UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Good neighbours : the role of non-lignified cells in Arabidopsis lignification Smith, Rebecca Anne 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_spring_smith_rebecca.pdf [ 178.44MB ]
Metadata
JSON: 24-1.0167289.json
JSON-LD: 24-1.0167289-ld.json
RDF/XML (Pretty): 24-1.0167289-rdf.xml
RDF/JSON: 24-1.0167289-rdf.json
Turtle: 24-1.0167289-turtle.txt
N-Triples: 24-1.0167289-rdf-ntriples.txt
Original Record: 24-1.0167289-source.json
Full Text
24-1.0167289-fulltext.txt
Citation
24-1.0167289.ris

Full Text

GOOD NEIGHBOURS: THE ROLE OF NON-LIGNIFIED CELLS IN ARABIDOPSIS LIGNIFICATION  by Rebecca Anne Smith  B.Sc., The University of Manitoba, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Botany)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2014  © Rebecca Anne Smith, 2014   ii Abstract Lignin is a critical structural component of plants, providing vascular integrity and mechanical strength. Lignin precursors, monolignols, must be exported to the extracellular matrix where random oxidative coupling produces a complex lignin polymer. The objectives of this study were twofold: to determine the timing of lignification, with respect to programmed cell death during Arabidopsis thaliana primary xylem development, and to determine which cells are contributing to the lignification of tracheary elements and fibres. This thesis demonstrates that lignin deposition is not exclusively a post-mortem event, but also occurs prior to programmed cell death. Radiolabelled monolignols were not detected in the cytoplasm or vacuoles of tracheary elements or neighbours. To experimentally define which cells in lignifying tissues contribute to lignification in intact plants, a microRNA against CINNAMOYL CoA-REDUCTASE1, driven by the promoter from CELLULOSE SYNTHASE 7 (proCESA7:miRNA CCR1), was used to silence monolignol biosynthesis in cells developing secondary cell walls. When monolignol biosynthesis was knocked down specifically in the cells with thickened secondary cell walls, but not in the neighbouring cells, lignin was still deposited in the xylem secondary cell walls. This indicates that “good neighbour” cells are sufficient to produce lignin in the vascular bundles. Surprisingly, this was not the case in the interfascicular fibres, where a dramatic reduction in cell wall lignification demonstrates that these extra-xylary fibers undergo cell autonomous lignification. When a fibre-specific promoter (proAtPEROXIDASE64) was used to drive the miRNA, autonomous extraxylary fibre lignification was again observed, as was non-cell autonomous lignification between xylary fibres and neighbouring tracheary elements. These effects may have reflected compensatory mechanisms in response to lignin downregulation, so to demonstrate that discrete cell populations, such as xylem parenchyma, do contribute to lignification, genes encoding enzymes catalyzing the synthesis of novel monolignol conjugates were introduced into wild-type Arabidopsis using cell population-specific promoters. The detection of novel monolignol conjugates in the cell wall by chemical analysis and fluorescence microscopy supported the contribution of tracheary elements and fibres to lignification and also revealed that xylary parenchyma cells are producing monolignol substrates and acting as “good neighbours” to tracheary elements and xylary fibres during lignification.    iii Preface  Some sections of text from Chapter 1, with modifications, and Figures 1.2 and 1.5, are from Schuetz et al. (2013), a manuscript that I co-authored: Schuetz, M., Smith, R., and Ellis, B.E. 2013. Xylem tissue specification, patterning and differentiation mechanisms. The Journal of Experimental Botany 64: 11-31 © Copyright Society for Experimental Biology, 2013 (http://jxb.oxfordjournals.org), used with permission.  The majority of Chapter 2 has been published in: Smith, R.A., Schuetz, M., Roach, M., Mansfield, S.D., Ellis, B.E., and Samuels, A.L. 2013. Neighboring parenchyma cells contribute to Arabidopsis xylem lignification, while lignification of interfascicular fibers is cell autonomous. The Plant Cell 25: 3988-3999 © Copyright American Society of Plant Biologists, 2013 (http://www.plantcell.org). Most of the text from the results and images from Figures 2.4, 2.5 and 2.8 were reproduced with permission. Rebecca Smith, Dr. Lacey Samuels and Dr. Brian Ellis identified the research question. Rebecca Smith, Dr. Lacey Samuels, Dr. Brian Ellis and Dr. Shawn Mansfield designed the research. Rebecca Smith performed the autoradiography on roots and stems. Rebecca Smith and Dr. Shawn Mansfield performed HPLC on root extracts. All authors contributed to writing the manuscript. Sections 3.2.1 and 3.2.2 of Chapter 3 have also been published in: Smith, R.A., Schuetz, M., Roach, M., Mansfield, S.D., Ellis, B.E., and Samuels, A.L. 2013. Neighboring parenchyma cells contribute to Arabidopsis xylem lignification, while lignification of interfascicular fibers is cell autonomous. The Plant Cell 25: 3988-3999 © Copyright American Society of Plant Biologists, 2013 (http://www.plantcell.org). These results sections, the majority of the discussion, Figures 3.1-3.7 and 3.9-3.11, and Tables 3.1, 3.3 and 3.4 have been reproduced with permission. Dr. Lacey Samuels, Dr. Brian Ellis, Rebecca Smith and Dr. Mathias Schuetz designed the research. Rebecca Smith and Mathias Schuetz performed the research. Dr. Shawn Mansfield and Melissa Roach provided analytical tools for chemical cell wall analysis. All authors analyzed the data and wrote the manuscript. For Chapter 4, Dr. Lacey Samuels, Dr. Brian Ellis and Rebecca Smith identified the research question and designed the research. Dr. Curtis Wilkerson (Michigan State University)   iv and Dr. Shawn Mansfield provided FMT and PMT entry vector constructs. Dr. John Ralph, Dr. Steven Karlen, Dr. Dharshana Padmakshan and Sarah Liu (all from University of Wisconsin, Madison) assisted Rebecca Smith with the DFRC and 2D-NMR experiments. Rebecca Smith performed all other experiments and analyzed the data.   v Table of Contents  Abstract .......................................................................................................................................... ii!Preface ........................................................................................................................................... iii!Table of Contents ......................................................................................................................... iii!List of Tables ................................................................................................................................ ix!List of Figures ................................................................................................................................. x!List of Symbols and Abbreviations .......................................................................................... xiii!Acknowledgements .................................................................................................................. xviii!Chapter 1: Introduction ................................................................................................................1!1.1! Xylem development ........................................................................................................ 1!1.2! Lignin: biosynthesis to polymerization ........................................................................... 9!1.3 Good neighbours during lignification ………………………………………………..16 1.4  Research objectives and significance ………………………………………………...22 Chapter 2: Lignification before and after programmed cell death in tracheary element development ..................................................................................................................................25!2.1! Introduction ................................................................................................................... 25!2.1.1! Using autoradiography to label monolignols ............................................................ 25 2.1.2    The Arabidopsis root as a model system for lignification studies ………………....27 2.2! Results ........................................................................................................................... 28!2.2.1! Secondary cell wall labelling in living tracheary elements ...................................... 28 2.2.2    Phenylpropanoid metabolites do not accumulate within the tracheary element cells ……………………………………………………………………………………………...34   vi 2.2.3     Radiolabel appears in the cell wall of tracheary elements that have just undergone PCD but not those that are long dead ……………………………………………………...40 2.2.4     Lignification in the Arabidopsis stem also begins pre-mortem and continues post-mortem ……………………………………………………………………………………..41 2.3! Discussion ..................................................................................................................... 43!2.4! Methods......................................................................................................................... 47 2.4.1! Plant growth conditions and phenylpropanoid labelling .......................................... 47 2.4.2! High pressure freezing .............................................................................................. 48 2.4.3! Autoradiography - light microscopy ......................................................................... 48 2.4.4! Transmission electron microscopy ........................................................................... 48 2.4.5! HPLC - soluble phenolic analysis ............................................................................. 49 Chapter 3: Good neighbours in lignification .............................................................................51!3.1! Introduction ................................................................................................................... 51!3.1.1! Evidence for post-mortem lignification .................................................................... 51 3.1.2!!!!!Introduction to the "good neighbours" in Arabidopsis …………………………….52 3.1.3!!!!!Cell-specific silencing of lignification using artificial microRNAs ……………….53 3.1.4    Cell-specific promoters for specific artificial microRNA targeting ………………..54       3.2! Results ........................................................................................................................... 55!3.2.1! "Good neighbours" in root tracheary element lignification ...................................... 55 3.2.2    "Good neighbours" in stem xylem and extra-xylary fibre lignification ……………57 3.2.3! Silencing monolignol biosynthesis specifically in xylary parenchyma cells or fibres does not affect lignification of xylem tissue ......................................................................... 72!3.3! !Discussion ……………………………………………………………………………76 3.4! !Methods ……………………………………………………………………………...80   vii 3.4.1! Plant growth conditions and staining ........................................................................ 80 3.4.2! Confocal imaging ...................................................................................................... 80 3.4.3! Transgenic lines ........................................................................................................ 81 3.4.4! Q-RT-PCR ................................................................................................................ 83 3.4.5! Chemical analysis ..................................................................................................... 83 Chapter 4: Using “novel” monolignol biosynthetic enzymes to investigate the good neighbours during lignification ..……………………………………………………………...85 4.1! Introduction ................................................................................................................... 85!4.1.1! FMT and PMT from Angelica sinensis and Oryza sativa ......................................... 85 4.1.2!!!!!Detecting the novel monolignols …………………………………………………..88 4.1.3!!!!!Ferulate-monolignol conjugates can be localized by fluorescence microscopy …...89 4.2! Results ........................................................................................................................... 90!4.2.1! Cells with secondary cell walls (tracheary elements and fibres) are contributing to lignification ........................................................................................................................... 90 4.2.2    Xylary parenchyma cells contribute to the lignification of tracheary elements    and xylary fibres…...……………………………………………………………………………93 4.3! !Discussion ……………………………………………………………………………97 4.4! !Methods ….………………………………………………………………………...…..100 4.4.1! Transgenic lines ...................................................................................................... 100 4.4.2! Plant growth conditions .......................................................................................... 100 4.4.3! Cell wall chemical analysis ..................................................................................... 100 4.4.4! Fluorescence microscopy and spectral analysis ...................................................... 101 Chapter 5: Conclusions and future directions ……………………………………………...102   viii 5.1! Main findings of this thesis ......................................................................................... 102!5.2! Background for the timing of lignification and "good neighbours" ........................... 105 5.3! Broad significance of this research to the process of lignification ............................. 106 5.3.1! What specifies xylary fibre and xylary parenchyma cell fates? ............................. 107 5.3.2! Differences between the timing of lignification in vitro and in planta ................... 109 5.3.3! Mechanisms of monolignol export ......................................................................... 110 5.3.4! "Good neighbours" in lignin polymerization .......................................................... 112 5.3.5! Implications for engineering plants with reduced or altered lignin ........................ 113 References ...................................................................................................................................116!   ix List of Tables  Table 3.1. Chemical analysis of lignin in different miRNA lines ………………………………65 Table 3.2. Soluble sugar analysis from miRNA lines …………………………………………..67 Table 3.3. Lignin monomer content analysis of different miRNA lines ………………………..68 Table 3.4. Primer sequences …………………………………………………………………….82 Table 4.1. The target cell populations of the promoters used in this study ……………………..87     x List of Figures  Figure 1.1. Developmental schematics of protoxylem and metaxylem tracheary elements (TE) and fibres in procambial cells ……………………………………………………….....................2  Figure 1.2. The transcriptional cascade regulating tracheary element and fibre differentiation and secondary cell wall formation …………………………………………………………………….4  Figure 1.3. Current view of the monolignol biosynthesis pathway …………………………….12  Figure 1.4. Radical coupling of a monolignol to a lignin polymer (endwise polymerization) ……………………………………………………………………………………………………15  Figure 1.5. The “good neighbour” model of lignification in Arabidopsis inflorescence stems ……………………………………………………………………………………………………18  Figure 2.1. A schematic of an Arabidopsis root cross-section ………………………………….29  Figure 2.2. Cellulose deposition precedes lignification in Arabidopsis root protoxylem ………30  Figure 2.3. A wild-type root stained with the lignin indicator Basic Fuchsin ………………….31  Figure 2.4. Root microautoradiography on 3H-Phe-treated roots ………………………………33  Figure 2.5. Monolignols do not accumulate within vacuoles or cytoplasm of root tracheary elements during lignification ……………………………………………………………………35  Figure 2.6. HPLC chromatographs of soluble phenolic extracts from WT and MYB63/58 overexpressing roots …………………………………………………………………………….37   xi  Figure 2.7. Treatment with inhibitors or cold temperature did not slow monolignol export ……………………………………………………………………………………………………39  Figure 2.8. Lignification does not continue indefinitely after programmed cell death …...........42  Figure 2.9. Stem microautoradiography showing label in living and dead fibres and metaxylem tracheary elements ………………………………………………………………………………44  Figure 3.1. Neighbouring cells can rescue tracheary element lignification in Arabidopsis roots ………………………………………………………………………………………....................56  Figure 3.2. CCR1 miRNA effectively silences YFP:CCR1 expression .………………………...58  Figure 3.3. Cell specific silencing of CCR1 in Arabidopsis roots ………………………...........59  Figure 3.4. Stem phenotypes of 2-month-old miRNA-expressing plants ………………………61  Figure 3.5. qRT-PCR results for CCR1 expression level in miRNA lines stems ………………62  Figure 3.6. Arabidopsis stem tissues show diverse degrees of cell-autonomous lignification ………………………………………………………………………………................................63  Figure 3.7. Localization of proCESA7::GFP in inflorescence stems of Arabidopsis ………….64  Figure 3.8. Whole cell wall analysis revealed changes in lignin content and composition in the miRNA line stems ………………………………………………………………….....................66  Figure 3.9. CCR1 miRNA silences YFP:CCR1 expression only in lignified cells in the stem ………………………………………………………………………………………....................70   xii  Figure 3.10. Anatomy of a vascular bundle in Arabidopsis thaliana …………………………71  Figure 3.11. Expression pattern of cell-specific promoters in stem longitudinal sections ……73  Figure 3.12. Phloroglucinol-HCl staining of stem developmental stages reveals differences in lignification through xylem development in the miRNA lines …………………………………74  Figure 3.13. Analysis of intrinsic lignin fluorescence in mature stems confirms the miRNA stem histochemical staining results …………………………………………………………………...75  Figure 4.1. The amount of monolignols and monolignol-conjugates released through DFRC analysis …………………………………………………………………………………………..91  Figure 4.2. 2D-NMR spectra of promoter-PMT lines ………………………………………….92  Figure 4.3. Localization of ferulate-monolignol conjugates in secondary cell walls …………..95  Figure 4.4. Average maximum fluorescence intensities at the wavelength of maximum emission for the different cell populations in WT (negative control), pro35S::miRNA CCR1 (positive control) and FMT-expressing plants …………………………………….....................................96  Figure 5.1 The potential good neighbours during xylem development throughout the Arabidopsis plant body.…………………………………….......................................................103    xiii List of Symbols and Abbreviations 1˚CW    Primary cell wall 13C    Carbon-13 1H    Hydrogen/proton 2˚CW    Secondary cell wall 2D-NMR   2-Dimensional-Nuclear Magnetic Resonance 35S    Constitutive promoter from the cauliflower mosaic virus 3H-Phe   Tritiated phenylalanine 4CL    4-COUMARATE:CoA LIGASE ABC    ATP-binding cassette ABCG   ABC transporter G subfamily AGO    ARGONAUTE amiRNA   Artificial microRNA ATP    Adenosine triphosphate ATPase   ATP phosphohydrolase BFA    Brefeldin A BGLU45/46   beta-glucosidase 45/46 C-C bonds   Carbon-carbon bonds C3H    p-COUMARATE 3-HYDROXYLASE C4H    CINNAMATE-4-HYDROXYLASE CA    Coniferyl alcohol CAD    CINNAMYL ALCOHOL DEHYDROGENASE CCCP    Carbonyl cyanide m-chlorophenyl hydrazine CCoAOMT   CAFFEOYL COA 3-O-METHYLTRANSFERASE  CCR    CINNAMOYL-COA REDUCTASE ccr1g    cinnamoyl-CoA reductase 1g mutant CESA    CELLULOSE SYNTHASE CH    Cycloheximide CoA    Coenzyme A Col-0    Columbia-0 Arabidopsis ecotype   xiv COMT   CAFFEIC ACID 3/5-O-METHYLTRANSFERASE  CSE    CAFFEOYL SHIKIMATE ESTERASE cyt    Cytosol d6-pyridine   Pyridine with deuterium (2H) replacing 1H on carbon 6 DAPI    4',6-diamidino-2-phenylindole DCL1    DICER-LIKE1 DFRC    Derivatization Followed by Reductive Cleavage DMSO   Dimethyl sulfoxide e    Endodermis e.g.     exempli gratia (latin) EG    Endoglucanase  et al.    et alia (latin) EXLA    EXPANSIN-LIKE A family EXLB    EXPANSIN-LIKE B family EXPA    EXPANSIN A family EXPB    EXPANSIN B family f    Fibre F5H    FERULATE-5-HYDROXYLASE FITC    Fluorescein isothiocyanate FMT    FERULOYL-CoA-MONOLIGNOL TRANSFERASE FT-IR    Fourier transform-infrared spectroscopy g    grams G-lignin   Guiacyl lignin G-units   Guiacyl lignin units GC-MS   Gas chromatography-Mass spectroscopy GFP    Green fluorescent protein GUS    Glucuronidase H-lignin   Hydroxycinnamyl lignin H+-ATPase   Proton ATP phosphohydrolase HA    p-Coumaryl alcohol   xv HCl    Hydrochloric acid HCT   HYDROXYCINNAMOYL-COA SHIKIMATE/QUINATE HYDROXYCINNAMOYLTRANSFERASE HD-ZIP III   Class III homeodomain leucine zipper transcription factors HEN1    HUA ENHANCER1 HGS    Total H-, G-, and S-lignin HPLC    High performance liquid chromatography HYL1    HYPONASTIC LEAVES i.e.    id est (latin) IFF    Interfascicular fibres IRX    IRREGULAR XYLEM irx4    irregular xylem 4 mutant LAC17   LACCASE17 LAC4    LACCASE4 LC    Lignified cell m    meters m    milli (prefix) – 10-3  miR139a/JAW  MicroRNA139a/JAW miRNA   MicroRNA mRNA   Messenger RNA MS    Murashige and Skoog medium  mx    Metaxylem tracheary element MYB    Myeloblastosis  (family) transcription factor n    nano (prefix) -10-9 N    Nucleus NAC    NAM, ATAF, and CUC transcription factors NST    NAC SECONDARY WALL THICKENING PROMOTING FACTOR OMe    O-methyl OPLS-DA   Orthogonal projections to latent structures discriminant analysis P-type    Proton-type    xvi PA    Piperonylic acid PAL    PHENYLALANINE AMMONIA LYASE pCA    p-Coumarate-monolignol conjugates PCD    Programmed cell death PCR    Polymerase chain reaction pDONR   Marks gateway entry vector pH    power of Hydrogen pkGW    Gateway destination vector PM    Plasma membrane pMDC107   Marks gateway destination vector pMDC99   Marks gateway destination vector PMT    p-COUMAROYL-CoA-MONOLIGNOL TRANSFERASE pro    Promoter proAtPRX47   Promoter for PEROXIDASE 47 proAtPRX64   Promoter for PEROXIDASE 64 proCESA7   Promoter for CELLULOSE SYNTHASE 7 proUBQ10   Promoter for UBIQUITIN 10 (UBQ10) px    Protoxylem tracheary element Q-RT-PCR   Quantitative real-time PCR REF3    REDUCED EPIDERMAL FLUORESCENCE 3  REV    REVOLUTA/INTERFASCICULARFIBRELESS RISC    RNA INDUCED SILENCING COMPLEX RNAi    RNA interference S-lignin   Syringyl lignin S-units   Syringyl lignin units SA    Sinapyl alcohol SCoA    S-Coenzyme A SD    Standard deviation SND1    SECONDARY WALL-ASSOCIATED NAC DOMAIN PROTEIN 1 st    Stele   xvii TE    Tracheary element TEM    Transmission Electron Microscopy tt4-2    transparent testa 4 mutant UBQ10   UBIQUITIN 10 gene UV    Ultraviolet V    Vacuole vb    Vascular bundle VE    Vessel element VND    VASCULAR NAC DOMAIN  VNI2    VND-INTERACTING 2 WMD    Web miRNA designer WoW    Working on Walls WT    Wild type  XCP    XYLEM CYSTEINE PROTEASE XF    Xylary fibre XP    Xylary parenchyma cell YFP    Yellow fluorescent protein Zn2+    Zinc ion !    alpha  "    beta  µ    micro (prefix) -10-6     xviii Acknowledgements  I would like to thank my supervisors Dr. Lacey Samuels and Dr. Brian Ellis. Lacey and Brian have been so supportive and encouraging throughout my entire PhD and have been exceptional role models and mentors.   I would also like to acknowledge my committee members, Dr. Carl Douglas, Dr. Shawn Mansfield, and Dr. Fred Sack, for their insights and new perspectives throughout my research. The professors and graduate students in the Working on Walls group have provided helpful research suggestions and have shared their expertise with me. The members of the Samuels lab past and present, as friends and colleagues, have made this experience enjoyable. All of my lab mates have been very supportive and helpful, but I would especially like to thank Dr. Mathias Schuetz who has been my mentor through the WoW program. I am also grateful for the valuable technical support provided by the Bioimaging facility at UBC.   I would like to thank Dr. John Ralph who kindly hosted me in his lab at the University of Wisconsin, Madison, and the people in his lab who helped me learn DFRC and 2D-NMR cell wall analysis.    NSERC CGS M and CGS D3 scholarships and the Working on Walls CREATE grant provided funding for this research.    Last, but not least, I am eternally grateful for the love and support of my parents, siblings, family and friends.   1 Chapter 1: Introduction 1.1 Xylem development Xylem is the main water conducting tissue of the plant and the development of xylem was one of the key evolutionary adaptations of plants to terrestrial life. Xylem tissue is composed of numerous cell types, including water conducting tracheary elements and non-conducting parenchyma cells and fibres. Due to their essential roles in plant survival and structural support, the differentiation of tracheary elements and fibres, principally induced by polar auxin transport (Aloni, 1987; Lev-Yadun and Flaishman, 2001), has been extensively studied. Tracheary elements and fibres differentiate through a series of defined developmental stages, including: transcriptional commitment to cell fate, cell expansion, secondary cell wall deposition and lignification, and programmed cell death (Figure 1.1; Fukuda, 1996; Roberts and McCann, 2000; Pesquet et al., 2005). The differentiation process for tracheary elements and fibres is similar for cells developing from the procambium in primary xylem and from the vascular cambium in secondary xylem. However, there is currently more information about primary xylem development because of the relative ease with which it can be studied compared to secondary xylem development in woody plants.   Transcriptional commitment to tracheary element or fibre cell fate is preceded by the activity of transcription factors that promote the formation of procambium from the apical meristem in primary tissues. Transcriptional control of xylem and phloem precursor development from procambium depends on the involvement of a number of HD-ZIPIII transcription factors and KANADI transcription factors (Sieburth and Deyholos, 2006; Jung and Park, 2007). Tracheary element cell fate is determined by two transcription factors of the NAC-domain family: VASCULAR-RELATED NAC-DOMAIN6 (VND6) and VND7 (Kubo et al., 2005; Yamaguchi et al., 2008). VND6 and VND7 are considered “master regulators” of vessel, or tracheary element, cell fate (Kubo et al., 2005). VND7 is expressed in protoxylem vessel elements and overexpression of VND7 can ectopically induce protoxylem tracheary element-like formation with deposition of annular or helical secondary wall thickenings (Kubo et al., 2005; Yamaguchi et al., 2010a). The expression of VND7 is restricted to protoxylem tracheary    2  Figure 1.1 Developmental schematics of protoxylem and metaxylem tracheary elements (TE) and fibres from procambial cells.  V=vacuole, N=nucleus, PM=plasma membrane, 1˚CW=primary cell wall, 2˚CW=secondary cell wall, green circles represent organelles and black dots represent microtubules.    3 elements because of the activity of a VND-interacting protein, VNI2 (Yamaguchi et al., 2010b). VNI2 is a repressor of VND7 expression that is first active in the procambium and is then degraded when VND7 must become active in order to initiate protoxylem tracheary element formation. VNI2 remains active in neighbouring parenchyma cells, however, and represses VND7 transcription in all cells neighbouring the protoxylem tracheary elements (Yamaguchi et al., 2010b), thus ensuring the spatial specificity of VND7 expression.  VND6 is expressed in metaxylem vessel elements and ectopic expression of VND6 confers metaxylem vessel identity, accompanied by the formation of reticulate or pitted secondary wall thickenings (Kubo et al., 2005; Yamaguchi et al., 2010a). Unlike VND7, VND6 does not interact significantly with VNI2 (Yamaguchi et al., 2010b), and therefore the mechanism(s) controlling the spatial pattern of VND6 expression domain remain unknown. Further evidence for the role of VND6 and VND7 as master regulators of vessel element formation is that dominant repression of VND6 and VND7 inhibits metaxylem and protoxylem formation, respectively (Yamaguchi et al., 2008). In committing vessel elements to protoxylem or metaxylem cell fate, VND6 and VND7 initiate a transcriptional cascade that ultimately results in the up-regulation of genes involved in all the subsequent stages of xylogenesis (Figure 1.2).!  Key transcription factors that function as master regulators for the differentiation of fibres, such as the interfascicular fibres in the Arabidopsis stem, have also been identified. While all fibres in the Arabidopsis stem are grouped together as the same cell type, the differentiation and development of xylary fibres versus interfascicular fibres may have some key differences. However, the abundance of interfascicular fibres and the homogeneity of the tissue, relative to those found interspersed with tracheary elements in the xylem, has made the interfascicular fibres a useful model system for studying fibre differentiation and development. Loss of function at the HD-ZIP gene REVOLUTA/INTERFASCICULAR FIBRELESS (REV) locus largely eliminates interfascicular fibre formation at some developmental stages. Ectopic expression of REV, however, does not result in ectopic differentiation of fibre-like cells, and is therefore thought to act as an indirect regulator of fibre differentiation, perhaps by promoting polar auxin transport (Zhong et al., 1997; Zhong and Ye, 1999). In contrast, ectopic expression of the NAC domain-containing transcription factor SND1/NST3 is sufficient to elicit ectopic differentiation of    4   Figure 1.2 The transcriptional cascade regulating tracheary element and fibre differentiation and secondary cell wall formation.  Reproduced with permission from Schuetz et al. (2013) Copyright Society for Experimental Biology © 2013 (http://jxb.oxfordjournals.org).    5 non-vascular cell types into fibre-like cells (Zhong et al., 2006; Zhong et al., 2007; Mitsuda et al., 2007; Ohashi-Ito et al., 2010). Like VND6 and VND7, SND1 regulates a transcriptional cascade that ultimately activates the genes specific to secondary cell wall biosynthesis (Figure 1.2). However, in contrast to VND6 and VND7, SND1 does not appear to activate the expression of genes involved in programmed cell death, a feature that is consistent with the prolonged period of fibre development relative to that of tracheary elements (Zhong et al., 2006; Ohashi-Ito et al., 2010; Courtois-Moreau et al. 2009). Interestingly, ectopic overexpression of SND1 results in uniform deposition of secondary cell walls, in contrast to the helical or pitted secondary cell walls observed in protoxylem and metaxylem tracheary elements (Zhong et al., 2006; Ohashi-Ito et al., 2010). SND1 functions redundantly with two other NAC transcription factors, NST1 and NST2, to control secondary cell wall formation in all fibre types in Arabidopsis (Mitsuda et al., 2005; Zhong et al., 2006; Mitsuda et al., 2007; Zhong et al., 2007). Single loss-of-function mutants of snd1 or nst1 have no observable phenotype, but developing fibres in the snd1nst1 double mutant retain the general shape of fibre cells and undergo typical apical intrusive growth without developing secondary cell walls (Mitsuda et al., 2007; Zhong et al., 2007). This double mutant phenotype is also consistent with the fibre-specific expression observed for SND1 and NST1 (Mitsuda et al., 2005; Zhong et al., 2006; Mitsuda et al., 2007).  The development of inducible expression/activation systems for the VND6, VND7 and SND1 genes, coupled to comparative transcriptome analysis, has uncovered both overlapping and distinct target genes for these three transcription factors, including numerous other transcription factors (Figure 1.2)(Yamaguchi et al., 2008; Ohashi-Ito et al., 2010; Yamaguchi et al., 2010a; Zhong et al., 2010). In particular, several members of the MYB family of transcription factors appear to be crucial targets whose activity amplifies the transcriptional network and thereby promotes secondary cell wall formation. Although VND6, VND7 and SND1 share significant functional overlap in their ability to activate downstream transcriptional networks and the general metabolic machinery required to form secondary cell walls (Ohashi-Ito et al., 2010; Zhong et al., 2010), specific factors that establish the pattern of deposition of the secondary cell wall in different xylem cell types have yet to be identified.   In secondary xylem, specific xylem cell types are often not distinguishable from one another prior to radial cell expansion (Courtois-Moreau et al., 2009), after which, some of the   6 cell types can be identified based on size and shape. Radial cell expansion in tracheary elements, as in all other plant cell types, relies on high turgor pressure within the cells and the extensibility of the cell wall (Lee and Roberts, 2004). The period of maximum cell growth during cell development is vacuole-driven growth, which is concurrent with, but independent from, primary cell wall deposition (Cosgrove, 2000). In order for turgor within the tracheary elements to drive growth, the cell wall must loosen to accommodate the increase in cell size. It has been hypothesized that both non-covalent and covalent wall loosening agents in the cell wall participate in the regulation of cell wall extensibility. Non-covalent wall loosening agents act directly on the cell wall to promote extensibility (Cosgrove, 2000).  The best-studied example of a non-covalent wall loosening agent is expansin, which belongs to the family of hydroxyproline-rich glycoproteins found in the cell wall. Expansins were found to function in the acid-induced growth of cell walls, not by acting as endo- or transglycosylases as originally hypothesized (McQueen-Mason et al., 1992; Cosgrove, 2005), but by breaking non-covalent hydrogen bonds in the cell wall (McQueen-Mason and Cosgrove, 1994), such as hydrogen bonds between cellulose and hemicellulose. Expansins are divided into ! (EXPA), " (EXPB), expansin-like A (EXLA) and expansin-like B (EXLB) subfamilies, based on substrate specificity and sequence similarity, and the presence of expansin genes in ancestrally distant species suggests that the evolution of all four expansin subfamilies predates the evolution of land plants (Li et al., 2002; Cosgrove, 2005; Sampedro et al., 2006). Members of the EXPB subfamily were originally identified as allergens of grass pollen, and later determined to have cell wall-loosening activity (Cosgrove et al., 1997; Sampedro et al., 2006). The function of the smaller EXLA and EXLB subfamilies is not yet known (Sampedro et al., 2006). The proposed structure of expansins includes an N-terminal signal peptide to direct the proteins into the secretory pathway, a domain to cleave non-covalent hydrogen bonds within the cell wall, and a putative C-terminal carbohydrate-binding domain that is hypothesized to anchor the expansins within specific cell walls to prevent the migration of the secreted expansins to neighbouring cells (Cosgrove, 2000; Cosgrove, 2005).  The mechanism of fibre elongation is unique, as it involves apical intrusive growth (Lev-Yadun, 1997). As in tracheary elements, expansins act as non-covalent cell wall loosening agents for fibre elongation and mRNA encoding expansins from the !-subfamily have been localized at   7 the tips of fibres undergoing apical intrusive growth (Gray-Mitsumune et al., 2004).  The bonds broken by expansins will be reformed following turgor-driven growth. Another putative wall loosening agent is endo-1,4-"-D-glucanase (EG), which hydrolyses the covalent bonds of xyloglucans (Cosgrove, 2005; Eklöf et al., 2012; Eklöf et al., 2013). This hydrolysis may increase the susceptibility of the wall to expansin activity, and therefore, to cell expansion (Cosgrove, 2000). When the cell turgor pressure decreases, due to increased extracellular osmolarity, cell expansion ceases and secondary wall deposition begins (Lee and Roberts, 2004).  Secondary cell walls in tracheary elements are highly organized, with cellulose microfibrils deposited in S1, S2, and S3 layers (Bailey and Kerr, 1935; Barnett and Bonham, 2004). The three layers differ in the orientation of cellulose microfibrils (Bailey and Kerr, 1935; Barnett and Bonham, 2004). The deposition of secondary walls is often initiated prior to cessation of primary cell wall expansion, leading to stretching in the first layer of secondary wall (Roelofsen, 1959). The chemical composition of the secondary wall differs from that of the primary wall, as the secondary wall has a higher concentration of cellulose (about 40-50%) and hemicellulose (about 20-30%), with the addition of lignin to the cell wall (about 25%), little to no pectic material, and fewer structural proteins or enzymes (Mellerowicz and Sundberg, 2008; Dejardin et al., 2010). There is a specific group of cellulose synthase (CESA) enzymes involved in the synthesis of secondary cell wall cellulose. In Arabidopsis, CESA4, CESA7 and CESA8 appear to function non-redundantly in a complex to catalyze the production of secondary cell wall cellulose (reviewed in Taylor, 2008). Secondary cell wall hemicelluloses are synthesized in the Golgi and trafficked to the cell wall. Several glycosyltransferases have been shown to be involved in the synthesis of the backbone (IRX9, IRX10, IRX10-like, IRX14) and others have been associated with synthesis of the reducing end of the polymer (IRX7, IRX8, PARVUS) (reviewed in Scheller and Ulvskov, 2010).  Secondary cell walls are also usually lignified (Figure 1.1; see section 1.2), with lignin composing about 18-25% of the cell wall biomass (Dharmawardhana et al., 1992; Dejardin et al., 2010). The deposition of the lignin component of the secondary cell wall is the main focus of this thesis, and lignin biosynthesis, export and polymerization are reviewed in section 1.2, below.  As the lignin infiltrates the wall matrix, following the deposition of wall polysaccharides (Terashima et al., 1993), it displaces water in the walls, conferring hydrophobicity to lignified areas of the   8 wall. The deposition and lignification of the secondary cell wall is of particular importance during xylogenesis because the secondary cell wall enables the tracheary elements to maintain both structure and function after the cell undergoes programmed cell death.  Programmed cell death (PCD), the final stage of xylogenesis (Figure 1.1), is marked by enzymatic degradation of the protoplast. Unlike other programmed cell death events in plants that involve an oxidative burst of reactive oxygen species, such as the hypersensitive response, PCD in tracheary elements and fibres is driven by the release of the contents of the cell vacuole (Groover et al., 1997). The current model of tracheary element PCD has been heavily influenced by studies involving the Zinnia elegans cell culture system in which mesophyll cells are induced to differentiate into tracheary elements (Fukuda, 1997). By examining differentiating cells in the Zinnia in vitro system, it has been determined that PCD is initiated by collapse of the central vacuole, which releases hydrolytic enzymes, such as cysteine proteases and nucleases, and ultimately results in the autolysis of all cell contents (Groover et al., 1997; Fukuda, 2000). This mega-autolysis of all cell contents is preceded by micro-autolysis within the vacuole prior to vacuolar implosion (Avci et al., 2008). In tracheary element mega-autolysis, single membrane structures first are degraded, followed by rapid nuclear degradation and loss of plasma membrane integrity (Groover et al., 1997). Degradation of the chloroplasts and mitochondria occurs before general autolysis to remove all remaining cell contents (Groover et al., 1997). Studies in poplar secondary xylem have demonstrated that, once initiated, tracheary element programmed cell death occurs rapidly, leading to the formation of a zone of dead, but functional, water-conducting cells within ca. 400 µm of the cambium (Courtois-Moreau et al., 2009). Xylem cysteine proteases play an important role in executing tracheary element programmed cell death. In xcp1 single mutants and xcp1xcp2 double mutants, incomplete degradation of the cytoplasm during PCD was observed (Avci et al., 2008). XCP1 and XCP2 are localized to living tracheary elements and persist in the space formerly occupied by the vacuole following vacuolar collapse (Avci et al., 2008). Together with gene transcript profiling studies from Zinnia cell culture systems (Fukuda and Komamine, 1980; Groover et al., 1997), the specific pattern of XCP expression and accumulation supports a model of cell-autonomous programmed cell death in tracheary elements. Transcript profiling in the Zinnia cell culture system has also identified other   9 nucleases, proteases and lytic enzymes putatively stored in the vacuole that are likely to be involved in the vacuole-mediated tracheary element PCD (Groover et al., 1997; Fukuda, 2000).  It is hypothesized that some of the signals that initiate PCD are generated during secondary wall formation. Such signals may include shifts in gene expression at the onset of secondary wall formation (Courtois-Moreau et al., 2009) and increases in cellular calcium ion uptake during secondary cell wall formation (Fukuda, 2000). On the other hand, additional genes involved in programmed cell death, such as two XYLEM CYSTEINE PROTEASE genes from Arabidopsis (Avci et al., 2008), are up-regulated directly by secondary cell wall master transcription factors such as VND6 and VND7 (Zhong et al., 2006; Ohashi-Ito et al., 2010). The fibre master regulator, SND1, however, does not directly regulate genes involved in programmed cell death, which is consistent with a more specific role for SND1 in secondary cell wall formation (Zhong et al., 2006; Ohashi-Ito et al., 2010; Zhong et al., 2010). Programmed cell death in fibres and in tracheary elements presumably occurs in response to similar signals, but the progression of events leading to cell death varies slightly in fibres. In fibres, collapse of the vacuole is preceded by autophagy of cellular contents surrounding the vacuole, which gives the cytoplasm a less dense appearance (Courtois-Moreau et al., 2009). Nuclear degradation in fibres is also a slow process that is ultimately followed by vacuole implosion and autolysis of remaining cell contents by hydrolytic enzymes (Courtois-Moreau et al., 2009). Fibre programmed cell death seems to occur synchronously around the stem within the secondary xylem, and it occurs further away from the cambium (approximately 650-1000 µm from the cambium) than does vessel programmed cell death (Courtois-Moreau et al., 2009).  1.2 Lignin: biosynthesis to polymerization Lignification of tracheary elements or fibres is important for the maintenance of cell structure and function in the mature cells. Lignin is a complex three-dimensional phenolic polymer that strengthens the secondary cell wall and increases its hydrophobicity. These characteristics increase the efficiency of water transport through the cells, and help protect the plant against abiotic and biotic stress (Humphreys and Chapple, 2002). While the presence of lignin is beneficial to plants, the difficulty associated with degrading and extracting lignin from plant-derived biomass presents challenges in the pulp and paper industry, in the development of   10 forage crops, and in the development of biofuels (Humphreys and Chapple, 2002; Weng et al., 2008; Mansfield et al., 2012). The procedures currently available for lignin removal on an industrial scale are expensive and energy-intensive, which has led to the search for methods to develop plants with altered, or decreased, lignin that would allow easier post-harvest lignin removal (Weng et al., 2008) without compromising plant viability or survival. In order to successfully manipulate lignin content, the process of lignification must be fully understood, from the formation of monolignols, to their export to the secondary cell wall and polymerization within the cell wall.  Lignin precursors (monolignols) are generated through the general phenylpropanoid pathway and then diverted to the monolignol-specific biosynthetic pathway (Vanholme et al., 2010). The shikimate pathway produces the immediate upstream precursor for the phenylpropanoid pathway, the amino acid L-phenylalanine. The enzymes involved in modifying the structure of phenylalanine to ultimately generate monolignols include phenylalanine ammonia lyase (PAL), cinnamate-4-hydroxylase (C4H), 4-coumarate: CoA ligase (4CL), hydroxycinnamoyl-CoA shikimate/quinate hydroxycinnamoyltransferase (HCT), p-coumarate 3-hydroxylase (C3H), caffeoyl shikimate esterase (CSE), caffeoyl CoA 3-O-methyltransferase (CCoAOMT), cinnamoyl CoA reductase (CCR), ferulate 5-hydroxylase (F5H), caffeic acid 3/5-O-methyltransferase (COMT) and cinnamyl alcohol dehydrogenase (CAD) (reviewed in Boerjan et al., 2003; Vanholme et al., 2008; Vanholme et al., 2013). The first modification to phenylalanine is a deamination step performed by PAL. The next step is the 4-hydroxylation of the phenyl ring by C4H, a modification that is reiterated at the 3 and 5 positions of the ring by C3H and F5H, respectively, later in the pathway. 4CL then activates the carboxyl group on the propane chain by esterifying it with the SCoA leaving group, which facilitates the subsequent addition of shikimic/quinic acid by HCT.  Vanholme et al. (2013) recently reported a new enzyme in the lignin biosynthetic pathway called CSE (caffeoyl shikimate esterase), which, acting with the previously characterized 4CL enzyme, may play an important role in the formation of caffeoyl CoA from caffeoyl shikimic acid. Another major modification is the methylation of the ring hydroxyl groups by CCoAOMT and COMT.  The SCoA group at the end of the propane chain is also reduced to an aldehyde by CCR. The final modification involves the conversion of the aldehyde group into an alcohol by the   11 CAD enzyme, to generate the monolignol alcohols (reviewed in Humphreys and Chapple, 2002). The main end-products of the monolignol biosynthetic pathway are three main phenylpropanoid alcohols that can be polymerized to form lignin: p-coumaryl alcohol, which produces H-lignin; coniferyl alcohol, which produces G-lignin; and sinapyl alcohol, which produces S-lignin (Figure 1.3; reviewed in Boerjan et al., 2003; Vanholme et al., 2008). In addition to these well-established monolignols, intermediates from the lignin biosynthetic pathway, such as various aldehydes and acids, can also be incorporated into lignin. In Arabidopsis, these non-traditional monolignols are incorporated into lignin in significant amounts only in mutants of the core lignin biosynthetic pathway (Do et al., 2007; Vanholme et al., 2010; Thevenin et al., 2011).  The lignin composition of all lignified cells is not uniform either within a plant, or among plant taxa. Gymnosperms only have G-lignin in the tracheids of their wood, accompanied by small amounts of H-lignin. Angiosperms, on the other hand, have both G- and S-lignin and only trace amounts of H-lignin. Tracheary elements generally have G-rich lignin in their cell walls, while extraxylary fibres, such as interfascicular fibres in Arabidopsis, have S-rich lignin, based on the different reactions that fibres and tracheary elements produce when they are exposed to the Mäule histochemical stain (Chapple et al., 1992; Meyer et al., 1998). Xylary fibres might therefore be expected to also have S-rich lignified cell walls, but a recent study in poplar suggests by FT-IR spectroscopy coupled with OPLS-DA (orthogonal projections to latent structures discriminant analysis) that the lignin composition of fibres adjacent to tracheary elements, such as xylary fibres, is intermediate between that of tracheary elements and that of fibres (Gorzsas et al., 2011). This indicates that the xylary fibre lignin may have elements of both the G-rich tracheary elements and S-rich fibres.  The monolignol biosynthetic reactions occur within the cytosol, or in close proximity to the endoplasmic reticulum, of monolignol-producing cells (Bonawitz and Chapple, 2010; Chen et al., 2011; Bassard et al., 2012), but the mechanism of monolignol export from the site of synthesis to the region of polymerization remains unclear. Recent studies do not support the model of vesicle-mediated monolignol export, because lignifying tissues exposed to tritiated phenylalanine and phenylpropanoid inhibitors did not display alterations in the distribution of Golgi radiolabel, and disruption of vesicular trafficking through the Golgi using the ATPase inhibitor Brefeldin A (BFA) does not alter lignification (Kaneda et al., 2008; Samuels lab,    12  Figure 1.3 Current view of the monolignol biosynthesis pathway. Modifications to the propane chain or phenyl ring are highlighted in red. PAL, phenylalanine ammonia lyase; C4H, cinnamate-4-hydroxylase; 4CL, 4-coumarate CoA ligase; HCT, hydroxycinnamoyl-CoA shikimate/quinate hydroxycinnamoyltransferase; C3H, p-coumarate 3-hydroxylase; CSE, caffeoyl shikimate esterase; CCoAOMT, caffeoyl CoA 3-O-methyltransferase; CCR, cinnamoyl CoA reductase; F5H, ferulate 5-hydroxylase; COMT, caffeic acid 3/5-O-methyltransferase; CAD, cinnamyl alcohol dehydrogenase.  NH2OHO Modifications to the phenyl ringModifications to the propane chainOHO OHOOHOS-CoAOH OHOOOHOHOHOOOOHHOOHOHOHOOS-CoAOHHOOS-CoAOHOCH3OHOHOCH3OHOCH3CH2OHOHOHOHCH2OHOHOHOCH3OHOHOCH3CH2OHOHOHOHH3CO OCH3OHH3COCH2OHOCH3OHOOHHOL-phenylalanineCinnamic acid p-Coumaric acidp-Coumaroyl CoAp-Coumaroylshikimic acidCaffeoylshikimic acidCaffeic acidCaffeoyl CoA Feruloyl CoAConiferaldehydeConiferyl alcohol5-Hydroxy- coniferaldehyde5-Hydroxy-coniferyl alcoholSinapaldehydeSinapyl alcoholp-Coumaraldehydep-Coumaryl alcoholPALC4H4CLHCTC3HHCTCCoAOMTCCRCAD CAD CADF5H COMTF5H COMTCCRCADC4H/C3H4CLCSEFigure 1.3. Current view of the monolignol biosynthesis pathway.Modifications to the propane chain or phenyl ring are highlighted in red.PAL, phenylalanine ammoia lyase; C4H, cinnamate-4-hydroxylase; 4CL, 4-coumarate CoA ligase; HCT, hydroxy-cinnamoyl-CoA shikimate/quinate hydroxycinnamoyltransferase; C3H, p-coumarate 3-hydroxylase; CSE, caffeoyl shikimate esterase; CCoAOMT, caffeoyl CoA 3-O-methyltransferase; CCR, cinnamoyl CoA reductase; F5H, ferulate 5-hydroxylase; COMT, caffeic acid 3/5-O-methyltransferase; CAD, cinnamyl alcohol dehydrogenase.   13 unpublished data).!The small size of monolignols, and their demonstrated ability to partition into the membrane of synthetic lipid disks, supports the idea that monolignols could potentially exit the cell by passive diffusion (Boija and Johansson, 2006; Boija et al., 2007). In this model, monomer export would be driven by the concentration gradient between the cytosol, where monolignols are being actively synthesized, and the cell wall matrix, where they are rapidly polymerized into lignin. The rate of diffusion of the monolignols across the plasma membrane would therefore have to be very high to account for the rapid and extensive lignification occurring in the maturing secondary cell wall. However, only low levels of monolignol diffusion across the membrane of plasma membrane vesicles have been reported (Miao and Liu, 2010).  The best model for monolignol export suggests that monolignol export to the cell wall should occur via plasma membrane-localized transporters, such as ABC (ATP-binding cassette) transporters (Kaneda et al., 2008; Li and Chapple, 2010; Simmons et al., 2010; Miao and Liu, 2010; Alejandro et al., 2012).!A set of candidate ABC transporters for monolignol export was previously identified based on their co-expression with phenylpropanoid biosynthesis genes in developing Arabidopsis inflorescence stems (Ehlting et al., 2005), but subsequent loss-of-function mutant analysis did not uncover phenotypes associated with defective lignification (Kaneda et al., 2011). Miao and Liu (2010) tested the ability of plasma membrane vesicles derived from Arabidopsis seedlings to export monolignols. Transport of the coniferyl alcohol monolignol into these vesicles was shown to be primarily energy-dependent. Disruption of transmembrane pH or potential gradients with pharmacological inhibitors did not affect the observed monolignol transport, but treatment with chemicals known to act as ABC transporter inhibitors, such as vanadate or nifedipine, greatly reduced monolignol accumulation in these vesicles (Miao and Liu, 2010).   Co-expression analysis between phenylpropanoid biosynthetic genes and members of the ABCG ABC transporter family identified ABCG29 as a candidate monolignol transporter (Alejandro et al., 2012). In Arabidopsis roots, ABCG29 is plasma membrane-localized, and expressed in endodermal cells and vascular tissue, the primary locations of lignification in the root (Alejandro et al., 2012). Transport assays in yeast microsomes later showed that this transporter was capable of exporting p-coumaryl alcohol from the microsomes, but not coniferyl alcohol or sinapyl alcohol (Alejandro et al., 2012). Knock-out mutations in the ABCG29 gene   14 resulted in slightly decreased H-, G- and S-lignin in planta. Together these data suggest that ABCG29 may be a candidate monolignol exporter for p-coumaryl alcohol units. However, since most angiosperm lignin has only trace amounts of H-lignin, and therefore a low requirement for p-coumaryl alcohol transport, further studies are required to identify transporters capable of efficiently exporting coniferyl alcohol and sinapyl alcohol.  A recent study examined the possible role of coniferin (the beta-glucosyl derivative of coniferyl alcohol) in lignification, by examining coniferin transport mechanisms operating across the membrane of vesicles prepared from hybrid poplar (Populus sieboldii # Populus grandidentata) and Japanese cypress (Chamaecyparis obtusa) xylem tissue (Tsuyama et al., 2013). Coniferin transport across the tonoplast and endomembrane system membranes in vesicles derived from actively lignifying tissue involved H+/coniferin antiport. The observed transport of coniferin was inhibited by the addition of bafilomycin A1, a reagent that blocks vacuolar-ATPase activity (Tsuyama et al., 2013), which provides further support for the dependence of coniferin transport on an ATPase and an associated proton gradient. This study, however, did not provide direct support for the involvement of coniferin in lignification as a lignin precursor, nor did it demonstrate a mechanism for the movement of coniferin across the plasma membrane into the cell wall. Another study, performed in Pinus contorta, demonstrated that a "-glucosidase has activity towards coniferin and syringin and is localized to differentiating xylem cell walls (Dharmawardhana et al., 1995). This may indicate that, in conifers, coniferin can be metabolized in the xylem cell walls, but the mechanism for transport to the cell wall is still unknown. The site of monolignol polymerization is the cell wall and the oxidative polymerization reaction involves combinatorial radical coupling of the monolignols (reviewed in Ralph et al., 2004). The oxidation of the monolignols is catalyzed by peroxidases and/or laccases (reviewed by Novo-Uzal et al., 2013 and Berthet et al., 2012; Figure 1.4), but as large gene families encode these enzymes in plants, the specific peroxidase and/or laccase that is involved in a given tissue is probably tissue-, and even cell type-dependent. While a number of peroxidases have been implicated in lignification, there is also evidence for the role of laccases in lignification, especially in the Arabidopsis inflorescence stem. LAC4 and LAC17 were shown to be required for normal lignification of fibres, and, to some extent, of tracheary elements (Berthet et al.,    15  Figure 1.4 Radical coupling of a monolignol to a lignin polymer (endwise polymerization).     16 2011). The lac4lac17 double mutants have decreased phloroglucinol-HCl staining in their secondary cell walls, and display collapsed xylem phenotypes, both of which are characteristic of decreased lignin content (Berthet et al., 2011). Transcription factors that up-regulate lignin biosynthesis genes, such as MYB58, have also been shown to up-regulate the expression of LAC4 (Zhou et al., 2009), further implicating laccases with lignin biosynthesis. The apparent catalytic promiscuity of these oxidases, and the large size of the peroxidase and laccase gene families, suggest that a diversity of specific laccases and peroxidases may participate in different stages of lignification and/or in lignification of specific cell types.  The oxidized monolignols are coupled to the growing lignin chain via a phenoxy radical through a process of endwise polymerization (Figure 1.4; Morreel et al., 2004; Mechin et al., 2007). Autoradiography studies in Magnolia kobus, Japanese cedar (Cryptomeria japonica) and Japanese black pine (Pinus thumbergii) have provided the most information about the deposition of lignin in the different cell wall layers. The first step in lignification of secondary cell walls involves the incorporation of H-lignin and G-lignin into the middle lamella and cell corners (Fujita and Harada, 1979; Takabe et al., 1981; Takabe et al., 1985; Terashima and Fukushima, 1988). The next stage involves the lignification of the primary cell wall and outer secondary cell wall with predominantly G-lignin (Terashima and Fukushima, 1988). Lastly, the inner secondary cell wall, a layer that is prominent in fibres, becomes impregnated with S-lignin, in taxa producing sinapyl alcohol (Fujita and Harada, 1979; Takabe et al., 1981; Takabe et al., 1985; Terashima and Fukushima, 1988). Many other aspects of the lignin polymerization process remain unclear, including the physical nature of the association between lignin and other cell wall polymers, control of the spatial patterning of lignin deposition in the wall, and the relationship between the metabolic supply of G- and S-type monolignols and the composition of the final polymer.! 1.3 Good neighbours during lignification Secondary cell wall lignification in differentiating tracheary elements (TEs) and programmed cell death are overlapping processes. Because the lignification process is thought to continue after the tracheary elements are dead (Hosokawa et al., 2001; Pesquet et al., 2010; Pesquet et al., 2013), non-conducting xylary parenchyma cells surrounding the tracheary   17 elements could potentially be involved in the production of monolignols destined for polymerization in the walls of the TEs. The concept of parenchyma cell involvement in the lignification of tracheary elements is referred to as the co-operative model or the “good neighbour” hypothesis (Figure 1.5). One of the first occasions in which evidence for co-operation was found was in the continued deposition of radiolabelled compounds, derived from tritiated cinnamic acid, in the cell walls of mature tracheary elements of wheat coleoptiles after programmed cell death had occurred (Pickett-Heaps, 1968). The conclusion drawn from that early study was that lignification was able to continue after programmed cell death through the contribution of phenylpropanoid precursors from adjacent cells.  Following this preliminary study, investigations have indirectly inferred the site of lignin biosynthesis by using promoter reporter assays, such as glucuronidase (GUS), to identify the site of promoter activity for various phenylpropanoid pathway genes. Phenylalanine ammonia-lyase (PAL) catalyzes the initial irreversible step in the phenylpropanoid pathway, and is therefore thought to play a role in regulating flux through the phenylpropanoid pathway (Bevan et al., 1989). Transgenic tobacco expressing a Phaseolus promoter PAL2::GUS construct showed GUS activity in differentiating xylem elements of primary xylem, as well as in more mature tracheary elements and fibre cells. These latter cells also stained positively with phloroglucinol-HCl, which suggests the presence of lignin. However, GUS staining was also observed in secondary xylem parenchyma ray cells, which are characterized by their dense cytoplasm and organization into radial files of cells. These parenchyma cells also contained lignin deposits, based on histochemical staining, but did not undergo programmed cell death. The results from this study may support the idea that xylem parenchyma can be involved in lignification, but it might also reflect the involvement of PAL in the synthesis of other phenylpropanoid compounds (Bevan et al., 1989).  As well as the lack of specificity in PAL’s contribution to lignin biosynthesis, an additional problem with the conclusions presented in this study is that the images showing promoter PAL2::GUS localization do not seem to support the researchers’ description of GUS activity in parenchyma cells adjacent to tracheary elements, or in files of parenchyma ray cells.  A more credible report was found in a study using the 4-coumarate:Coenzyme A ligase (4CL) promoter to drive GUS expression (Hauffe et al., 1991). 4CL catalyzes the formation of      18  Figure 1.5 The “good neighbour” model of lignification in Arabidopsis inflorescence stems. Xylary parenchyma cells (XP) are potentially contributing to the lignification of neighbouring vessel elements (VE) and xylary fibres (XF) pre- and post-mortem. Reproduced with permission from Schuetz et al. (2013) Copyright Society for Experimental Biology © 2013 (http://jxb.oxfordjournals.org).   CH2OHCH2OHCH2OH  19 CoA esters from cinnamate derivatives (Figure 1.3). As with the promoter PAL2 expression, promoter 4CL1::GUS activity was highest in differentiating tracheary elements, but no activity was observed in fully differentiated and mature xylem. In the secondary xylem of older tobacco stems, GUS activity was restricted to the files of ray parenchyma cells between lignified tracheary elements (Hauffe et al., 1991). This suggests that the xylem parenchyma cells are capable of synthesizing phenylpropanoids. Subsequent studies have used similar promoter::GUS assays to analyze the expression pattern of genes encoding enzymes specific to the lignin biosynthetic pathway. One such study examined the eucalyptus cinnamyl alcohol dehydrogenase promoter (CAD)::GUS activity in transgenic poplar plants (Feuillet et al., 1995). CAD promoter activity was noted in actively lignifying cells, such as lignifying tracheary elements, but was absent in mature tracheary elements. The GUS activity was, instead, observed in parenchyma cells located between vessel elements in primary xylem, and in the primary and secondary ray parenchyma cells. Similarly, a study using poplar promoter CCoAOMT::GUS constructs reported the expression of the CCoAOMT promoter in differentiating vessel elements prior to programmed cell death and in ray parenchyma cells adjacent to the vessels (Chen et al., 2000). The parenchyma ray cells are therefore thought to play a role in the lignification of surrounding tracheary elements. In a more recent study, eucalyptus CAD::GUS and cinnamoyl coenzyme A reductase (CCR)::GUS promoter activities were analyzed in Arabidopsis (Baghdady et al., 2006). GUS activity was observed in immature xylem elements until programmed cell death occurred and the promoters were also expressed in paratracheal parenchyma cells through all tracheary element developmental stages. These data suggest a refinement of the co-operative model of lignification into two stages of lignification. In the first stage, lignification is primarily cell autonomous, but with some help from surrounding cells. Following programmed cell death, the second step of co-operation is initiated as parenchyma cells adjacent to lignified tracheary elements provide the substrates for the remainder of lignin biosynthesis in the TEs (Baghdady et al., 2006). Many studies relevant to the co-operative model have been performed in the Zinnia elegans cell culture system. In this system, leaf mesophyll cells are induced to differentiate into tracheary elements and the differentiation of these tracheary elements mirrors the development of TEs in vivo, making the Zinnia cell culture a valuable model system. The synchronous nature of   20 TE differentiation, and the presence of only two cell types (Fukuda and Komamine, 1982), simplifies studies on the lignification of tracheary elements. Even after tracheary elements in Zinnia cell culture have died, lignin biosynthesis genes, such as C4H and CCoAOMT, continue to be expressed by undifferentiated parenchyma cells (Demura et al., 2002; Pesquet et al., 2005), and the lignin content of the tracheary element cell walls continues to increase after cell death (Hosokawa et al., 2001). Differentiating and non-differentiating cells in Zinnia cell culture appear to differ on a molecular level, insofar as non-differentiating cells may not have the same ability as differentiating cells to perceive (or respond to) hormones such as auxin and cytokinin, which are necessary for differentiation (Pesquet et al., 2005). Of the two cell types present in the cell culture, only tracheary elements develop a lignified secondary cell wall.  It has been proposed that both tracheary elements and parenchyma-like cells in the Zinnia culture system secrete monolignols into the culture medium and, as lignification progresses, the precursors from the medium become incorporated into the cell walls of developing tracheary elements (Hosokawa et al., 2001; Tokunaga et al., 2005). Tokunaga et al. (2005) provided support for this hypothesis when they detected four coniferyl alcohol-derived di-lignols in the culture medium and determined that levels of coniferyl alcohol in the medium decline as secondary wall thickening occurs. Based on these findings, lignification was proposed to occur in three steps in the Zinnia cell culture: 1) coniferyl alcohol is secreted from the parenchyma-like cells and tracheary elements before secondary wall thickening occurs, 2) peroxidase enzymes dimerize the coniferyl alcohol, and 3) secondary wall thickening occurs followed by the accumulation of coniferyl alcohol and dilignols in the medium, and subsequent polymerization of monomers and dimers from the medium into the cell wall (Tokunaga et al., 2005). Hosokawa et al. (2001) also presented data that supported the secretion and re-uptake model by demonstrating that continuous exchange of the culture medium during tracheary element differentiation inhibited lignification. One caveat to this study, however, is that while the exchange of media would have removed monolignols, it would also remove other differentiation signals such as hormones and metabolites. While the lignin precursors would be secreted into the apoplast in planta, versus into the culture medium, these studies did demonstrate that tracheary elements have the potential to incorporate exogenous lignin precursors into their cell walls.    21 While the Zinnia TE transdifferentiation system has provided important insights into TE development and the associated lignification of secondary cell walls, it suffers from one major drawback – the limited genetic resources available in this species for analysis and manipulation of genes of interest. To bypass this limitation, an analogous Arabidopsis-based TE transdifferentiation system has now been developed (Oda et al., 2005). The results of a recent study performed with such transdifferentiating Arabidopsis cell cultures have suggested an alternative to the model of apoplastic secretion followed by re-uptake of monolignols. This study demonstrated, based on the level of lignin autofluorescence in the tracheary element-like cells, that lignification of the secondary cell walls occurred predominantly after programmed cell death (Pesquet et al., 2010). Because the tracheary elements cannot synthesize monolignols post-mortem, it was hypothesized that monolignols are stored in the vacuole until programmed cell death, at which point vacuolar collapse releases the monolignols and lignification proceeds rapidly through an oxidative burst in the apoplast (Pesquet et al., 2010). It has been suggested previously that monolignols are stored in the vacuole in their glucoside form (eg. coniferin and syringin) (Leinhos and Savidge, 1993; Pesquet et al., 2010), and in woody gymnosperm stems, coniferin accumulates to high levels in areas close to the cambium that are actively undergoing radial expansion (Savidge, 1989). It is possible, however, that coniferin does not play as significant a role in angiosperm wood or primary xylem. HPLC analyses of light-grown Arabidopsis root extracts have demonstrated that detectable levels of monolignol glucosides do accumulate (Hemm et al., 2004; Bednarek et al., 2005; Lanot et al., 2006). Chapelle et al. (2012) showed that loss-of-function mutations in BGLU45 and BGLU46, genes encoding "-glucosidases, resulted in coniferin and syringin accumulation. However, the single mutants did not have decreased lignin, indicating that monolignol glucosides are unlikely to serve as the primary pool of lignin precursors (Chapelle et al., 2012). Further studies involving BGLU45 and BGLU46 double mutants are required to further examine the role of "-glucosidases in Arabidopsis lignification. This finding is further supported by recent studies examining endodermis lignification, where lignin deposition is also localized to a specific domain, the Casparian strip. During this process, which is analogous to protoxylem tracheary element lignification, the Casparian strip lignifies gradually without the prior accumulation of monolignols (Naseer et al., 2012; Lee et al., 2013).    22 In light of these recent findings, it seems likely that the post-mortem model of lignification involves monolignols being provided by neighbouring cells rather than being released from the vacuole during programmed cell death. Tracheary element-like cells in Zinnia or Arabidopsis cultures still incorporated exogenously applied monolignols after treatment with a phenylpropanoid biosynthesis inhibitor, indicating that non-cell autonomous lignification can occur (Hosokawa et al., 2001; Pesquet et al., 2013). Since only 60-70% of the cells in Zinnia mesophyll cell cultures differentiate into tracheary elements, the non-differentiating cells may be playing an essential supportive role during lignification of the differentiating cells (McCann et al., 2001).  While studies on the Zinnia cell culture system seem to provide substantial evidence for some degree of co-operation between cells during lignification, it is important to emphasize that these studies were performed in vitro and not in planta. As such, discrepancies remain with respect to the origin of monolignols and the timing of lignification relative to programmed cell death. The production and secretion of monolignols from non-differentiating cells within organized plant tissues has not been directly demonstrated, and thus the evidence for existence of such metabolic co-operation in the intact plant remains indirect. In planta autoradiography studies that examined the timing of lignification have concluded that lignification begins as a pre-mortem process (Terashima et al., 1986; Terashima and Fukushima 1988; Kaneda et al., 2008), which introduces a discrepancy between the in vitro data and in planta data.  1.4 Research objectives and significance The main objectives of this PhD research were two-fold: 1. To determine the timing of lignification relative to the onset of programmed cell death during primary xylem development. 2. To assess the ability of non-lignifying neighbouring cells to contribute to lignification of both tracheary elements and fibres.  The experimental system used for these studies was Arabidopsis thaliana; specifically, the protoxylem in the developing primary root as well as lignifying cells in the stem. Chapter 2 addresses the first objective of the research through an autoradiographic investigation of the   23 timing of lignification. Seedling roots and inflorescence stems were subjected to treatment with tritiated phenylalanine, the metabolic precursor for monolignols, as well as a participant in de novo protein synthesis. In tissues that were co-incubated with radiolabeled phenylalanine and the protein synthesis inhibitor cycloheximide, radiolabel was observed in the secondary cell walls but not in the cytosol or vacuole, indicating rapid transport of phenylpropanoid metabolites to the cell wall. Based on the incorporation of radiolabel into the cell walls, lignification appeared to begin pre-mortem and to continue post-mortem, but not indefinitely. Long dead tracheary elements did not have radiolabeled cell walls, indicating that monolignols were no longer being polymerized to lignin in these cell walls. This part of my work clarifies the timing of lignification in planta and demonstrates that the secondary cell wall lignification process begins before programmed cell death and continues immediately after cell death within a defined developmental window.   Chapter 3 examines the second objective of this PhD research, namely whether non-lignifying cells can act as “good neighbours” by providing lignification substrates to cells with lignifying secondary cell walls. To address this question, a novel targeted genetic approach was used in which an artificial microRNA was designed to suppress the expression of the CCR1 monolignol biosynthetic gene, and this expression of the miRNA was driven in specific cell populations using different promoters.  One promoter used to drive the miRNA was proCESA7, which is specifically expressed in all cells that form thickened secondary cell walls (tracheary elements and fibres). This construct allowed me to determine the contribution of neighbouring xylary parenchyma cells, whose walls do not thicken, to xylem lignification. In addition, this strategy allowed me to assess the metabolic autonomy of extra-xylary fibres, such as interfascicular fibres in Arabidopsis stems. My results provided the first strong in planta evidence that xylary parenchyma cells can contribute to the lignification of the neighbouring xylary fibres and tracheary elements. Interfascicular fibres, on the other hand, did not have good neighbours and experienced drastic reductions in lignification in the proCESA7::miRNA CCR1 lines. Two other promoters were used to further assess the contribution of xylary parenchyma cells (proAtPRX47) and fibres (proAtPRX64) to tracheary element lignification. While knocking down monolignol biosynthesis specifically in xylem parenchyma with proAtPRX47::miRNA CCR1 did not produce detectable loss of lignin (with phloroglucinol-HCl staining), loss of    24 monolignol synthesis in fibres in plants expressing proAtPRX64::miRNA CCR1 produced dramatic effects, and suggests a role for xylary fibres in tracheary element lignification. This represents the first study in which specific populations of cells have been genetically manipulated to determine which cells are actively involved in monolignol production during lignification.  Chapter 4 also addresses the objective of testing the good neighbour hypothesis. It is possible that the contribution of xylary parenchyma cells, in particular, to lignification occurs as compensation in response to the miRNA activity in tracheary elements and fibres. To test this point, we engineered Arabidopsis plants to express enzymes that generate “novel” monolignols within specific cell types, and then examined the cell walls of these plants for the presence of the novel monolignols, to provide further evidence that the targeted cell type is involved in lignification. The enzymes of interest for this experiment are a feruloyl-monolignol transferase (FMT) from Angelica sinensis and a p-coumaroyl-CoA-monolignol transferase (PMT) from Oryza sativa. The corresponding genes were placed under the control of the previously described promoters (proCESA7, proAtPRX47, proAtPRX64). The presence of unusual monolignol-conjugates in the transformed plants was analyzed by 2D-NMR, derivatization followed by reductive cleavage (DFRC) and confocal microscopy. These enzymes and the related cell wall conjugates have only recently been discovered and described, and this chapter explored the feasibility of using these enzymes, and the resulting conjugates, as a tool to test hypotheses about the roles of specific cell populations in the process of lignification.    25 Chapter 2: Lignification before and after programmed cell death in tracheary element development  2.1  Introduction During xylem development, tracheary elements and fibres synthesize their secondary cell walls, and then undergo programmed cell death as the final stage of development.  Early autoradiography studies on xylem development demonstrated that radiolabelled precursors or intermediates of the phenylpropanoid pathway were metabolized into products that radiolabelled the cell wall of tracheids and vessels during the developmental stages when the lignified cell retained its cytoplasmic contents. More recent studies in cell culture, however, have argued for post-mortem lignification (Pesquet et al., 2010, 2013). These studies have suggested that programmed cell death may actually trigger the release of stored monolignols, and is therefore required for lignification. The question, therefore, is when does lignification occur relative to programmed cell death during the development of lignified cells? This chapter addresses the issue of the timing of lignification during Arabidopsis protoxylem tracheary element formation.  2.1.1 Using autoradiography to label monolignols Unlike the polysaccharides found in the cell wall, such as cellulose, hemicelluloses and pectin, the phenolic polymer lignin does not have a regular or predictable structure. The variety of monolignols, and linkages formed between the monolignols, makes it difficult to predict the structure of lignin, as the structure will vary from one plant to another and between lignified cell types. Moreover, the structural irregularity of lignin is problematic when developing antibodies to specifically label lignin. Recently, fluorescently-tagged monolignols have been developed (Tobimatsu et al., 2011, 2013), which makes it easier to localize the monolignols within the cell wall but these are only suitable for tracking monolignols that have polymerized in the wall, and not those within the cells, as the fluorescent tag changes the solubility characteristics of the monolignols.  In contrast to antibody approaches, autoradiography is a technique in which a radiolabelled compound is fed to a tissue and is metabolized by that tissue. The radiolabelled tissue is then sectioned, the section is coated with a photographic emulsion and the decay events   26 from the radiolabelled compounds within the section emit beta-particles that react with the silver ions in the emulsion to produce metallic silver. When the emulsion layer is developed after a set incubation period, unreacted silver ions are removed, but the path of the beta-particles through the emulsion is revealed by the accumulation of reacted silver grains. The decay paths associated with each radiolabelled molecule are spatially diffuse, but it is possible to approximate the origin of the beta-emission, based on probability analysis.  Tritiated phenylalanine is a natural choice to feed to lignifying plants for autoradiography, as phenylalanine is the main precursor to the entire phenylpropanoid pathway and amino acid permease transporters function in the uptake of phenylalanine (Okumoto et al., 2002; Lee et al., 2007). Phenylalanine can be metabolized into monolignols (and other phenolic compounds) and therefore into lignin, but can also be used in protein synthesis. However, the incorporation of phenylalanine into proteins can be blocked during the feeding experiment through the simultaneous application of protein synthesis inhibitors, such as cycloheximide (Kerridge, 1958), which permits the observation of de novo phenolic production. High pressure freezing and freeze substitution of tissue samples that have been incubated with 3H-phenylalanine (3H-Phe) allows for the radiolabelling of metabolized phenylalanine and the immobilization of the radiolabel within the cell and cell wall, while maintaining excellent preservation of the cellular ultrastructure (Kaneda et al., 2008).  The basis for using autoradiography in this study is an earlier study performed in lodgepole pine (Pinus contorta var latifolia) (Kaneda et al., 2008). In this study, high performance liquid chromatography (HPLC) analysis showed that 3H-Phe is metabolized into monolignols, but not monolignol glucosides, during lignification in pine, although unlabelled conferin was readily detected in the P. contorta samples. Spatial distribution of the radiolabel within the P. contorta wood was tracked with autoradiography, which demonstrated that the cell wall had the strongest label, with lower levels of labelling in the cytosol, Golgi, and vacuole (Kaneda et al., 2008). Cytosol, Golgi and vacuole labelling was decreased with the application of cycloheximide, but addition of piperonylic acid, an inhibitor of the phenylpropanoid pathway (Schalk et al., 1998), only significantly decreased the label in the cytosol, indicating that the label in the Golgi and vacuole was predominantly from proteins. The major finding from that   27 study was therefore that monolignol export from the site of synthesis to the cell wall is probably not Golgi-mediated (Kaneda et al., 2008).  A similar study in poplar wood examined not only the localization of radiolabel within one cell type, but within different cells in the wood. In this study, ray cells treated with 3H-Phe and cycloheximide did not display as large a decrease in cytosolic label as expected (Kaneda, unpublished data). The persistence of radiolabel despite protein synthesis inhibition suggests that the ray cells are synthesizing some phenylpropanoid compounds from the radiolabelled phenylalanine. Ray cells are known to produce some phenolic compounds (Ranocha et al., 2002), and since they appear to have some of the necessary enzymes for monolignol biosynthesis (Chen et al., 2000), it is possible that they are contributing to the lignification of neighbouring lignified tracheary elements and fibres in the wood. To test this intriguing possibility, and to learn more about the timing of lignification, a simpler model system, Arabidopsis thaliana, was chosen.   2.1.2 The Arabidopsis root as a model system for lignification studies Zinnia elegans and Arabidopsis thaliana cell culture systems are popular systems for the study of lignification for a number of reasons. One of the major advantages of these culture systems is the ability to synchronize protoxylem differentiation by using appropriate hormone treatments (Fukuda, 1997). Large populations of cells can be induced to synchronously differentiate into tracheary elements, which enables the analyses of gene expression patterns across different developmental stages of “transdifferentiating” tracheary element-like cells (Demura et al., 2002; Milioni et al., 2002; Oh et al., 2003; Kubo et al., 2005; Pesquet et al., 2005). Another major advantage of this system is that only two cell types remain in the culture after hormone-induced transdifferentiation. One cell type is the tracheary element-like cell and the other is the parenchyma-like cell (McCann et al., 2001; Pesquet et al., 2010). The parenchyma-like cells respond differently to signals from the hormones, and therefore remain in a more undifferentiated state. The fact that such cells exist, and are retained in the culture, has led to speculation that these parenchyma-like cells may have a role in lignification of the tracheary element-like cells (McCann et al., 2001). The synchronicity and restricted number of cell types present are two key reasons why the relatively simple cell culture system has been   28 valuable in the study of lignification, but the simplicity of this system is also its major weakness, as it cannot fully emulate what actually occurs within the whole plant. An alternative model system is the Arabidopsis seedling root, where the in planta developmental context is preserved, and more cell types are present compared to the cell culture systems.  The Arabidopsis root is a very simple organ that possesses a limited number of distinct cell types (Figure 2.1). In the young 7-day-old root, there are very few lignified cells. These cells include the two protoxylem tracheary elements (Dolan et al., 1993) and the Casparian strip region of the endodermal cell wall (Naseer et al., 2012). The only differentiated cells present within the xylem in the region proximal to the root tip are therefore the tracheary elements and xylary parenchyma cells, which makes this region in the root system analogous to the cell culture systems. The onset of differentiation and lignification of the two protoxylem tracheary elements is not entirely synchronous, but there are only slight temporal differences in the development of the two cells. Another advantage of the root model system is that, from the root tip to regions distal from the root tip, all stages of tracheary element differentiation, such as secondary cell wall deposition (Figure 2.2) and lignification (Figure 2.2, 2.3) can be observed sequentially along the length of the root. Thus, the objective of this study was to use the Arabidopsis seedling root as a model for lignification to determine the timing of lignification with respect to programmed cell death. To demonstrate when lignification occurred during tracheary element development, I used cryofixation and autoradiography to label phenylalanine-derived compounds in living and dead tracheary elements, which were identified based on the presence or absence of cytosolic contents.  2.2 Results  2.2.1 Secondary cell wall labelling in living tracheary elements In order to determine whether tracheary elements are living or dead during lignification, Arabidopsis seedling roots were treated with 3H-phenylalanine (3H-Phe), with or without cycloheximide, for two hours. The treated roots were then high pressure frozen, freeze-substituted, and embedded in epoxy resin. Thick and thin resin sections were examined by    29  Figure 2.1 A schematic of an Arabidopsis root cross-section.  Asterisks indicate the location of protoxylem tracheary elements.    EpidermisCortexEndodermisPericyclePhloemPhloemXylemResidual ProcambiumResidual Procambium* *  30                                        Figure 2.2 Cellulose deposition precedes lignification in Arabidopsis root protoxylem. A) Brightfield image of a root with one visible protoxylem tracheary element with helical secondary cell wall thickenings (arrow); B) Calcofluor staining of the same root showing cellulose staining the secondary cell wall (arrow); C) Basic Fuchsin staining of the root showing intense secondary cell wall lignin staining (arrow) in more mature region of the root, relative to the cellulose staining. Scale bar = 10µm. root tipABC  31  Figure 2.3 A wild-type root stained with the lignin indicator Basic Fuchsin. A) Lignified protoxylem tracheary elements (asterisk) fluoresce more intensely than surrounding tissue with the Basic Fuchsin stain; B) Higher magnification of the boxed region showing that the intensity of lignin staining changes drastically from one tracheary element to the next. Scale bars = 10µm.   TE1TE2TE3B*A*  32 microautoradiography for the localization of radiolabelled phenylpropanoids within the cells, as well as in their cell walls.  To view the distribution of silver grains associated with phenylpropanoid label among the different root tissues in control 3H-Phe samples (no cycloheximide), light microscopy was employed. Very dense radiolabel was apparent at the root tip and throughout the elongation and differentiation zones in all cell types. For each autoradiograph, a similar reference image, stained with toluidine blue, from the same sample was captured to provide information about the protoxylem cell structure (Figure 2.4A and B). In control 3H-Phe samples, radioactivity was found throughout the cytoplasm of all cell types, as expected from protein translation incorporating radiolabelled 3H-Phe (Figure 2.4A and B), although label was most intense in the lignifying tracheary element cell walls.  To further examine the tracheary element radiolabel pattern, transmission electron microscopy (TEM) was used. TEM analysis detected label both in the cytoplasm and in the secondary cell wall thickenings of tracheary elements (Figure 2.4C and D). While some of the tracheary elements observed with secondary cell wall label no longer had cytoplasmic contents (i.e. PCD had already occurred), others displayed intact cytoplasmic contents and were therefore presumably still living and metabolically active. Specifically, of 54 3H-labelled developing tracheary elements observed in the correct developmental stage and plane of section, 18 appeared to be living and 36 were dead. This result suggests that tracheary elements are living when lignification of the cell wall is occurring.  To ensure that the label detected in the cell wall was associated with phenolic compounds, protein synthesis inhibitors were included in the 3H-Phe treatments. When cycloheximide was used to block protein synthesis, radiolabel was still retained in the secondary cell walls of tracheary elements (Figure 2.4E and F). On the other hand, when piperonylic acid, an inhibitor of the C4H enzyme, was included as an inhibitor in the treated seedlings, cell wall label from 3H-Phe in tracheary elements was no longer observed (Figure 2.4G and H). These results indicate that the labelled metabolites in the cell wall represent a downstream product of the general phenylpropanoid pathway, and, given the developmental context of the differentiating xylem cells, that this labelling pattern is likely a reflection of the incorporation of 3H-labelled monolignols into lignin.      33 Figure 2.4 Root microautoradiography on 3H-Phe-treated roots.  A) Reference light microscopy image of toluidine blue stained section through differentiating protoxylem tracheary elements from control 3H-Phe-treated roots (asterisks); B) Sections from the same 3H-Phe-treated sample prepared for autoradiography, with radioactive decay events in all cell types, especially tracheary element cell walls (asterisks); C) TEM of 3H-Phe-treated control root with radioactive decay; D) Enlargement of boxed area in C, showing tracheary element and neighbouring cells with dense label; E) Reference light microscopy image of toluidine blue stained section through differentiating protoxylem tracheary elements from 3H-Phe and cycloheximide (CH)-treated roots; F) Sections from the same 3H-Phe and CH-treated sample prepared for autoradiography, with more restricted pattern of decay events in tracheary elements (asterisks) and in cortical cells; G) Reference light microscopy image of toluidine blue stained section through differentiating protoxylem tracheary elements from 3H-Phe and piperonlyic acid (PA)-treated roots; H) Corresponding autoradiograph from the same 3H-Phe and PA-treated root with dense label in all cells except tracheary element cell walls.  (A and B scale bars=15µm, C and D scale bars=2µm, E and F scale bars=10µm and G and H scale bars=15µm). Asterisks indicate protoxylem tracheary elements, TE = tracheary element, 2˚CW= secondary cell wall, 1˚CW = primary cell wall.  Images reproduced with permission from Smith et al. (2013). Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org). Figure 2.4. Root microautoradiography o 3 -Phe-treated roots. AE FC DH**TE2oCW1oCW2oCW1oCW* * * **B**G*  34  2.2.2 Phenylpropanoid metabolites do not accumulate within the tracheary element cells Tritiated phenylalanine-treated roots had both cytosolic and vacuolar label, as well as cell wall label, in the tracheary elements. However, in this experiment, any labelling of phenolics, such as monolignols, would be indistinguishable from label incorporated into proteins. With the addition of the protein synthesis inhibitor, cycloheximide, most of the cytosolic label in the tracheary elements was lost (Figure 2.4F, Figure 2.5A and B). Label in the cytosol of living cells was at the background levels, as was vacuolar label, even after extended exposure (2-3 weeks) of the labelled sections to the emulsion (Figure 2.5B). Radiolabel above background levels was only observed in the lignified secondary cell walls of the tracheary elements. This result was interesting, as a number of previous studies have suggested the possibility that monolignols may be converted to monolignol glucosides and stored within the vacuole prior to lignification of the cell wall. The low level of radiolabel observed in the cytosol of the living tracheary elements suggests that phenylalanine is efficiently metabolized to form monolignols, and that those monolignols are then exported rapidly from the site of synthesis to the secondary cell walls. As non-lignified cells adjacent to tracheary elements have been suggested to play a role in lignification, the subcellular distribution of radiolabel in parenchyma cells adjacent to the tracheary elements was also examined in 3H-Phe and cycloheximide-treated samples. In these cells, as in the tracheary elements, 3H-labelled compounds were not observed to accumulate within the vacuole or within the cytosol (Figure 2.5C). This study therefore could not offer any conclusive evidence as to whether neighbouring cells are involved in the lignification of tracheary elements or not.  Because sample preparation involves an extensive dehydration in acetone at -80oC during the freeze substitution, followed by room-temperature resin infiltration steps, there was a possibility that 3H-labelled phenolic compounds might have been extracted from the vacuole, especially as levels of radioactivity above background were detected by liquid scintillation counting of resin samples recovered from the resin infiltration process. Such an extraction of radiolabelled metabolites from within the cells would affect the interpretation of autoradiography results. However, using transmission electron microscopy, cell types outside the stele, such as cortical cells, were observed to possess vacuolar label at levels above background in 3H-Phe and    35  Figure 2.5 Monolignols do not accumulate within vacuoles or cytoplasm of root tracheary elements during lignification.  A) TEM image of living tracheary element with secondary cell wall thickenings; B) TEM autoradiograph, with outlines superimposed over key cellular features, of a 3H-Phe and cycloheximide-treated living tracheary element showing dense label at secondary cell wall thickenings but not in the cytosol or vacuole; C) TEM image of a cells neighbouring the tracheary elements with low levels of intracellular label; D) TEM image of a cortical cells with vacuolar label above background levels. V=vacuole, cyt=cytosol, 2˚CW=secondary cell wall, 1˚CW=primary cell wall. Scale bars in A), B) and D) 500 nm, scale bar in C) 2µm. Images reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).    Vcyt2ÝCW2ÝCW1ÝCWAVB2ÝCWcyt2ÝCW1ÝCWCDTETE2ÝCW1ÝCWVcytcytV1ÝCW  36 cycloheximide-treated roots (Figure 2.5D). The presence of vacuolar label in cells not undergoing lignification provides an internal control, and indicates that phenolic compounds are not extensively extracted during sample preparation. Therefore, the lack of vacuolar label in tracheary elements indicates that phenylpropanoid compounds did not accumulate in the vacuole during the two-hour 3H-Phe incubation period of this experiment. Similar results were found for developing lodgepole pine wood, following a four-hour incubation period with 3H-Phe and cycloheximide (Kaneda et al., 2008).  There is a possibility that labelling of the roots for such a short period of time may result in a lack of vacuolar label if monolignols are stored in the vacuole early in tracheary element development, and de novo monolignols synthesized after the 3H-Phe treatment are not routed through the vacuole. To address this possibility, HPLC analyses of methanol-soluble phenolic root extracts were investigated. A number of controls were first performed to determine which soluble phenolic metabolites could be detected in untreated roots, in roots incubated in 0.2M sucrose (the solution used for 3H-Phe treatments), and in roots treated with cycloheximide. The phenolic profiles of roots after each of these treatments appeared similar, and therefore roots treated with 3H-Phe without added inhibitor were subjected to phenolic extraction and HPLC analysis.  Based on UV detection at 280 and 320 nm, the roots were found to accumulate some phenolic compounds, but no peaks could be detected for the monolignol glucosides or monolignols (Figure 2.6). This contrasts with other studies that have observed monolignol glucoside peaks, but not monolignol peaks, in extracts from root tissue (Hemm et al., 2004; Lanot et al., 2006). As the production of monolignol glucosides is induced by light (Hemm et al., 2004), and the roots used in this analysis were grown hydroponically, the roots may have been shielded from light by the leaves, thus preventing the production of glucosides. More importantly, when samples of the column effluent were collected during HPLC fractionation of the radiolabelled samples and analyzed by liquid scintillation counting, no radioactivity above background was observed at the monolignol elution times. These data argue against newly synthesized monolignols accumulating within the vacuole of tracheary elements, although it is possible that, given the short developmental window for tracheary elements and the low number    37   Figure 2.6 HPLC chromatographs of soluble phenolic extracts from WT and MYB63/58 overexpressing roots at 280 nm. Lines indicate the elution time of standards of interest: 1, coniferin (coniferyl alcohol 4-O-glucoside); 2, syringin (sinapyl alcohol 4-O-glucoside); 3, coniferyl alcohol; 4, sinapyl alcohol.   WTpro35S::MYB63pro35S::MYB581 2 3 4  38 of cells in the Arabidopsis root, the pool of radiolabelled monolignols would be below our limits of detection.  To determine if the detection of monolignols in roots was possible, I analyzed the seedling roots of transgenic Arabidopsis lines that were over-expressing two transcription factors (MYB58 and MYB63), known to be activators of the monolignol biosynthetic pathway (Zhou et al., 2009). Overexpression of these transcription factors causes extensive ectopic cell wall lignification without inducing programmed cell death or ectopic deposition of cell wall polysaccharides. I hypothesized that the increased flux of carbon through the phenylpropanoid pathway in these MYB-overexpressing plants would improve the chance of capturing a pool of monolignols prior to their polymerization. In these seedling roots, no peaks were observed at the coniferyl alcohol and sinapyl alcohol elution times (Figure 2.6). This experiment demonstrated that tissue accumulation of monolignols could not be detected in Arabidopsis primary roots when using an appropriate extraction and analysis protocol. The accumulation of monolignols may therefore be too low to be quantified accurately. The phenomenon of small metabolic pool sizes for monolignols is consistent with a model in which monolignols are rapidly exported after synthesis, a model that was also suggested by the autoradiography demonstration that monolignols are not sequestered within the cell cytosol or vacuole. To further test the hypothesis that the rapidity of monolignol movement out of the cell was the reason for low cytosolic radiolabel, radiolabelling treatments with 3H-phenylalanine and cycloheximide were carried out at 4°C in an effort to slow the monolignol export, and thereby capture more cytosolic label. This treatment did not, however, slow monolignol export sufficiently to cause an accumulation of monolignols within tracheary elements or neighbouring cells (Figure 2.7A and B). As an alternative approach, I attempted to block monolignol export from the cell into the cell wall by using the inhibitors vanadate and CCCP, together with cycloheximide. Vanadate is a general inhibitor of all P-type transporters and is therefore expected to block all the active transporters in the plasma membrane, such as ABC transporters (Goodno, 1979). However, because vanadate treatment is lethal to seedlings, seedlings were only treated with this inhibitor for an hour. In addition, the inhibitor was applied after a one-hour treatment with 3H-Phe and cycloheximide, as it was possible that vanadate might disrupt      39  Figure 2.7 Treatment with inhibitors or cold temperature did not slow monolignol export.  A) Reference image of a toluidine-stained root, treated with 3H-Phe and cycloheximide at 4˚C, with a developing tracheary element; B) Microautoradiograph from the same root showing dense tracheary element cell wall label; C)Toluidine blue-stained root treated with 3H-Phe, cycloheximide and CCCP; D) Corresponding autoradiograph showing little to no label within the root; E) Toluidine blue-stained root treated with 3H-Phe, cycloheximide and vanadate showing two protoxylem tracheary elements; F) Microautoradiograph of the same root showing dense cell wall label and no accumulation of monolignols within the cytosol (see TEM inset). Asterisks indicate protoxylem tracheary elements. Scale bars = 15 µm, TEM inset scale bar = 5µm.     A BC D* *****E* **2oCW*F*  40 phenylalanine uptake into the root. Carbonyl cyanide m-chlorophenyl hydrazine (CCCP), on the other hand, acts as a proton gradient uncoupler (Mahler and Cordes, 1966) and would therefore be predicted to suppress monolignol export if the export were dependent upon a proton-coupled transporter.  The results obtained with both inhibitors were inconclusive. Roots treated with CCCP failed to take up the 3H-Phe and therefore very little labelling was observed anywhere in the root (Figure 2.7C and D). The roots treated with vanadate did have cell wall labelling, but still displayed only background levels of intracellular label in the tracheary elements (Figure 2.7E and F). This might indicate that monolignol export across the plasma membrane is not an ATP-mediated process, but it is also possible that the incubation time with vanadate was insufficient to cause an accumulation of monolignols, or that treating the roots with vanadate led to membrane damage that allowed monolignols to leak from the cell interior into the cell wall.  Surprisingly, the cells in the 3H-Phe and cycloheximide autoradiographic study with the strongest cytosolic label were cortical cells, which are not predicted to have a role in lignification. The label accumulating in these cells may represent soluble phenolics, as it has been reported that Arabidopsis seedling roots accumulate elevated levels of soluble phenolics when grown in the light (Lanot et al., 2006; Lanot et al., 2008).   2.2.3 Radiolabel appears in the cell wall of tracheary elements that have just undergone PCD but not those that are long dead Autoradiography revealed 3H-radiolabelling in secondary cell walls of both living and dead tracheary elements. One of the questions raised by this observation is: if lignification begins in living cells, does it continue after programmed cell death, and if so, for how long? As mentioned earlier, one of the major advantages of the seedling root model system is that all stages of tracheary element development and lignification can be observed sequentially along the root axis. Thus, we hypothesized that evaluating the amount of cell wall radiolabel in tracheary elements at different developmental stages along the root, (assuming that each tracheary element follows approximately the same developmental trajectory), would illustrate the duration of lignification in situ. To test this hypothesis, seedlings were treated with 3H-Phe and cycloheximide for two hours, which is sufficient for extensive labelling (Kaneda and Rensing,    41  unpublished data), and therefore any label observed represents de novo monolignol biosynthesis and lignification. Living tracheary elements had dense secondary cell wall label, as did tracheary elements that had just undergone programmed cell death (Figure 2.8). The “just dead” tracheary elements were identified based on the absence of cytoplasmic contents and their proximity to living tracheary elements and the root tip, relative to other dead tracheary elements. The label in the cell wall of these newly dead tracheary elements could come from the protoplast that had recently undergone programmed cell death, or it could be deposited during or after programmed cell death.  As the distance from the region of programmed cell death increased, the amount of labelling in the secondary cell walls of dead tracheary elements decreased to background levels (Figure 2.8). While the label in the cell walls of newly dead tracheary elements may have been deposited while the tracheary element was still living, it is very unlikely that the cell wall label in the older tracheary elements was deposited prior to programmed cell death given the short incubation period with 3H-Phe. This label must therefore be coming from outside the tracheary elements. One hypothesis is that parenchyma cells neighbouring the tracheary elements are synthesizing monolignols and exporting them to the neighbouring cell wall. Such co-operative lignification could potentially continue indefinitely, because parenchyma cells do not undergo programmed cell death during their development, but the lack of 3H-Phe label in long dead tracheary elements argues against this theory. Rather, it seems that lignification might begin pre-mortem with autonomous monolignol production, continue post-mortem with contributions from neighbouring parenchyma cells and then cease at some point later in development. What cues might trigger this cessation of lignification, whether it is a simple case of lack of space in the cell wall for further monolignol polymerization, or some other more tightly regulated signal, are unknown.  2.2.4 Lignification in the Arabidopsis stem also begins pre-mortem and continues post-mortem Arabidopsis roots have only a few lignified cells and therefore autoradiography was also performed on Arabidopsis stem sections to examine a greater diversity of lignified cell types.  Metaxylem tracheary elements and xylary fibres can be found in the Arabidopsis stem xylem, in    42 Figure 2.8 Lignification does not continue indefinitely after programmed cell death. A model of the primary root developmental gradient showing differences in the amount of radiolabel incorporated in secondary cell walls of tracheary elements (red lines) during the two hour incubation period: i) strong label in living TEs, ii) strong label in TEs that have just undergone PCD, iii) less label in slightly older TEs, and iv) no/background label in TEs that are long dead. Scale bars = 15 µm. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).   i ii iii ivLiving Long DeadJust Deadi iiiii iv  43 addition to protoxylem tracheary elements. One-centimeter stem segments located 5cm from the shoot apical meristem of mature plants (about 2 months old) were treated with 3H-Phe for 2 hours, high pressure frozen, freeze-substituted and resin-embedded for subsequent microautoradiography analysis. At this stem developmental stage, the majority of living cells that possessed thick secondary cell walls were fibres, but dead protoxylem and metaxylem tracheary elements were also observed. All of the living fibres had dense secondary cell wall labelling, which I attribute to lignification of the cell wall (Fi gure 2.9A and B). The cell walls in the tracheary elements, on the other hand, were labelled only at background levels, most likely indicating that these cells were long dead and no longer undergoing lignification (Figure 2.9A and B). To remove protein rad iolabel, 3H-Phe-treated stems were also incubated with cycloheximide, and the cytosolic labelling observed in the autoradiographs was reduced, compared to non-cycloheximide-treated samples (Figure 2.9B, D). Again, all the living fibres had dense cell wall label, but a metaxylem tracheary element lacking cell contents also displayed strong cell wall labelling (Figure 2.9C and D). This suggests that tracheary elements in the stem, like those in the root, have some capacity for post-mortem lignification. Fibres had dense cell wall label when living, suggesting pre-mortem lignification, but the base of the stem, where the fibres may have undergone programmed cell death, would also have to be examined to determine if lignification also occurred post-mortem in fibres.  2.3 Discussion Cryofixation coupled with autoradiography revealed the in situ distribution of phenylpropanoid metabolites in developing xylem. From the autoradiography data, two conclusions can be inferred: first, lignification precedes and overlaps with  programmed cell death; and secondly, phenylpropanoids are concentrated in secondary cell walls of lignifying cells with little intracellular 3H-labelling, suggesting that monolignols or monolignol glucosides do not accumulate inside cells.  Only background signals were detected in the cytoplasm of both lignifying cells and their non-lignifying neighbours, indicating that the synthesis and transport of monolignols in these cells must be rapid. There was also no evidence of vacuolar or endomembrane accumulation of monolignols in developing tracheary elements prior to programmed cell death.  Accumulation of monolignol glucosides in the vacuole, followed by    44  Figure 2.9 Stem microautoradiography showing label in living and dead fibres and metaxylem tracheary elements.  A) Reference light microscopy image of a toluidine blue-stained 3H-Phe-treated stem in longitudinal section showing metaxylem tracheary elements and fibres; B) Microautoradiograph of the same 3H-Phe-treated stem showing label in the cell wall of living fibres; C) Reference image of a 3H-Phe and cycloheximide-treated stem; D) Microautoradiograph of the same stem with dense cell wall label in living and dead lignified cells. mx, metaxylem tracheary elements; px, protoxylem tracheary elements; f, fibres. Scale bars = 15 µm.   fff ffffmxfffffmxffffACBDmx mxmxpx mxmx mxpx mxmx  45 monolignol release during protoplast lysis, has been hypothesized as a lignification mechanism (Leinhos and Savidge, 1993; Pesquet et al., 2010). Contrary to one recent report (Pesquet et al., 2013), lignification in Arabidopsis cells in this study can be considered a pre-mortem process that does not require programmed cell death for initiation of lignification. It is possible that the radiolabel observed in the secondary cell wall of living tracheary elements is from radiolabelled monolignols and not from the polymerized product, but the early expression of peroxidase and laccase genes during the developmental phase of secondary cell wall deposition (Hall and Ellis, 2013) suggests that these putative polymerization catalysts may be present and functional in the cell wall at this stage of tracheary element development. The monolignols rapidly exported from the cell to the cell wall would then be immediately polymerized within the cell wall. Overall, my results are consistent with a previous autoradiographic study from developing pine wood where 3H-Phe products were prominent in the cell wall but not in vacuoles, as assessed by microscopy, nor in the extractable phenolic pool when monitored by HPLC and scintillation counting (Kaneda et al., 2008).   My data support a lignification model where monolignols or monolignol glucosides are rapidly exported across the plasma membrane for crosslinking in the cell wall.  A recent study in Arabidopsis characterized a subset of !-glucosidases that display the ability to hydrolyze monolignol glucosides, and demonstrated that, while their loss of function did influence the size of coniferin and coniferyl alcohol pools, it did not change the extent of tissue lignification, suggesting that monolignol glucosides are not the direct precursors of lignin (Chapelle et al., 2012).  The direct use of monolignols, and not monolignol glucosides, in lignification is also supported by another lignification model system, the Casparian strip in the endodermis of roots, in which lignification occurs gradually and without prior accumulation of monolignols or their glucosides (Naseer et al., 2012; Lee et al., 2013).  ABC transporters in the plasma membrane are hypothesized to facilitate the rapid export of monolignols (Miao and Liu, 2010; Alejandro et al., 2012), although vanadate treatment in my experimental system did not lead to monolignol accumulation or suppression of monolignol export to the cell wall.  However, it is difficult to interpret the potential effects of such a broad effect inhibitor.    46 While this study used phenylalanine as a precursor for monolignols and lignin, early microautoradiography studies of in planta lignification made use of a number of precursors and intermediates in the phenylpropanoid pathway to label the lignin. One of the first studies used tritiated cinnamic acid as a precursor for lignin and used chemical fixation of the tissue to immobilize the metabolized precursor (and any remaining free precursor) after a set incubation time (Pickett-Heaps, 1968). The incorporation of radiolabelled cinnamic acid into the lignified secondary cell wall of developing xylem with intact protoplasts in that study indicated that phenylpropanoids were being deposited in secondary cell walls prior to cell death. However, grasses such as wheat are known to bind phenolic acids into non-lignified parts of the cell wall and therefore the labelling pattern observed by Pickett-Heaps (1968) may not have been an accurate reflection of lignin deposition in the cell wall (Harris and Hartley, 1976). An even bigger problem, though, is that this study used chemical fixation, a technique that results in vesiculation of the plasma membrane and led the author to the incorrect conclusion that monolignol export is likely vesicle-mediated. Analogous studies in pine (Pinus thumbergii) used radiolabelled coniferin, a monolignol glucoside, as a precursor for lignin (Terashima et al., 1988a,b), and these authors were able to show that incorporation of radiolabelled metabolites occurs while the protoplast of tracheids is still intact (Takabe et al., 1985; Terashima et al., 1986). However, the size of the coniferin pool that accumulated prior to lignification was considered to be insufficient to account for all the lignification that occurred in the pine tissue during the experiment (Fukushima et al., 1997). The question of which upstream precursors should be used to label the pool of endogenous monolignols in developing xylem, and ultimately lignin, is important. Re-metabolism of phenylpropanoid intermediates can potentially lead to label reaching lignin through detoxification pathways that may or may not be physiologically relevant.  For example, coniferin may be converted into coniferaldehyde glucoside, which can be deglucosylated into coniferaldehyde, an intermediate in the monolignol biosynthesis pathway (Tsuji et al., 2005). Feeding exogenous monolignols in high concentrations could lead to toxicity responses in cells. One compromise between feeding a precursor and preventing toxicity is the use of 3H-phenylalanine (Kaneda et al., 2008), which proved successful in this study.     47 2.4 Methods  2.4.1 Plant growth conditions and phenylpropanoid radiolabelling Arabidopsis thaliana (Col-0) seeds were grown on !  MS media (Sigma) under continuous light until seedlings were 7-days-old and then seedling roots were used for treatment with radiolabel and subsequent high pressure freezing. For autoradiography of seedling roots, 7-day-old whole seedlings were placed in a solution containing 0.25mL of 0.2M sucrose and 12.5 µCi of L-[2,6-3H]Phe (0.9 µM Phe; GE Healthcare) in 2 mL Eppendorf tubes and incubated at room temperature for 2 hours.  Separate treatments were set up containing 3H-Phe with or without 10µM cycloheximide (Sigma),10µM piperonylic acid (Aldrich), 2µM CCCP (Sigma) or 100µM vanadate (Sigma).  After the 2-hour incubation, the radiolabelled roots were excised and high pressure frozen for subsequent sectioning and autoradiography. Alternatively, 7-day-old seedling roots were used for histochemical staining. Whole seedlings were stained in 0.0001% Basic Fuchsin in 95% ethanol for 5 minutes, followed by de-staining for 2 minutes in 70% ethanol.  Seedling roots double-stained with Calcofluor White and Basic Fuchsin stains were first stained with Basic Fuchsin, as described above, and then stained with 3.5%(w/v) Calcofluor White in 0.1M Tris-HCl, pH 9.0 for 5 minutes. Seedlings were mounted in 50% glycerol, and the fluorescence was observed using a Leica DMR light microscope with a DAPI filter set (excitation 340-380nm, emission 450nm) for Calcofluor White cellulose fluorescence and a red filter set (excitation 560nm, emission 645 nm) for Basic Fuchsin lignin fluorescence.  Images were captured using a Q-CAM digital camera (Q-Imaging) using identical imaging conditions for all genotypes.  For stem autoradiography, 7-day-old seedlings were transplanted to soil (Sunshine mix 4, Sungrow Horticulture) and grown in a growth chamber at 21°C under 16/8 hours light/dark cycles.  Plant inflorescence stems were destructively sampled after 2 months and the region 5cm from the shoot apical meristem was treated with 3H-phenylalanine with or without 10µM cycloheximide (Sigma). After a 2-hour radiolabel treatment, the stems were high pressure frozen.     48 2.4.2 High pressure freezing Seedling roots were high pressure frozen in 1-hexadecene (as a cryoprotectant) using a Leica EM HPM 100 high pressure freezer.  Samples were freeze substituted with 2% osmium tetroxide in 8% dimethoxypropane in acetone for 120 hours in a dry ice-acetone bath at approximately -80oC as previously described (Young et al., 2008).  Acetone was gradually replaced with Spurr’s resin (Spurr, 1969).  Samples were embedded in fresh Spurr’s resin in flat-bottomed beam capsules and allowed to polymerize overnight at 60°C.  Extraction of radiolabel during fixation and embedding was assessed by collecting 50-100 µL of the fixative or resin solution, adding it to 3 mL Fisher ScintiVerse scintillation cocktail and counting the number of decay events per minute in a Beckman LS600IC liquid scintillation counter.  2.4.3 Autoradiography – light microscopy Sections (300-350 nm) were made using a Leica Ultracut microtome and heat-fixed onto glass slides.  To control for differences in emulsion thickness, and to facilitate comparison between treatments, sections from both 3H-Phe and 3H-Phe plus cycloheximide samples were placed on the same slides.  For every three sections prepared for autoradiography, one section was placed on a separate glass slide and stained with 1% Toluidine Blue-O in 1% sodium borate as a reference for the autoradiography sections.  Slides for autoradiography were dipped in a 50% v/v aqueous solution of Ilford L4 emulsion (PolySciences, Inc.) at 40°C under a sodium safelight with closed filters (Thomas Duplex Super Safelight).  Dipped slides were stored in a lightproof box at 4°C for 2 days.  The emulsions were developed in Kodak D19 developer solution (1:1 dilution in water) for 5 minutes, rinsed in distilled water, fixed in a 10% (v/v) aqueous solution of Ilford Multigrade Paper Fixer and gently rinsed under cold running water for 15 minutes.  The autoradiographs, and the Toluidine Blue-stained reference sections, were mounted in 50% glycerol and observed using a Leica DMR light microscope and Q-CAM digital camera (Q-Imaging).    2.4.4 Transmission electron microscopy Thin sections (70-75 nm) were made on a Leica Ultracut microtome using a diamond knife and placed on copper Athene polyslot grids (Canemco) coated with 0.3% Formvar in 1,2-  49 dichloroethane. The grids were stained for 20 minutes with 2% uranyl acetate (w/v) in 70% methanol and 5 minutes with Reynold’s lead citrate and then were carbon coated (Bal-Tec MED 010 Evaporator and Glow Discharge Apparatus). Reference grids remained uncoated and were examined and photographed using a Hitachi H7600 Transmission Electron Microscope with AMT Advantage CCD camera. Under a sodium safe-light, 7 mm diameter wire loops were dipped in 1:1.7 Ilford L4 emulsion:distilled water solutions at 40°C, placed horizontally until almost dried and then stained and carbon-coated autoradiography grids were placed sample side down on the emulsion. The emulsion was allowed to dry completely before the grid plus emulsion were removed from the loop, placed in a grid box, wrapped in aluminum foil and placed in a light-tight photographic storage bag. Grids were stored at 4°C for 2-3 weeks before emulsion development. Emulsions were developed as described above for light microscopy thick sections.  2.4.5 HPLC – soluble phenolic analysis Root tissue from wild type or pro35S::MYB58/MYB63 plants for soluble phenolic analysis was grown hydroponically in a nutrient solution containing 1 mm KH2PO4, 0.5 mm MgSO4, 0.25 mm CaSO4, 20 µm Fe-EDTA, 25 µm H3BO3, 2 µm ZnSO4, 2 µm MnSO4, 0.5 µm CuSO4, 0.5 µm Na2MoO4 and 1 mm NH4NO3 (Yong et al., 2010).  Samples of 2-week-old root tissue (100-200 mg) were ground in a “methanol water” solution (49.5% methanol:1% acetic acid in water) using Lysing Matrix Tubes in a FastPrep FP20 (Thermo Electron Corporation).  Samples were incubated at 45°C for 4 hours to extract soluble phenolic compounds, centrifuged at 15 000 rpm for 15 minutes, and the supernatant was transferred to a glass vial. Phase partitioning with 1 mL diethyl ether was performed three times and pooled ether phases were allowed to evaporate to dryness overnight before re-suspension in 50% methanol, sonication for 15-20 minutes and incubation at 35°C for 1 hour.  Re-dissolved samples were filtered (0.22 micron) into HPLC vials and the identity of soluble phenolic compounds was determined by separating samples on a Summit HPLC (Dionex) fit with a Symmetry C18 column (Waters) held at 400C and a PDA-100 Photodiode Array Detector (Dionex).  Sample components were eluted from the column at a flow rate of 1 mL/minute using a gradient from 95% A (100% water:0.1% trifluoroacetic acid (TFA)) to 45% B (75% acetonitrile:25% methanol: 0.1% TFA) over 50   50 minutes, followed by a 10-minute wash with 75% B and re-acclimation of the column with 95% A for 10 minutes.     51 Chapter 3: Good neighbours in lignification 3.1 Introduction The objective of this investigation was to test if lignifying cells themselves can provide the monolignols for lignification, or if adjacent non-lignifying cells can act as “good neighbours”. More specifically, the goal was to use the combination of an artificial miRNA targeting the CCR1  lignin biosynthetic gene together with cell population-specific promoters to determine whether tracheary elements, fibres and/or xylary parenchyma cells are contributing to lignification.  3.1.1 Evidence for post-mortem lignification  The autoradiography results described in the previous chapter demonstrated that lignification of tracheary elements, while initially a pre-mortem event, concludes after the tracheary elements have undergone programmed cell death. The idea of post-mortem lignification is not a novel one. Hosokawa et al. (2001) found that exogenously applied coniferyl alcohol stimulated lignification in tracheary element-like cells that had already undergone programmed cell death in the Z innia  culture system. Although the in vitro  system does not accurately depict the lignification scenario in situ , this study demonstrated that tracheary elements could accept monolignols from external sources and effectively incorporate them into the lignin polymer post-mortem. A more recent report examined tracheary element post-mortem lignification in a combination of in vitro and in situ  studies. Exogenous application of monolignols was found to rescue the lignification process in Arabidopsis cell culture tracheary element-like cells in which the phenylpropanoid pathway had been blocked by chemical inhibitors (Pesquet et al., 2013). This study also suggested that programmed cell death is a requirement for lignification, as culture cells that were arrested at the secondary cell wall deposition phase of development did not appear to form lignin in their cell walls (Pesquet et al., 2013). There is, therefore, evidence for post-mortem lignification of tracheary elements. However, the question remains as to where the monolignols required for such post-mortem lignification originate and which cells are actively producing these monolignols.    52 3.1.2  Introduction to the “good neighbours” in Arabidopsis  One popular model for post-mortem lignification is called the co-operative model, or the “good neighbour” hypothesis (McCann et al., 2001). The premise for this hypothesis is that living, but non-lignifying, cells adjacent to lignifying cells could be synthesizing monolignols and exporting them to the cell walls of the neighbouring lignifying cells, thereby allowing lignification to continue even after programmed cell death.  The evidence supporting the “good neighbour” hypothesis in plants has been indirect. The best in planta evidence comes from a study demonstrating that the promoters of key lignin biosynthetic genes (CCR and CAD) from Eucalyptus guneii are active not only in lignifying cells in Arabidopsis, such as tracheary elements and fibres, but also in xylary parenchyma cells adjacent to tracheary elements (Baghdady et al., 2006). The xylary parenchyma cells were shown to have promoter activity even after the neighbouring tracheary elements had undergone programmed cell death (Baghdady et al., 2006). In assessing such promoter activity studies, however, the assumption must be made that the promoter activity assays accurately reflect the activity of their gene, the activity of the encoded enzyme, and the cellular production of monolignols, which may not necessarily be the case.  In the Arabidopsis stem xylem, xylary parenchyma cells are found interspersed with the tracheary elements and fibres. In addition to the roles in general metabolism and storage common to most parenchyma cells, these cells are also the best candidates for “good neighbours” within the xylem. Interfascicular fibres form between the vascular bundles of Arabidopsis and, while non-cell autonomous lignification has not been predicted for these cells (Baghdady et al., 2006), it is possible that cortical or pith cells could be producing monolignols and supplying them to neighbouring fibres. Selaginella moellendorffii deposits S-lignin within the cell walls of its cortical cells (Weng et al., 2008) and it is possible that the ability of cortical cells to produce monolignols is evolutionarily conserved. The Arabidopsis root has fewer types of lignified cells (only tracheary elements), but there is more ambiguity regarding their potential “good neighbours”. Xylary parenchyma cells, if present, are few in number because the root xylem typically consists of approximately five tracheary elements (Dolan et al., 1993). It is more likely that co-operative lignification would involve cells from the pericycle abutting the protoxylem tracheary elements, which in woody species ultimately forms part of the vascular cambium,   53 and/or cells in the residual cambial layer located between the xylem and phloem tissues (see Figure 2.1).  There is no direct evidence for the roles of any of these non-lignifying cell types in lignification, but advances in cell-specific targeting and genetic manipulation may allow us to discern exactly which cells are involved in lignification.  3.1.3  Cell - specific silencing of lignification using artificial microRNAs  Cell-specific manipulation of lignification can be used to directly test the “good neighbour” hypothesis. Silencing or down-regulating lignin biosynthesis in specific cell types enables the assessment of the contribution of subpopulations of cells to lignification. In Arabidopsis, one method for gene silencing is the expression of artificial microRNAs. Endogenous microRNAs (miRNA) occur naturally in plants and play roles in such key developmental processes as establishment of polarity and patterning, and organ morphogenesis and identity (reviewed by Wu, 2013). These small 20-24 base oligonucleotide non-coding regions are transcribed by a polymerase and the resulting pri-miRNAs are processed to their final hairpin structure through the activity of a complex of proteins, including DICER-LIKE1 (DCL1) and HYPONASTIC LEAVES (HYL1) (reviewed by Ossowski et al., 2008; Pashkovskiy and Ryazansky, 2013; Wu, 2013). DCL1 further processes the pre-miRNAs by cleaving the hairpin and releasing the 20-24 nucleotide fragment, a miRNA/miRNA* duplex, which becomes methylated on the 3! end by the HUA ENHANCER 1 (HEN1) protein to ensure that the duplex remains stable and is not targeted for degradation. The miRNAs are exported from the nucleus to the cytoplasm by an exportin (HASTY), after which the guide miRNA strand from the duplex (miRNA) is incorporated into a RNA-induced silencing complex (RISC). The passenger miRNA strand (miRNA*) is typically degraded. The RISC complex contains an ARGONAUTE (AGO) protein that facilitates mRNA cleavage, transcriptional repression or epigenetic modification of the target gene (reviewed by Ossowski et al., 2008; Pashkovskiy and Ryazansky, 2013; Wu, 2013). The relative specificity of miRNAs to their target mRNAs, compared to RNAi for example, has inspired further study into how miRNAs ensure specificity. The set of parameters that has been described for miRNA target specificity includes such conditions as limited mismatches between the mRNA and miRNA,   54 especially between nucleotides 2-12, and no mismatches at bases 10 and 11, which form the mRNA cleavage site (Schwab et al., 2005).  Armed with this information, platforms (such as Detlef Weigel’s artificial microRNA designer, WMD; http://wmd3.weigelworld.org) have been established to facilitate the design of miRNAs that can target any gene of interest. These artificial miRNAs (amiRNAs) are created by using PCR to substitute key nucleotides in an endogenous plant miRNA with nucleotides that will permit the modified miRNA to match the mRNA of interest (Ossowski et al., 2008). The endogenous miRNA backbone used in the WMD system is miR139a/JAW, which is normally involved in leaf morphogenesis and represents one of the first examples of miRNA-induced mRNA cleavage identified in plants (Palatnik et al., 2003). Apart from its ability to target specific genes, an additional advantage to the use of artificial miRNAs is that tissue-specific expression of the amiRNA does not cause off-target silencing effects (Schwab et al., 2006). It has been demonstrated that an amiRNA has limited mobility beyond the tissue in which it is initially expressed, which creates opportunities for tissue-type-specific gene silencing.  Using this amiRNA system, I targeted specific xylem cells for lignin biosynthesis silencing. By silencing lignin biosynthesis in tracheary elements and/or fibres or xylary parenchyma cells, I predicted that I should be able to evaluate the role of each cell type in lignification. An amiRNA was designed to target the Arabidopsis CINNAMOYL-CoA REDUCTASE (CCR1) gene, whose encoded protein catalyzes an essential step in lignin biosynthesis (Jones et al., 2001). CCR1 is the first monolignol-specific enzyme in the lignin biosynthetic pathway and previously characterized ccr1 mutants (irx4, ccr1g) have drastic reductions in lignification, confirming the importance of this gene to lignification (Jones et al., 2001; Mir Derikvand et al., 2008; Thevenin et al., 2011).   3.1.4  Cell - specific promoters for specific artificial microRNA targeting  The effectiveness of a miRNA in silencing its target mRNA is dependent upon the extent of miRNA expression relative to the target gene (Schwab et al., 2006). For this reason, promoters that were not only cell-specific, but also relatively highly expressed were required for CCR miRNA expression in this study. The promoter of the CELLULOSE SYNTHASE7 gene (Taylor et al., 1999) is strongly expressed in Arabidopsis cells that produce thickened secondary cell walls,   55  such as root tracheary elements (Wightman and Turner, 2010), and tracheary elements and interfascicular fibres in stems (Mitsuda et al., 2007). To assess the role of oth er xylem cell types in lignification, a fibre-specific promoter, from the PEROXIDASE64 (AtPRX64) gene, was identified from a study examining the expression pattern of different Arabidopsis peroxidase genes (Tokunaga et al., 2009). Likewise, to assess the c ontribution of xylary parenchyma cells to lignification, a xylary parenchyma-specific promoter, from the PEROXIDASE47 (AtPRX47) gene, reported in the same peroxidase expression study (Tokunaga et al., 2009), was used.    3.2 Results  3.2.1 “Good neighbours” in root tracheary element lignification The effect of miRNA CCR1 on lignification was first examined in the Arabidopsis root protoxylem tracheary elements. As stated earlier, the simplicity of the root and the limited number of lignified cells present makes it an ideal model system. To ensure that the microRNA was capable of down-regulating monolignol biosynthesis, miRNA CCR1 was overexpressed in the roots (pro35S::miRNA CCR1). These roots were stained with Basic Fuchsin, a fluorescent stain that binds to phenolic polymers. Wild-type Arabidopsis roots had bright fluorescence in the mature tracheary elements and in the Casparian strip (Figure 3.1A). The most severely affected pro35S::miRNA CCR1 lines, however, had little to no observable fluorescence in the tracheary elements or Casparian strip, although they did develop unlignified spiral secondary cell wall thickenings typical of protoxylem tracheary elements (Figure 3.1B, C). These results indicate that the miRNA CCR1 was effective in down-regulating lignin biosynthesis in planta.   To assess if root protoxylem tracheary elements can supply their own monolignols, the CCR1 miRNA was expressed specifically in developing cells that deposit thickened secondary cell walls, in this case, tracheary elements. The promoter for the CELLULOSE SYNTHASE 7 (CESA7), which is strongly expressed specifically during secondary cell wall synthesis in protoxylem tracheary elements in the root (Figure 3.1D), was used to drive the expression of the miRNA. In the proCESA7::miRNA CCR1-expressing roots, Basic Fuchsin staining of the tracheary elements revealed a wild-type pattern of lignification (Figure 3.1E ). The tracheary elements therefore appeared to be lignifying normally despite the fact that the activity of the    56  Figure 3.1 Neighbouring cells can rescue tracheary element lignification in Arabidopsis roots. A) WT root stained with Basic Fuchsin. Arrows indicate the protoxylem tracheary elements and asterisks denote the location of the Casparian Strip; B) and C) Loss of lignification when monolignol biosynthesis is downregulated in pro35S:miRNA CCR1 expressing roots; D) Tracheary elements are the only cell types labelled in the late elongation zone by expression of proCESA7:GFP; E) If downregulation of monolignol biosynthesis is restricted to tracheary elements by expression of tracheary element-specific proCESA7:miRNA CCR1, Basic Fuchsin stain indicates lignification similar to WT. Scale bars = 10 µm. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).   A !" # $!" #$%&'()*+,-./0(+1234)5&&36 #$%&'()*+1234)5&&36!"#$%&'()*''+&"#,-.$%"/#'0&112'03/'%&20$&'4%30,&3%5'&1&6&/4'1"#/"7"034"./'"/'8%3-"9.:2"2'%..42)%&'()'*++,'-,./012'3/,4'5.-/6'7864-/09'"&'.02'#&':+--'+7';/<0/7/6.,/+0'3410'=+0+;/<0+;'5/+->0,41-/-'/-'2+30*1<8;.,12'/0'/0(+1234)5&&36'1?@*1--/0<'*++,-9'!&')*.641.*>'1;1=10,-'.*1',41'+0;>'61;;',>@1-';.51;12'/0',41';.,1'1;+0<.,/+0'A+01'5>'1?@*1--/+0'+7'#$%&'()*+,-.9'$&'B7'2+30*1<8;.,/+0'+7'=+0+;/<0+;'5/+->0,41-/-'/-'*1-,*/6,12',+',*.641.*>'1;1=10,-'5>'1?@*1--/+0'+7',*.641.*>'1;1=10,C-@16/7/6'#$%&'()*+1234)5&&36D'5.-/6'7864-/0'-,./0'/02/6.,1-';/<0/7/6.,/+0'-/=/;.*',+'()E'6FDOHEDUV ȝP****  57 miRNA CCR1 prevents the tracheary elements from contributing to their own lignification. This suggests that there are other cells capable of contributing to protoxylem tracheary element lignification in the root. Parenchyma cells adjacent to the tracheary elements would be strong candidates for these monolignol-providing cells. One concern with the miRNA was that it has been reported that miRNAs can move from one cell to another within a tissue type through plasmodesmata (Kidner and Martienssen, 2004; de Felippes et al., 2011; Vaten et al., 2011). Cells within the stele of the root arise from the same meristem and are symplastically connected (Lucas et al., 2009), thus providing a potential avenue for miRNA movement from the tracheary elements to neighbouring cells. To test whether CCR1 silencing was tracheary element-specific, proCESA7::miRNA CCR1 plants were crossed with plants constitutively expressing a yellow fluorescent protein-tagged CCR1 (proUBQ10::YFP:CCR1), which showed strong YFP:CCR1 expression in the cytoplasm of all cell types. We postulated that the CCR1 miRNA would silence both the endogenous CCR1 and the YFP:CCR1 translational fusion construct.  The ability of the CCR1 miRNA to silence CCR1 was first tested by constitutively expressing the pro35S::miRNA CCR1 together with YFP:CCR1, and monitoring the effect of this co-expression on the YFP:CCR1 signal using confocal laser scanning microscopy.  In the presence of the pro35S::miRNA CCR1, YFP fluorescence was barely detectable anywhere in the root (Figure 3.2).  However, when the proCESA7::miRNA CCR1 construct was expressed in the YFP:CCR1 lines, strong YFP:CCR1 signal was observed in cells neighbouring the tracheary element cells but not in living tracheary elements undergoing secondary cell wall deposition, such as the tracheary elements in primary roots (Figure 3.3).  To ensure that the loss of YFP:CCR1 signal reflected miRNA-targeted degradation of CCR1 transcripts, rather than programmed cell death in these cells, developmentally matched samples were compared, using the living, developing tracheary element most proximal to the root tip.   3.2.2 “Good neighbours” in stem xylem and extra - xylary fibre lignification  The miRNA-expressing roots, whether with constitutive or cell-specific expression, did not display any obvious developmental or morphological defects resulting from the down-regulation of lignin biosynthesis. The stems of the mature plants, however, displayed striking phenotypes. The pro35S::miRNA CCR1 stems were dwarf with thin, pendant stems, more    58                                         Figure 3.2 CCR1 miRNA effectively silences YFP:CCR1 expression.  A) Brightfield image of a root showing tracheary elements (arrows); B) Pro35S::miRNA CCR1 ! proUBQ10::YFP:CCR1 showing very little YFP fluorescence. Scale bar = 150 µm. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).  AB  59  Figure 3.3 Cell specific silencing of CCR1 in Arabidopsis roots. A) Fluorescence image of wild-type Arabidopsis roots expressing proUBQ10::YFP:CCR1. B) Fluorescence image overlayed with corresponding brightfield image. YFP signal is observed in the late elongation zone of roots in endodermis (e) and throughout the stele (st), including tracheary elements, which can be identified by their cell wall pattern (inset, arrow). C) Fluorescence image of YFP:CCR1 in the proCESA7::miRNA CCR1  plant lines. D) Fluorescence image overlayed with corresponding brightfield image. CCR1 is still expressed in endodermis and stele, but not in the tracheary elements (arrow, inset). Scale bars A-D = 150 µm and insets = 75 µm. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).  ! !"#! "# !$ %& '!"#$%%&'!"#$%%&'()*+,$  60 susceptible to fungal attack and had smaller siliques with few viable seeds (Figure 3.4). These observed phenotypes are consistent with those seen in previously characterized genetic ccr knockout mutants (Jones et al., 2001; Thevenin et al., 2011). Quantitative RT-PCR analysis confirmed that CCR1 gene expression was severely reduced or absent in these plants (Figure 3.5), and phloroglucinol-HCl staining of lignin in cross-sections of inflorescence stems revealed only low levels of lignin in tracheary elements, xylary fibers and interfascicular fibers (Figure 3.6). Stem cross sections also revealed a collapsed tracheary element (irregular xylem) phenotype (Figure 3.6), characteristic of mutants with severe defects in secondary cell wall components or lignin deposition (Turner and Somerville, 1997; Jones et al., 2001).  In the stem, the effect of silencing monolignol production using the proCESA7::miRNA CCR1 construct was particularly striking because secondary cell walls of cells from different tissues were affected differently. As in the root, proCESA7 expression was specific to cells producing secondary cell walls, such as tracheary elements (Figure 3.7A, B) and fibres (Figure 3.7C, D). Phloroglucinol-HCl staining of inflorescence stem sections from proCESA7::miRNA CCR1 plants revealed that lignified cells within the xylem, such as tracheary elements and xylary fibers, appeared normal and displayed wild-type levels of lignification, whereas the interfascicular fibers (IFF) between the vascular bundles displayed drastic reductions in lignification (Figure 3.6). Quantification of Klason acid-insoluble lignin confirmed this pattern (Table 3.1). Stems of pro35S::miRNA CCR1 plants had the strongest reduction in lignin, with 37-49% of wild-type lignin levels. This drastic decrease in lignin was also observed in 2D-NMR spectra of stems, which showed relative decreases in S- and G-lignin and an increase in H-lignin (Figure 3.8).  Stems of proCESA7::miRNA CCR1 plants had intermediate lignin levels, with 57-73% of wild-type lignin (Table 3.1). Down-regulating lignin biosynthesis also had an effect on polysaccharides in the cell wall; pro35S::miRNA CCR1 lines had decreased cellulose and an increase in hemicellulose, while proCESA7::miRNA CCR1 lines displayed increases in both cellulose and hemicellulose content (Table 3.2). The monolignol composition of the proCESA7::miRNA CCR1 lignin exhibited minor changes, as thioacidolysis revealed slightly higher levels of S-units and reduced levels of G-units compared to wild-type stems (Table 3.3). A similar shift in lignin composition has been observed in other lignin-deficient plants (Berthet et al., 2011; Thevenin et al., 2011). The slight increase in S-units and decrease in G-units was    61                                       Figure 3.4  Stem phenotypes of 2 -month-old miRNA -expressing plants.  Growth habit of wild-type and miRNA plants. A) WT; B) pro35S::miRNA CCR1 , C) proCESA7::miRNA CCR1 , inset is a close-up of a WT inflorescence stem with normal siliques (i) and a pro35S::miRNA CCR1  inflorescence stem showing fertility defects (ii). Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).    A B Ci ii  62   Figure 3 .5  qRT - PCR results for CCR1 expression level in miRNA lines stems .  Error bars indicate standard deviation. Sample size for each line was 5-7 stems. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).      WT pro35S::CCR miRNAproCESA7::CCR miRNARelative Gene Expression1.21.00.80.60.40.20  63                                 Figure 3.6 Arabidopsis stem tissues show diverse degrees of cell-autonomous lignification.  A) Arabidopsis stem cross section (WT) stained with phloroglucinol-HCl; B) WT interfascicular fibres with secondary cell wall thickenings (white arrow); C) Loss of stem lignification in all cell types demonstrated with low phloroglucinol-HCl staining when monolignol biosynthesis is knocked down with pro35S::miRNA CCR1, arrows indicate collapsed tracheary elements; D) Inner secondary cell wall layers (white arrow) of interfascicular fibers detach in these miRNA lines; E) When monolignol biosynthesis is downregulated in cells with thickened secondary cell walls using proCESA7::miRNA CCR1 Line A, the vascular bundles of the stem still stained with phloroglucinol-HCl while interfascicular fibers had reduced staining; F) Secondary cell wall thickenings remain intact (white arrow). IFFs = interfascicular fibers, stars = tracheary element, triangles= xylary fibers, Scale bars A,C,E = 15 µm and B,D,F = 5 µm. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).  AIFFsCIFFsEIFFsBDF  64  Figure 3 .7  Localization of proCESA7::GFP in inflorescence stems of Arabidopsis.  proCESA7::GFP expression in xylem tracheary elements (A and B) and fibers (C and D), as identified by the tapered end walls (arrow). Images on the left are GFP fluorescence and images on the right are the corresponding brightfield. Scale bars = 100 µm. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).   ACBDFigure 3.7. Localization of proCESA7:GFP in inflorescence stems of Arabidop-sis. proCESA7:GFP expression in xylem tracheary elements (A and B) and fibers (C and D). Images on the left are GFP fluorescence and images on the ULJKWDUHWKHFRUUHVSRQGLQJEULJKWILHOG6FDOHEDU ѥP  65      Plant Klason (acid insoluble) lignin (mg/100 mg dry cell walls ± SD) Acid soluble lignin (mg/100 mg dry cell walls ± SD) WT  19.58±0.19 (100) 2.09±0.13 pro35S::miRNA CCR1 Line A  7.30±0.17 (37) 3.57±0.08 pro35S::miRNA CCR1 Line B  9.66±0.03 (49) 3.38±0.04 proCESA7::miRNA CCR1 Line A 11.22±0.03 (57) 2.46±0.08 proCESA7::miRNA CCR1 Line B 14.35±0.24 (73) 2.02±0.04  Table 3.1 Chemical analysis of lignin in different miRNA lines.  SD indicates standard deviation for 3 technical replicates. Values in parentheses are the percent of WT lignin. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).     66    Figure 3 .8  Whole cell wall analysis revealed changes in lignin content and composition in the miRNA line stems.  A) 2D- NMR spectrum of WT stems showing the presence of S -  (S2/6), G -  (G2, G5+G6), and H -  (H2/6) lignin as well as cinnamyl alcohol (X1). The relative percent of each type of lignin was determined through peak integration. B) pro35S::miRNA CCR1 stems displayed a significant reduction in lignin, with decreased S-  and G - lignin relative to WT but increased H - lignin; C) proCESA7::miRNA CCR1 stems had a slight increase in S- lignin and decrease in G - lignin relative to WT.   !"#$" $#%#%"&#!'#!(#!)#!!#!##*#+,-./-01234#,5)6$7)7"-8-7$9!:;)6$<=>?@?AB<=>?@?AB9!CS 15.8%G 83.5%H 0.8%S/G 0.19$" $#%#%"&#1,-!"#$%&'())*+,-'.$$,/-5)6$7)7"-8-7$9!:;)6$<=>?@?AB<=>?@?AB9!CS 16.6%G 79.8%H 3.5%S/G 0.21!"#$" $#%#%"&#!'#!(#!)#!!#!##*#!"#!'#!(#!)#!!#!##*#D,-!"#01&))*+,-'.$$,/-5)6$7)7"-8-7$9!:;)6$<=>?@?AB<=>?@?AB9!CS 4.8%G 31.7%H 63.6%S/G 0.15  67 Plant Arabinose (ug/mg sample±SD) Rhamnose (ug/mg sample±SD) Galactose (ug/mg sample±SD) Glucose (ug/mg sample±SD) Xylose (ug/mg sample±SD) Mannose (ug/mg sample±SD) Cellulose  (% of total) Hemicellulose  (% of total) WT 9.6 ± 0.18 7.4 ± 0.05  13.7 ± 0.13  334.7 ± 1.65  124.7 ± 0.42  21.2 ± 0.18  33.5 ± 0.17  17.7 ± 0.08  pro35S::miRNA CCR1 Line A 16.3 ± 0.30 11.1 ± 0.29 19.5 ± 0.37 280.4 ± 6.49 143.8 ± 3.69 28.3 ± 1.39 28.0 ± 0.65 21.9 ± 0.58 pro35S::miRNA CCR1 Line B 15.3 ± 0.37 11.2 ± 0.34 19.7 ± 0.34 284.5 ± 3.83 137.1 ± 1.41 26.2 ± 0.66 28.5 ± 0.38 21.0 ± 0.12 proCESA7::miRNA CCR1 Line A 9.6 ± 0.40 11.9 ± 0.55 14.6 ± 0.75 363.3 ± 6.37 148.3 ± 2.48 29.5 ± 2.16 36.3 ± 0.64 21.4 ± 0.47 proCESA7::miRNA CCR1 Line B 9.6 ± 0.27 10.3 ± 0.41 14.2 ± 0.26 359.0 ± 1.78 139.4 ± 0.25 25.2 ± 0.15 35.9 ± 0.18 19.9 ± 0.08  Table 3 .2 Soluble sugar analysis from miRNA lines.   SD indicates standard deviation for 3 technical replicates.  68  Plant  %G Lignin  %S Lignin  WT 71 ± 1%  29 ± 2%  pro35S:CCR miRNA1 Line A 64.4 ± 7%  35.6 ± 8%  proCESA7:CCR miRNA1 Line A 62.8 ± 3%  37.2 ± 3%  proCESA7:CCR miRNA1 Line B 67.4 ± 2%  32.6 ± 2%   Table 3 .3  Lignin monomer content analysis of different miRNA lines.  Values are the percentage of each type of lignin ± the range of values from 3 biological replicates. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org) .    69 also confirmed by integrating peaks from a solution-state 2D-NMR spectrum of the stem tissue (Figure 3.8).  The specific decrease in IFF lignification in proCESA7::miRNA CCR1 plants was correlated with a macroscopic stem phenotype in which stems were more pendant and flexible than wild-type stems, although in contrast to the pro35S:miRNA CCR1 plants, the  transgenic proCESA7::miRNA CCR1 plants appeared normal in height, growth and fertility (Figure 3.4). Quantitative RT-PCR confirmed a decrease in CCR1 gene expression in proCESA7::miRNA CCR1 plants compared to wild-type, but CCR1 mRNA was not completely absent, consistent with cell type-specific silencing of CCR1 (Figure 3.5).   To ensure that the miRNA was not moving from the lignified cells to neighbouring cells, longitudinal stem sections were examined from progeny of proCESA7::miRNA CCR1 ! proUBQ10::YFP:CCR1 crosses. The sections were taken from the top centimeter of the stem to ensure that the tracheary elements would still be alive. Confocal microscopy revealed that YFP:CCR1 signal was absent from tracheary elements, while the signal was strong in neighbouring cells that lacked secondary cell walls (Figure 3.9). This indicates that, as in the roots, the miRNA expression was specific to the targeted cells, with no detectable movement to adjacent cells. To investigate which xylem cell types could potentially act as neighbors during lignification, inflorescence stem sections were also examined using light microscopy and high-resolution TEM.  Xylem parenchyma cells, which could be seen interspersed among the tracheary elements and fibers of the vascular bundle (Figure 3.10), displayed thin primary cell walls and the presence of persistent cytoplasmic contents, indicative of metabolically functional cells.  In contrast, no parenchyma cells are interspersed among the interfascicular fibers outside the xylem, which were unable to lignify when their endogenous monolignol production was suppressed through CCR1 silencing.    70   Figure 3 .9  CCR1 miRNA  silences YFP:CCR1  expression only in lignified cells in the stem.   A) ProUBQ10::YFP:CCR1 expression in proCESA7:miRNA CCR1  stem longitudinal sections showing expression throughout the vascular bundle (vb, area defined by arrows) but not the tracheary elements (TE) or fibres (IFF); B) Brightfield overlayed with YFP expression.  Scale bars = 30 µm. Reproduced with permission from Smith et al. (2013) C opyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).  ABTEIFFsvbTEIFFsvb  71  Figure 3 .10  Anatomy of a vascular bundle in Arabidopsis thaliana.  A) Toluidine blue-stained stem cross-section showing the xylem parenchyma, tracheary elements, and xylem fiber cells. Scale bar = 10 µm; B) TEM image of xylary parenchyma cell and neighbouring lignified cell (scale bar = 2 µm) with inset high magnification image showing the difference in primary and secondary cell wall thickness (scale bar = 500 nm). XP=xylary parenchyma cell; LC=lignified cell. Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).    A BLCLCLCXP1 ÝCW2 ÝCW  72  3.2.3  Silencing monolignol biosynthesis specifically in xylary parenchyma cells or fibres does not affect lignification of xylem tissue  Control of amiRNA CCR1  expression by the CESA7  promoter silenced lignin biosynthesis in all cells with secondary cell wall thickenings, but to further test the non-cell autonomous lignification hypothesis, I also wanted to direct the miRNA specifically to xylary  fibres and xylary parenchyma cells. Fibre-specific silencing of lignin biosynthesis would reveal if xylary fibres act as “good neighbours” to adjacent tracheary elements, and provide further evidence for whether interfascicular fibre lignification is cell autonomous. On the other hand, xylary parenchyma-specific lignin silencing was expected to reveal the degree to which lignification could proceed without contributions from parenchyma cell “good neighbours”. Promoters for two peroxidase genes were confirmed to be fibre-specific (proAtPRX64 ) and xylary parenchyma-specific (proAtPRX47 ), based on analysis of promoter::GFP  expression for each promoter in the stem (Figure 3.11).  When the CCR miRNA was expressed in fibres (proAtPRX64::miRNA CCR1 lines), lignin biosynthesis in both xylary and extra-xylary fibres was low compared to wild-type, based on phloroglucinol-HCl staining. In the stem developmental stage when xylary fibre lignified secondary cell walls first start to appear (about 15 -21 cm below the shoot apical meristem), the amount of lignin deposited in the cell walls appears diminished relative to wild-type stems (Figure 3.12). In addition, metaxylem tracheary elements neighbouring the developing fibres appeared to have slightly less lignin in their cell walls relative to wild-type stems (Figure 3.12). This suggests that xylary fibres normally play a role in the lignification of neighbouring tracheary elements during xylem development. In xylem tissue of the mature stem of proAtPRX64::miRNA CCR1 lines, however, lignification appeared wild type, suggesting that the tracheary elements and xylary parenchyma cells are eventually capable of compensating for the lack of monolignols coming from the xylary fibres. As the interfascicular fibres in the proAtPRX64::miRNA CCR1 lines began to develop lignified secondary cell walls, it was apparent that there was a significant reduction in lignification of these extra-xylary fibres, based on the phloroglucinol-HCl staining and 2 -photon intrinsic lignin fluorescence, similar to that observed in the proCESA7::miRNA CCR1  plants (Figures 3.6, 3.12, 3.13).       73  Figu re 3 .11 Expression pattern of cell - specific promoters in stem longitudinal sections.  A) proAtPRX47::GFP has GFP signal specifically in xylary parenchyma cells adjacent to tracheary elements (TEs); B) proAtPRX64::GFP has GFP signal in fibres, as identified based on the tapered end of the cell. vb = vascular bundle. Scale bars = 40µm.     A BproAtPRX47::GFP proAtPRX64::GFPTEsvb  74  Figure 3.12 Phloroglucinol-HCl staining of stem developmental stages reveals differences in lignification through xylem development in the miRNA lines. Cross-sections were taken at 6cm (A, E, I), 9cm (B, F, J), and 21 cm (C, G, K) from the shoot apical meristem and at the base of the mature (approx. 30cm) stem (D, H, L). WT stems (A-D) and proAtPRX47::miRNA CCR1  stems (E-H) appeared similar throughout development based on the Phloroglucinol-HCl staining. proAtPRX64::miRNA CCR1  stems (I-L) displayed reductions in xylary fibre lignification (K) and interfascicular fibre lignification (K, L). Scale bar s = 15µm. WT proAtPRX47::miRNA CCR1 proAtPRX64::miRNA CCR16cm 6cm6cm9cm 9cm9cm21cm 21cm21cmmature maturematureAACBDEGFHIKJL  75  Figure 3 .13  Analysis of intrinsic lignin fluorescence in mature stems confirms the miRNA stem histochemical stain ing results.  WT stems show bright cell wall fluorescence in the xylem and IFFs (A - C); proCESA7::miRNA CCR1 stems have WT xylem fluorescence (D, E) but decreased IFF fluorescence (F); Fluores cence in proAtPRX47::miRNA CCR1 appears WT (G - I); IFF fluorescence  is decreased in proAtPRX64::miRNA CCR1 stems (L) but the xylem fluorescence appears WT (J, K), even in the xylary fibres. Scale bars =  500µm.   WTproCESA7::miRNA CCR1proAtPRX47::miRNA CCR1proAtPRX64::miRNA CCR1WTproCESA7::miRNA CCR1proAtPRX47::miRNA CCR1proAtPRX64::miRNA CCR1WTproCESA7::miRNA CCR1proAtPRX47::miRNA CCR1proAtPRX64::miRNA CCR1Whole stem Xylem Interfascicular FibresADGJ K LH IE FB C  76  Given the ability of xylary fibres to act as “good neighbours” later in development, the role of xylary parenchyma cells in lignification in the early stages of stem development ( i.e. before xylary fibres start to develop and lignify) may be important. To test the role of xylary parenchyma cells during tracheary element and xylary fibre lignification, stem sections from proAtPRX47::miRNA CCR1 plants, in which  the miRNA is targeted to xylary parenchyma cells, were stained with phloroglucinol-HCl at different developmental stages. I hypothesized that early in xylem development, protoxylem and early metaxylem tracheary  elements may have drastic reductions in lignification or display a collapsed xylem phenotype if the xylary parenchyma cells are the main source of monolignols, as they appear to be in the proCESA7::miRNA CCR1 plants. However, the first -formed tracheary elements in the proAtPRX47::miRNA CCR1 plants appeared wild type, based on the intensity and distribution of phloroglucinol-HCl staining (Figure 3.12). Quantitative chemical analysis of lignin at this stage of development has yet to be performed. The qualita tively wild -type tracheary element lignification observed in these plants suggests that the tracheary elements can contribute to their own lignification, and that the contribution of monolignols from xylary parenchyma, while sufficient to support normal levels of xylem lignification in proCESA7::miRNA CCR1 plants, may not be necessary. Later in stem development, and in the mature stem, tracheary elements and fibres appeared to possess wild-type levels of lignification, based on phloroglucinol -HCl staining (Figure 3.12). Analysis of sections from the base of the stem using 2 -photon microscopy to examine the intrinsic lignin fluorescence confirmed that the lignification of the xylem and extra-xylary fibres appeared normal (Figure 3.13). This normal xylem ligni fication could be due to the ability of xylary fibres to also act as “good neighbours” and provide monolignols to neighbouring tracheary elements while also lignifying their own cell walls.  3.3 Discussion The objective of this study was to test the good neighbor hypothesis, i.e. to determine which subpopulations of xylary and extra-xylary cells are active in monolignol production. By reducing monolignol biosynthesis in specific cell types, with and without thick secondary cell walls, I was able to distinguish between the diverse mechanisms of monolignol contributions   77 operating within the complex xylem tissue and the simpler cell-autonomous mechanism that operates in fibres located outside of the vascular bundles. Cell-to-cell communication (signal transduction between cells) in plants facilitates a wide range of essential processes. In the shoot apical meristem, for example, the CLAVATA 3 protein is synthesized in the outer layer of the meristem and moves to an inner layer where it binds to the CLAVATA 1 receptor and promotes the differentiation of meristematic cells (Haecker and Laux, 2001). Similarly, the SHORTROOT protein transcribed in the stele of the root must move to the endodermis, promote the transcription of SCARECROW, which encodes another transcription factor that binds to SHORTROOT , and sequester it in the nucleus of endodermal cells, to promote endodermal cell fate (Cui et al., 2007). As illustrated by these examples, many instances of cell-cell co-operation in plants involve intercellular protein signalling.  There are examples of intra-organ and intra-tissue metabolite, rather than protein, movement, as well, as exemplified in alkaloid biosynthesis. Nicotine, an alkaloid produced by Nicotiana species, is synthesized in the roots and transported via the xylem to the vacuoles of leaf cells for storage (Hashimoto and Yamada, 1994; De Luca and St Pierre, 2000). Intra -tissue metabolite transport is important in two prominent, non-model plant systems, Papaver somniferum (opium poppy) and Catharanthus roseus (Madagascar periwinkle), for the synthesis and storage of morphine and vinblastine, respectively. The enzymes involved in the synthesis of these complex alkaloids are located in several different cell types. In the P. somniferum stem, intermediates in the morphine biosynthesis pathway move from parenchyma cells in the vascular bundle and companion cells in the phloem to laticifers, the location of morphinan alkaloid storage (Bird et al., 2003; Weid et al., 2004). Intermediates in vinblastine synthesis in C. roseus move from epidermal cells to mesophyll cells and finally to laticifers and idioblasts, where the alkaloid products are stored (Murata and De Luca 2005).  The “good neighbour” hypothesis, by con trast, presents a novel form of metabolic co-operation in which two cell types contribute precursors to the synthesis of a secondary cell wall, rather than transducing a signal or coordinating the synthesis and storage of metabolites.  Expression of proCESA7::miRNA CCR1 specifically in cells with thickened cell walls effectively silenced CCR1 gene expression in these cells, yet xylem cells were still able to lignify their secondary cell walls. This suggests that non-cell autonomous lignification is occurring in   78 the xylem tissue. In contrast, lignification of the interfascicular fibers was strongly reduced, as revealed spatially by the pattern of phloroglucinol-HCl staining and quantitatively by Klason analysis. These two contrasting cell populations, xylem and lignifying interfascicular fibers, are developmentally specified by different transcriptional networks controlled by VASCULAR NAC-DOMAIN (VND) and SECONDARY WALL-ASSOCIATED NAC DOMAIN (SND1/NST3) transcription factors, respectively (Kubo et al., 2005; Zhong et al., 2010). In poplar wood, the vessel and xylary fibre cell populations display different lignin chemistry, depending on their proximity to one another (Gorzsas et al., 2011). In addition, developmentally non-cell autonomous lignification in xylem tissues is consistent with tracheary elements undergoing programmed cell death, the final stage in their development, much earlier than fibers (Courtois-Moreau et al., 2009). Perhaps it is not surprising that these tissues have distinct lignification mechanisms, given their diverse transcriptional networks and the developmental differences between them.  The expression of proAtPRX64::miRNA CCR1 in fibres resulted in reduced interfascicular fibre lignification but normal xylem lignification, based on phloroglucinol-HCl staining. This data further supports the idea that non-cell autonomous lignification occurs in the xylem and cell autonomous lignification occurs in the interfascicular fibres, as seen in proCESA7::miRNA CCR1 plants. Phloroglucinol-HCl staining of immature proAtPRX64::miRNA CCR1 stems showed not only decreased lignification in lignifying fibres, but also in tracheary elements surrounded by xylary fibres, suggesting that xylary fibres can also act as good neighbours during xylem lignification. The ability of xylary fibre lignification to proceed non-cell autonomously, and the capacity of xylary fibres to act as neighbours, distinguishes them from interfascicular fibres. Therefore, the fibre cell population can be further divided into the xylary fibre and extra-xylary fibre subsets. While the transcriptional network specifying interfascicular fibres has been well studied (SND1/NST3), the transcriptional cues necessary for xylary fibre cell fate remain unknown. Little is understood about the development of xylary fibres, an oversight that should be amended given the important role of xylary fibres in non-cell autonomous xylem lignification. A recent study (Yang et al., 2012) exploited the different transcriptional networks in xylem versus extra-xylary fibers and used the promoter for a key VND transcription factor to   79  rescue lignification only in xylem tissues by expression of proVND6:C4H in the c4h ref3  background.  The resulting stem phenotype appeared identical to that observed in the mature proCESA7:miRNA CCR1 and proAtPRX64::miRNA CCR1  stems.  Although the stem phenotypes described by Yang et al. (2012) and observed in this CESA7/AtPRX64  study appear identical, the conclusions that can be drawn about lignification are different.  The VND6  promoter, unlike the CESA7 and AtPRX64  promoters, has a broad expression pattern in Arabidopsis stem xylem and is expressed in all xylem cell types, with or without secondary cell walls (Yamaguchi et al., 2010a). The use of the VND6 promoter to perform a targeted re scue of lignification succeeded in improving saccharification, which was the goal of the study, but the authors did not comment on the role of any specific cell type in the xylem during lignification.  The CESA7 and AtPRX64  promoters, by contrast, are expressed only in cells with secondary cell walls (tracheary elements and/or fibres), and their use here has allowed the assessment of the role of non -lignifying neighbours during xylem lignification.  In the proCESA7::miRNA CCR1 plants, the non-lignified xylem cells played an integral role in maintaining xylem lignification. It was therefore hypothesized that by specifically down-regulating monolignol biosynthesis in xylary parenchyma cells, the only non-lignified cell type in the xylem tissue, an effect would be observed on tracheary element and xylary fibre lignification. Unlike tracheary elements and fibres, there is no known unique transcriptional cue that specifies the cell fate of the xylary parenchyma cell population. However, these non -lignified cells do express a repressor, VND -INTERACTING 2 (VNI2), which inhibits the activity of VND transcription factors, necessary for tracheary element cell fate, in the parenchyma cells (Yamaguchi et al., 2010b). Rather than promoting differentiation, the repressor ensures that xylary parenchyma cells remain in a more undifferentiated state. The fact that such a system exists to safeguard the presence of parenchyma cells in xylem tissue speaks to their putative importance in xylem development. However, the xylem lignif ication in proAtPRX47::miRNA CCR1  plants, where the miRNA was targeted specifically to xylary parenchyma cells, appeared normal. This suggests that while monolignols from xylary parenchyma cells are sufficient for normal xylem lignification, they are not necessary for lignification when tracheary elements and fibres also are synthesizing monolignols. One potential problem with the use of the proAtPRX47::miRNA CCR1  construct in plants is that the   80 activity of the AtPRX47 promoter is relatively weak, and it is therefore possible that the promoter was not strong enough to express the miRNA at a high enough level to adequately suppress lignin biosynthesis. The full impact of silencing lignin biosynthesis in xylary parenchyma cells therefore may not have been revealed.   3.4  M ethods  3.4.1  Plant growth conditions and staining  Arabidopsis thaliana (Col-0) seeds were grown on ! MS media (Sigma) under continuous light until seedlings were 7-days-old and then seedling roots were used for histochemical staining.  Whole seedlings were stained in 0.0001% Basic Fuchsin in 95% ethanol for 5 minutes, followed by de-staining for 2 minutes in 70% ethanol.  Seedlings were mounted in 50% glycerol and the lignin fluorescence in roots was observed using a Leica DMR light microscope using a red filter set (excitation 560nm, emission 645 nm).  Images were captured using a Q-CAM digital camera (Q-Imaging).  Seven-day-old seedlings were transplanted to soil (Sunshine mix 4, Sungrow Horticulture) and grown in a growth chamber at 21°C under 16/8 hours light/dark cycles.  Plant inflorescence stems were destructively sampled after 2 months and thin hand sections were taken from the base of the stem for histochemical staining and 2-photon imaging. Sections were stained with phloroglucinol-HCl for 5 minutes, mounted in water and analyzed using brightfield microscopy (Leica DMR microscope). For TEM analysis of stem xylem cell walls, the region of the Arabidopsis stem 7-8 cm from the shoot apical meristem was high pressure frozen, freeze substituted, sectioned and prepared for TEM as previously described (Section 2.4.4).  3.4.2  Confocal imag ing Live cell imaging of proCESA7::GFP and proUBQ10::YFP:CCR1 seedling roots was performed on a Perkin-Elmer UltraView VoX spinning disk confocal mounted on a Leica DMI6000 inverted microscope using a Hamamatsu 9100-02 CCD camera with the following excitation/emission filters; GFP (488/525), and YFP (514/540).  For imaging YFP:CCR1, or   81 proCESA7::GFP, 7-day-old seedlings were mounted in ! MS solution on slides and imaged using Leica oil immersion 63X or 20X objectives.  All images were captured and processed using Volocity image analysis software (Improvision).  Imaging of proCESA7/proAtPRX64/proAtPRX47::GFP, proUBQ10::YFP:CCR1, and proCESA7/proAtPRX64/proAtPRX47::miRNA CCR1 stem longitudinal sections was performed on an Olympus FV1000 Multiphoton Laser Scanning Microscope using conventional (GFP or YFP) or 2-photon settings. For 2-photon, the tunable MaiTai BB DeepSee (710-990 nm) laser was adjusted to 740 nm. Simultaneous absorption of two 740 nm photons results in excitation of lignin at approximately 350-370 nm. During imaging, two emission channels were simultaneous collected using 420-460 nm (Channel 1/Blue) and 495-540 nm (Channel 2/ Red) filter sets. Stem longitudinal sections were mounted in water on a glass slide and imaged using an Olympus water dipping 25" objective. In conventional confocal mode, the following excitation/emission filters were used: eGFP (488/509) and eYFP (514/527). Longitudinal sections from the top cm (YFP plants) or 5 cm from the top (GFP plants) of the mature stems were mounted on slides in water and imaged using Olympus 10" or 20" air objectives. The images were captured using Fluoview FV1000 software (Olympus) and processed using Volocity image analysis software (Improvision).  3.4.3 Transgenic lines Gateway cloning technology (Invitrogen) was used to generate the following constructs related to CELLULOSE SYNTHASE 7  (CESA7/IRX3; AT5G17420), PEROXIDASE64 (AtPRX64; AT5G42180), PEROXIDASE47 (AtPRX47; AT4G33420) and CINNAMOYL CoA REDUCTASE 1 (CCR1; AT1G15950).  The proCESA7::GFP (pMDC107; Curtis and Grossniklaus, 2003) was generated using 1127 base pairs upstream of the transcriptional start site of CESA7. The proAtPRX64::GFP and proAtPRX47::GFP (pMDC107) were generated using 900 and 1593 base pairs, respectively, upstream of the translational start sites for the AtPRX64 and AtPRX47 genes. The other Gateway constructs generated were pro35S::miRNA CCR1 (pK2GW7; Karimi et al., 2002), proCESA7::miRNA CCR1, proAtPRX64::miRNA CCR1, proAtPRX47::miRNA CCR1 (pKGW; Karimi et al., 2002) and pUBQ10::YFP:CCR1 (pUBN:YFP;Grefen et al., 2010) expression vectors (primers in Table 3.4).  Artificial    82 proCESA7 F AAAAAGCAGGCTGGCTCCAACGTTTTCAGTTT proCESA7 R AGAAAGCTGGGTCGGTGATCAATGAGAGACGA proAtPrx47 F AAAAAGCAGGCTGAGAGATGATGTCTTATATCCCTCATTTT proAtPrx47 R AGAAAGCTGGGTCACTATATTTGCCCTAACCATCTTATTAT proAtPrx64 F AAAAAGCAGGCTATGTTCATTTAGATTAAATAGATTCAA proAtPrx64 R AGAAAGCTGGGTCATTTTAACAAACTTTTCGAAAT CCR miRNA-I GATATTTGTATGGCTTGGCCCTTTCTCTCTTTTGTATTCC CCR miRNA-II GAAAGGGCCAAGCCATACAAATATCAAAGAGAATCAATGA CCR miRNA-III GAAAAGGCCAAGCCAAACAAATTTCACAGGTCGTGATATG CCR miRNA-IV GAAATTTGTTTGGCTTGGCCTTTTCTACATATATATTCCT RT CCR1 F TGTGGATGTTCGCGATGTC RT CCR1 R GAGGAGCAAGATGGCCTTTC YFP CCR1 F CGGGAAATATGCCAGTCGA YFP CCR1 R AAGGGGGATACTTGAGGGAAT RT ACTIN F CCAGAAGGATGCATATGTTGGTGA RT ACTIN R GAGGAGCCTCGGTAAGAAGA  Table 3.4 Primer sequences.  Reproduced with permission from Smith et al. (2013) Copyright American Society of Plant Biologists © 2013 (http://www.plantcell.org).     83 microRNAs were designed using the Web microRNA Designer (http://wmd2.weigelworld.org) and amplified using primers listed in Table 3.4, and as described in Schwab et al. (2006).  Plant expression constructs were introduced into Agrobacterium tumefaciens strain GV3101 and transformed into Arabidopsis thaliana Col-0 using the floral dip method (Clough and Bent, 1998) to generate transgenic plants.  3.4.4  Q - R T - PCR  RNA extraction was performed using TRIzol (Invitrogen) after grinding stems from 6-week-old Arabidopsis plants in liquid nitrogen.  RNA was quantified using a Nanodrop spectrophotometer (Thermo Fisher Scientific Inc.) and cDNA was generated using SuperScript III Reverse Transcriptase (Invitrogen) and oligo dT primers as per manufacturer’s instructions. Q-RT-PCR was performed using IQ SYBR Green supermix (Biorad) and cycled using a CFX Connect Real-time System (Biorad) according to manufacturer’s specification. Relative gene expression was calculated using the delta delta CT method (Livak and Schmittgen, 2001).  3.4.5  C hemical a nalysis  Dried Arabidopsis stems (WT; pro35S::miRNA CCR1; proCESA7::miRNA CCR1) were ground to a fine powder using a mortar and pestle, or a Wiley mill coupled to a 40 mesh screen. Stem tissue was acetone extracted in a Soxhlet apparatus for 7-16 hours followed by drying at 50°C overnight.  Extractive-free tissue (10 mg) was used to determine relative S and G monolignol abundance using thioacidolysis (Robinson and Mansfield, 2009) with tetracosane (200 µL at a concentration of 5mg/mL methylene chloride) as the internal standard.  Gas chromatography was performed on a ThermoFinnigan Trace GC-PolarisQ instrument equipped with a DB-5 column, as previously described (Coleman et al., 2008).  Acid-insoluble lignin was determined using Klason lignin analysis of 200 mg extractive-free tissue, as previously described (Coleman et al., 2008).  Acid-soluble lignin was determined by measuring the absorbance of the filtrate after isolation of acid insoluble lignin at 205nm and calculated using an extinction coefficient of 110 L/g-cm.  Solid-state 2D-HSQC-NMR analysis of mature (about 2 months old) WT, pro35S::miRNA CCR1, and proCESA7::miRNA CCR1 stems was performed as described in   84  Mansfield et al. (2012). Briefly, stems were cut into small pieces and then milled (Retsch MM400 mill) to a coarse powder. The ground samples were then solvent extracted in 80% ethanol four times and freeze-dried for about 2 days. Approximately 500 mg of the freeze-dried tissue was ball milled at 600 rpm for 5 minutes ! 11 (Fritsch Pulverisette 7 premium line mill) and 50 mg of the milled tissue was placed along the side of a 5 mm NMR tube. DMSO -d6 pyridine was used as a solvent for the cell wall samples. The samples were loaded into a Bruker 700 MHz NMR and characterized by HSQC spectroscopy. Data processing and analysis was performed using TopSpin v.3.1. software (Bruker).        85  Chapter 4: Using “novel” monolignol biosynthetic enzymes to investigate good neighbours during lignification  4.1 Introduction One of the main goals of this thesis was to determine which cell types are involved in lignification, and specifically if xylary parenchyma cells can act as “good neighb ours” during the lignification of adjacent cells that possess thickened secondary cell walls. The CCR miRNA plants created provided an effective tool for determining which cell types are capable of contributing to lignification. However, because the miRNA -expressing plants are mutated with respect to lignification in specific cell types, the results from examining cell-specific silencing of lignification in this genetic background might not provide an accurate picture of which cell types are actually contributing to lignification in wild-type plants. The extent to which each cell type produces monolignols and deposits them in the cell wall for lignification therefore remained an outstanding question. Unfortunately, live cell imaging of the metabolic fate of fluorescently-tagged monolignols, a method that could potentially demonstrate which cells are involved in wild-type lignification, is not yet possible. As an alternative strategy, I chose to introduce the capacity for synthesis of monolignol-conjugates, no t normally found in Arabidopsis, into specific Arabidopsis cell populations (tracheary elements, fibres, and xylary parenchyma cells) by combining ectopic expression of the necessary enzyme with promoters specifically expressed in each cell population. The whole cell wall composition in these transgenic plants was then analyzed for the presence of the conjugate. The presence of the conjugate would represent the contribution of the different cell populations to wild-type lignification, as only the cell types targeted by the promoters should produce the novel monolignol conjugates.   4.1.1 FMT and PMT from Angelica sinensis and Oryza sativa  Many plants produce and incorporate alternative or unusual monolignols into their cell walls. In Arabidopsis, this phenomenon often occurs as a result of disrupting crucial enzymatic reactions in the lignin biosynthesis pathway (Vanholme et al., 2012). For  example, mutants in which the COMT gene has been knocked out accumulate 5-hydroxyconiferyl alcohol, which is typically not a product of the biosynthesis pathway, and incorporate it into the lignin polymer   86 (Goujon et al., 2003). Other plant species incorporate alternative phenolic monomers into their cell walls even without genetic manipulation of the lignin biosynthesis pathway. Grasses, for example, incorporate ferulic acid into their cell walls, sometimes as ferulate esters (Harris and Hartley, 1976; Carpita, 1996). The discovery of “novel” monolignols in non-model plants is ongoing, as is the discovery of the enzymes that synthesize these monolignols.  Chinese Angelica (Angelica sinensis) was recently discovered to produce and incorporate monolignol-conjugates into the lignin in its cell walls (Xie et al., 2009). These conjugates are produced as a result of an ester linkage formed between a monolignol (p-coumaryl alcohol, coniferyl alcohol, or sinapyl alcohol) and feruloyl-CoA (Xie et al., 2009; Wilkerson et al., 2013). The acyltransferase enzyme that mediates the conjugate formation is called FERULOYL-CoA-MONOLIGNOL TRANSFERASE (FMT) (Wilkerson et al., 2013). Similarly, in Oryza sativa, a p-COUMAROYL-CoA-MONOLIGNOL TRANSFERASE (PMT) acyltransferase enzyme was found (Withers et al., 2012). The PMT enzyme preferentially couples p-coumaroyl-CoA with a monolignol through an ester linkage and the resulting conjugate is incorporated into the lignified cell walls of O. sativa (Withers et al., 2012). These “novel” monolignol conjugates are particularly interesting from a plant metabolic engineering perspective. Unlike many of the linkages formed among monolignols (e.g. ether bonds, C-C bonds, etc.), mild alkaline pre-treatments can readily cleave the ester linkages that couple the monolignol to the p-coumarate/ferulate (Grabber et al., 2008; Ralph, 2010). Integration of ester linkages throughout the lignin polymer would therefore ease the degradation of lignin and improve the efficiency of saccharification of the cell wall for industrial application, such as the emerging biofuel industry.  These “novel” enzymes and their resulting conjugates also have the potential to answer questions regarding which cell types are contributing to lignification in wild-type plants. Arabidopsis does not normally produce these monolignol conjugates, nor does it have the FMT or PMT enzymes. It does, however, produce the substrates necessary for the enzymes (monolignols, p-coumaroyl-CoA, feruloyl-CoA). I therefore introduced FMT and PMT into Arabidopsis, but restricted expression of the encoded enzyme to specific cell types by using cell-specific promoters: proCESA7 for tracheary elements (TEs) and fibres, proAtPRX64 for fibres, and proAtPRX47 for xylary parenchyma cells (Table 4.1).  Detection of monolignol      87    Promoter Target cell population Figure(s) showing promoter expression pattern Reference proCESA7 (CELLULOSE SYNTHASE 7 ) Tracheary elements Xylary fibres Interfascicular fibres Figure 3.1 (root)  Figure 3.7 (stem) Wightman and Turner, 2010 Mitsuda et al., 2007 proAtPRX64  ( PEROXIDASE 64 ) Xylary fibres Interfascicular fibres Figure 3.11 (stem) Tokunaga et al., 2009 proAtPRX47 ( PEROXIDASE 47 ) Xylary parenchyma cells Figure 3.11 (stem) Tokunaga et al., 2009  Table 4.1 The target cell populations of the promoters used in this study.           88 conjugates during whole stem cell wall analysis of these transgenic plants would provide direct evidence that a specific cell type(s) has the ability to contribute to lignification.  4.1.2  Detecting the novel monolignols  An effective way to detect the accumulation of monolignol conjugates in the cell wall is through cell wall chemical analysis. Although the currently available technologies report the chemical profile of the whole stem and do not differentiate between different cell types, the presence of any monolignol conjugates in the whole stem analyses would imply that they could only originate from a specific cell type, based on the specificity of the promoter driving the expression of FMT or PMT .  One of the methods for detecting monolignol conjugates is Derivatization Followed by Reductive Cleavage (DFRC) (Lu and Ralph, 1997a, b). This method targets the !- and "-ether linkages in the lignin polymer and releases monomers where two ether linkages are found side-by-side or where an ether bond links a monolignol to the end of the polymer. In DFRC, the !-hydroxyl group on the propane chain is initially brominated. This allows a Zn2+-facilitated reductive cleavage of the ether bond on the neighbouring "-carbon. The zinc donates two electrons to the bromine, which is thereby released from the polymer, and the subsequent rearrangement leads to the cleavage of the ether bond. The products obtained from the DFRC-treated cell wall sample are analyzed by gas chromatography-mass spectroscopy (GC-MS) to quantify the amount of monolignols released. It should be noted that this method does not allow direct quantification of the total amount of each monolignol in the original cell wall, as the DFRC-released monomers represent only about 12-15% of the total lignin (Mansfield et al., 2012). The presence of monolignol conjugates can be detected by this method in the form of dihydro-coumarate derivatives, however, because DFRC specifically targets ether linkages and does not disrupt the ester linkages that link the monolignols to the p -coumarate or ferulate moiety (Petrik et al., 2014; Wilkerson et al., 2013).  An alternative method for detecting the presence of coumarate- or ferulate-monolignol conjugates is the use of 2-dimensional nuclear magnetic resonance (2D-NMR). This method facilitates direct analysis of the chemical composition of the whole cell wall, both in terms of lignin and all other cell wall polysaccharides (Mansfield et al., 2012). In this ‘solution-state’   89 NMR, dried, solvent-extracted plant cell wall material is ground to a fine powder and then suspended in d6 -pyridine in DMSO. The naturally occurring 13C and 1H nuclear magnetic resonance signals are detected and compiled into a 2-dimensional spectrum. Integration of the signal intensities associated with individual monolignol-derived signals provides an approximate measure of the relative amounts of H -, G- and S-lignin in the whole cell wall (Mansfield et al., 2012). Because the NMR spectral pr operties of coumarate- and ferulate-monolignol conjugates are already known, the spectra obtained from the cell walls of transgenic PMT/FMT Arabidopsis plants could be examined to determine if monolignol conjugates are present (Withers et al., 2012; Petrik et al., 2013).  4.1.3  Ferulate - monolignol conjugates can be localized by fluorescence microscopy  As previously mentioned, neither of the cell wall chemical analysis techniques described above can localize the conjugates to the cell walls of specific cell types. CINNAMOYL - CoA REDUCTASE (CCR) mutants accumulate ferulic acid and some of it is incorporated into the cell wall as ferulate esters, much like the ferulate-monolignol conjugates synthesized by FMT (Leplé  et al., 2007; Vanholme et al., 2012). Lepl é  et al. (2007) reported that this incorporation of ferulate into the cell wall altered the intrinsic fluorescence of the cell wall such that greater fluorescence intensity was observed when the cell walls of ccr  mutants were exposed to blue light, relative to the fluorescence observed in wild-type plants. I reasoned that plants over-expressing FMT  might therefore display increased green fluorescence in their lignified cell walls. As such, we should be able to determine, by use of fluorescence microscopy, which cell walls have the highest fluorescence. A spectral analysis could permit the quantification of the amount of fluorescence in tracheary element and fibre lignified cell walls.   The objective of these experiments was therefore to direct different cell populations in Arabidopsis stems, using specific promoters as a proxy for the cell types (Table 4.1), to make novel monolignol conjugates, whose pattern of cell wall deposition within the tissue could then act as markers for the contribution of each cell population to lignification. The incorporation of the monolignol conjugates into the cell walls was documented by chemical analysis (DFRC and 2D-NMR) and fluorescence spectral analysis.   90  4.2 Results   4.2.1  Cells with seco ndary cell walls (tracheary elements and fibres) are contributing to lignification The monolignol conjugates produced by specific cell populations, as defined by the activities of different promoters, could act as signatures to demonstrate which cell types are contributing to lignification in wild-type plants. The promoter for CESA7  was used to drive the expression of FMT  and PMT  in cells with thickened secondary cell walls (tracheary elements and fibres), and because it was a strong promoter expressed in a large number of cells (Figure 3.7), this promoter was also used to determine if the FMT and PMT enzymes were functional in Arabidopsis. Mature (2-month-old) stems were subjected to DFRC and 2D-NMR cell wall analysis to assess the presence of ferulate- or p-coumarate-monolignol conjugates in plants expressing FMT or PMT , respectively. No ferulate-monolignol conjugates could be detected by either DFRC or 2D-NMR analysis in these or any other plants expressing the FMT gene, and therefore the focus of the chemical analysis was the plants expressing PMT . p-Coumarate-monolignol conjugates could be detected by DFRC and represented 1.4% of the total monolignols released from the lignin polymer (Figure 4.1). This indicates that tracheary elements and fibres were expressing the PMT  gene, the enzyme encoded by the gene was functional in Arabidopsis, and the resulting monolignol conjugates were incorporated into the cell wall. In terms of the “good neighbour” hypothesis, these results support the idea that cells with secondary cell walls are contributing to their own lignification.  Non-destructive whole plant cell wall solution-state 2D-NMR was also used to analyze the cell wall for p-coumarate-monolignol conjugates. 2D-NMR spectra from proCESA7::PMT plants had peaks corresponding to p-coumarate monolignol conjugates (Figure 4.2). Integration of the lignin peaks revealed that the p-coumarate conjugates represented approximately 2.3% of the total cell wall lignin (Figure 4.2). The higher percent of monolignol conjugates detected by NMR, compared to DFRC and GC-MS, may be the result of greater sensitivity, and the analysis of the whole cell wall rather than just the degraded portions of the cell wall. The presence of      91 Figure 4.1  The amount of monolignols and monolignol- conjugates released through DFRC analysis. The amount of p-coumaryl alcohol (HA), coniferyl alcohol (CA), sinapyl alcohol (SA) and p-coumarate-monolignol conjugate (pCA) released from the lignin polymer through the DFRC reactions are expressed as percents of the sum total of H-, G- and S-lignin (HGS) and as µg of monomer or conjugate per mg of cell wall material.  WT proCESA7::PMT proAtPrx64::PMT proAtPrx47::PMT% CA:HGS% SA:HGS% pCA:HGS% HA:HGSPercent of total lignin  92   Figure 4.2 2D -NMR spec tra of promoter-PMT lines. A)WT NMR spectrum showing S - , G - ,  H - lignin and cinnamyl alcohol (X1) ; B) proCESA7::PMT  spectrum showing the presence of p- coumarate monolignol conjugates (arrows); C) proAtPRX64::PMT  spectrum with monolignol conjugates not visibl e; D) proAtPRX47::PMT  spectrum without detectable monolignol conjugates.  15065 6070758014013012011010090A) WT (Col-0)S2/6G2G5 + G6X1aH2/6PyridinePyridineX1bS 5.04%G 65.8%H 29.2%S/G 0.0815065 6070758014013012011010090S2/6G2G5 + G6X1aH2/6PyridinePyridineC) proAtPRX64::PMTS 11.2%G 82.6%H 6.2%S/G 0.1415065 6070758014013012011010090D) proAtPRX47::PMTS2/6G2G5 + G6X1aH2/6PyridinePyridineX1bS 13.1%G 83.3%H 3.6%S/G 0.1615065 6070758014013012011010090B) proCESA7::PMTS2/6G2G5 + G6X1aH2/6PyridinePyridineX1bS 11.0%G 70.0%H 19.0%pCA 2.3%S/G 0.16pCA2/6  93 conjugates in the spectrum provides another line of evidence to support the model that tracheary elements and fibres are contributing to their own lignification.  The most abundant form of the conjugates appeared as coniferyl-dihydro-coumarate conjugates rather than sinapyl-dihydro-coumarate conjugates. This result is in direct contrast to reports of naturally occurring coumarate-monolignol conjugates produced by the PMT enzyme in rice (Oryza sativa ), which preferentially use sinapyl alcohol instead of coniferyl alcohol as a substrate (Withers et al., 2012).   To further separate the contribution of fibres to lignification from that of tracheary elements, we used the fibre-specific promoter for the PEROXIDASE 64  gene ( proAtPRX64 ; Figure 3.11). The PMT  gene, and therefore enzyme, was expressed in xylary fibres and interfascicular fibres in these stems. p -Coumarate-monolignol conjugates were present with the coniferyl-dihydro-coumarate conjugate as the most abundant form (Figure 4.1). The detection of conjugates in proAtPRX64::PMT  lines (about 0.6% of the total monolignols released by DFRC) indicates that fibres contribute to lignification (Figure 4.1).  Although the presence of monolignol conjugates in the cell walls of proAtPRX64::PMT  plants could be detected through DFRC analysis, they were not  observed in the 2D-NMR spectrum (Figure 4.2). The S-, G-, and H-lignin peaks were present, as in the wild-type and proCESA7::PMT  plants, but the p -coumarate monolignol conjugate peaks near the H-lignin peaks and at the base of a G5 + G6 peak were not visible (Figure 4.2). The conjugates only represented 0.6% of the lignin in the DFRC analysis, but with the greater sensitivity of the NMR, it is unclear why these conjugates could not be detected.   4.2.2 Xylary parenchyma cells contribute to the lignification of tracheary elements and xylary fibres In genetically modified miRNA-expressing plants, I obtained evidence that xylary parenchyma cells could contribute substantially to xylem lignification. The contribution of xylary parenchyma cells to lignification in wild-type plants, however, has not been assessed. To examine this question, the promoter for PEROXIDASE 47 ( proAt PRX47 ), which drives expression specifically in xylary parenchyma cells (Figure 3.11), was used to express the PMT gene in these cells. The expression of PMT  in xylary parenchyma cells did not yield monolignol-  94  conjugate accumulation detectable with DFRC or  2D-NMR analysis (Figures 4.1, 4.2). There are a number of possible reasons for the lack of cell wall-bound monolignol conjugates in these plants, the most likely being that the level of conjugates produced and incorporated into the lignin polymer is below the detection limit of the GC-MS and NMR. Because neither ferulate nor coumarate-monolignol conjugates were detectable in these plants with chemical analysis, an alternative technique was needed. Ferulate esters in planta have been reported to fluoresce green when exposed to blue wavelength excitation (450 -490 nm) (Lepl é  et al., 2007). The lignin polymer does not fluoresce strongly under blue wavelength, having optimal fluorescence under UV light. The difference in excitation and emission spectra between ferulate esters and the lignin polymer therefore provides a way to localize ferulate esters in the secondary cell wall. As the ferulate-monolignol conjugates are linked by an ester bond, I hypothesized that irradiating the cell walls with blue light would allow me to determine not only which cell populations are contributing to lignification, but the localization of the conjugates in specific secondary cell walls.   As a positive control, pro35S::miRNA CCR1 mature (2-month old) plant stems were imaged using a 473 nm laser. Plants defective in the CCR gene are known to accumulate ferulic acid and incorporate some of the pool into the cell wall as ferulate esters (Leplé  et al., 2007; Vanholme et al., 2012). Spectral analysis was performed to measure fluorescence intensity of tracheary element, xylary fibre and interfascicular fibre cell walls in pro35S::miRNA CCR1 tissues over a range of emission wavelengths. I chose the regions of interest for the spectral scan by outlining the cell wall of the specified cell type. The analysis revealed higher fluorescence in tracheary element and fibre cell walls in plants expressing CCR1 miRNA than in wild-type plants (Figure 4.3, 4.4).   Monolignol conjugates were detected in the cell walls of proCESA7::PMT and proAtPRX64::PMT plants by chemical analysis and therefore the corresponding lines expressing FMT under the control of these promoters were used as another control for the detection of ferulate esters by fluorescence microscopy. The green fluorescence emitted from the lignified cell walls in proCESA7::FMT stems appeared intermediate in intensity between wild-type plants and pro35S::miRNA CCR1 plants (Figure 4.3, 4.4). The spectral scan of tracheary element,    95                       Figure 4.3 Localization of ferulate-monolignol conjugates in secondary cell walls. A) Blue - excited fluorescence of a mature wild - type stem and (B) a spectral scan of the fluorescence intensity from a tracheary element (TE), xylary fibr e (XF) and interfascicular fibre (IFF) cell wall; C) Fluorescence of a pro35S::miRNA CCR1 stem and (D) the corresponding spectral scan of fluorescence intensity; Intermediate levels of fluorescence intensity were observed in proCESA7::FMT  stems (E) and in the spectral scan (F); Xylary fibre fluores cence (G) and intensity (H) was greater than that of TEs in proAtPRX64::FMT  lines; proAtPRX47::FMT  lines had stronger fluorescence in tracheary elements than fibres (I, J). Scale bars =  50 µm.  ACEGIWTpro35S::miRNA CCR1proCESA7::FMTproAtPRX64::FMTproAtPRX47::FMT050100150200250300490 500 510 520 530 540 550 560 570 580 590 600 610 620 630Average intensityWavelengths (nm)Interfascicular FibresT racheary elementsXylary FibresB1400Average intensity020040060080010001200490 500 510 520 530 540 550 560 570 580 590 600 610 620 630Wavelength (nm)IFFTEXFD01002003004005006007008009001000490 500 510 520 530 540 550 560 570 580 590 600 610 620 630Average IntensityWavelength (nm)IFFTEXFF050100150200250300350400450500490 500 510 520 530 540 550 560 570 580 590 600 610 620 630Average IntensityWavelength (nm)IFFTEXFJ0100200300400500600700800490 500 510 520 530 540 550 560 570 580 590 600 610 620 630Average IntensityWavelength (nm)IFFTEXFH  96  Figure 4.4 Average maximum fluorescence intensities at the wavelength of maximum emission for the different cell populations in WT (negative control), pro35S::miRNA CCR1 (positive control) and FMT-expressing plants. The averages were taken from 2-5 independent lines for each sample. Errors bars represent standard deviation.   WT pro35S::miRNA CCR1 proCESA7::FMT proAtPRX64::FMT proAtPRX47::FMTTracheary elements 291.69 1159.16 696.25 472.69 487.87;\ODU\ÀEUHV 331.56 1254.23 692.49 504.60 462.38,QWHUIDVFLFXODUÀEUHV 257.20 952.72 568.75 337.71 304.520.00200.00400.00600.00800.001000.001200.001400.001600.001800.00AYHUDJHÁXRUHVFHQHLQWHQVLW\DWPD[LPXPHPLVVLRQTracheary elements;\ODU\ÀEUHV,QWHUIDVFLFXODUÀEUHV  97  xylary fibre and interfascicular fibre cell walls revealed fluorescence intensities higher than wild -type, but lower than the positive control miRNA lines. This result makes sense, given that the  proCESA7::FMT  construct was placed in the wild- type backgro und, which may not accumulate large amounts of feruloyl- CoA. The intermediate fluorescence intensity levels may therefore reflect the incorporation of ferulate esters into the cell wall of plants that produce wild- type levels of the feruloyl- CoA substrate. Of the three cell types examined, tracheary elements fluoresced most strongly in these lines.  The proAtP RX 64::FMT  plants were designed to target ferulate- monolignol conjugate synthesis only to fibres (xylary fibres and interfascicular fibres) and not to tracheary elements. A consistent level of fluorescence was observed in all lignified cell walls of these plants, but some cell walls had brighter fluorescence. As predicted, the spectral scan revealed higher fluorescence intensity in xylary fibres and inte rfascicular fibres than wild- type stems (Figure 4.3. 4.4). Interestingly, tracheary element cell wall fluorescence was also above background wild - type levels (Figure 4.4), which may support the idea of xylary fibres acting as “good neighbours” to tracheary elements during lignification.  The  most interesting results from the fluorescence study came from the analysis of the proAtPRX47::FMT  lines, where the promoter is active only in the xylary parenchyma cell population. The cell population with the highest cell wall fluorescence was the tracheary elements, indicating the presence of ferulate esters in the wall (Figure 4.3; 4.4). This resul t supports the idea that xylary parenchyma cells are acting as “good neighbours” to tracheary elements during lignification. The fluorescence of xylary fibres was also above background wild -type levels while the interfascicular fibre cell wall fluorescence was not (Figure 4.4). This supports the role of xylary parenchyma cells as “good neighbours” to all cells with secondary cell walls within the xylem.   4.3 Discussion The objective of this study was to determine which cell types are involved in the lignification of tracheary elements and fibres in wild- type plants. FMT and PMT  genes were introduced into wild- type plants and expressed in specific cell types to assess the ability of that cell type to contribute to lignification. p- Coumarate- monolignol conjugates were detected in   98 plants where the PMT gene was directed to tracheary elements and fibres (proCESA7::PMT ) or just fibres (proAtPRX64::PMT ). This indicates that tracheary elements and fibres are contributing to lignification. Fluorescence analysis of FMT-expressing plants provided further support for the ability of tracheary elements and fibres to contribute to lignification, and also showed that xylary parenchyma cells are contributing to lignification. Spectral analysis of specific cell types revealed that xylary fibres are contributing monolignol conjugates to neighbouring tracheary elements, and that xylary parenchyma cells are “good neighbours” to tracheary elements and xylary fibres. This data also presents direct evidence that xylary parenchyma cells produce monolignol substrates.   The p-coumarate-monolignol conjugates produced by tracheary elements and fibres (proCESA7::PMT ) represented 1.4-2.3% of the total lignin. In contrast, p-coumarate conjugates from fibres only represented 0.6% of the total lignin (proAtPRX64::PMT ). The difference in the amount of conjugates could be attributed to the contribution of tracheary elements to lignification; however, the discrepancy could also be attributed to variations in promoter strength. The CESA7 promoter is expressed briefly, but very strongly, in tracheary elements and fibres. The AtPRX64  promoter is expressed very specifically in fibres, but the expression is not as strong as the CESA7 promoter. The strength of the promoters affects the expression of the PMT  gene, which may determine how much enzyme is present and functioning in the production of monolignol conjugates. A stronger promoter may therefore lead to the production of more monolignol conjugates and, as a result, while the presence of monolignol conjugates can demonstrate which cells are contributing to lignification, the difference in promoter strength makes it difficult to ascertain the relative contribution of each cell type to lignification. The contribution of xylary parenchyma cells to lignification could be determined through localization of ferulate conjugates, but not by cell wall chemical analysis. This is possibly because the contributions of xylary parenchyma cells to lignification may be below the detection limit of the techniques being used. The inability to detect xylary parenchyma-produced conjugates may not only be due to low contributions from these cells. The xylary parenchyma-specific promoter, proAtPRX47 , is a weak promoter and may not express the PMT  strongly enough to produce significant PMT enzyme activity in the xylary parenchyma cells. There are   99 also relatively few xylary parenchyma cells in the vascular tissue compared to tracheary elements and fibres, which may further hinder the detection of any contributions to lignification.  The function of monolignol conjugates in plants that naturally produce them is unknown, but it is hypothesized that association with a coumarate moiety may improve radical formation of sinapyl alcohol by peroxidases (Grabber et al., 2008; Hatfield et al., 2008). This process is called radical transfer/exchange and refers to the fact that p-coumarates are more easily converted to radicals than sinapyl alcohol, but when p-coumarate is coupled to sinapyl alcohol, the radical preferentially migrates to the sinapyl alcohol (Hatfield et al., 2008). This transfer reaction explains why p-coumarate-monolignol moieties are found pendent off the lignin polymer and not found interspersed within the polymer. In rice and Brachypodium distachyon, PMT favours the acyltransferase reaction between sinapyl alcohol and p-coumaroyl-CoA. However, the most prevalent p-coumarate conjugates formed in the proCESA7::PMT and proAtPRX64::PMT plants involved coniferyl alcohol. As coniferyl alcohol was more abundant than sinapyl alcohol in these plants (Figure 4.1), the preference for coniferyl-dihydro-coumarate conjugates is most likely a reflection of what monolignol is most readily available as a substrate for the PMT enzyme.  The amount of p-coumarate conjugates produced relative to the total lignin appears small, but is actually an overestimation of the conjugates in the lignin polymer. p-Coumarate-monolignol conjugates do not become incorporated into the lignin polymer, but rather, hang off the polymer as decorations. Conjugates on the end of the polymer are more readily detected by NMR and by techniques that degrade the cell wall, like DFRC, so the number of monolignol conjugates produced and incorporated into the cell wall is therefore actually smaller than these data might indicate. Monolignol conjugates incorporated throughout the lignin polymer, such as the ferulate-monolignol conjugates, may therefore be present at levels much lower than 0.6% of the total lignin and appear undetectable by cell wall chemical analysis. One of the substrates for the PMT or FMT enzymes is an intermediate in the monolignol biosynthesis pathway that likely does not accumulate to a significant degree in wild-type plants. This further limits the efficiency of the PMT or FMT enzyme in Arabidopsis and our ability to detect the enzyme product in the cell wall.      100 4.4  M ethods  4.4.1  Transgenic lines  pDONR::FMT and pDONR::PMT lines were kindly provided by Dr. Curtis Wilkerson (Michigan State University) and Dr. Shawn Mansfield (Department of Wood Science, University of British Columbia) (Withers et al., 2012; Wilkerson et al., 2013). Gateway cloning technology (Invitrogen) was used to generate the following constructs: proCESA7::FMT, proCESA7::PMT, proAtPRX47::FMT, proAtPRX47::PMT (pKGW; Karimi et al., 2002), proAtPRX64::F MT, and proAtPRX64::PMT (pMDC99 ; Curtis and Grossniklaus, 2003). The specific promoters are as described in Chapter 3 (section 3.4).  4.4.2  Plant growth conditions  Transgenic seeds were plated on !  MS media (Sigma) with appropriate antibiotics (25 µg/mL hygromycin for proAtPRX64::FMT/PMT lines and 50 µg/mL kanamycin for proCESA7/proAtPRX47::FMT/PMT lines) and grown under continuous light for two weeks. Antibiotic-resistant seedlings were transferred to soil (Sunshine mix 4, Sungrow Horticulture) and grown in a growth chamber at 21°C under continuous light. Mature (approximately 2-month old) plants were destructively sampled for chemical analysis and fluorescence microscopy.  4.4.3  Cell wall chemical analysis  The harvested stems were dried, cut into small pieces and subjected to pre-grinding (Retsch MM400 mill) until the stems were ground to a coarse powder. The ground tissue was solvent extracted in 80% ethanol four times and freeze-dried for two days. For DFRC analysis, approximately 20 mg of freeze-dried stem tissue from each line was used. DFRC was performed as described in Petrik et al. (2013). 2D-HSQC-NMR was performed as described previously (Chapter 3, section 4.4; Mansfield et al., 2012).    101 4.4.4 Fluorescence microscopy and spectral analysis Imaging of WT (two plants), pro35S::miRNA CCR1 (two independent lines), proCESA7::FMT (two independent lines), proAtPRX64::FMT (5 independent lines), and proAtPRX47::FMT (4 independent lines) plants was performed on the bas e of dried, mature stems. Fluorescence of the stems was first analyzed by epifluorescence using a Leica DMR light microscopy with a FITC filter set (excitation 450 -490 nm, emission 515 nm). Images were captured using a Q -CAM digital camera (Q -Imaging).  Spectral analysis of the same stems was performed using an Olympus FV1000 Multiphoton Laser Scanning Microscope with conventional confocal settings. A 473 nm laser was used to excite the stem cross -sections and sections were observed using an Olympus 20! air objective. The lambda scans of the stem sections were performed over the emission wavelengths 490 nm to 650 nm. Tracheary element, xylary fibre and interfascicular fibre cell walls were chosen as the regions of interest following the scan and the Fluoview  FV1000 software (Olympus) generated an emission profile for each chosen cell wall. The maximum fluorescence emission intensity for each cell wall type in each line was averaged over all independent lines.      102 Chapter 5: Conclusions and future directions  5.1 Main findings of this thesis The main objective of this thesis was to address two outstanding questions regarding plant cell wall lignification: when does lignification occur relative to the occurrence of programmed cell death, and which cells are involved in synthesi zing monolignols for lignification? The microautoradiographic investigation examining the timing of lignification revealed that lignification of protoxylem tracheary elements begins pre - mortem and continues post- mortem. Post- mortem lignification does not, however, continue indefinitely. Phenolic radiolabel did not accumulate within the cytosol or vacuole of tracheary elements or neighbouring cells, but strongly label led the secondary cell wall. This suggests that monolignols are rapidly exported from their site of synthesis into the cell wall without cytosolic or vacuolar accumulation.  Post- mortem lignification was hypothesized to involve xylary parenchyma cells neighbouring tracheary elements, as described in the “good neighbour” hypothesis. Plants express ing a miRNA targeting CCR1 specifically to cells developing thickened secondary cell walls (tracheary elements and fibres) revealed the contribution of non - lignified xylary parenchyma cells to the lignification of cells with secondary cell walls. Xylary parenchyma cells were capable of contributin g to xylem lignification and maintaining normal tracheary element and xylary fibre lignification in the miRNA line  (Figure 5.1) . Interfascicular fibre lignification, on the other hand, was a cell autonomous process. Tracheary elements and fibres were deter mined to also contribute to lignification, not only in the genetically modified miRNA -expressing plants, but also in wild - type plants (Figure 5.1) . Engineering the production of novel monolignol conjugates to tracheary elements and fibres resulted in detec table levels of conjugates in the cell walls, indicating that cells developing thickened secondary walls are themselves contributing to lignification. The contribution of xylary parenchyma cells, while sufficient for lignification in the miRNA - expressing p lants, was not necessary for lignification. However, monolignol conjugates were detected in secondary cell walls of tracheary elements and xylary fibres by fluorescence microscopy when monolignol conjugate production was      103    104  Figure 5.1 The potential good neighbours during xylem development throughout the Arabidopsis plant body. A) In early stem xylem development when protoxylem and metaxylem tracheary elements are forming, xylary parenchyma cells may be important for the lignification of neighbouring tracheary elements, while later in stem xylem development (B) the xylary fibres can also act as good neighbours to neighbouring metaxylem tracheary elements. C) In the root, the good neighbours are most likely pericycle parenchyma cells neighbouring the pr otoxylem tracheary elements. Scale bars in A) and B) = 15µm and C) = 10µm. TE, tracheary element; XP, xylary parenchyma; XF, xylary fibre; dTE, developing tracheary element; P, parenchyma cell; Pe, pericycle; Px, protoxylem; Mx, metaxylem.      105 targeted specifically to xylary parenchyma cells, indicating that xylary parenchyma cells are also involved in lignification in wild-type plants.  5.2  Background for the timing of lignification and “good neighbours”  The two main questions addressed in this thesis were identified as gaps in our understanding of the process of lignification. Early studies examined lignification in wheat and various tree species using autoradiography to answer questions about the timing of lignification and the deposition of H-, G- and S-lignin into the cell wall (Pickett-Heaps, 1968; Takabe et al., 1985; Terashima et al., 1986). In all these studies, radiolabel was detected in the cell walls of living tracheids, and as a result, the idea arose that lignification is a pre-mortem process. Kaneda et al. (2008) expanded on these earlier studies by coupling radiolabelling with the use of different inhibitors to independently block protein or phenylpropanoid biosynthesis. Although the main goal of the Kaneda study was to determine if monolignol export  was Golgi-mediated, the authors also presented lignification as a pre-mortem process. Significant reductions in Golgi label resulted from protein biosynthesis inhibition but not with phenylpropanoid biosynthesis inhibition, suggesting that monolignol export is not Golgi-mediated (Kaneda et al., 2008). The most likely monolignol export alternative hypothesis is some form of ATP-dependent plasma membrane transport (Miao and Liu , 2010; Alejandro et al., 2012).  Another interesting observation gleaned from this autoradiography study was that pine and poplar ray cells retained intracellular radiolabel even with protein synthesis inhibited, potentially implicating ray cells in lignin biosynthesis (Kaneda, unpublished data). This raised the possibility that ray parenchyma cells are contributing to the lignification of neighbouring xylem cells with thickened secondary cell walls (Ranocha et al., 2002).  Non-cell autonomous lignification, or the “good neighbour” hypothesis, was first proposed in the 1950’s by Freuden berg (1959) and support for this hypothesis has slowly been building since. A number of tracheary element-like cell culture studies have concluded that monolignols from sources external to the tracheary elements contribute to lignification of the tracheary elements even after they have died, and that undifferentiated cells in the culture may be important for lignification in this system (Hosokawa et al., 2001; McCann et al., 2001). In the cell culture system, the medium is viewed as a proxy for the extracellular matrix between cells in   106  the vascular system, while the differentiated and undifferentiated cells represent the various xylem tissue components, but the comparative complexity of the extracellular matrix and the xylem tissue in planta raises the quest ion of how accurately such a culture system can replicate the process of lignification. In planta, the activity of promoters for genes involved in lignin biosynthesis has been localized to xylary parenchyma cells and ray parenchyma cells adjacent to cells with secondary cell walls, indicating that these cells may be actively involved in monolignol biosynthesis (Chen et al., 2000; Baghdady et al., 2006). However, since these observations have provided only indirect evidence supporting the “good neighbour” hy pothesis, it is difficult to understand why the hypothesis has persisted for so long. It would not be the simplest explanation for the process of lignification, and if lignification proceeds before programmed cell death, the biological relevance of non- cell autonomous monolignol contributions is unclear.  The potential significance of “good neighbours” participating in lignification became of greater interest when it was reported that, in cell cultures, tracheary element lignification is apparently a post- mortem process (Hosokawa et al., 2001; Pesquet et al., 2010). Despite the fact that the Pesquet study dismissed a wealth of literature describing pre - mortem lignification in planta, it did present a potentially biologically relevant role for parenchyma cells in post- mortem lignification of tracheary elements. In light of these contrasting views in the current literature, I sought to resolve the pre- mortem versus post- mortem lignification debate, and to establish the timing of lignification relative to the timing of programmed cell death, and thus to directly test the “good neighbour” hypothesis.   5.3  Broad significan ce of this research to the process of lignification  One of the major contributions that this research makes to the general field of lignification is that it broadens the context in which we should consider the process of lignification. Cell culture studies, while useful as a simplified way to study tracheary element development, have perpetuated the idea that all lignified cells are equal. To some exte nt this is true, because the end result is a cell with a lignified secondary cell wall. However, when considering the broader context of all the lignified cell types present in plants such as Arabidopsis, and focusing on the actual process of lignification, rather than just the end result,   107  the picture is distinctly more complex. Extra -xylary fibres, such as interfascicular fibres, are unique among cells that develop a lignified cell wall because, as shown in this thesis, the process of lignification in these cells is cell autonomous. Xylary fibres are more similar to tracheary elements during lignification because both cell types can be lignified by “good neighbour” xylary parenchyma cells. Xylary fibres are further characterized by their ability to also act as “good neighbours” to tracheary elements. Treating all lignified cells as one biological end-point, or grouping the cells based on cell type (e.g. fibres versus tracheary elements) is therefore a gross oversimplification. The results from this thesis demonstrate that extra-xylary versus xylary lignification is a better approximation of the contrasts within the actual process. Xylary lignification can be further divided into those cells that act as “source” neighbours (xylary parenchyma cells and xylary fibres) and those that are “sink” neighbours (tracheary elements and xylary fibres). Therefore, one of the most significant contributions that my research makes to the study of lignification is that discussions of mechanisms of lignification can no longer be restricted to those cells that develop lignified cell walls, but must also include neighbouring xylary parenchyma cells. The empirical evidence presented in this thesis demonstrates that metabolic transfers from xylary parenchyma cells are sufficient for xylem lignification and contribute to tracheary element and xylary fibre lignification in wild-type plants. My findings also influence the current view of, and potential future research in, the regulation of lignification, the timing of lignification, the mechanisms of monolignol export, lignin polymerization and plant bioengineering.   5.3.1  What specifies xylary fibre and xylary parenchyma cell fates?  The results of my research also generate new questions regarding the specification and development of cells within the xylem. The transcriptional commitment of cells to tracheary element cell fate, mediated by the master transcription factors VND6 and VND7, and the role of the NST family transcription factors (SND1/NST1) in interfascicular fibre cell fate have bee n well studied (Kubo et al., 2005; Zhong et al., 2010). However, the transcriptional regulation of xylary fibre formation is most likely not identical to that of interfascicular fibres. This proposal is based not only on the differences in the mechanisms of lignification (non-cell autonomous versus cell autonomous), but also on their location in different tissues and the difference in lignin   108 composition between xylary and extra-xylary fibres reported in poplar (Gorzsas et al., 2011). The identity of the master switch for the transcriptional commitment to xylary fibre cell fate is therefore likely still unknown.  Also unknown are the cues regulating xylary parenchyma cell development. These were not previously considered important because xylary parenchyma cells were not recognized as a component of the lignification process. Xylary parenchyma cells express the VNI2 transcriptional repressor, which prevents them from committing to tracheary element cell fate (Yamaguchi et al., 2010b), and it is unlikely that t here is a master switch similar to those for tracheary elements or fibres because parenchyma cells appear to be the most undifferentiated cell type in the plant body. However, there must be some transcriptional circuitry or other set of cues inducing the transcription of monolignol biosynthesis genes within these cells in order for them to produce monolignols and contribute to the lignification of their neighbouring cells.  One mechanism that could trigger monolignol biosynthesis in parenchyma cells, independent of secondary cell wall thickening, is the expression of the key transcription factors MYB63 and MYB58, which up -regulate monolignol biosynthesis genes (Zhou et al., 2009).  Usually these transcription factors are turned on as part of the larger secon dary cell wall transcription factor cascade, but, if present in parenchyma cells, which most likely lack that secondary cell wall transcription factor network, how they might be activated requires further investigation.  Another point of interest arising from my studies is the question of the timing of monolignol biosynthesis gene transcription, and ultimately monolignol production, in tracheary elements relative to xylary parenchyma cells. The timing of xylary parenchyma contribution to tracheary elements (whether post-mortem only or pre-mortem and post-mortem) could be determined by using gene expression, or enzyme expression, as a proxy for monolignol biosynthesis. The radiolabel detected in cell walls during tracheary element development (Chapter 2) demonstrated that post-mortem lignification does not continue indefinitely. There must therefore be some cue or feedback mechanism in xylary parenchyma cells to stop monolignol production when the tracheary element and fibre cell walls are fully lignified, but, again, this cue or mechanism is unknown. Perhaps when the neighbouring cell wall is fully lignified, the export of monolignols into its cell wall becomes blocked and the resulting   109  accumulation of monolignols within the parenchyma cell could act as negative feedback on enzymes within the monolignol biosynthesis pathway.   5.3.2 Differences between the timing of lignification in vitro and in planta  The pre- mortem versus post- mortem debate was reconciled by the results in Chapter 2 of this thesis, which established that lignification begins pre- mortem and continues post- mortem, but not indefinitely. This conclusion can be viewed within the context of previous reports in this field. Earlier autoradiography studies reporting pre- mortem lignification did not examine t he lignifying cells for intact protoplasts, but instead the authors inferred from the continued deposition of radiolabel in the cell walls across the xylem developmental gradient that the tracheids must still be living  (Takabe et al., 1985; Terashima et a l., 1986). The cell culture studies, on the other hand, were correct in their report of post- mortem lignification, but the authors focused on lignification occurring around the time of programmed cell death and therefore may not have observed the significant pre- mortem lignification (Pesquet et al., 2010; Pesquet et al., 2013). Both the timing of lignification and reduced cell type complexity in the cell culture system may also not accurately represent the process in planta, which is why in vitro studies should be interpreted with care when extrapolating the data to a  living organism. In cell cultures, the study object is the single cell type, tracheary elements, while in a vascular bundle comprised of both tracheary elements and fibres, lignification of shared cell walls is certainly a more complex process.  Considering the timing of lignification within a broader developmental context encompassing a number of lignifying cell types provides a new viewpoint from which to re- assess many hypotheses made about the process of lignification. In addition, many plants, of which Arabidopsis is just one, have fibres that are still living at the base of the stem. If programmed cell death were an a priori requirement for lignification, these fibres would never develop lignified cell walls. There are other aspects to the problems associated with use of a simple model system to describe the entire process of lignification. Pesquet et al. (2013) argued that lignification is dependent upon programmed cell death, based on the results of treating cultured tracheary elements with programmed cell death inhibitors, which inhibited lignification. In my autoradiography work, I saw no evidence for monolignol s or monolignol glucosides   110 accumulating in the vacuole of tracheary elements, suggesting that, if Pesquet’s model is correct and polymerization does not occur until programmed cell death, the monolignols must be free (unpolymerized) in the cell wall until programmed cell death. However, methanol extraction of roots should capture both the intracellular monolignols and monolignols free in the cell wall, yet no monolignols were detected by HPLC analysis of root methanol extracts. These data suggest that monolignols do not accumulate, but are rapidly synthesized, exported and polymerized both pre- and post-mortem.  5.3.3  Mechanisms of monolignol export  Monolignols, or their glucosides, did not accumulate in the vacuole of living tracheary elements, which led to the conclusion that monolignols are exported rapidly from the cytosol to the cell wall for polymerization (Chapter 2). This data may appear to only provide another line of evidence for transporter-mediated monolignol export, but the results from the vanadate treatment would appear to contradict this interpretation. Vanadate is a phosphate analog that binds to and inhibits P-type ATPases, including the H+-ATPase at the plasma membrane and ABC transporters (Palmgren and Harper, 1999). Treatment with vanadate did not result in intracellular monolignol accumulation, nor did the treatment reduce the amount of radiolabel detected in the secondary cell walls. From this, I concluded that inhibition of ABC transporter activity, at least for short time periods, did not affect cell wall lignification. Similarly, vanadate did not reduce lignification of Arabidopsis root protoxylem tracheary elements (Kaneda, 2008). Vanadate is a crude tool, as it inhibits both ABC transporters and the H+-ATPase transporters responsible for maintaining the plasma membrane electrochemical gradient. However, taken at face value, these data indicate that an ABC transporter(s) may not be the monolignol exporter, or at least not the only type of transporter capable of exporting monolignols. Tsuyama et al. (2013) recently reported that coniferin transport was inhibited by bafilomycin A1, a more specific vacuolar ATPase inhibitor, but not by vanadate. Their study suggested that a H+/coniferin antiporter, dependent upon an ATPase, could be the mechanism for coniferin export (Tsuyama et al., 2013), thus providing an alternative to classical ABC-transporter mediated monolignol export. However, the authors did not report the effect of bafiolmycin A1 on monolignol aglycone transport and at this point there is no direct evidence for the involvement of coniferin in   111 lignification (Chapelle et al., 2012). Nevertheless, these inhibitor studies are not definitive. Higher concentrations of inhibitors, or treatment for longer times may yield different effects on tracheary element lignification. Such studies should be performed to provide more information about what controls monolignol export in planta , and open the door to further research on the transporters potentially involved in monolignol export. Future monolignol exporter studies should also be broadened to examine transporters that are expressed not only in cells with thickened secondary cell walls, but also transporters that are expressed in xylary parenchyma cells. Presumably a similar, or the same, transporter would be functioning in lignifying cells and in their “good neighbours”. Al l the cells present in the xylem tissue have the same developmental origin, the procambium. The developmental origin of the interfascicular fibres in Arabidopsis, however, is not clear, nor is the mechanism of monolignol export in these cells that appear to lack “good neighbours”. The interfascicular fibres are superficially similar to the sclerenchyma sheath that surrounds closed vascular bundles in monocots and some herbaceous eudicots that do not undergo secondary growth. These sclerenchyma cells originate, in part, from the procambium and may recruit parenchyma cells from ground tissue located between the vascular bundles (Esau, 1943). In some instances, the sclerenchyma sheath forms a continuous cylinder around the stem (Esau, 1943), much like the interfascicular fibres in Arabidopsis. Given that the interfascicular fibres may have heterogenous meristematic origins, the fibres might be split into developmentally distinct groups that would be defined by how they develop and how monolignols are exported. This possibility requires further exploration.  In this dissertation, monolignol conjugates were used to test aspects of the “good neighbour” hypothesis, but this research also produced an interesting observation regarding monolignol export. Earlier studies  examining monolignol export across the plasma or vacuolar membranes have focused on the export of the aglycone monolignols such as p -coumaryl alcohol, coniferyl alcohol and sinapyl alcohol, or their corresponding glucosides (coniferin and syringin) (Miao and Liu, 2010; Alejandro et al., 2012; Tsuyama et al., 2013). My detection of monolignol conjugates in the cell wall indicates that phenolic dimers can also be exported across the plasma membrane into the cell wall. How such export might be facilitated is unknown. It has been proposed that lignin precursors may diffuse across the plasma membrane, however, while   112 monolignol diffusion across membranes has been observed, the rates of diffusion were too low to account for the extent of lignification (Miao and Li u, 2010). The current hypothesis for monolignol export therefore involves ATP-dependent transport through plasma membrane-localized proteins, such as ABC transporters. Beyond the ability of one such transporter to translocate p-coumaryl alcohol in yeast (Alejandro et al., 2012), however, there is little known about the transport mechanism. The observation that dimers, as well as monomers, can be moved out of the cell and into the cell wall raises the question of how promiscuous the binding site in a plasma membrane-localized transporter would have to be to transport monomers and dimers. The wide variety of phenolic monomers that can be incorporated into the lignin polymer in wild type and mutant plants makes it unlikely that there is a unique transporter for each monomer. The PMT -expressing plants in my study produced only low levels of p-coumarate-monolignol conjugates and the transport of dimers is therefore presumably an anomaly rather than the norm. In grasses, such as rice, however, these conjugates are produced in larger quantities and may play an important role in the lignification process (Hatfield et al., 2008; Withers et al., 2012). The plasticity of phenolic export, seen here in the monolignol conjugates as well as in mutant phenotypes, provides important information on the flexible nature of the monolignol transport system.  5.3.4  “Good neighbours” in lignin polymerization  Another aspect of lignification with many unanswered questions is lignin polymerization. The research from this thesis adds important information to this field of study, because now it is known that the monolignols for oxidative polymerization are not just produced by the lignifying cells. The ability of xylary parenchyma cells and xylary fibres to act as “good neighbours” to tracheary elements during lignification indicates that monolignols will not be entering the cell wall from one direction, but from multiple sources. Rather than simplifying the process of lignin polymerization, this adds another level of complexity to the process and raises many more questions. Does lignification begin cell autonomously in tracheary elements, with xylary parenchyma and xylary fibre cells only contributing substrates once a polymer is established in the cell wall, or do monolignols from all cells participate in the establishment of the polymer from the outset? The answer to this question could provide useful information about the location   113 of the initiation site, because if all cells contribute to lignification early in the process, the initiation site would most likely be in the outermost part of the secondary cell wall, closest to the other cell types.  In addition, what prevents the lignification of the xylary parenchyma cell walls? It is possible that xylary parenchyma cells lack oxidizing enzymes suc h as laccases or peroxidases in their cell walls, which may ensure that monolignols move through the primary cell wall without binding or polymerizing. However, there are xylary parenchyma -specific peroxidases (Tokunaga et al., 2009), whose function, if no t in lignification, is unknown.  Do xylary fibres lignify their own cell walls and those of tracheary elements concomitantly or sequentially? Is the contribution from xylary fibres regulated or do monolignols diffuse from the fibres to neighbouring tracheary elements? If the contribution to tracheary element lignification is an important function of fibres, it is more logical that the tracheary elements would become lignified first because there would be less hindrance to the movement of the monolignols through the xylary fibre wall to the tracheary element wall. If the monolignols diffuse from the xylary fibres into neighbouring tracheary element cell walls, the contribution could occur at any time during xylary fibre lignification.  Finally, what signals t he cessation of lignification in the tracheary element and fibre cell wall? The answer to this question could be as simple as a lack of further space for lignin polymerization in the cell wall matrix. The point of real interest is how the completion of cel l wall lignification signals the cessation of intracellular monolignol production in the contributing cell(s).  5.3.5  I mplications for engineering plants with reduced or altered lignin  Although the focus of this study was to dissect the mechanisms of lignification by examining specifically which cells in the xylem are contributing to lignification, understanding that xylary and extra-xylary lignified cells employ different lignification mechanisms has implications for strategies to engineer plants for altered lignin content.  Morphological defects common to many genetically modified plants with decreased lignin content include dwarf plant stature, collapsed xylem, reduced fertility, and increased susceptibility to fungi and other pathogens (Jones et al., 2001; Sch och et al., 2001; Franke et al., 2002a,b; Schilmiller et al.,   114  2009).  Loss-of-function mutants with lesions in genes encoding NAC Secondary Wall Thickening Promoting Factor (NST) transcription factors have a pendant stem phenotype due to defective secondary cell wall formation in interfascicular and xylary fibres (Mitsuda et al., 2007; Zhong et al., 2007). Furthermore, snd1nst1 double mutants do not deposit cellulose, hemicellulose or lignin in fiber cells (Zhong et al., 2007). The proCESA7:miRNA CCR1 plants I produced in this study, by contrast, have reduced lignin levels and a less rigid stem than wild -type plants, yet these plants develop wild -type biomass, do not have collapsed xylem and are fertile.  The phenotypes of my proCESA7:miRNA CCR1, and proAtPRX64::miRNA CCR1, mature plants are interesting from a plant engineering perspective, as the lignification of their xylem is unaffected, but the lignin content of the whole plant is reduced, due to suppression of lignin deposition specifically in the extra-xylary (interfascicular) fibres. Saccharification assays of the stems of these plants would reveal whether this interfascicular fibre -specific lignin reduction translates into a more readily degradable cell wall biomass. Extra-xylary fibre cells make a substantial contribution to plant biomass, especially in potential biofuel feedstock taxa such as the grasses. There would therefore appear to be opportunities for independently manipulating two lignin pools, xylary and extra-xylary, to create plants with intact and still functional xylem but overall lower lignin levels. Esau (1943) reported that sclerenchyma sheaths surrounding vascular bundles in grasses and some eudicots are of heterogenous developmental origin, differentiating from both the procambium and ground meristem. This provides further possibilities for manipulating the lignification of a very specific subset of parenchyma -derived fibres in grasses without affecting the integrity of supporting fibres developing from the procambium. There are also further engineering possibilities with the FMT and PMT genes expressed in Arabidopsis. The low yield of monolignol conjugates from the FMT and PMT expressing lines in this thesis is potentially due to limitations in the intracellular supply of feruloyl-CoA or p-coumaroyl-CoA substrate, respectively. Increasing the amount of substrate available for the FMT/PMT enzyme to act upon may improve the chances of producing monolignol conjugates for polymerization in the cell wall. Therefore, Arabidopsis mutants that ac cumulate elevated levels of feruloyl -CoA or p-coumaroyl-CoA, and are also expressing the FMT or PMT gene in a cell-specific manner, may provide further insight into which cells are involved in the   115 lignification of tracheary elements and fibres, while also making it possible to enhance the incorporation of novel monolignol conjugates into the cell wall. Plants with mutations in the CCR gene, such as the irx4 , ccr1g or pro35S::miRNA CCR1  mutants, are known to accumulate feruloyl-CoA and ferulic acid, which can potentially be used with monolignols as the substrates for the FMT enzyme (Leplé et al., 2007; Vanholme et al., 2012). Increased substrate availability may increase monolignol conjugate production and make it easier to detect ferulate-monolignol conjugates in the cell wall, particularly those produced in small quantities, as is likely the case for xylary parenchyma cells.  To improve the production of coumarate-monolignol conjugates, the tt4- 2  mutant could be used. tt4- 2  is a null mutation in the CHALCONE  SYNTHASE  gene (Burbulis et al., 1996), a mutation that may cause the accumulation of p-coumaroyl CoA, the key intermediate at the branch point of the flavonoid and lignin biosynthesis pathways (Withers et al., 2012). With greater substrate availability and minimal impact to the lignin biosynthesis pathway (Li et al., 2010), the PMT enzymes may be capable of producing more p-coumaroyl-monolignol conjugates and incorporating them as decorations pendent off the lignin polymer.  Technology is developing at an incredible rate, and with its evolution, the opportunities for scientific exploration into questions once thought unanswerable become endless. This study, for example, was possible because recent developments in artificial microRNA technology could be used in Arabidopsis thaliana , a model system so well studied that cell population-specific promoters have been described. The advances in discovering “novel” monolignols and “novel” lignin enzymes for bioengineering also provided a previously unavailable strategy for answering a key basic biology question.  While my work provides answers to some central questions such as “when does lignification occur relative to programmed cell death?” and “are ‘good neighbours’ involved in the lignification of cells with secondary cell walls?”, the results have also highlighted the complexity of the process of lignification. Despite important advances in this field, we still do not understand how monolignols are exported, or how monolignols encounter oxidative enzymes and matrix polysaccharides during polymerization, both crucial pieces in the lignin puzzle. In order to most effectively manipulate lignification in plants from a bioengineering perspective, all aspects of the lignification process must be fully understood.    116  References  Alejandro S., Lee Y., Tohge T., Sudre D., Osorio S., Park J., Bovet L., Geldner N., Fernie A.R., and Martinoia E. (2012). AtABCG29 is a monolignol transporter involved in lignin biosynthesis. Curr Biol 22, 1207 - 12. Aloni R. (1987). Differentiation of Vascular Tissues. Annu. Rev. Plant Physiol. Plant Mol. Biol.  38, 179 - 204. Avci U., Petzold H.E., Ismail I.O., Beers E.P., and Haigler C.H. (2008). Cysteine proteases XCP1 and XCP2 aid micro - autolysis within  the intact central vacuole during xylogenesis in Arabidopsis roots. Plant J  56, 303 - 15.  Baghdady A., Blervacq A.S., Jouanin L., Grima-Pettenati J., Sivadon P., and Hawkins S. (2006). Eucalyptus gunnii  CCR and CAD2  promoters are active in lignifying cells during primary and secondary xylem formation in Arabidopsis thaliana . Plant Physiol Biochem 44, 674 - 83.  Bailey I.W. and Kerr T. (1935). The visible structure of the secondary wall and its significance in physical and chemical investigations of tracheary ce lls and fibers.  J. Arnold Arb. 16, 273 - 300.  Barnett J. and Bonham V. (2004). Cellulose microfibril angle in the cell wall of wood fibres. Biological Reviews 79, 461 - 472.  Bassard J., Richert L., Geerinck J., Renault H., Duval F., Ullmann P., Schmitt M., Meyer E., Mutterer J., Boerjan W., De Jaeger G., Mely Y., Goossens A., and Werck-Reichhart D. (2012). Protein- protein and protein - membrane associations in the lignin pathway. Plant Cell  24, 4465 - 4482.   117  Bednarek P., Schneider B., Svatos A., Oldham N., and Hahlbrock K. (2005). Structural complexity, differential response to infection, and tissue specificity of indolic and phenylpropanoid secondary metabolism in Arabidopsis roots. Plant Physiol. 138, 1058 -1070.  Berthet S., Demont-Caulet N., Pollet B., Bidzinski P., Cezard L., Le Bris P., Borrega N., Herve J., Blondet E., Balzergue S., Lapierre C., and Jouanin L. (2011). Disruption of LACCASE4 and 17 results in tissue-specific alterations to lignification of Arabidopsis thaliana stems. Plant Cell 23, 1124-37. Berthet S., Thevenin J., Baratiny D., Demont-Caulet N., Debeaujon I., Bidzinski P., Leplé J., Huis R., Hawkins S., Gomez L., Lapierre C., and Jouanin L. (2012). Role of plant laccases in lignin polymerization. Lignins: Biosynthesis, Biodegradation and Bioengi neering 61, 145 -172.  Bevan M., Shufflebottom D., Edwards K., Jefferson R., and Schuch W. (1989). Tissue- and cell-specific activity of a phenylalanine ammonia-lyase promoter in transgenic plants. Embo J  8, 1899-906.  Bird D., Franceschi V., and Facchini P. (2003). A tale of three cell types: Alkaloid biosynthesis is localized to sieve elements in opium poppy. Plant Cell 15, 2626 -2635.  Boerjan W., Ralph J., and Baucher M. (2003). Lignin biosynthesis. Annu Rev Plant Biol  54, 519 -46.  Boija E. and Johansson G. (2006). Interactions between model membranes and lignin -related compounds studied by immobilized liposome chromatography. Biochim Biophys Acta 1758, 620 -6.  Boija E., Lundquist A., Edwards K., and Johansson G. (2007). Evaluation of bilayer disks as plant cell membrane models in partition studies. Anal Biochem 364, 145 -52.  Bonawitz N.D. and Chapple C. (2010). The genetics of lignin biosynthesis: Connecting genotype to phenotype. Annual Review of Genetics 44, 337 -363.    118 Burbulis I.E., Iacobucci M., and Shirley B.W. (1996). A null mutation in the first enzyme of flavonoid biosynthesis does not affect male fertility in Arabidopsis. Plant Cell 8, 1013 -1025.  Carpita N. (1996). Structure and biogenesis of the cell walls of grasses. Annu. Rev. Plant Physiol. Plant Mol. Biol.  47, 445 -476.  Chapelle A., Morreel K., Vanholme R., Le-Bris P., Morin H., Lapierre C., Boerjan W., Jouanin L., and Demont-Caulet N. (2012). Impact of the absence of stem-specific beta -glucosidases on lignin and monolignols. Plant Physiol. 160, 1204-1217.  Chapple C.C., Vogt T., Ellis B.E., and Somerville C.R. (1992). An Arabidopsis mutant defective in the general phenylpropanoid pathway. Plant Cell 4, 1413 -24. Chen C., Meyermans H., Burggraeve B., De Rycke R.M., Inoue K., De Vleesschauwer V., Steenackers M., Van Montagu M.C., Engler G.J., and Boerjan W.A. (2000). Cell-specific and conditional expression of caffeoyl-coenzyme A-3 -O-methyltransferase in poplar. Plant Physiol 123, 853 -67.  Chen H., Li Q., Shuford C.M., Liu J., Muddiman D.C., Sederoff R.R., and Chiang V.L. (2011). Membrane protein complexes catalyze both 4-and 3 -hydroxylation of cinnamic acid derivatives in monolignol biosynthesis. Proc. Natl. Acad. Sci. U. S. A.  108, 21253 -21258.  Clough S. and Bent A. (1998). Floral dip: a simplified me thod for Agrobacterium -mediated transformation of Arabidopsis thaliana . Plant J 16, 735 -743.  Coleman H.D., Park J., Nair R., Chapple C., and Mansfield S.D. (2008). RNAi-mediated suppression of p-coumaroyl-CoA 3 ' -hydroxylase in hybrid poplar impacts lignin deposition and soluble secondary metabolism. Proc. Natl. Acad. Sci. U. S. A.  105, 4501 -4506.  Cosgrove D. (1997). Assembly and enlargement of the primary cell wall in plants. Annu. Rev. Cell Dev. Biol.  13, 171 -201. Cosgrove D. (2000). Expansive growth of p lant cell walls. Plant Physiol Biochem 38, 109-124.   119 Cosgrove D. (2005). Growth of the plant cell wall. Nature Reviews Molecular Cell Biology 6, 850 -861. Courtois-Moreau C.L., Pesquet E., Sjodin A., Muniz L., Bollhoner B., Kaneda M., Samuels L., Jansson S., and Tuominen H. (2009). A unique program for cell death in xylem fibers of Populus stem. Plant J  58, 260-274.  Cui H., Levesque M.P., Vernoux T., Jung J.W., Paquette A.J., Gallagher K.L., Wang J.Y., Blilou I., Scheres B., and Benfey P.N. (2007). An evoluti onarily conserved mechanism delimiting SHR movement defines a single layer of endodermis in plants. Science  316, 421-425.  Curtis M. and Grossniklaus U. (2003). A gateway cloning vector set for high -throughput functional analysis of genes in planta. Plant Physiol. 133, 462-469. de Felippes F.F., Ott F., and Weigel D. (2011). Comparative analysis of non-autonomous effects of tasiRNAs and miRNAs in Arabidopsis thaliana. Nucleic Acids Res. 39, 2880-2889. De Luca V. and St Pierre B. (2000). The cell and developmental biology of alkaloid biosynthesis. Trends Plant Sci.  5, 168-173.  Dejardin A., Laurans F., Arnaud D., Breton C., Pilate G., and Leplé J.C. (2010). Wood formation in Angiosperms. C R Biol  333, 325 -34.  Demura T., Tashiro G., Horiguchi G., Kishimoto N., Kubo M., Matsuoka N., Minami A., Nagata-Hiwatashi M., Nakamura K., Okamura Y., Sassa N., Suzuki S., Yazaki J., Kikuchi S., and Fukuda H. (2002). Visualization by comprehensive microarray analysis of gene expression programs during transdifferentiation of mesophyll cells into xylem cells. Proc Natl Acad Sci U S A  99, 15794 -9. Derikvand M.M., Sierra J.B., Ruel K., Pollet B., Do C., Thevenin J., Buffard D., Jouanin L., and Lapierre C. (2008). Redirection of the phenylpropanoid pathway to feruloyl malate in Arabidopsis mutants deficient for cinnamoyl-CoA reductase1. Planta 227, 943 -956.    120 Dharmawardhana D., Ellis B., and Carlson J. (1992). Characterization of vascular lignification in Arabidopsis thaliana. Can. J. Bot. 70, 2238-2244. Dharmawardhana D., Ellis B., and Carlson J. (1995). A beta-glucosidase from lodgepole pine xylem specific for the lignin precursor coniferin. Plant Physiol. 107, 3339. Do C., Pollet B., Thevenin J., Sibout R., Denoue D., Barriere Y., Lapierre C., and Jouanin L. (2007). Both caffeoyl Coenzyme A 3-O -methyltransferase 1 and caffeic acid O -methyltransferase 1 are involved in redundant functions for lignin, flavonoids and sinapoyl malate biosynthesis in Arabidopsis. Planta 226, 1117-1129. Dolan L., Janmaat K., Willemsen V., Linstead P., Poethig S., Roberts K., and Scheres B. (1993). Cellular organization of the Arabidopsis thaliana root. Development 119, 71-84. Ehlting J., Mattheus N., Aeschliman D.S., Li E., Hamberger B., Cullis I.F., Zhuang J., Kaneda M., Mansfield S.D., Samuels L., Ritland K., Ellis B.E., Bohlmann J., and Douglas C.J. (2005). Global transcript profiling of primary stems from Arabidopsis thaliana identifies candidate genes for missing links in lignin biosynthesis and transcriptional regulators of fiber differentiation. Plant J 42, 618-40. Eklof J.M., Ruda M.C., and Brumer H. (2012). Distinguishing xyloglucanase activity in endo-beta(1 -> 4)glucanases. Cellulases 510, 97-120. Eklof J.M., Shojania S., Okon M., McIntosh L.P., and Brumer H.  (2013). Structure-function analysis of a broad specificity Populus trichocarpa endo-beta-glucanase reveals an evolutionary link between bacterial licheninases and plant XTH gene products. J. Biol. Chem.  288, 15786-15799. Esau K. (1943). Ontogeny of the vascular bundle in Zea mays. Hilgardia 15, 327-68. Evert R.F.  (2006). Cell Wall.  In: Esau’s Plant Anatomy Third Edition, p. 65 -101, John Wiley & Sons, Inc., Hoboken, New Jersey.     121 Feuillet C., Lauvergeat V., Deswarte C., Pilate G., Boudet A., and Grima-Pettenati J. (1995). Tissue - and cell-specific expression of a cinnamyl alcohol dehydrogenase promoter in transgenic poplar plants. Plant Mol Biol  27, 651-67.  Franke R., Hemm M., Denault J., Ruegger M., Humphreys J., and Chapple C. (2002a). Changes in secondary metabolism and deposition of an unusual lignin in the ref8 mutant of Arabidopsis. Plant J 30, 47 -59.  Franke R., Humphreys J., Hemm M., Denault J., Ruegger M., Cusumano J., and Chapple C. (2002b). The Arabidopsis REF8 gene encodes the 3-hydroxylase of phenylpropanoid metabolism. Plant J 30, 33-45.  Freudenberg K. (1959). Biosynthesis and constitution of lignin. Nature  183, 1152-1155. Fujita M., Harada H. (1979) Autoradiographic investigations of cell wall development. II. Tritiated phenylalanine and ferulic acid assimilation in relation to lignification. Mokuzai Gakkaishi  25, 89 – 94.  Fukuda H. (1996). Xylogenesis: Initiation, progression, and cell death. Annu. Rev. Plant Physiol. Plant Mol. Biol.  47, 299 -325. Fukuda H. (1997). Tracheary element differentiation. Plant  Cell 9, 1147 -1156. Fukuda H. (2000). Programmed cell death of tracheary elements as a paradigm in plants. Plant Mol Biol  44, 245 -253. Fukuda H. and Komamine A. (1980). Establishment of an experimental system for the study of tracheary element differentiation from single cells isolated from the mesophyll of Z innia elegans . Planta 65, 57 -60. Fukuda H. and Komamine A. (1982). Lignin synthesis and its related enzymes as markers of tracheary-element differentiation in single cells isolated from the mesophyll of Z innia elegans . Planta 155, 423 -430.    122 Fukushima K., Taguchi S., Matsui N., and Yasuda S.  (1997). Distribution and seasonal changes of monolignol glucosides in Pinus thunbergii . Mokuzai Gakkaishi 43, 254-259. Goodno C. (1979). Inhibition of myosin ATPase by vanadate ion. Proc. Natl. Acad. Sci. U. S. A. 76, 2620-2624. Gorzsas A., Stenlund H., Persson P., Trygg J., and Sundberg B.  (2011). Cell-specific chemotyping and multivariate imaging by combined FT-IR microspectroscopy and orthogonal projections to latent structures (OPLS) analysis reveals the chemical landscape of secondary xylem. Plant J 66, 903-14. Goujon T., Sibout R., Pollet B., Maba B., Nussaume L., Bechtold N., Lu F., Ralph J., Mila I., Barriere Y., Lapierre C., and Jouanin L.  (2003). A new Arabidopsis thaliana  mutant deficient in the expression of O-methyltransferase impacts lignins and sinapoyl esters. Plant Mol. Biol. 51, 973-989. Grabber J.H., Hatfield R.D., Lu F., and Ralph J. (2008). Coniferyl ferulate incorporation into lignin enhances the alkaline delignification and enzymatic degradation of cell walls. Biomacromolecules 9, 2510-2516. Gray-Mitsumune M., Mellerowicz E., Abe H., Schrader J., Winzell A., Sterky F., Blomqvist K., McQueen -Mason S., Teeri T., and Sundberg B.  (2004). Expansins abundant in secondary xylem belong to subgroup a of the alpha-expansin gene family. Plant Physiol. 135, 1552-1564. Grefen C., Donald N., Hashimoto K., Kudla J., Schumacher K., and Blatt M.R. (2010). A ubiquitin-10 promoter-based vector set for fluorescent protein tagging facilitates temporal stability and native protein distribution in transient and stable expression studies. Plant J 64, 355-365. Groover A., DeWitt N., Heidel A., and Jones A. (1997). Programmed cell death of plant tracheary elements: Differentiating in vitro. Protoplasma 196, 197-211.   123 Grunwald C., Ruel K., Kim Y., and Schmitt U. (2002). On the cytochemistry of cell wall formation in poplar trees. Plant Biology 4, 13- 21. Haecker A. and Laux T. (2001). Cell- cell signaling in the shoot meristem. Curr. Opin. Plant Biol. 4, 441- 446. Hall H. and Ellis B. (2013). Transcriptional programming during cell wall maturation in the expanding Arabidopsis stem. BMC Plant Biology 13, 14. Harris P.J. and Hartley R.D. (1976). Detection of bound ferulic acid in cell walls of the Gramineae by ultraviolet fluorescence microscopy . Nature  259, 508- 510.  Hashimoto T. and Yamada Y. (1994). Alkaloid biogenesis -  Molecular aspects. Annu. Rev. Plant Physiol. Plant Mol. Biol.  45, 257- 285. Hatfield R., Ralph J., and Grabber J.H. (2008). A potential role for sinapyl p- coumarate as a radical transfer mechanism in grass lignin formation. Planta 228, 919- 928. Hauffe K.D., Paszkowski U., Schulze-Lefert P., Hahlbrock K., Dangl J.L., and Douglas C.J. (1991). A parsley 4CL- 1 promoter fragment specifies complex expression patterns in transgenic tobacco. Plant Cell 3, 435- 43. Hemm M., Rider S., Ogas J., Murry D., and Chapple C. (2004). Light induces phenylpropanoid metabolism in Arabidopsis roots. Plant J 38, 765- 778. Hosokawa M., Suzuki S., Umezawa T., and Sato Y. (2001). Progress of lignification mediated by intercellular transportation of monolignols during tracheary element differentiation of isolated Z innia  mesophyll cells. Plant Cell Physiol 42, 959- 68. Humphreys J. and Chapple C. (2002). Rewriting the lignin roadmap. Curr. Opin. Plant Biol.  5, 224- 229.   124 Jones L., Ennos A.R., and Turner S.R. (2001). Cloning and characterization of irregular xylem4  (irx4 ): a severely lignin-deficient mutant of Arabidopsis. Plant J 26, 205-16. Jung J.H. and Park C.M. (2007). Vascular development in plants: Specification of xylem and phloem tissues. Journal of Plant Biology 50, 301-305. Kaneda M. (2008). Elucidation of secondary cell wall secretion mechanisms of Arabidopsis thaliana , poplar (Populus deltoides x P. trichocarpa ) and pine (Pinus contorta ). PhD Thesis, University of British Columbia. Kaneda M., Rensing K.H., Wong J.C., Banno B., Mansfield S.D., and Samuels A.L. (2008). Tracking monolignols during wood development in lodgepole pine. Plant Physiol 147, 1750-60. Kaneda M., Schuetz M., Lin B.S., Chanis C., Hamberger B., Western T.L., Ehlting J., and Samuels A.L. (2011). ABC transporters coordinately expressed during lignification of Arabidopsis stems include a set of ABCBs associated with auxin transport. J Exp Bot 62, 2063-77. Karimi M., Inze D., and Depicker A. (2002). GATEWAY((TM)) vectors for Agrobacterium -mediated plant transformation. Trends Plant Sci. 7, 193-195. Kerridge D. (1958). The effect of actidione and other antifungal agents on nucleic acid and protein synthesis in Saccharomyces carlsbergensis . J. Gen. Microbiol. 19, 497-506 Kidner C.A. and Martienssen R.A. (2004). Spatially restricted microRNA directs leaf polarity through ARGONAUTE1. Nature 428, 81-4. Kubo M., Udagawa M., Nishikubo N., Horiguchi G., Yamaguchi M., Ito J., Mimura T., Fukuda H., and Demura T. (2005). Transcription switches for protoxylem and metaxylem vessel formation. Genes and Development 19, 1855-60.   125 Lanot A., Hodge D., Jackson R.G., George G.L., Elias L., Lim E., Vaistij F.E., and Bowles D.J. (2006). The glucosyltransferase UGT72E 2 is responsible for monolignol 4-O -glucoside production in Arabidopsis thaliana. Plant J 48, 286-295. Lanot A., Hodge D., Lim E., Vaistij F.E., and Bowles D.J.  (2008). Redirection of flux through the phenylpropanoid pathway by increased glucosylation of soluble intermediates. Planta 228, 609-616. Lee S. and Roberts A. (2004). Tracheary element differentiation is correlated with inhibition of cell expansion in xylogenic mesophyll suspension cultures. Plant Physiol Biochem 42, 43-48. Lee Y., Foster J., Chen J., Voll L.M., Weber A.P.M., and Tegeder M.  (2007). AAP1 transports uncharged amino acids into roots of Arabidopsis. Plant J 50, 305-319. Lee Y., Rubio M.C., Al assimone J., and Geldner N. (2013). A mechanism for localized lignin deposition in the endodermis. Cell 153, 402-412. Leinhos V. and Savidge R.A.  (1993). Isolation of protoplasts from developing xylem of Pinus banksiana and Pinus strobus . Canadian Journal of Forest Research 23, 343-348. Leplé J., Dauwe R., Morreel K., Storme V., Lapierre C., Pollet B., Naumann A., Kang K., Kim H., Ruel K., Lefebvre A., Joseleau J., Grima-Pettenati J., De Rycke R., Andersson-Gunneras S., Erban A., Fehrle I., Petit-Conil M., Kopka J., Polle A., Messens E., Sundberg B., Mansfield S.D., Ralph J., Pilate G., and Boerjan W. (2007). Downregulation of cinnamoyl-Coenzyme A reductase in poplar: Multiple -level phenotyping reveals effects on cell wall polymer metabolism and structure. Plant Cell 19, 3669-3691. Lev-Yadun S. and Flaishman M.  (2001). The effect of submergence on ontogeny of cambium and secondary xylem and on fiber lignification in inflorescence stems of Arabidopsis. Iawa Journal 22, 159-169. LevYadun S.  (1997). Fibres and fibre -sclereids in wild-type Arabidopsis thaliana. Ann. Bot.  80, 125-129.   126 Li X. and Chapple C. (2010). Understanding lignification: challenges beyond monolignol biosynthesis. Plant Physiol 154, 449-52. Li X., Bonawitz N.D., Weng J., and Chapple C. (2010). The growth reduction associated with repressed lignin biosynthesis in Arabidopsis thaliana  is independent of flavonoids. Plant Cell 22, 1620-1632. Li Y., Darley C., Ongaro V., Fleming A., Schipper O., Baldauf S., and McQueen-Mason S. (2002). Plant expansins are a complex multigene family with an ancient evolutionary origin. Plant Physiol. 128, 854-864. Livak K.J., Schmittgen T.D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402-8. Lu F. and Ralph J. (1997a). DFRC method for lignin analysys. 1. New method for beta aryl ether cleavage: Lignin model studies. J. Agric. Food Chem. 45, 4655-4660. Lu F. and Ralph J. (1997b). Derivatization followed by reductive cleavage (DFRC method), a new method for lignin analysis: Protocol for analysis of DFRC monomers. J. Agric. Food Chem. 45, 2590-2592. Lucas W.J., Ham L., and Kim J. (2009). Plasmodesmata - bridging the gap between neighboring plant cells. Trends Cell Biol. 19, 495-503. Mahler H.R. and Cordes E.H. (1966) Biological Chemistry. Harper & Row, Publishers, Inc. New York, New York. pp.697. Mansfield S.D., Kang K., and Chapple C. (2012). Designed for deconstruction - poplar trees altered in cell wall lignification improve the efficacy of bioethanol production. New Phytol. 194, 91-101. Mansfield S.D., Kim H., Lu F., and Ralph J. (2012). Whole plant cell wall characterization using solution-state 2D NMR. Nature Protocols 7, 1579-1589.   127  McCann M.C., Stacey N.J., Dahiya P., Milioni D., Sado P.E., and Roberts K. (2001). Zinnia. Everybody needs good neighbors. Plant Physiol  127, 1380-2. McQueen-Mason S. and Cosgrove D. (1994). Disruption of hydrogen -bonding between plant-cell wall polymers by proteins that induce wall extension. Proc. Natl. Acad. Sci. U. S. A.  91, 6574 -6578.  McQueen-Mason S., Durachko D., and Cosgrove D. (1992). Two endogenous proteins that induce cell-wall extension in plants. Plant Cell 4, 1425 -1433.  Mechin V., Baumberger S., Pollet B., and Lapierre C. (2007). Peroxidase activity can dictate the in vitro lignin dehydrogenative polymer structure. Phytochemistry 68, 571 -579.  Mellerowicz E.J. and Sundberg B. (2008). Wood cell walls: biosynthesis, developmental dynamics and their implications for wood properties. Curr Opin Plant Biol  11, 293-300. 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. Proc Natl Ac ad Sci U. S. A.  95, 6619 -23. Miao Y. and Liu C. (2010). ATP-binding cassette-like transporters are involved in the transport of lignin precursors across plasma and vacuolar membranes. Proc. Natl. Acad. Sci. U. S. A.  107, 22728 -22733.  Milioni D., Sado P., Stacey N., Roberts K., and McCann M. (2002). Early gene expression associated with the commitment and differentiation of a plant tracheary element is revealed by cDNA -amplified fragment length polymorphism analysis. Plant Cell 14, 2813-2824.  Mitsuda N., Iwase A., Yamamoto H., Yoshida M., Seki M., Shinozaki K., and Ohme-Takagi M. (2007). NAC transcription factors, NST1 and NST3, are key regulators of the formation of secondary walls in woody tissues of Arabidopsis. Plant Cell 19, 270 -80.   128 Mitsuda N., Seki M., Shinozaki K., and Ohme-Takagi M.  (2005). The NAC transcription factors NST1 and NST2 of Arabidopsis regulate secondary wall thickenings and are required for anther dehiscence. Plant Cell 17, 2993 -3006.  Morreel K., Ralph J., Kim H., Lu F., Goeminne G., Ralph S., Messens E., and Boerjan W. (2004). Profiling of oligolignols reveals monolignol coupling conditions in lignifying poplar xylem. Plant Physiol. 136, 3537 -3549.  Murata J. and De Luca V. (2005). Localization of tabersonine 16 -hydroxylase and 16 -OH tabersonine-16 -O-methyltransferase to leaf epidermal cells defines them as a major site of precursor biosynthesis in the vindoline pathway in Catharanthus roseus . Plant J 44, 581-594.  Naseer S., Lee Y., Lapierre C., Franke R., Nawrath C., and Geldner N.  (2012). Casparian strip diffusion barrier in Arabidopsis is made of a lignin polymer without suberin. Proc. Natl. Acad. Sci. U. S. A. 109, 10101-10106.  Novo-Uzal E., Fernandez -Perez F., Herrero J., Gutierrez J., Gomez-Ros L.V., Angeles Bernal M., Diaz J., Cuello J., Pomar F., and Angeles Pedreno M.  (2013). From Zinnia to Arabidopsis: approaching the involvement of peroxidases in lignification. J. Exp. Bot.  64, 3499 -3518. Oda Y., Mimura T., and Hasezawa S.  (2005). Regulation of secondary cell wall development by cortical microtubules during tracheary element differentiation in Arabidopsis cell suspensions. Plant Physiol. 137, 1027 -1036.  Oh S., Park S., and Han K. (2003). Transcriptional regulation of secondary growth in Arabidopsis thaliana.  J. Exp. Bot.  54, 270 9 -2722.  Ohashi-Ito K., Oda Y., and Fukuda H.  (2010). Arabidopsis VASCULAR-RELATED NAC-DOMAIN6 directly regulates the genes that govern programmed cell death and secondary wall formation during xylem differentiation. Plant Cell 22, 3461 -73.    129 Okumoto S., Schmidt R., Tegeder M., Fischer W., Rentsch D., Frommer W., and Koch W. (2002). High affinity amino acid transporters specifically expressed in xylem parenchyma and developing seeds of Arabidopsis. J. Biol. Chem. 277, 45338-45346. Ossowski S., Schwab R., and Weigel, D. (2008). Gene silencing in plants using artificial microRNAs and other small RNAs. Plant J 53, 674–690. Palatnik J., Allen E., Wu X., Schommer C., Schwab R., Carrington J., and Weigel D. (2003). Control of leaf morphogenesis by microRNAs. Nature 425, 257-263. Palmgren M.G. and Harper J.F. (1999). Pumping with plant P-type ATPases. J. Exp. Bot. 50, 883-893. Pashkovskiy P.P. and Ryazansky S.S. (2013). Biogenesis, evolution, and functions of plant microRNAs. Biochemistry-Moscow 78, 627-637. Pesquet E., Korolev A.V., Calder G., and Lloyd C.W. (2010). The microtubule-associated protein AtMAP70-5 regulates secondary wall patterning in Arabidopsis wood cells. Curr Biol 20, 744-9. Pesquet E., Ranocha P., Legay S., Digonnet C., Barbier O., Pichon M., and Goffner D. (2005). Novel markers of xylogenesis in zinnia are differentially regulated by auxin and cytokinin. Plant Physiol. 139, 1821-1839. Pesquet E., Zhang B., Gorzsás A., Puhakainen T., Serk H., Escamez S., Barbier O., Gerber L., Courtois-Moreau C., Alatalo E., Paulin L., Kangasjärvi J., Sundberg B., Goffner D., and Tuominen H. (2013). Non-cell-autonomous postmortem lignification of tracheary elements in Zinnia elegans . Plant Cell 25, 1314-1328. Petrik D.L., Karlen S.D., Cass C.L., Padmakshan D., Lu F., Liu S., Le Bris P., Antelme S., Santoro N., Wilkerson C.G., Sibout R., Lapierre C., Ralph J., and Sedbrook J.C. (2013). p-Coumaroyl-CoA:Monolignol Transferase (PMT) acts specifically in the lignin biosynthetic pathway in Brachypodium distachyon . Plant J (Accepted Article, doi: 10.1111/tpj.12420).   130  Pickett -Heaps J.D. (1968). Radiographic investigations using lignin precursors. Protoplasma 65, 181-205.  Ralph J. (2010). Hydroxycinnamates in lignification. Phytochemistry Reviews 9, 65 -83.  Ralph J., Lundquist K., Brunow G., Lu F., Kim H., Schatz P., Marita J., Hatfield R., Ralph S., Christensen J., et al. (2004). Lignins: natural polymers from oxidative coupling of 4 -hydroxyphenylpropanoids. Phytochemistry 3 , 29– 60. Ranocha P., Chabannes M., Chamayou S., Danoun  S., Jauneau A., Boudet A., and Goffner D. (2002). Laccase down -regulation causes alterations in phenolic metabolism and cell wall structure in poplar. Plant Physiol. 129, 145 -155.  Roberts K. and McCann M.C.  (2000). Xylogenesis: the birth of a corpse. Curr  Opin Plant Biol  3, 517 -522.  Robinson A.R. and Mansfield S.D.  (2009). Rapid analysis of poplar lignin monomer composition by a streamlined thioacidolysis procedure and near-infrared reflectance-based prediction modeling. Plant J 58, 706 -714.  Roelofsen, P. A. (1959). The plant cell wall. H erb. Pflanzenanct. 111 Teil 4. Abteilung Cytologie Gebruder Bomtregger, Berlin.  Sampedro J., Carey R., and Cosgrove D.  (2006). Genome histories clarify evolution of the expansin superfamily: New insights from the poplar ge nome and pine ESTs. J. Plant Res.  119, 11-21. Savidge R.  (1989). Coniferin, a biochemical indicator of commitment to tracheid differentiation in conifers. Can J Bot 67, 2663 -2668. Schalk M., Cabello -Hurtado F., Pierrel M., Atanossova R., Saindrenan P., and  Werck -Reichhart D. (1998). Piperonylic acid, a selective, mechanism-based inactivator of the trans-  131 cinnamate 4-hydroxylase: A new tool to control the flux of metabolites in the phenylpropanoid pathway. Plant Physiol. 118, 209-218. Scheller H.V. and Ulvsko v P. (2010). Hemicelluloses. Annu Rev Plant Biol  61, 263-89. Schilmiller A.L., Stout J., Weng J., Humphreys J., Ruegger M.O., and Chapple C.  (2009). Mutations in the cinnamate 4-hydroxylase gene impact metabolism, growth and development in Arabidopsis. Plant J 60, 771 -782.  Schoch G., Goepfert S., Morant M., Hehn A., Meyer D., Ullmann P., and Werck -Reichhart D. (2001). CYP98A3 from Arabidopsis thaliana is a 3 ' -hydroxylase of phenolic esters, a missing link in the phenylpropanoid pathway. J. Biol. Chem. 276, 36566-36574.  Schuetz M., Smith R., and Ellis B.  (2013). Xylem tissue specification, patterning, and differentiation mechanisms. J. Exp. Bot . 64, 11-31. Schwab R., Ossowski S., Riester M., Warthmann N., and Weigel D.  (2006). Highly specific gene silencing by artificial microRNAs in Arabidopsis. Plant Cell 18, 1121-1133. Schwab R., Palatnik J., Riester M., Schommer C., Schmid M., and Weigel D. (2005). Specific effects of microRNAs on the plant transcriptome. Developmental Cell 8, 517 -527.  Sieburth L. and Deyholos M. (2006). Vascular development: the long and winding road. Curr Opin Plant Biol 9, 48-54. Sieburth L.E., Muday G.K., King E.J., Benton G., Kim S., Metcalf K.E., Meyers L., Seamen E., and Van Norman J.M. (2006). SCARFACE encodes an ARF -GAP that is required for normal auxin efflux and vein patterning in Arabidopsis. Plant Cell 18, 1396-411. Simmons B.A., Loque D., and Ralph J.  (2010). Advances in modifying lignin for enhanced biofuel production. Curr Opin Plant Biol 13, 313-20.   132 Spurr A. (1969). A low -viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res.  26, 31-43. Takabe K., Fujita M., Harada H., and Saiki H. (1985). Autoradiographic investigation of lignification in the cell walls of Cryptomeria (Cryptomeria japonica  D Don). Mokuzai Gakkaishi 31, 613 -619.  Takabe K., Fujita M., Harada H., Saiki H. (1981). Lignification process of Japanese black pine ( Pinus thu nbergii  Parl.) tracheids. Mokuzai Gakkaishi 11, 813-820. Taylor N., Scheible W., Cutler S., Somerville C., and Turner S. (1999). The irregular xylem 3  locus of Arabidopsis encodes a cellulose synthase required for secondary cell wall synthesis. Plant Cell 11, 769 -779. Taylor N.G. (2008). Cellulose biosynthesis and deposition in higher plants. New Phytol 178, 239-52. Terashima N. and Fukushima K. (1988). Heterogeneity in formation of lignin. XI. An autoradiographic study of the heterogeneous formation and structure of pine lignin. Wood Sci Technol 22, 259-270. Terashima N., Fukushima K., and Takabe K. (1986). Heterogeneity in formation of lignin. VIII. An autoradiographic study on the formation of guaiacyl and syringyl lignin in Magnolia kobus  DC. Wood Sci Technol 40, 101-105. Terashima N., Fukushima K., He L.-., and Takabe K. (1993). Comprehensive model of the lignified plant cell wall. In Forage Cell Wall Structure and Digestibility, H. G. Jung, D. R. Buxton, R. D. Hatfield and J. Ralph, eds (Madison, WI: A SA, CSSA, and SSSA) pp. 247-270. Thevenin J., Pollet B., Letarnec B., Saulnier L., Gissot L., Maia-Grondard A., Lapierre C., and Jouanin L. (2011). The simultaneous repression of CCR and CAD, two enzymes of the lignin biosynthetic pathway, results in sterility and dwarfism in Arabidopsis thaliana . Molecular Plant 4, 70-82.   133 Tobimatsu Y., Davidson C.L., Grabber J.H., and Ralph J. (2011). Fluorescence-tagged monolignols: synthesis, and application to studying in vitro  lignification. Biomacromolecules 12, 1752 -1761.  Tobimatsu Y., Wagner A., Donaldson L., Mitra P., Niculaes C., Dima O., Kim J.I., Anderson N., Loque D., Boerjan W., Chapple C., and Ralph J. (2013). Visualization of plant cell wall lignification using fluorescence-tagged monolignols. Plant J 76, 357 -366. Tokunaga N., Kaneta T., Sato S., and Sato Y. (2009). Analysis of expression profiles of three peroxidase genes associated with lignification in Arabidopsis thaliana . Physiol. Plantarum 136, 237 -249.  Tokunaga N., Sakakibara N., Umezawa T., Ito Y., Fukuda H., and Sato Y. (2005). Involvement of extracellular dilignols in lignification during tracheary element differentiation of isolated Zinnia  mesophyll cells. Plant Cell Physiol 46, 224 -232. Tsuji Y., Chen F., Yasuda S., and Fukushima K. (2005). Unexpect ed behavior of coniferin in lignin biosynthesis of Ginkgo biloba  L. Planta 222, 58 -69.  Tsuyama T., Kawai R., Shitan N., Matoh T., Sugiyama J., Yoshinaga A., Takabe K., Fujita M., and Yazaki K. (2013). Proton-dependent coniferin transport, a common major tr ansport event in differentiating xylem tissue of woody plants. Plant Physiol.  162, 918 -926.  Turner S. and Somerville C. (1997). Collapsed xylem phenotype of Arabidopsis identifies mutants deficient in cellulose deposition in the secondary cell wall. Plant Cell 9, 689 -701.  Vanholme R., Cesarino I., Rataj K., Xiao Y., Sundin L., Goeminne G., Kim H., Cross J., Morreel K., Araujo P., Welsh L., Haustraete J., McClellan C., Vanholme B., Ralph J., Simpson G.G., Halpin C., Boerjan W. (2013). Caffeoyl shikimate esterase (CSE) is an enzyme in the lignin biosynthetic pathway in Arabidopsis. Science 341, 1103-6. Vanholme R., Demedts B., Morreel K., Ralph J., and Boerjan W. (2010). Lignin Biosynthesis and Structure. Plant Physiol. 153, 895 -905.    134 Vanholme R., Morreel K., Ralph J., and Boerjan W. (2008). Lignin engineering. Curr Opin Plant Biol  11, 278 -85.  Vanholme R., Storme V., Vanholme B., Sundin L., Christensen J.H., Goeminne G., Halpin C., Rohde A., Morreel K., and Boerjan W. (2012). A systems biology view of responses to lignin biosynthesis perturbations in Arabidopsis. Plant Cell 24, 3506 -3529. Vaten A., Dettmer J., Wu S., Stierhof Y., Miyashima S., Yadav S.R., Roberts C.J., Campilho A., Bulone V., Lichtenberger R., Lehesranta S., Mahonen A.P., Kim J., Jokitalo E., Sauer N., Scheres B., Nakajima K., Carlsbecker A., Gallagher K.L., and Helariutta Y.  (2011). Callose biosynthesis regulates symplastic trafficking during root development. Developmental Cell 21, 1144-1155. Weid M., Ziegler J., and Kutchan T.  (2004). The roles of latex and the vascular bundle in morphine biosynthesis in the opium poppy, Papaver somniferum . Proc. Natl. Acad. Sci. U. S. A.  101, 13957-13962.  Weng J., Li X., Bonawitz N.D., and Chapple C. (2008). Emerging  strategies of lignin engineering and degradation for cellulosic biofuel production. Curr. Opin. Biotechnol.  19, 166 -172. Weng J., Li X., Stout J., and Chapple C. (2008). Independent origins of syringyl lignin in vascular plants. Proc. Natl. Acad. Sci. U. S. A.  105, 7887 -7892.  Wightman R. and Turner S. (2010). Trafficking of the cellulose synthase complex in developing xylem vessels. Biochem Soc Trans  38, 755-60.  Wilkerson C.G., Mansfield S.D., Lu F., Karlen S.D., Rencoret J., Withers S., Park J. - Y., Gonzales- Vigil E., and Ralph J. (2013). Introduction of “zip -ligninsTM” into plants. Science (submitted December 30, 2013).    135 Withers S., Lu F., Kim H., Zhu Y., Ralph J., and Wilkerson C.G.  (2012). Identification of grass-specific enzyme that acylates monolignols with p-coumarate. J. Biol. Chem.  287, 8347 -8355.  Wu G. (2013). Plant microRNAs and development. Journal of Genetics and Genomics 40, 217 -230. Xie J., Lu J., Qian Z., Yu Y., Duan J., and Li S.  (2009). Optimization and comparison of five methods for extraction of coniferyl ferulate from Angelica sinensis. Molecules 14, 555-565. Yamaguchi M., Goue N., Igarashi H., Ohtani M., Nakano Y., Mortimer J.C., Nishikubo N., Kubo M., Katayama Y., Kakegawa K., Dupree P., and Demura T. (2010a). VASCULAR -RELATED NAC -DOMAIN6  and VASCULAR -RELATED NAC -DOMAIN7 effectively induce transdifferentiation into xylem vessel elements under control of an induction system. Plant Physiol 153, 906-14. Yamaguchi M., Kubo M., Fukuda H., and Demura T. (2008). Vascular -related NAC-DOMAIN7 is in volved in the differentiation of all types of xylem vessels in Arabidopsis roots and shoots. Plant J  55, 652-64. Yamaguchi M., Ohtani M., Mitsuda N., Kubo M., Ohme- Takagi M., Fukuda H., and Demura T. (2010b). VND -INTERACTING2, a NAC domain transcription fa ctor, negatively regulates xylem vessel formation in Arabidopsis. Plant Cell 22, 1249-63. Yang F., Mitra P., Zhang L., Prak L., Verhertbruggen Y., Kim J.S., Sun L., Zheng K., Tang K., Auer M., Scheller H.V., and Loqué D.  (2012). Engineering secondary cell wall deposition in plants. Plant Biotech J 11, 325-335. Yong Z., Kotur Z., and Glass A.D.M.  (2010). Characterization of an intact two-component high-affinity nitrate transporter from Arabidopsis roots. Plant J 63, 739 -74 8.    136 Young R.E., McFarlane H.E., Hahn M.G., Western T.L., Haughn G.W., and Samuels A.L. (2008). Analysis of the Golgi apparatus in Arabidopsis seed coat cells during polarized secretion of pectin-rich mucilage. Plant Cell 20, 1623-1638. Zhong R. and Ye Z.H. (1999). IFL1, a gene regulating interfascicular fiber differentiation in Arabidopsis, encodes a homeodomain-leucine zipper protein. Plant Cell  11, 2139-52. Zhong R., Demura T., and Ye Z.H. (2006). SND1, a NAC domain transcription factor, is a key regulator of secondary wall synthesis in fibers of Arabidopsis. Plant Cell 18, 3158-70.  Zhong R., Lee C., and Ye Z. (2010). Evolutionary conservation of the transcriptional network regulating secondary cell wall biosynthesis. Trends Plant Sci.  15, 625-632. Zhong R., Lee C., and Ye Z.H. (2010). Global analysis of direct targets of secondary wall NAC master switches in Arabidopsis. Molecular Plant 3, 1087 -103. Zhong R., Richardson E., and Ye Z. (2007). Two NAC domain transcription factors, SND1 and NST1, function redu ndantly in regulation of secondary wall synthesis in fibers of Arabidopsis. Planta 225, 1603-1611. Zhong R., Taylor J.J., and Ye Z.H. (1997). Disruption of interfascicular fiber differentiation in an Arabidopsis mutant. Plant Cell 9, 2159-70.  Zhou J., Lee C., Zhong R., and Ye Z.H. (2009). MYB58 and MYB63 are transcriptional activators of the lignin biosynthetic pathway during secondary cell wall formation in Arabidopsis. Plant Cell 21, 248-66. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0167289/manifest

Comment

Related Items