Journal of the Science of Food and Agriculture 82, 1216–1223. Kawaoka, A. and Ebinuma, H. (2001). Transcriptional control of lignin biosynthesis by tobacco LIM protein. Phytochemistry 57, 1149–1157. Kersey, R., Inoue, K., Schubert, K. R. and Dixon, R. A. (1999). Immunolocalization of two lignin O-methyltransferases in stems of alfalfa (Medicago sativa L.). Protoplasma 209, 46–57. Kim, H., Ralph, J., Lu, F. C., Pilate, G., Leple, J. C., Pollet, B. and Lapierre, C. (2002). Identification of the structure and origin of thioacidolysis marker compounds for cinnamyl alcohol dehydrogenase deficiency in angiosperms. Journal of Biological Chemistry 277, 47412–47419. Kirst, M., Johnson, A. F., Baucom, C., Ulrich, E., Hubbard, K., Staggs, R., Paule, C., Retzel, E., Whetten, R. and SederoV, R. (2003). Apparent homology of expressed genes from wood-forming tissues of loblolly pine (Pinus taeda L.) with Arabidopsis thaliana. Proceedings of the National Academy of Sciences of the United States of America 100, 7383–7388. Korth, K. L., Blount, J. W., Chen, F., Rasmussen, S., Lamb, C. and Dixon, R. A. (2001). Changes in phenylpropanoid metabolites associated with homology-dependent silencing of phenylalanine ammonia-lyase and its somatic reversion in tobacco. Physiologia Plantarum 111, 137–143. Kuc, J. and Nelson, O. E. (1964). The abnormal lignins produced by the brownmidrib mutants of maize. I. The brown-midrib-1 mutant. Archives of Biochemistry and Biophysics 105, 103–113. Lacombe, E., Van Doorsselaere, J., Boerjan, W., Boudet, A. M. and Grima-Pettenati, J. (2000). Characterization of cis-elements required for vascular expression of the Cinnamoyl CoA Reductase gene and for protein-DNA complex formation. The Plant Journal 23, 663–676.


101

Lapierre, C., Pollet, B., Petit-Conil, M., Toval, G., Romero, J., Pilate, G., Leple, J. C., Boerjan, W., Ferret, V., de Nadai, V. and Jouanin, L. (1999). Structural alterations of lignins in transgenic poplars with depressed cinnamyl alcohol dehydrogenase or caVeic acid O-methyltransferase activity have an opposite impact on the eYciency of industrial kraft pulping. Plant Physiology 119, 153–163. Lauvergeat, V., Rech, P., Jauneau, A., Guez, C., Coutos-Thevenot, P. and GrimaPettenati, J. (2002). The vascular expression pattern directed by the Eucalyptus gunnii cinnamyl alcohol dehydrogenase EgCAD2 promoter is conserved among woody and herbaceous plant species. Plant Molecular Biology 50, 497–509. Lee, D., Meyer, K., Chapple, C. and Douglas, C. J. (1997). Antisense suppression of 4-coumarate:Coenzyme A ligase activity in Arabidopsis leads to altered lignin subunit composition. Plant Cell 9, 1985–1998. Lewis, N. G. (1999). A 20th century roller coaster ride: A short account of lignification. Current Opinion in Plant Biology 2, 153–162. Leyva, A., Liang, X. W., Pintortoro, J. A., Dixon, R. A. and Lamb, C. J. (1992). Cis-element combinations determine phenylalanine ammonia-lyase gene tissue-specific expression patterns. Plant Cell 4, 263–271. Li, L. G., Popko, J. L., Umezawa, T. and Chiang, V. L. (2000). 5-Hydroxyconiferyl aldehyde modulates enzymatic methylation for syringyl monolignol formation, a new view of monolignol biosynthesis in angiosperms. Journal of Biological Chemistry 275, 6537–6545. Li, L. G., Cheng, X. F., Leshkevich, J., Umezawa, T., Harding, S. A. and Chiang, V. L. (2001a). The last step of syringyl monolignol biosynthesis in angiosperms is regulated by a novel gene encoding sinapyl alcohol dehydrogenase. Plant Cell 13, 1567–1585. Li, L., Zhou, Y. H., Cheng, X. F., Sun, J. Y., Marita, J. M., Ralph, J. and Chiang, V. L. (2003). Combinatorial modification of multiple lignin traits in trees through multigene cotransformation. Proceedings of the National Academy of Sciences of the United States of America 100, 4939–4944. Li, Y., Baldauf, S., Lim, E. K. and Bowles, D. J. (2001b). Phylogenetic analysis of the UDP-glycosyltransferase multigene family of Arabidopsis thaliana. Journal of Biological Chemistry 276, 4338–4343. Liljegren, S. J., Ditta, G. S., Eshed, H. Y., Savidge, B., Bowman, J. L. and Yanofsky, M. F. (2000). Shatterproof MADS-box genes control seed dispersal in Arabidopsis. Nature 404, 766–770. Lim, E. K., Li, Y., Parr, A., Jackson, R., Ashford, D. A. and Bowles, D. J. (2001). Identification of glucosyltransferase genes involved in sinapate metabolism and lignin synthesis in Arabidopsis. Journal of Biological Chemistry 276, 4344–4349. Loake, G. J., Faktor, O., Lamb, C. J. and Dixon, R. A. (1992). Combination of H-box [CCTACC(N)7CT] and G-box (CACGTG) cis elements is necessary for feed-forward stimulation of a chalcone synthase promoter by the phenylpropanoid-pathway intermediate p-coumaric acid. Proceedings of the National Academy of Sciences of the United States of America 89, 9230–9234. Lois, R., Dietrich, A., Hahlbrock, K. and Schulz, W. (1989). A phenylalanine ammonia-lyase gene from parsley—structure, regulation and identification of elicitor and light responsive cis-acting elements. EMBO Journal 8, 1641–1648. Lu, F. C. and Ralph, J. (2003). Non-degradative dissolution and acetylation of ballmilled plant cell walls: High-resolution solution-state NMR. The Plant Journal 35, 535–544.

102

C. HALPIN

MacKay, J. J., O’Malley, D. M., Presnell, T., Booker, F. L., Campbell, M. M., Whetten, R. W. and SederoV, R. R. (1997). Inheritance, gene expression, and lignin characterization in a mutant pine deficient in cinnamyl alcohol dehydrogenase. Proceedings of the National Academy of Sciences of the United States of America 94, 8255–8260. MacKay, J., Presnell, T., Jameel, H., Taneda, H., O’Malley, D., SederoV, R., MacKay, J., Presnell, T., Jameel, H., Taneda, H., O’Malley, D. and SederoV, R. (1999). Modified lignin and delignification with a CADdeficient loblolly pine. Holzforschung 53, 403–410. Maher, E. A., Bate, N. J., Ni, W. T., Elkind, Y., Dixon, R. A. and Lamb, C. J. (1994). Increased disease susceptibility of transgenic tobacco plants with suppressed levels of preformed phenylpropanoid products. Proceedings of the National Academy of Sciences of the United States of America 91, 7802–7806. Marita, J. M., Ralph, J., Hatfield, R. D. and Chapple, C. (1999). NMR characterization of lignins in Arabidopsis altered in the activity of ferulate 5-hydroxylase. Proceedings of the National Academy of Sciences of the United States of America 96, 12328–12332. Marita, J. M., Ralph, J., Hatfield, R. D., Guo, D. J., Chen, F. and Dixon, R. A. (2003a). Structural and compositional modifications in lignin of transgenic alfalfa down-regulated in caVeic acid 3-O-methyltransferase and caVeoyl coenzyme A 3-O-methyltransferase. Phytochemistry 62, 53–65. Marita, J. M., Vermerris, W., Ralph, J. and Hatfield, R. D. (2003b). Variations in the cell wall composition of maize brown midrib mutants. Journal of Agricultural and Food Chemistry 51, 1313–1321. Martin, C. and Paz-Ares, J. (1997). MYB transcription factors in plants. Trends in Genetics 13, 67–73. Mavandad, M., Edwards, R., Liang, X. O., Lamb, C. J. and Dixon, R. A. (1990). EVects of trans-cinnamic acid on expression of the bean phenylalanine ammonia-lyase gene family. Plant Physiology 94, 671–680. McDougall, G. J., Stewart, D. and Morrison, I. M. (1994). Cell-wall-bound oxidases from tobacco (Nicotiana-tabacum) xylem participate in lignin formation. Planta 194, 9–14. Mele, G., Ori, N., Sato, Y. and Hake, S. (2003). The knotted1-like homeobox gene BREVIPEDICELLUS regulates cell diVerentiation by modulating metabolic pathways. Genes and Development 17, 2088–2093. Meyer, K., Cusumano, J. C., Somerville, C. and Chapple, C. C. S. (1996). Ferulate-5-hydroxylase from Arabidopsis thaliana defines a new family of cytochrome P450-dependent monooxygenases. Proceedings of the National Academy of Sciences of the United States of America 93, 6869–6874. Meyer, K., Shirley, A. M., Cusumano, J. C., Bell-Lelong, D. A. and Chapple, C. (1998). Lignin monomer composition is determined by the expression of a cytochrome P450-dependent monooxygenase in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 95, 6619–6623. Meyermans, H., Morreel, K., Lapierre, C., Pollet, B., De Bruyn, A., Busson, R., Herdewijn, P., Devreese, B., Van Beeumen, J., Marita, J. M., Ralph, J., Chen, C. Y., Burggraeve, B., Van Montagu, M., Messens, E. and Boerjan, W. (2000). Modifications in lignin and accumulation of phenolic glucosides in poplar xylem upon down-regulation of caVeoyl-coenzyme A O-methyltransferase, an enzyme involved in lignin biosynthesis. Journal of Biological Chemistry 275, 36899–36909.


103

Nakashima, J., Awano, T., Takabe, K., Fujita, M. and Saiki, H. (1997). Immunocytochemical localization of phenylalanine ammonia-lyase and cinnamyl alcohol dehydrogenase in diVerentiating tracheary elements derived from Zinnia mesophyll cells. Plant and Cell Physiology 38, 113–123. Newman, L. J., Perazza, D. E., Juda, L. and Campbell, M. M. (2004). Involvement of the R2R3-MYB, AtMYB61, in the ectopic lignification and dark-photomorphogenic components of the det3 mutant phenotype. The Plant Journal 37, 239–250. Ni, W. T., Paiva, N. L. and Dixon, R. A. (1994). Reduced lignin in transgenic plants containing a caVeic acid O-methyltransferase antisense gene. Transgenic Research 3, 120–126. O’Connell, A., Holt, K., Piquemal, J., Grima-Pettenati, J., Boudet, A., Pollet, B., Lapierre, C., Petit-Conil, M., Schuch, W. and Halpin, C. (2002). Improved paper pulp from plants with suppressed cinnamoyl-CoA reductase or cinnamyl alcohol dehydrogenase. Transgenic Research 11, 495–503. ¨ nnerud, H., Zhang, L. M., Gellerstedt, G. and Henriksson, G. (2002). PolymerizaO tion of monolignols by redox shuttle-mediated enzymatic oxidation: A new model in lignin biosynthesis I. Plant Cell 14, 1953–1962. Osakabe, K., Tsao, C. C., Li, L. G., Popko, J. L., Umezawa, T., Carraway, D. T., Smeltzer, R. H., Joshi, C. P. and Chiang, V. L. (1999). Coniferyl aldehyde 5-hydroxylation and methylation direct syringyl lignin biosynthesis in angiosperms. Proceedings of the National Academy of Sciences of the United States of America 96, 8955–8960. Parvathi, K., Chen, F., Guo, D. J., Blount, J. W. and Dixon, R. A. (2001)). Substrate preferences of O-methyltransferases in alfalfa suggest new pathways for 3-O-methylation of monolignols. The Plant Journal 25, 193–202. PatzlaV, A., McInnis, S., Courtenay, A., Surman, A., Newman, L. J., Smith, C., Bevan, M. W., Mansfield, S., Whetten, R. W., SederoV, R. R. and Campbell, M. M. (2003). Characterization of a pine MYB that regulates lignification. The Plant Journal 36, 743–755. Pilate, G., Guiney, E., Petit-Conil, M., Lapierre, C., Leple, J.-C., Pollet, B., Mila, I., Webster, E. A., Marstrop, H. G., Hopkins, D. W., Jouanin, L., Boerjan, W., Schuch, W., Cornu, D. and Halpin, C. (2002). Field and pulping performances of transgenic trees with altered lignification. Nature Biotechnology 20, 607–612. Pillonel, C., Mulder, M. M., Boon, J. J., Forster, B. and Binder, A. (1991). Involvement of cinnamyl-alcohol dehydrogenase in the control of lignin formation in Sorghum-bicolor l Moench. Planta 185, 538–544. Pinc,on, G., Maury, S., HoVmann, L., GeoVroy, P., Lapierre, C., Pollet, B. and Legrand, M. (2001a). Repression of O-methyltransferase genes in transgenic tobacco aVects lignin synthesis and plant growth. Phytochemistry 57, 1167–1176. Pinc,on, G., Chabannes, M., Lapierre, C., Pollet, B., Ruel, K., Joseleau, J. P., Boudet, A. M. and Legrand, M. (2001b). Simultaneous down-regulation of caVeic/ 5-hydroxy ferulic acid-O-methyltransferase I and cinnamoyl-coenzyme A reductase in the progeny from a cross between tobacco lines homozygous for each transgene. Consequences for plant development and lignin synthesis. Plant Physiology 126, 145–155. Piquemal, J., Lapierre, C., Myton, K., O’Connell, A., Schuch, W., Grima-Pettenati, J. and Boudet, A. M. (1998). Down-regulation of cinnamoyl-CoA reductase induces significant changes of lignin profiles in transgenic tobacco plants. The Plant Journal 13, 71–83.

104

C. HALPIN

Rae, A. L., Manners, J. M., Jones, R. J., McIntyre, C. L. and Lu, D. Y. (2001). Antisense suppression of the lignin biosynthetic enzyme, caVeate O-methyltransferase, improves in vitro digestibility of the tropical pasture legume, Stylosanthes humilis. Australian Journal of Plant Physiology 28, 289–297. Raes, J., Rohde, A., Christensen, J. H., Van de Peer, Y. and Boerjan, W. (2003). Genome-wide characterization of the lignification toolbox in Arabidopsis. Plant Physiology 133, 1051–1071. Ralph, J., MacKay, J. J., Hatfield, R. D., O’Malley, D. M., Whetten, R. W. and SederoV, R. R. (1997). Abnormal lignin in a loblolly pine mutant. Science 277, 235–239. Ralph, J., Hatfield, R. D., Piquemal, J., Yahiaoui, N., Pean, M., Lapierre, C. and Boudet, A. M. (1998). NMR characterization of altered lignins extracted from tobacco plants down-regulated for lignification enzymes cinnamylalcohol dehydrogenase and cinnamoyl-CoA reductase. Proceedings of the National Academy of Sciences of the United States of America 95, 12803–12808. Ralph, J., Lapierre, C., Marita, J. M., Kim, H., Lu, F. C., Hatfield, R. D., Ralph, S., Chapple, C., Franke, R., Hemm, M. R., Van Doorsselaere, J., SederoV, R. R., O’Malley, D. M., Scott, J. T., Mackay, J. J., Yahiaoui, N., Boudet, A. M., Pean, M., Pilate, G., Jouanin, L. and Boerjan, W. (2001). Elucidation of new structures in lignins of CAD- and COMT-deficient plants by NMR. Phytochemistry 57, 993–1003. Rasmussen, S. and Dixon, R. A. (1999). Transgene-mediated and elicitor-induced perturbation of metabolic channeling at the entry point into the phenylpropanoid pathway. Plant Cell 11, 1537–1551. Richardson, A., Stewart, D. and McDougall, G. J. (1997). Identification and partial characterization of a coniferyl alcohol oxidase from lignifying xylem of Sitka spruce (Picea sitchensis). Planta 203, 35–43. Ro, D. K., Mah, N., Ellis, B. E. and Douglas, C. J. (2001). Functional characterization and subcellular localization of poplar (Populus trichocarpa x Populus deltoides) cinnamate 4-hydroxylase. Plant Physiology 126, 317–329. Ruegger, M., Meyer, K., Cusumano, J. C. and Chapple, C. (1999). Regulation of ferulate-5-hydroxylase expression in Arabidopsis in the context of sinapate ester biosynthesis. Plant Physiology 119, 101–110. Ruelland, E., Campalans, A., Selman-Housein, G., Puigdomenech, P. and Rigau, J. (2003). Cellular and subcellular localization of the lignin biosynthetic enzymes caVeic acid-O-methyltransferase, cinnamyl alcohol dehydrogenase and cinnamoyl-coenzyme A reductase in two monocots, sugarcane and maize. Physiologia Plantarum 117, 93–99. Sablowski, R. W. M., Moyano, E., Culianezmacia, F. A., Schuch, W., Martin, C. and Bevan, M. (1994). A flower-specific Myb protein activates transcription of phenylpropanoid biosynthetic genes. EMBO Journal 13, 128–137. Samaj, J., Hawkins, S., Lauvergeat, V., Grima-Pettenati, J. and Boudet, A. (1998). Immunolocalization of cinnamyl alcohol dehydrogenase 2 (CAD 2) indicates a good correlation with cell-specific activity of CAD 2 promoter in transgenic poplar shoots. Planta 204, 437–443. Samuels, A. L., Rensing, K. H., Douglas, C. J., Mansfield, S. D., Dharmawardhana, D. P. and Ellis, B. E. (2002). Cellular machinery of wood production: DiVerentiation of secondary xylem in Pinus contorta var. latifolia. Planta 216, 72–82.


105

Savidge, R. and Udagama-Randeniya, P. (1992). Cell wall-bound coniferyl alcohol oxidase associated with lignification in conifers. Phytochemistry 31, 2959–2966. Schoch, G., Goepfert, S., Morant, M., Hehn, A., Meyer, D., Ullmann, P. and WerckReichhart, D. (2001). CYP98A3 from Arabidopsis thaliana is a 30 -hydroxylase of phenolic esters, a missing link in the phenylpropanoid pathway. Journal of Biological Chemistry 276, 36566–36574. Sewalt, V. J. H., Ni, W. T., Blount, J. W., Jung, H. G., Masoud, S. A., Howles, P. A., Lamb, C. and Dixon, R. A. (1997a). Reduced lignin content and altered lignin composition in transgenic tobacco down-regulated in expression of L-phenylalanine ammonia-lyase or cinnamate 4-hydroxylase. Plant Physiology 115, 41–50. Sewalt, V. J. H., Ni, W. T., Jung, H. G. and Dixon, R. A. (1997b). Lignin impact on fiber degradation: Increased enzymatic digestibility of genetically engineered tobacco (Nicotiana tabacum) stems reduced in lignin content. Journal of Agricultural and Food Chemistry 45, 1977–1983. Sibout, R., Baucher, M., Gatineau, M., Van Doorsselaere, J., Mila, I., Pollet, B., Maba, B., Pilate, G., LapieVe, C., Boerjan, W. and Jouanin, L. (2002). Expression of a poplar cDNA encoding a ferulate-5-hydroxylase/coniferaldehyde 5-hydroxylase increases S lignin deposition in Arabidopsis thaliana. Plant Physiology and Biochemistry 40, 1087–1096. Sibout, R., Eudes, A., Pollet, B., Goujon, T., Mila, I., Granier, F., Seguin, A., Lapierre, C. and Jouanin, L. (2003). Expression pattern of two paralogs encoding cinnamyl alcohol dehydrogenases in Arabidopsis. Isolation and characterization of the corresponding mutants. Plant Physiology 132, 848–860. Smith, C. G., Rodgers, M. W., Zimmerlin, A., Ferdinando, D. and Bolwell, G. P. (1994). Tissue and subcellular immunolocalization of enzymes of lignin synthesis in diVerentiating and wounded hypocotyl tissue of French bean (Phaseolus-vulgaris l). Planta 192, 155–164. Stewart, D., Yahiaoui, N., McDougall, G. J., Myton, K., Marque, C., Boudet, A. M. and Haigh, J. (1997). Fourier-transform infrared and Raman spectroscopic evidence for the incorporation of cinnamaldehydes into the lignin of transgenic tobacco (Nicotiana tabacum L) plants with reduced expression of cinnamyl alcohol dehydrogenase. Planta 201, 311–318. Suzuki, S., Lam, T. B. T. and Iiyama, K. (1997). 5-hydroxyguaiacyl nuclei as aromatic constituents of native lignin. Phytochemistry 46, 695–700. Takabe, K., Takeuchi, M., Sato, T., Ito, M. and Fujita, M. (2001). Immunocytochemical localization of enzymes involved in lignification of the cell wall. Journal of Plant Research 114, 509–515. Takeuchi, M., Takabe, K. and Fujita, M. (2001). Immunolocalization of O-methyltransferase and peroxidase in diVerentiating xylem of poplar. Holzforschung 55, 146–150. Tamagnone, L., Merida, A., Parr, A., Mackay, S., Culianez-Macia, F. A., Roberts, K. and Martin, C. (1998a). The AmMYB308 and AmMYB330 transcription factors from antirrhinum regulate phenylpropanoid and lignin biosynthesis in transgenic tobacco. Plant Cell 10, 135–154. Tamagnone, L., Merida, A., Stacey, N., Plaskitt, K., Parr, A., Chang, C. F., Lynn, D., Dow, J. M., Roberts, K. and Martin, C. (1998b). Inhibition of phenolic acid metabolism results in precocious cell death and altered cell morphology in leaves of transgenic tobacco plants. Plant Cell 10, 1801–1816.

106

C. HALPIN

Teutonico, R. A., Dudley, M. W., Orr, J. D., Lynn, D. G. and Binns, A. N. (1991). Activity and accumulation of cell division-promoting phenolics in tobacco tissue-cultures. Plant Physiology 97, 288–297. Tsai, C. J., Popko, J. L., Mielke, M. R., Hu, W. J., Podila, G. K. and Chiang, V. L. (1998). Suppression of O-methyltransferase gene by homologous sense transgene in quaking aspen causes red-brown wood phenotypes. Plant Physiology 117, 101–112. Van Doorsselaere, J., Baucher, M., Chognot, E., Chabbert, B., Tollier, M. T., PetitConil, M., Leple, J. C., Pilate, G., Cornu, D., Monties, B., Van Montagu, M., Inze, D., Boerjan, W. and Jouanin, L. (1995). A novel lignin in poplar trees with a reduced caVeic acid 5-hydroxyferulic acid O-methyltransferase activity. Plant Journal 8, 855–864. Vignols, F., Rigau, J., Torres, M. A., Capellades, M. and Puigdomenech, P. (1995). The brown midrib3 (Bm3) mutation in maize occurs in the gene encoding caVeic acid O-methyltransferase. Plant Cell 7, 407–416. Wagner, G. J. and Hrazdina, G. (1984). Endoplasmic-reticulum as a site of phenylpropanoid and flavonoid metabolism in Hippeastrum. Plant Physiology 74, 901–906. Whetten, R. and SederoV, R. (1995). Lignin biosynthesis. Plant Cell 7, 1001–1013. Winkel-Shirley, B. (1999). Evidence for enzyme complexes in the phenylpropanoid and flavonoid pathways. Physiologia Plantarum 107, 142–149. Yahiaoui, N., Marque, C., Myton, K. E., Negrel, J. and Boudet, A. M. (1998). Impact of diVerent levels of cinnamyl alcohol dehydrogenase downregulation on lignins of transgenic tobacco plants. Planta 204, 8–15. Zhong, R. Q., Morrison, W. H., Negrel, J. and Ye, Z. H. (1998). Dual methylation pathways in lignin biosynthesis. Plant Cell 10, 2033–2045. Zhong, R. Q., Ripperger, A. and Ye, Z. H. (2000a). Ectopic deposition of lignin in the pith of stems of two Arabidopsis mutants. Plant Physiology 123, 59–69. Zhong, R. Q., Morrison, W. H., Himmelsbach, D. S., Poole, F. L. and Ye, Z. H. (2000b). Essential role of caVeoyl coenzyme A O-methyltransferase in lignin biosynthesis in woody poplar plants. Plant Physiology 124, 563–577. Zhong, R. Q., Kays, S. J., Schroeder, B. P. and Ye, Z. H. (2002). Mutation of a chitinase-like gene causes ectopic deposition of lignin, aberrant cell shapes, and overproduction of ethylene. Plant Cell 14, 165–179.

Application of Thermal Imaging and Infrared Sensing in Plant Physiology and Ecophysiology

HAMLYN G. JONES

University of Dundee, Plant Science Research Group, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Leaf Temperature in Plant Physiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Improved Sensor Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Thermal Radiation and Remote Temperature Measurement Basics. . . . . A. Black Body Radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Remote Measurement of Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Errors in Estimation of Temperature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Image Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Plant Energy Balance and Leaf Temperature. . . . . . . . . . . . . . . . . . . . . . . . . . A. Energy Balance Basics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Stomatal Conductance and Evaporation As Functions of Leaf Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Water loss, Transpiration, Stomatal Conductance, and Stress Indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Biophysical and Aerodynamic Properties . . . . . . . . . . . . . . . . . . . . . . . . . . C. Metabolic Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Disease and Infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Pollution and Agronomic EVects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Genetic Screening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Frost Tolerance and Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Concluding Comments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Advances in Botanical Research, Vol. 41 Incorporating Advances in Plant Pathology 0065-2296/04 $35.00

108 108 110 111 111 111 113 119 124 124 127 131 131 144 145 148 150 151 153 155 155 155


108

H. G. JONES

ABSTRACT The applications of remote temperature sensing of plants by infrared thermography and infrared thermometry are reviewed and their advantages and disadvantages for various purposes discussed. The great majority of applications of thermography and of infrared thermometry depend on the sensitivity of leaf temperature to evaporation rate (and hence to stomatal aperture). In most applications, such as in the early or pre-symptomatic detection of disease or water deficits, what is actually being studied is the effect of the disease on stomatal behaviour or membrane permeability to water. Other applications of thermography in plant physiology include the study of thermogenesis as well as the characterisation of boundary layer transfer processes. Thermography is shown to be more than just a method for obtaining pretty pictures; it has particular advantages for the quantitative analysis of spatial and dynamic physiological information. Its capacity for large throughput has found application in screening approaches, such as in the selection of stomatal or hormonal mutants. The use of wet and dry reference surfaces for the enhancement of the power of thermal imaging approaches, especially in the field is reviewed, and the problems and potential solutions when applying thermography in the field and in the laboratory discussed.

I. INTRODUCTION A. LEAF TEMPERATURE IN PLANT PHYSIOLOGY

Leaf temperature is important to plants both because of the subtle eVects of small temperature changes on the rates of key physiological processes such as biochemical reactions and cell growth and division and because of the damaging eVects of extreme temperatures. Any study of physiological processes needs to take account of the temperature sensitivity of the process in relation to the likely natural variation (spatial and temporal) of temperature. Although many biochemical processes are fairly insensitive to temperature changes of several degrees around the temperature optimum, at the extreme, diVerences of a degree or less may become crucial. This should be put in the context of the potential spatial temperature variation over a single leaf, which may be as much as 3–5 8C (Raschke, 1956; Roth-Nebelsick, 2001) under appropriate conditions. Leaf temperature is also important as an indicator of aspects of physiological function, especially those related to evaporation rate; this will form the basis of much of the review in this chapter. It has been well known for many years (e.g., from the classical studies of Brown and Escombe, 1905; but also see reviews by Huber, 1935; Jackson, 1982; Raschke, 1960) that leaf temperature depends on stomatal opening, with temperatures decreasing as stomata open and as evaporation rates

THERMAL IMAGING AND INFRARED SENSING

109

increase. The use of leaf temperature as an indicator of stomatal conductance or transpiration, however, is confounded by the fact that leaf temperature is also aVected by a wide range of other plant and environmental characters according to the leaf energy balance. Furthermore, as the environment is constantly changing, at least for plants in the field, it also becomes necessary to consider the dynamic behavior of leaf temperature in any precise study of leaf temperature. Until the development of remote infrared sensing of leaf or canopy temperature, applications involving the measurement of leaf temperature had been limited by the diYculty of using thermocouples in any large-scale field or even laboratory study. Following Raschke’s (1960) review of the principles of leaf energy exchanges in the early 1960s and the availability of new infrared remote sensors that could be used for the sensing of canopy temperature, there was a rapid development of their use for the study of plant water relations, especially to provide guidance for irrigation scheduling (see Jones, 1999b). Tanner (1963) and Fuchs and Tanner (1966) were among the earliest to publish papers describing the remote measurement of leaf temperature by infrared thermometry for the study of plant water relations in the field. A key milestone in the application of thermal sensing for irrigation management was the definition of what was termed a stress degree day (SDD) by Jackson et al. (1977) as the diVerence between the canopy temperature and air temperature. This provided a powerful way of normalizing for day-to-day and regional diVerences in environmental conditions. By measuring canopy-air temperature diVerences daily these researchers derived an integrated measure of ‘‘stress’’ experienced by the crop, the cumulative SDD. The next important step was the further normalization to take account of diVerences in atmospheric humidity and the comparison with a notional reference (well-watered) crop by means of the crop water stress index (Idso, 1982; Idso et al., 1981; Jackson et al., 1981). Further improvements involved the more general introduction of wet and dry reference surfaces to normalize for all aspects of current environmental conditions (Jones, 1999b; Jones et al., 1997) and allow absolute estimates of conductance or evaporation. An extensive literature developed over 30 years or so, making use of the temperature rise as stomata close under drought as an indicator of crop ‘‘stress.’’ This was greatly facilitated by the increasing availability of infrared thermometers. Most studies in this field have been based on infrared thermometry rather than on thermography, which is the process of obtaining thermal images. It is only relatively recently that thermography has become feasible for many laboratories, with the development of a new generation of

110

H. G. JONES

uncooled imagers. In this chapter we concentrate on thermography but will refer to the literature on infrared thermometry wherever relevant advances have been made in that area.

B. IMPROVED SENSOR TECHNOLOGY

The availability of appropriate equipment limited the use of thermographic techniques until thermal scanners became available in the 1960s. Some early studies used aerial thermal scanners for the detection of water stress in plant canopies and for the estimation of crop evaporation rates (Bartholic et al., 1972; Heilman et al., 1976). Other studies used similar equipment for the study of temperature distributions and boundary layer transfer processes across plant leaves (Clark and Wigley, 1975). The earliest applications of thermal imagery in plant physiology were in the 1970s and early 1980s when Omasa and colleagues (Hashimoto et al., 1984; Omasa et al., 1981a, b) started using laboratory systems for the study of distributions of temperature and, by implication stomatal conductances, over plant leaves in the laboratory. In these early days instrumentation was heavy and the cameras required the sensor to be cooled using liquid nitrogen. The opportunities to develop methods based on infrared thermography have been greatly increased by the introduction of uncooled, handheld cameras with thermal resolution of better than 0.1 8C and by more aVordable pricing. For example, it is now possible to purchase such a camera for less than £7,000 sterling ($9,000). Advances in thermography have occurred in parallel with advances in a wide range of other imaging technologies that are or could be used to complement information from thermography. These additional techniques range from those that give information of internal structure such as x-ray or nuclear magnetic resonance (NMR) tomography, through conventional visible reflectance (red/green/blue) images familiar from standard digital cameras to fluorescence images. In this chapter we concentrate on a consideration of what information can and cannot be obtained from thermal imagery. Although our emphasis is on plant physiological and ecophysiological applications at the leaf to plant or crop scales, for completeness we also mention applications of thermal imagery at the larger scale (e.g., use of satellite imaging for studies of plant stress and evapotranspiration). Applications at the microscopic scale are limited by the lateral thermal conductivity of plant leaves (see the section on spatial resolution, later in this chapter).


111

II. THERMAL RADIATION AND REMOTE TEMPERATURE MEASUREMENT BASICS A. BLACK BODY RADIATION

All bodies with a temperature above absolute zero (273.15 8C) emit thermal radiation as a function of their temperature. For a perfect emitter, known as a ‘‘black body,’’ the wavelength dependence of the emitted energy is described by the Planck distribution function: Ll ¼ 2hc2 =ðl5 ðehc=lkT 1ÞÞ

ð1Þ

in which Ll is the spectral radiance (W m2 sr1 mm1), defined as the radiant flux density emanating from a surface per unit solid angle per unit wavelength interval (centered on l), h is the Planck constant (6.6256 1034 J s), k is the Boltzmann constant (1.38054 1023 J K1), c is the speed of light (2.998 108 m s1) and l refers to the wavelength (mm). The wavelength (lm) with the peak emittance decreases with increasing temperature according to Wein’s displacement law: lm ¼ 2897=T

ð2Þ

The total radiant flux density emitted (radiant excitance) from unit area of surface (R, W m2) over a hemisphere is p multiplied by the integral over all wavelengths of black body spectral radiance: Z 1 R ¼ ep Ll dl ¼ esT 4 ð3Þ 0

in which s is the Stefan-Boltzmann constant (5.6697 108 W m2 K4) and e is known as the emissivity. The concept of emissivity is introduced to take account of the fact that most real surfaces are not perfect emitters of thermal radiation, so that the actual radiant excitance may be less than the theoretical value defined for a black body. Emissivity relates the actual radiance of a body at a given temperature to that of a black body and falls between 0 and 1. Although the emissivity varies somewhat as a function of wavelength, for many purposes it is assumed that real surfaces approximate ‘‘gray bodies,’’ having a constant emissivity across the thermal infrared region. Some typical values for the emissivities of diVerent materials are summarized in Table I. B. REMOTE MEASUREMENT OF TEMPERATURE

The aforementioned relationships allow one to estimate the temperature of a surface from measurements of the amount of thermal radiation emitted.

112

H. G. JONES

TABLE I Broad-band Thermal Emissivities a of DiVerent Surfaces b,c Plant leaves Plant canopies Dry leaves Dry grass Wood Bark Dry soil Wet soil Sand Distilled water Water

Emissivity

References

0.95 (0.92–0.99) 0.98–0.99 0.96 0.88 0.90 0.94–0.97 0.92 0.95 0.87–0.92 0.96 0.98–0.99

1, 2, 4, 5 4 1 1 1 1 1, 5 1, 5 2, 4, 5 5 1, 3, 5

a

8–14 mm unless otherwise stated. These values are guidelines only because values vary markedly with surface condition. c Spectral emissivities may be found in the MODIS emissivity library (www.icess.ucsb.edu/ modis/EMIS/html/em.html (Zhengmin Wan, University of California, Santa Barbara). 1 ¼ Rees, 2001; 2 ¼ Sutherland, 1986; 3 ¼ Campbell and Norman, 1998; 4 ¼ Idso 1969; 5 ¼ MODIS emissivity library. b

Because the amount of energy emitted in the thermal wave band is rather small, extremely sensitive detectors are necessary. With recent advances in sensor technology it is no longer necessary to cool the detector (e.g., with liquid nitrogen) to attain adequate accuracy and a range of room-temperature sensors are becoming available. The particular sensor technology depends on the wavelength and precision required. Indium-Gallium-Arsenide sensors and Indium-Antimonide sensors are often used for the shorter wavelengths of thermal infrared (up to approximately 5 mm; primarily used for sensing hightemperature objects in industry), whereas quantum well photodetectors and microbolometers are used for the longer wavelengths required for measurement outdoors. Pyroelectric sensors generate current, whereas bolometer sensors are based on changing electrical resistance of the sensing element. Because thermal sensors do not respond equally to all wavelengths of thermal radiation, conversion from sensor output to temperature is achieved in the built-in software by means of integration of Eq. 1 over the appropriate wavelengths. Most cameras use one of two atmospheric windows where the atmosphere does not absorb radiation markedly (3–5 mm or 8–14 mm). Only the latter of these is of much use to an ecophysiologist for daytime measurements outdoors because there is too much interference by the upper


113

tail of the highly energetic solar radiation spectrum in the shorter of these two wavelength windows. An alternative remote measurement technique is to detect colour changes of calibrated liquid crystals painted on leaves (Smith et al., 2004). C. ERRORS IN ESTIMATION OF TEMPERATURE

1. Background temperature and emissivity errors The algorithms used to estimate surface temperatures must not only correct for the spectral sensitivity of the detector and the emissivity of the surface, but they must also determine the fraction of the total thermal radiation received at the detector, which originates from the object being measured. The radiation emitted is attenuated somewhat by the atmosphere through which it passes (as a function of atmospheric path length and humidity) and is supplemented by the thermal radiation originating in the surroundings that is reflected by the object, as well as any thermal radiation emitted toward the detector by the intervening air. The influence of environmental temperature on the accuracy of radiometric estimates of temperature has been discussed in some detail by Omasa et al. (1984) and Jones et al. (2003). These and other authors have pointed out that infrared thermography or thermometry detects the total thermal radiation flux density leaving a surface: that is the sum of the emitted thermal radiation, Re, and the reflected thermal radiation, Rr, together with any transmitted radiation (this latter is usually assumed to be negligible, at least in plant canopies). To estimate the surface temperature it is necessary to determine only the emitted radiation component; this requires an estimate of Rr. One approach to correcting for this ‘‘background’’ radiation is to replace the surface of interest by a highly reflective (low emissivity) diVuse reflector surface such as crumpled aluminium foil, which reflects the incoming thermal radiation, and then to record the apparent temperature of this surface when e is set equal to 1. The subsequent correction is often incorporated within the camera software. This background temperature can vary from close to ambient air temperature when making measurements within a canopy (because the background is largely composed of other leaves at close to air temperature), in which case emissivity errors have only a small impact, to 270 K or lower. These lower, apparent environmental temperatures arise when measuring from above the canopy in clear sky conditions, because the background is then dominated by the sky, which may have a radiative temperature of lower than 250 K (Jones et al., 2003).

114

H. G. JONES

Substitution into Eq. 3 leads to the calculation that a 1% error in e would be expected to equate to approximately a 0.75 K error in the estimated temperature at 300 K. Errors in surface temperature estimates that arise from errors in emissivity (e) are often smaller than one might expect from this simple inversion because of the presence of reflected, incoming background thermal radiation, so that: R ¼ Rr þ Re ¼ ð1 eÞs T4background þ es T4s

ð4Þ

in which Tbackground is the eVective background temperature. Indeed for a leaf deep within a canopy, where the background temperature is close to that of the leaf itself, the apparent emissivity (i.e., the value of e required for substitution in Eq. 3 when the total outgoing radiation is substituted for Re) is close to unity. Therefore Eq. 4 reduces to the following equation: R ffi s T4s :

ð5Þ

In this case the brightness temperature (the apparent temperature assuming that the emitter has an emissivity equal to 1) is close to the actual temperature. If, however, one assumes a leaf temperature of 300 K with a typical leaf emissivity of 0.95 and calculates temperature by inversion of Eq. 3, one gets an apparent temperature of 303.9 K. As another example, when the background temperature is 260 K (clear sky) and the leaf temperature is 300 K, the brightness temperature (reflected and emitted radiation) is 298.4 K, but the apparent temperature calculated when using Eq. 3 is 302 K. In this latter case the apparent emissivity for substitution in Eq. 3 is 0.978 (when e ¼ 0:95). These examples illustrate the importance of obtaining both an accurate estimate of the emissivity and of the background temperature. Errors also arise as a result of thermal emission by the atmosphere between the object and the sensor and because the thermal transmissivity of the atmosphere is not perfect, declining with distance. These corrections are a strong function of humidity. The firmware or software provided with thermal imagers or thermometers makes some or all of these corrections automatically, so that temperature errors associated with errors in e are usually less than 0.3 K/ percentage error in e, whereas atmospheric eVects can usually be neglected for close-range viewing (within several tens of meters), although for satellite imagery the errors can be between 3 and 10 K (Campbell and Norman, 1998; Jones et al., 2004). Typical reported emissivities for diVerent surfaces are summarized in Table I. Most individual plant leaves have thermal emissivities between approximately 0.92 and 0.96. It is important to note, however, that the eVective emissivity (ecanopy ) of a closed plant canopy consisting of leaves


115

with e ¼ 0:95 will be between 0.98 and 0.99 because of the internal reflections between diVerent leaves. Campbell and Norman (1998) showed that this eVect may be approximated by the following equation: ecanopy ffi 1 ðð1

pffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffi eleaf Þ=ð1 þ eleaf ÞÞ:

ð6Þ

An interesting phenomenon related to the background radiation was observed by Lamprecht et al. (2002) when studying thermogenesis in the water lily Victoria cruziana. They observed that reflections of flower parts in the water could appear up to 4.5 K warmer than direct thermal images of the flowers. They attributed this to the fact that the reflection includes a fraction of the directly emitted radiation from the water, together with the fraction of the emitted radiation from the flower that is reflected by the water. As can be readily shown by substitution into Eq. 4, however, this overestimate of object temperatures should not occur if the instrument is set up with correct estimates of e for both the water and the flower tissue and of the background correction. Indeed the occurrence of such phenomena can be used as a diagnostic for incorrect camera setup. 2. Gaussian noise and cross-talk A potential error with thermographic arrays is possible cross-talk between adjacent sensor pixels, as has been noted by Omasa (2002). For longdistance imaging there can also be significant scattering of radiation from adjacent pixels, which gives a slight blurring eVect; this can be minimized by the use of a Laplacian filter (Mather, 1999). The inherent Gaussian noise dependent on sensor quality and integration time is another problem that needs to be considered. This can be reduced by increasing integration time (or replication); although this is at the expense of dynamic resolution, most sensors currently in use are sensitive enough that this is not a practical limitation in most plant physiological applications. Alternatively, noise may be reduced at the expense of spatial resolution by the application of a low-pass smoothing filter such as a moving-average filter or a median filter across the image (see Mather, 1999; Tukey, 1977). The latter is generally preferred because it is less influenced by outliers. In the medical field, attempts have been made to develop algorithms based on techniques such as mean-field annealing to remove noise while preserving image edges (Snyder et al., 2000). There is much scope for enhanced use of such artificial data improvement methods, although the existing algorithms may be of limited value for typical plant subjects. Further techniques are discussed in the section on image enhancement, later in this chapter).

116

H. G. JONES

3. Spatial errors in calibration across an image To achieve adequate accuracy for some plant water relations studies it has been found necessary, with at least one type of camera, to correct drifts in spatial calibration across the sensor array as the camera warms up. Jones et al. (2002) proposed a simple subtraction method to improve the comparability of data across the whole of an image. The procedure is to take a reference image of a homogeneous reference surface (in practice an outof-focus image with the lens cap on suYces with reasonable accuracy in the field), subtract this (pixel by pixel) from the sample image, and then add back to all pixels the average temperature of the reference image to obtain a corrected image. The fact that absolute temperature accuracy is not necessarily high is not usually a problem in plant physiological studies, in which most measurements are relative to reference areas within the image. 4. Radiometric versus aerodynamic temperature and other errors In the case of thermography of flat, single leaf surfaces it is only necessary to take account of the instrumental errors of the type outlined in the preceding paragraphs, because the suitably corrected radiative temperature equates well to the temperature required for substitution into heat and mass transfer equations. When studying plant canopies and other complex surfaces, however, the situation is more diYcult. The complexity of plant canopies results in significant temperature variation (e.g., between sunlit and shaded leaves), so that the apparent radiometric temperature (which depends on radiation from the visible upper layers of the canopy) may not truly reflect so-called aerodynamic temperature (the average canopy temperature of those surfaces exchanging heat and water vapor), a value that is required for calculation of heat and mass transfer (e.g., Campbell and Norman, 1998; Kimes, 1980). There has also been substantial discussion of the angular variation of emitted thermal radiation from a canopy and apparent emissivity. In addition to the instrumental errors outlined previously, there are many cases in which the apparent radiative temperature of a canopy varies with view zenith angle. In practice it is usually assumed that the individual canopy surfaces behave as Lambertian emitters (i.e., having the eVect that the apparent temperature is independent of view angle). The variation in apparent temperature of canopies with view angle (Franc,ois et al., 1997; Kimes, 1980; Otterman et al., 1999) is largely thought to result from varying fractions of soil and sunlit or shaded leaves as the zenith or azimuthal angles of view change. A less obvious error arises where the surface in the instantaneous field of view (a single pixel) is composed of an ensemble of surfaces at diVerent temperatures. In this case the wavelength distributions of the emissions from


117

each surface will diVer slightly, even if the emissivities equal one, and the wavelength distribution of the radiation emitted by the ensemble will not correspond exactly to that of a black body. Norman and Becker (1995) have suggested that for extreme temperature diVerences of 10 8C this can lead to errors in the radiometric estimate of surface temperature of the order of 1 8C. 5. Spatial resolution The potential image resolution is clearly set by the number of pixels and the area viewed, which in turn depends on angle of view (a) of the camera lens and distance to the object (z). The width of the imaged area (w) is given by the following equation: w ¼ 2z tanða=2Þ ð7Þ in which a is the total field of view of the lens and z is the distance to the object. For example, for a typical thermal imager with a 17-degree field of view and 256 pixels across the image the width of the object viewed at one meter ¼ 2 1 tan(17/2) ¼ 0.299 m. Therefore one pixel corresponds to 256/299 mm ¼ 1.17 mm. The size of pixels viewed clearly varies with lens focal length and distance to the object. Whether temperature variation at this spatial scale is physiologically meaningful depends on a number of factors, probably most important of which is the lateral thermal conductivity of the leaf or object being imaged. Spatial resolution is potentially much greater for thin leaves than for thick leaves because of the smaller lateral thermal conductivity of the former. For leaves in which the energy balance is varying spatially over the surface of the leaf, as happens where the stomata are closed in patches, one needs to take account of the three-dimensional heat flows rather than the one-dimensional equation used thus far. For typical thin leaves such as Phaseolus vulgaris, Jones (1999a) calculated that the half distance (i.e., the distance over which half the total temperature change between two steady values either side of a step change in surface conductance from 0 to 160 mmol m2 s1) was approximately 3.1 mm. Therefore the maximum temperature gradient likely to be observed across the lamina of such a leaf would be of the order of only approximately 0.4 8C/mm, which agreed well with the observed values. For small lesions or thicker leaves the gradient would be smaller and the ability of thermal imaging to detect localized stomatal variation more limited. It follows that the potential of thermal microscopy is somewhat limited, although with a sensitive detector, even quite small perturbations (

Advances in Botanical Research, Volume 41

Advances in Botanical Research, Volume 41

Advances in Botanical Research, Volume 41

Advances in Botanical Research, Volume 30

Advances in Botanical Research, Volume 24

Advances in Botanical Research, Volume 30

Advances in Botanical Research, Volume 55

Advances in Botanical Research, Volume 42

Advances in Botanical Research, Volume 33

Advances in Botanical Research, Volume 9

Advances in Botanical Research, Volume 39

Advances in Botanical Research, Volume 2

Advances in Botanical Research, Volume 16

Advances in Botanical Research, Volume 43

Advances in Botanical Research, Volume 29

Advances in Botanical Research, Volume 39

Advances in Botanical Research, Volume 46

Advances in Botanical Research, Volume 19

Advances in Botanical Research, Volume 48

Advances in Botanical Research, Volume 40

Advances in Botanical Research, Volume 54

Advances in Botanical Research, Volume 17

Advances in Botanical Research, Volume 25

Advances in Botanical Research, Volume 5

Advances in Botanical Research, Volume 43

Advances in Botanical Research, Volume 47

Advances in Botanical Research, Volume 1

Advances in Botanical Research, Volume 42

Advances in Botanical Research, Volume 56

Advances in Botanical Research, Volume 50

Advances in Botanical Research, Volume 49

Advances in Botanical Research, Volume 41