Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional analysis of KNOTTED-like homeobox and OVATE family proteins involved in secondary cell wall… Wang, Shumin 2018

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2018_september_wang_shumin.pdf [ 28.42MB ]
Metadata
JSON: 24-1.0366123.json
JSON-LD: 24-1.0366123-ld.json
RDF/XML (Pretty): 24-1.0366123-rdf.xml
RDF/JSON: 24-1.0366123-rdf.json
Turtle: 24-1.0366123-turtle.txt
N-Triples: 24-1.0366123-rdf-ntriples.txt
Original Record: 24-1.0366123-source.json
Full Text
24-1.0366123-fulltext.txt
Citation
24-1.0366123.ris

Full Text

FUNCTIONAL ANALYSIS OF KNOTTED-LIKE HOMEOBOX AND OVATE FAMILY PROTEINS INVOLVED IN SECONDARY CELL WALL DEVELOPMENT IN ARABIDOPSIS by  Shumin Wang  M.Sc., Nanjing Agricultural University, 2011  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 2018  © Shumin Wang, 2018  ii   Abstract The formation of plant secondary cell walls requires a complex network of transcriptional regulation, culminating in a coordinated suite of biosynthetic genes depositing walls, in a spatial and temporal fashion. The transcription factor KNOTTED ARABIDOPSIS THALIANA7 (KNAT7) is a Class II KNOTTED1-like homeobox (KNOX2) gene, that acts as a negative regulator of secondary cell wall biosynthesis in interfascicular fibers. Previously, members of Ovate Family Proteins (OFP1 and OFP4), were shown to interact with KNAT7 to negatively regulate wall formation. However, the function of other closely related KNOX2 and OFP genes in secondary wall formation remains unclear. Herein, I showed that knat3knat7 double mutants possess an enhanced irregular xylem (irx) phenotype relative to single mutants, and decreased interfascicular fiber cell wall thickness. Additionally, unlike the increased lignin content characteristic of knat7 mutants, knat3knat7 had no change in lignin content, while the monomeric lignin composition was substantially reduced relative to the wild-type plants.  In contrast, KNAT3 overexpression resulted in thicker interfascicular fiber secondary walls, suggesting a positive regulation of KNAT3 in wall development.  A thorough examination of OFP mutants showed that none of the single mutants revealed any wall defects, including ofp4, which was previously shown to interact with KNAT7. However, they do display leaf phenotypes.  In contrast, plants overexpressing OFP isoforms consistently exhibited cell swelling, disordered microtubules, and dark-grown de-etiolated phenotypes, resembling phenotypes common to brassinosteroid deficient mutants. Using yeast two-hybrid and bimolecular fluorescence complementation assays, I identified two genes that interacted with OFP4, NAP1;1 and NAP1;2, members of the Nucleosome Assembly Protein 1 (NAP1) family. Higher-order, loss-of-function NAP1 and OFP mutants also exhibit altered cotyledon shape and a reduced cotyledon width:length ratio. The kidney-shaped cotyledon phenotype apparent in OFP4 overexpressing plants was suppressed in the nap1;1 nap1;2 nap1;3 triple mutant background. Together, my research suggests that in addition to KNAT7, KNAT3 also contributes to cell wall deposition, and that a complex iii   network of positive and negative regulation governed by KNOX2 proteins regulates secondary wall formation. Moreover, the complex of OFP4 and NAP1 plays a significant role in the cotyledon development.   iv   Lay Summary Plant secondary cell walls form the foundation of fibers and wood, and understanding their formation has important biological and economic implications. Arabidopsis has proven a useful model for secondary wall biosynthesis, due to it short generation times, amenablity to transformation, substantive gene mutations, and have a plethora of genetic resources. Two proteins, KNAT7 and OFP4, were reported to negatively regulate secondary wall formation in interfascicular fibers. My research attempted to functionally characterize genes closely associated with KNAT7 and OFP4, and elucidate their involvement in secondary wall formation. I clearly showed that KNAT3 functions together with KNAT7 to activate xylem vessel secondary wall formation, while acting antagonistically during secondary wall formation in interfascicular fibers. Concurrently, although expressed in the appropriate developmental window, I showed that the OFP genes function to maintain plant hormone homeostasis and regulate cotyledon development instead of participating in secondary wall development as originally hypothesized.      v   Preface Figures in Chapter 1 were reproduced with permission from the publishers.  Experiments in Chapter 2 employed seeds of transgenic lines obtained graciously from Dr. John Bowman (Monash University, Australia). Dr. Etienne Grienenberger generated the knat3knat7 double mutants and first documented the weak stem phenotype. Dr. Masatoshi Yamaguchi generated the ProKNAT4::GUS and ProKNAT5::GUS transgenic lines. Shumin Wang, Dr. Carl Douglas, and Dr. Shawn Mansfield designed the experiments. Shumin Wang performed all the experiments and data analysis. Shumin Wang wrote the manuscript with the assistance of Drs. Lacey Samuels, Shawn Mansfield, and Etienne Grienenberger.  Experiments in Chapter 3 employed the seeds of nap1 mutants and pUBQ1::mRFP-TUB6 transgenic plants graciously obtained from Dr. Aiwu Dong (Fudan University, China) and Dr. Chris Ambrose (University of Saskatchewan, Canada), respectively. Shumin Wang, Dr. Carl Douglas, Dr. Lacey Samuels, and Dr. Shawn Mansfield designed the experiments. Shumin Wang performed all the experiments and data analysis.   vi   Table of Contents Abstract ............................................................................................................................... ii Lay Summary ..................................................................................................................... iv Preface ................................................................................................................................. v Table of Contents ............................................................................................................... vi List of Tables ...................................................................................................................... xi List of Figures ................................................................................................................... xii List of Abbreviations ........................................................................................................ xv Acknowledgements ........................................................................................................ xviii Dedication .......................................................................................................................... xx Chapter 1: Introduction ..................................................................................................... 1 1.1    Secondary cell wall component biosynthesis .......................................................... 2 1.2    Xylem development and interfascicular fiber differentiation .................................. 5 1.3    Molecular mechanisms of vascular development .................................................... 6 1.4    Secondary cell wall transcriptional regulation network .......................................... 8 1.5 KNOTTED-like homeobox (KNOX) family proteins ........................................... 11 1.6    Ovate Family Proteins (OFPs) ............................................................................... 13 1.7 Research objectives and Significance of Findings ................................................ 15 Chapter 2: The Class II KNOX genes KNAT3 and KNAT7 work cooperatively to regulate secondary cell wall deposition and provide mechanical support to Arabidopsis stems ............................................................................................................. 18 2.1 Introduction ............................................................................................................ 18 vii   2.2 Materials and methods ........................................................................................... 20 2.2.1 Plant material and growth condition ................................................................ 20 2.2.2 Cloning and plant transformation .................................................................... 21 2.2.3 GUS expression assay ...................................................................................... 22 2.2.4 Microscopy ...................................................................................................... 22 2.2.5 Physical tests .................................................................................................... 23 2.2.6 Cell-wall analysis ............................................................................................. 23 2.2.7 Total RNA isolation and quantitative RT-PCR ............................................... 24 2.2.8 RNA-seq analysis ............................................................................................ 24 2.3 Results .................................................................................................................... 25 2.3.1 KNOX2 genes are expressed in the inflorescence stems ................................. 25 2.3.2   knat3knat7 double knock-out mutants have enhanced irregular xylem (irx) phenotype .................................................................................................................... 26 2.3.3   Knockout of KNAT3 and KNAT7 affects stem mechanical properties and fiber wall thickness ............................................................................................................. 27 2.3.4 KNAT3 and KNAT7 expression affect secondary cell wall composition ......... 28 2.3.5   knat3knat7 phenotypes are complemented by ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP ............................................................................................... 29 2.3.6   Overexpression of KNAT3 increases interfascicular fiber wall thickness of stems ........................................................................................................................... 30 2.3.7   Genome-wide transcript profiling shows altered expression of secondary cell wall related genes in knat3knat7 plants ...................................................................... 31 viii   2.4    Discussion .............................................................................................................. 32 Chapter 3: Ovate Family Proteins are associated with Brassinosteroid homeostasis and function in cotyledon development by interacting with Nucleosome Assembly Protein 1 in Arabidopsis ................................................................................................... 63 3.1    Introduction ........................................................................................................... 63 3.2    Materials and methods ........................................................................................... 65 3.2.1   Plant materials and growth condition .............................................................. 65 3.2.2   RNA isolation and quantitative RT-PCR ........................................................ 65 3.2.3   Cloning and plant transformation .................................................................... 66 3.2.4   GUS expression assay ..................................................................................... 67 3.2.5   Transient expression in N. benthamiana and BiFC assay ............................... 67 3.2.6   Light and confocal microscopy ....................................................................... 67 3.2.7   BR treatment .................................................................................................... 68 3.2.8   Yeast two-hybrid assays .................................................................................. 68 3.3    Results ................................................................................................................... 69 3.3.1 OFP loss-of-function mutants have no secondary cell wall defects ................ 69 3.3.2   OFP genes are expressed in Arabidopsis seedlings ........................................ 70 3.3.3    OFP proteins are localized to the nucleus and cytoplasm .............................. 71 3.3.4    OFP overexpression plants have cell swelling and de-etiolated phenotypes in hypocotyls ................................................................................................................... 71 3.3.5    OFP overexpression plants show responses to exogenous BR treatment in hypocotyls ................................................................................................................... 72 ix   3.3.6    OFP4 interacts with Nucleosome Assembly Protein 1 ................................... 73 3.3.7    NAP1;1 and NAP1;2 proteins are localized to the ER membrane in epidermal cells of N. benthamiana leaves ................................................................................... 74 3.3.8    NAP1 and OFP4 loss-of-function mutants have altered cotyledon shapes .... 75 3.3.9    Genetic interactions between OFP4 and NAP1 .............................................. 77 3.4     Discussion ............................................................................................................. 77 Chapter 4: Conclusions .................................................................................................. 103 4.1    Major findings of the thesis ................................................................................. 103 4.1.1    KNAT3 and KNAT7 function together to activate xylem vessel secondary wall formation .................................................................................................................. 103 4.1.2    KNAT3 acts antagonistically with KNAT7 during secondary cell wall formation in interfascicular fibers ............................................................................ 104 4.1.3    OFP genes are not associated with Brassinosteroid signaling pathway in Arabidopsis ............................................................................................................... 105 4.1.4    OFP proteins regulate cotyledon development by interacting with NAP1 ... 105 4.2    Future directions .................................................................................................. 107 4.2.1    Identifying direct target genes of KNAT3 and KNAT7 in different cell tissues .................................................................................................................................. 107 4.2.2    Identifying KNAT3 and KNAT7 cell-specific interacting partners ............. 108 4.2.3    Investigating upstream signals of KNAT3 and KNAT7 involved in secondary cell wall development ............................................................................................... 109 4.2.4    Elucidating the roles of OFP in cotyledon development .............................. 109 x   4.2.5    Investigating the relationships between OFP and plant hormones ............... 110 4.2.6    Characterizing the functions of OFP4-TRM20 ............................................ 110 References ........................................................................................................................ 111   xi   List of Tables Table 2.1 Oligonucleotides used in Chapter 2 ........................................................................ 37 Table 2.2 Cell-wall monosaccharide content of WT, knat3, knat7 and knat3knat7 stems ..... 48 Table 2.3 Cellulose and lignin contents of WT, knat3, knat7 and knat3knat7 stem bases ..... 49 Table 2.4 Expression changes of secondary cell wall-related genes in knat3knat7 stems ..... 62 Table 3.1 Oligonucleotides used in Chapter 3 ........................................................................ 82 Table 3.2 Identification of the cDNA encoding interactors of OFP4 ..................................... 94   xii   List of Figures Figure 1.1 Phylogenetic analysis of Arabidopsis KNOX (a) and OFP (b) gene families ...... 16 Figure 1.2 Transcriptional network of secondary cell wall development in Arabidopsis ...... 17 Figure 2.1 The expression patterns of KNOX2 genes in Arabidopsis stems. ......................... 40 Figure 2.2 Xylem vessel morphology of wild-type (WT) and KNOX2 mutant plants. .......... 41 Figure 2.3 Cross-sections of stem vascular bundles in wild-type (WT) and different combinations of knox2 mutants. ..................................................................................... 43 Figure 2.4 Plant morphology and stem biomechanical properties of WT, knat3, knat7 and knat3knat7. ..................................................................................................................... 44 Figure 2.5 Plant morphology of WT, knat3knat7, knat4knat7, knat5knat7. .......................... 46 Figure 2.6 Interfascicular fibers of WT, knat3, knat7 and knat3knat7 inflorescence stems. . 47 Figure 2.7 Lignin composition of WT, knat3, knat7 and knat3knat7 stem cell walls. ........... 50 Figure 2.8 Complementation of knat3knat7 phenotypes with ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP. ................................................................................................... 51 Figure 2.9 Expression patterns of KNAT3 and KNAT7 in the root and stem. ........................ 54 Figure 2.10 Interfascicular fibers have thicker secondary cell walls in 35S:KNAT3 basal stems. .............................................................................................................................. 55 Figure 2.11 Expression of secondary cell wall-related genes in lower stems of WT, knat3, knat7 and knat3knat7 double mutants. ........................................................................... 56 Figure 2.12 Venn diagram and GO term enrichment analysis of differentially expressed genes (DEGs) of knat3 versus WT, knat7 versus WT, and knat3knat7 versus WT. ...... 58 xiii   Figure 2.13 MAPMAN schematic providing a metabolic overview of the differential gene expression between WT and knat3knat7 stems. ............................................................. 60 Figure 2.14 MAPMAN schematic providing a metabolic overview of the differential gene expression levels in knat3 and knat7 stems compared with WT. ................................... 61 Figure 3.1 Cross-sections of stem vascular bundles in wild type (WT) and ofp single mutants. ........................................................................................................................................ 85 Figure 3.2 Phenotypes of wild type (WT) and OFP overexpression (OX) seedlings. ........... 86 Figure 3.3 The expression patterns of OFP2, OFP3 and OFP5 in Arabidopsis seedlings. ... 87 Figure 3.4 Subcellular localization of OFP-YFP fusion proteins in epidermal cells of N. benthamiana leaves. ....................................................................................................... 88 Figure 3.5 Hypocotyl phenotypes of wild type (WT) and OFP overexpression (OX) plants. 90 Figure 3.6 BR-deficient related morphological phenotypes of OFP overexpression (OX) mutants. ........................................................................................................................... 91 Figure 3.7 Quantification of WT and OFP OX hypocotyl length under exogenous BR treatment. ........................................................................................................................ 93 Figure 3.8 OFP4 interacts with Nucleosome Assembly Protein 1;1 (NAP1;1) and NAP1;2 in vitro and in vivo. ............................................................................................................. 96 Figure 3.9 NAP1;1 and NAP1;2 are localized to the ER in epidermal cells of N. benthamiana leaves. ............................................................................................................................. 97 Figure 3.10 Cross-sections of stem vascular bundles in WT, nap1;1, nap1;2 and nap1;1 nap1;2 nap1;3 triple mutants. ........................................................................................ 98 Figure 3.11 Cotyledon phenotypes of NAP1 and OFP loss-of-function mutants .................. 99 xiv   Figure 3.12 Cotyledon vasculature development in NAP1 and OFP loss-of-function mutants. ...................................................................................................................................... 100 Figure 3.13 Phenotypes of 2 x 35S: GFP-OFP4 transgenic seedlings in WT and nap1;1 nap1;2 nap1;3 background. .......................................................................................... 101 Figure 3.14 GFP-OFP4 expression in nap1;1 nap1;2 nap1;3 triple mutants. ..................... 102   xv   List of Abbreviations 3-AT 3-Amino-1,2,4-triazole  4CL 4-Coumarate: Coenzyme A Ligase  ABA abscisic acid  BELL BELL-like  BES1  BRI1-EMS-SUPPRESSOR 1 BET9 Bromodomain and Extraterminal Domain Protein 9  bHLHZip helix-loop-helix leucine zipper  BiFC Bimolecular Fluorescence Complementation BIN2 BRASSINOSTEROID-INSENSITIVE 2  BLH6 BELL-LIKE HOMEODOMAIN6  BP BREVIPEDICELLUS BR brassinosteroid C catechyl C3H1 COUMARATE 3-HYDROXYLASE 1  C4H CINNAMATE-4-HYDROXYLASE  CAD5 CINNAMYL ALCOHOL REDUCTASE 5  CCoAOMT1 CAEEROYL-COA 3-O-METHYLTRANSFERASE 1  CCR Cinnamoyl-CoA Reductase  CCR1 CINNAMOYL-COA REDUCTASE 1  CESA cellulose synthase  ChIP-seq chromatin immunoprecipitation sequencing  CMU CELLULOSE SYNTHASE-MICROTUBULE UNCOUPLING  COMT Caffeic acid 3-O-methyltransferase  CSC cellulose synthase complex CSE Caffeol Shikimate Esterase  CSI1 CELLULOSE SYNTHASE INTERACTING1 DEGs differentially expressed genes xvi   DMCBH Djavad Mowafaghian Centre for Brain Health EMS ethylmethane sulphonate  EMSA electrophoretic mobility shift assays  epiBL epibrassinolide  ER endoplasmic reticulum F5H Ferulate-5-hydroxylase  FPKM fragments per kilobase of exon per million fragments mapped G guaiacyl GFP green fluorescent protein GO gene ontology GUS β-GLUCURONIDASE  H p-hydroxyphenyl  HCT hydroxycinnamoyl CoA: shikimate/quinate hydroxycinnamoyltransferase  HD homeodomain HD-ZIP III Class III homeodomain leucin zipper  His Histidine irx irregular xylem  KNAT7 KNOTTED ARABIDOPSIS THALIANA7  KNOX KNOTTED-like homeobox  KNOX2 Class II Knotted-like homeobox  Leu Leucine MP MONOPTEROS MS Murashige and Skoog mSILIP metabolic stable isotope labelling immuno-precipitation mass spectrometry  NAP1 Nucleosome Assembly Protein 1  NAP1;1 Nucleosome Assembly Protein 1;1  NAP1;2 Nucleosome Assembly Protein 1;2  xvii   NHEJ non-homologous end-joining  NST1 NAC SECONDARY WALL THICKENING PROMOTING FACTOR1  OFP Ovate Family Protein  OX overexpression PAL1 PHENYLALANINE AMMONIA LYASE 1  PCD programmed cell death  PHB PHABULOSA PHV PHAVOLUTA PME35 PECTIN METHYLESTERASE35  PXY PHLOEM INTERCALATED WITH XYLEM REV REVOLUTA RFP red fluorescent protein S syringyl SND1 SECONDARY WALL-ASSOCIATED NAC DOMAIN PROTEIN1 SPR2 Spiral 2 STM SHOOTMERISTEMLESS TALE Three Amino acid Loop Extension TRM TONNEAU1 Recruiting Motif TRM20 TON1 Recruiting Motif 20  Trp Trptophan VND6 VASCULAR-RELATED NAC-DOMAIN6  VNI2 VND-INTERACTING2  WT wild-type  X-Gluc 5-bromo-4-chloro-3-indolyl-β-D-glucuronide  XCP1 XYLEM CYSTEINE PEPTIDASE 1 XND1 XYLEM NAC DOMAIN1  Y1H yeast one-hybrid  YFP yellow fluorescent protein xviii   Acknowledgements Firstly, I would like to thank Dr. Lacey Samuels and Dr. Shawn Mansfield, my current supervisors. Without their continuous support, encouragement, and advice, I do not think I can finish my thesis. I am truly grateful for having them as my supervisors. Also, I want to express my sincere appreciation to Dr. Carl Douglas, my first supervisor at UBC, for his understanding, encouragement, inspiration, trust, support and care. My gratitude to them is beyond words. I am very fortunate to become their student.    To my committee members, Dr. Xin Li and Dr. Jae-Hyeok Lee, thank you so much for being available, helpful, and supportive all the time. Thank you for being always patient and giving me constructive feedbacks about my projects.   I am also grateful to Dr. Zhiyong Wang, a staff scientist in Carnegie Institution for Science, who hosted me to do my exchanges in his lab. I learned so much there and thank you for taking some time from your busy schedule to teach me some techniques in the lab. Besides, I would like to thank Wang lab’s people, Dr. Chanho Park, Dr. Veder Garcia, Dr. Thomas Hartwig and Bi Yang, for all the help, support, care and fun bringing to me.   Working in Douglas lab was very enjoyable. I would like to thank all the people who were working with me at the same time. I would like to give my special thanks to Dr. Etienne Grienenberger for his mentorship. I learned most of my experimental skills from him. Also, I would like to thank Dr. Masatoshi Yamaguchi, for giving me suggestions and help with my KNOX project. I would like to thank all the lab members in Dr. Douglas, Dr. Samuels, and Dr. Mansfiled labs, for being always so kind, helpful and supportive. Lan Tran, Yoichiro (Yoshi) Watanabe, Miranda Meents and Dr. Mathias Schuetz were always helpful with my questions and giving me a lot of support. Dr. Faride Unda, Dr. Eliana Gonzales, Francis de Araujo, Foster Hart, Grant McNair helped me a lot with my chemical analysis. I also would like to thank all people in Botany Department, for being always so nice and willing to help. xix   Kyra Janot from Dr. Patrick Martone’s lab helped me with the biomechanical assays. Dr. Chris Ambrose was trying to help me with my OFP project and Tongjun Sun taught me how to do the transcriptional activity assays using protoplasts. Sean Shang helped me with my RNA-seq analysis.   Funding for my projects was supported by China Scholarship Councils, NSERC (Natural Sciences and Engineering Research Council of Canada) CREATE Grant “Working on Walls”, and Dr. Carl Douglas’ NSERC Discovery Grant. The Support from of the University of British Columbia Faculty of Science following Dr. Carl Douglas’ death is gratefully acknowledged.  Thank you to my parents, Canming Wang and Guihua Tian, my sister Shuyun Wang and my best friend, Aiping Liu, for love, patience, and support.   xx   Dedication       In Memory of Dr. Carl Douglas  1   Chapter 1: Introduction Plant cell walls provide structural support for the “stately trees and other greenery that grace our planet” (Cosgrove, 2005). Primary and secondary cell walls are the two major structures surrounding almost all plant cells. The newly formed wall is first defined by the primary cell wall, which is thin, expandable, and has a relatively constant proportion of the three main cell wall polysaccharide polymers: cellulose, hemicelluloses, and pectin, as well as a minor fraction of protein (Albersheim et al., 2010). In specialized cells, when the primary wall has finished expanding, a secondary cell wall is deposited, which is a thicker layer composed predominantly of cellulose, hemicelluloses and lignin (Albersheim et al., 2010).  Secondary cell walls are the main constituents of plant fibers and wood, which are widely used for construction, paper, furniture, industrial pulp, energy and many other familiar products, so understanding the secondary cell wall development has significant meaning in biology and for the economy. Arabidopsis thaliana has proven to be a favorable model for secondary cell wall formation studies as the secondary cell walls are deposited abundantly in the xylem and interfascicular fiber cells in the inflorescence stems (Ehlting et al., 2005; Zhong et al., 2007).   In order to understand what triggers and controls the secondary cell wall deposition, Arabidopsis transcription factors have been intensively studied (Taylor-Teeples et al., 2014). One of the Arabidopsis Class II Knotted-like homeobox (KNOX2) genes, KNOTTED ARABIDOPSIS THALIANA7 (KNAT7), was reported to act as a repressor for secondary cell wall formation in interfascicular fibers, with thicker fiber walls observed in KNAT7 loss-of-function mutants (Li et al., 2012). Members of Ovate Family Proteins (OFPs), OFP1 and OFP4, were shown to interact with KNAT7 and negatively regulate secondary wall formation (Li et al., 2011; Liu and Douglas, 2015). Because of gene duplication and diversification, there are other KNOX2 and OFP genes that are closely related to KNAT7 and OFP4, respectively, and are in the same clade based on the reported phylogenetic 2   resconstructions (Figure 1.1) (Furumizu et al., 2015; Wang et al., 2016). However, we know little about their roles in secondary cell wall development. For my Ph.D. research, I used a combined molecular and genetic approach to characterize and elucidate the roles of other KNOX2 and OFPs in Arabidopsis secondary cell wall development.   1.1    Secondary cell wall component biosynthesis  Secondary cell walls are strong multilayered structures whose presence provides rigidity and strength to plant organs, and facilitates water transport in specialized cells of the xylem. The major components of secondary cell wall are cellulose, hemicellulose and lignin.   Cellulose consists of a single polymer of β-1, 4-linked glucan chains, and it makes up the main component of secondary cell walls. The hydrogen bonding and van der Waals interactions among the cellulose chains cause it to form a stable crystalline structure called cellulose microfibrils (Kim et al., 2013). Cellulose is synthesized in the plasma membrane by the cellulose synthase complex (CSC), which includes cellulose synthase (CESA) enzymes and accessory proteins (reviewed in McFarlane et al., 2014). In Arabidopsis, CESA4, CESA7, and CESA8, which correspond to the irregular xylem (irx) mutants irx5, irx3 and irx1, are required for cellulose synthesis in secondary cell wall development (Turner and Somerville, 1997; Taylor et al., 1999; 2000; 2003). The CSC has been found to have a rosette structure at the plasma membrane that forms a cellulose microfibril containing an estimated 18-24 glucan chains (Mueller and Brown, 1980; Oehme et al., 2015; Nixon et al., 2016). Cross-talk between CSCs movement and the cortical microtubules guides cellulose patterning, as CELLULOSE SYNTHASE INTERACTING1 (CSI1)/POM2 interacts with CESAs and guides CSCs moving along cortical microtubules in the early secondary cell wall developmental stage (Schneider et al., 2017). The stability of the cortical microtubule array depends on microtubule-associated proteins such as CELLULOSE SYNTHASE-MICROTUBULE UNCOUPLING (CMU) (Liu et al., 2016). In the loss-of-function cmu mutants, CSCs movement along the plasma membrane may lead to microtubule displacement (Liu et al., 2016).  3    Hemicelluloses are Golgi-synthesized polysaccharides that can bind to cellulose microfibrils by noncovalent bonds in the cell wall. Xyloglucan is the quantitatively most abundant hemicellulose in the primary cell walls of dicotyledons, while xylans and mannans are the dominant hemicelluloses during differentiation to secondary cell walls (reviewed in Scheller and Ulvskov, 2010). Xylans are composed of a backbone of β-(1,4)-linked xylose mainly with substitution of glucuronic acid residues, known as glucuronoxylans in eudicots, while monocot xylans are structurally varied with glucuronoxylan in non-grass and glucuronoarabinoxylan in grass walls (Peña et al., 2016). Mannans are the predominant hemicellulose of conifer secondary cell walls, and have β-(1,4)-linked backbones with all mannose, or with both mannose and glucose backbones (reviewed in Scheller and Ulvskov, 2010; Rodriguez-Gacio et al., 2012). Different levels of xylan and mannan acetylation and/or methylation exist in the secondary cell wall and affect its integrity (Pawar et al., 2013; Rennie and Scheller, 2014; Hao and Mohnen, 2014).  While mannan and xylan are both present in eudicot secondary cell walls, xylan is the most abundant (reviewed in Scheller and Ulvskov, 2010). Xylan biosynthesis requires a large number of enzymes such as glycosyltransferases, methyltransferases, acetyltransferases, and glycosyl hydrolases (reviewed in Rennie and Scheller, 2014), and the genes involved in xylan biosynthesis have been identified by numerous mutant studies (reviewed in Hao and Mohnen, 2014). For example, IRX9, IRX10 and IRX14 function redundantly with their homologs, IRX9-L, IRX10-L and IRX14-L, playing significant roles in xylan backbone synthesis (Brown et al., 2009; Lee et al., 2010; Wu et al., 2010). Hemicellulose biosynthetic complexes in the Golgi have been proposed for producing hemicelluloses (reviewed in Meents et al., 2018). IRX9, IRX10 and IRX14 in Asparagus were found to form a xylan biosynthesis complex in the Golgi by coimmunoprecipitation and bimolecular fluorescence complementation studies and the transient expression assay in tobacco leaves revealed that three proteins have to be co-expressed together to correctly locate at the Golgi or the ER localization would be observed (Zeng et al., 2016).  4    Lignin is a polymer of cross-linked phenolic monolignols that impregnates the polysaccharide matrix to strengthen secondary cell walls. There are mainly three monolignols, p-coumary alcohol, coniferyl alcohol and sinapyl alcohol, which produce p-hydroxyphenyl (H), guaiacyl (G) and syringyl (S) units in the lignin, respectively. Lignin in gymnosperms is highly enriched with G units, but angiosperm lignin is made up of both G and S units, with abundant G units in vessel secondary cell walls and dominant S units in the cell walls of fibers (Donaldson, 2001). The monolignols are derived from phenylalanine, which enters the phenylpropanoid and monolignol biosynthetic pathways in the cytosol through a series of hydroxylation and methylation reactions of the phenyl ring and reduction of the terminal carbon of the propane side chain (Albersheim et al., 2010). Mutant studies in Arabidopsis, alfalfa, and poplar as well as in vitro enzyme characterizations have contributed to our understanding of monolignol biosynthesis (reviewed in Bonawitz and Chapple, 2010; Dixon et al., 2014; Hao and Mohnen, 2014). For example, in the Arabidopsis 4-Coumarate: Coenzyme A Ligase (4CL) suppression lines, the total lignin content and G/S lignin ratio was significantly decreased (Lee et al., 1998). The Cinnamoyl-CoA Reductase (CCR) loss-of-function mutants has less than 50% of total lignin content compared with wild type, and display a stem phenotype with collapsed xylem vessels in Arabidopsis (Jones et al., 2001). Downregulation of Caffeic acid 3-O-methyltransferase (COMT) in alfalfa and poplar plants result in significant reductions in both S lignin and total lignin contents (Guo et al., 2001; Jouanin et al., 2000).   Although the monolignol biosynthetic pathway has been well characterized, the revision of the pathway is still ongoing (reviewed in Zhao, 2016). Bouchaud et al. (2013) found the enzyme hydroxycinnamoyl CoA: shikimate/quinate hydroxycinnamoyltransferase (HCT) alone is not enough to produce metabolite caffeoyl shikimate, as the caffeate ester is highly accumulated in the Caffeol Shikimate Esterase (CSE) loss-of-function mutants. The cse mutant has a decreased lignin content, but possesses a high enrichment in H lignin, which suggest that the CSE enzyme is also involved in the lignin biosynthesis, providing a new 5   insight to the original model (Bouchaud et al., 2013). After monolignols are produced by the cells undergoing lignification or possibly by other neighboring cells, they are exported to the cell wall. Secreted oxidative enzymes such as peroxidases and laccases catalyze monolignol oxidation, leading to randomly cross-coupling to generate the growing lignin polymer (reviewed in Zhao, 2016).   The lignin polymer is quite plastic, as it can adjust the composition of traditional three monomers and incorporate a variety of different unique units without affecting plant normal growth (reviewed in Mottiar et al., 2016; Zhao, 2016). For instance, the Ferulate-5-hydroxylase (F5H) loss-of-function mutants have lost almost all the S lignin, but the total lignin content is not changed (Marita et al., 1999). Overexpressing F5H in Arabidopsis comt mutants resulted in an enrichment of 5-hydroxyl-G units (Weng et al., 2010). The catechyl (C) monomer was detected in the seed coats of vanilla orchid, which coexisted with G and S lignin in seed coats (Chen et al., 2012; Tobimatsu et al., 2013).  1.2    Xylem development and interfascicular fiber differentiation Secondary cell walls are mainly deposited at the tracheary elements and fibers, where the vascular system is. Plant vascular systems include two main tissue types, phloem and xylem, which differentiate from two different meristematic tissues, the procambium and vascular cambium (Eames and MacDaniels, 1947). In herbaceous stems and the young stems of woody plants, the procambial initials form by the action of the apical meristems to produce primary xylem and primary phloem. When plants undergo secondary growth, vascular cambium initials, which originate from procambium and other parenchyma cells, give rise to secondary xylem known as wood and secondary phloem (Evert, 2006). Phloem tissues include fibers, parenchyma, sieve elements, and companion cells, and function to conduct organic compounds in a watery solution, and provide support. Xylem tissues from angiosperms include two tracheary element cell types, tracheids and vessel elements, which are able to facilitate water transport, as well as xylary fibers that function in physical support of the plant body. 6    Most xylem cells (e.g. tracheary elements) are dead at maturity, have secondary cell wall thickenings, and undergo a series of common developmental steps during their development, which is also defined as xylogenesis (reviewed in Fukuda, 1996). The first step in tracheary element or fiber differentiation is cell expansion, followed by secondary cell wall polysaccharide deposition and lignification (Samuels et al., 2006). For vessels of angiosperms, expansion is strongly in the radial direction, while for angiosperm fibers, expansion is axial as the cells elongate by intrusive growth (reviewed by Mellerowicz et al., 2001). The final stage of development for tracheary element cells is programmed cell death (PCD), which results in the cell becoming an empty tube to support water transport, such as the autolytic process in xylem fibers of the Populus stem (Courtois-Moreau et al., 2009).  In many woody and herbaceous plants, the interfascicular cambium differentiates relatively early, and it may develop from cells of ground meristem. If it differentiates later, interfascicular development may result from dedifferentiation of mature interfascicular parenchyma. The cambial activity starts initially in the vascular bundles. Upon the establishment of continuity of the cambium, its mitotic activity results in continuous production of secondary xylem and secondary phloem.  1.3    Molecular mechanisms of vascular development  While these patterns of vascular differentiation have long been observed, we are just beginning to discover the molecular mechanisms underlying their development. Auxin has been well studied to be an important signal for vascular procambium development in embryos, the primary roots and leaf vein development (reviewed in Greco et al., 2012; Ursache et al., 2013). In Arabidopsis inflorescence stems, Mazur et al. (2014) observed a high accumulation of auxin in vascular bundles, as well as in the interfascicular parenchyma cells differentiating into interfascicular cambium, which contributes to the cambial ring formation within the stems. PIN1 gene expression was always preceded by the high auxin 7   accumulation in differentiating cells, and polar PIN protein was observed in the basal plasma membrane of parenchyma cells and periclinal divisions (Mazur et al., 2014). One of the Class III homeodomain leucin zipper (HD-ZIP III) genes, AtHB8, coordinates the procambium formation within and between leaf veins, which is induced by the auxin-response transcription factor MONOPTEROS (MP) directly (Donner et al., 2009). In the auxin biosynthesis mutants, the expression of most of the HD-ZIP III genes, i.e., PHABULOSA (PHB), PHAVOLUTA (PHV), REVOLUTA (REV), AtHB15 and AtHB8, was downregulated and metaxylem patterning was affected (Ursache et al., 2014).  By analogy with other meristems, the balance between cell proliferation and differentiation is important for vascular development and controlled by secreted ligands, receptors, and HOMEOBOX transcription factors. Peptide ligands have been identified that lead to signals controlling xylem differentiation, regulating vascular cell division rate and determining the orientation of cell division (Zhang et al., 2011). By using the xylogenic culture system, CLE41, CLE42 and CLE44, encoding the dodecapeptide tracheary element differentiation inhibitory factors (TDIF), were originally identified and demonstrated to play a unique role in xylem differentiation (Ito et al., 2006). According to expression analysis, CLE41/44 peptide is mainly synthesized in the phloem and its neighboring cells, and then secreted from the phloem into the cambial tissue to suppress xylem differentiation by binding to the TDIF-receptor, which is the leucine-rich repeat receptor-like kinase PHLOEM INTERCALATED WITH XYLEM (PXY) (Fisher and Turner, 2007; Hirakawa et al., 2008). A WUSCHEL-related HOMEOBOX gene, WOX4, is activated by the TDIF/TDR signaling pathways, and its expression was highly elevated by TDIF application (Hirakawa et al., 2010). WOX14 functions redundantly with WOX4 in promoting vascular cell division, as wox4 wox14 double mutants presented fewer cells in vascular bundles than wild type and single mutants (Etchells et al., 2013). Disruption of CLE41 expression results in disordered vascular development, while increasing phloem-specific expression of CLE41 leads to well-ordered vascular tissue development, which reveals the function of CLE peptide in regulating the vascular tissue organization (Etchells and Turner, 2010).  8    BRASSINOSTEROID-INSENSITIVE 2 (BIN2) has been shown to be involved in the mechanism of mediating TDIF and TDIF RECEPTOR in the aspect of repressing xylem differentiation (Kondo et al., 2014). BIN2, encoding a GSK3/SHAGGY-like kinase, was identified as a negative regulator to destabilize BES1 (BRI1-EMS-SUPPRESSOR 1) in controlling the steroid signaling pathway in plants (Li and Nam, 2002; Yin et al., 2002). Kondo et al. (2014) found TDIF/TDR is directly associated with BIN2 at the plasma membrane in procambial cells, leading to the suppression of the transcription factor BES1, and subsequently inhibition of xylem differentiation. In the loss-of-function BES1 mutant, the ectopic xylem cell differentiation in cotyledons induced by bikinin was barely visible, indicating BES1 is the main downstream transcription factor of TDIF-TDR-GSK3 (Kondo et al., 2014; Kondo et al., 2015).  1.4    Secondary cell wall transcriptional regulation network  Secondary cell wall deposition must be tightly coordinated with differentiation of tracheary elements and fibers. Therefore, regulatory mechanisms exist to control the different biosynthetic pathways of secondary wall components in different cell types (Zhong and Ye, 2007). The formation of secondary cell wall requires a very complex network of transcriptional regulation that includes at least three layers of regulators (Figure 1.2) (reviewed in Yang and Wang, 2016).  A set of NAC domain transcription factors has been identified as key transcriptional activators in regulating the biosynthesis of the three main secondary cell wall components (cellulose, hemicellulose and lignin) in Arabidopsis. These positive regulators include VASCULAR-RELATED NAC-DOMAIN6 (VND6) and VND7 which are responsible for secondary cell wall deposition in xylem vessel elements and the NAC SECONDARY WALL THICKENING PROMOTING FACTOR1 (NST1), NST2, NST3/SECONDARY WALL-ASSOCIATED NAC DOMAIN PROTEIN1 (SND1), that are associated with secondary cell wall formation in fibers (Kubo et al., 2005; Ko et al., 2007; Zhong et al., 2006; Ruiqin 9   Zhong et al., 2007; Mitsuda et al., 2007; Zhong and Ye, 2015). In Arabidopsis, VND6 and VND7 have been reported to induce the differentiation of metaxylem- and protoxylem- like vessels respectively, and a dominant repression of VND6 and VND7 specifically inhibits metaxylem and proxylem formation in roots (Kubo et al., 2005; M. Yamaguchi et al., 2010). Global transcriptome analysis reveals that VND7 can regulate genes involved in a broad range of processes in xylem vessel differentiation directly, such as XYLEM CYSTEINE PEPTIDASE 1 (XCP1), an enzyme involved in autolytic processing following cell death (Funk et al., 2002; Yamaguchi et al., 2011). The other five additional Arabidopsis VND genes, VND1 to VND5, were also found to be able to regulate secondary cell wall biosynthesis in vessels (Zhou et al., 2014). In Arabidopsis stems, loss-of-function both SND1 and NST1 results in the reduction of secondary wall formation in fibers, and overexpression of SND1 or NST1 induces ectopic secondary wall thickenings in various tissues (Zhong et al., 2006; Ruiqin Zhong et al., 2007; Mitsuda et al., 2007).  Zhong and Ye (2015) found that NST2 functions together with SND1 and NST1 regulating secondary wall formation in fibers of stems, as triple mutants snd1 nst1 nst2 caused a complete loss of secondary wall thickening in fibers.   The secondary wall NAC transcription factors directly activate the expression of MYB46 and MYB83, central regulators of secondary wall formation in both xylem vessels and fibers (Zhong et al., 2007; McCarthy et al., 2009; Zhong and Ye, 2012). It was reported that SND1 binds to MYB46 promoter and activates its transcription (Zhong et al., 2007). Dominant repression of MYB46 results in a dramatic reduction in secondary cell wall thickening of both fibers and vessels (Zhong et al., 2007). MYB83 functions redundantly with MYB46, promoting secondary cell wall biosynthesis in Arabidopsis (McCarthy et al., 2009). Simultaneous downregulation of MYB46 and MYB83 leads to a more severe phenotype than nst1 nst3 double mutants, with reduced growth, severe deformation of vessels and reduction in fiber secondary wall thickening (McCarthy et al., 2009; Mitsuda et al., 2007).  10   There are a lot of transcription factors functioning downstream of the NAC and MYB domain master switches (Zhong et al., 2008a; Ko et al., 2009; Zhong and Ye, 2012). SND2, SND3, MYB103, MYB85, MYB52 and MYN54 are positively regulated by SND1, and dominant repression of these genes leads to thinner secondary cell walls thickening in fibers (Zhong et al., 2008a). Overexpressing SND2, SND3 and MYB103 upregulates the expression of cellulose synthase genes and results in thicker fiber secondary cell walls, while overexpression of MYB85 results in ectopic lignin deposition in epidermal and cortical cells in stems by specifically increasing 4CL1 expression (Zhong et al., 2008). The promoters of most monolignol pathway genes contain AC-rich elements (Douglas, 1996; Zhao and Dixon, 2011). These AC-rich elements are binding sites for the transcription factors MYB58 and MYB63, which are specific activators of lignin during secondary wall formation (Zhou et al., 2009). However, the lignin biosynthetic gene encoding F5H is an exception to this rule, as it lacks AC-rich elements and is not regulated by MYB58 and MYB63 (Zhou et al., 2009). However, the fiber-specific transcription factor SND1/NST1 regulates F5H expression (Zhao et al., 2010). Ohman et al. (2013) found that MYB103 is also required for F5H expression and S lignin biosynthesis, as the myb103 mutant has very low transcript abundance of F5H gene and significant reduction in S lignin deposition in Arabidopsis.  Some transcription factors are characterized as repressors, fine-tuning the whole secondary cell wall regulatory network. KNAT7 was identified as a negative regulator of secondary cell wall formation, with thicker secondary cell walls in interfascicular fibers in the loss-of function KNAT7 mutant (Li et al., 2012). MYB75 and BLH6 interact with KNAT7, acting as transcriptional repressors to contribute to the regulation of secondary wall formation in interfascicular fibers, and the interaction between BLH6 and KNAT7 enhances the repression activity of BLH6 or KNAT7 alone (Bhargava et al., 2010; Y., Liu et al., 2014). The promoter activity of REV, one of the HD-ZIP III genes, is directly repressed by KNAT7 and BLH6, modulating interfascicular fiber secondary cell wall formation (Liu et al., 2014). A large scale of yeast one-hybrid assay revealed that REV negatively regulates the lignin biosynthetic gene PAL4 by binding to its promoter directly, and the PAL4 expression is 11   significantly increased in the rev-5 loss-of-function mutants (Taylor-Teeples et al., 2015). So KNAT7 can be a potential activator by repressing a repressor, such as REV. KNAT7 is a direct target of several NAC master switches and MYB 46 ( Zhong et al., 2008; Ko et al., 2009). XYLEM NAC DOMAIN1 (XND1), a member of the NAC domain family, is highly expressed in the xylem and inhibits the secondary cell wall deposition and programmed cell death in xylem cells (Zhao et al., 2008). MYB4, MYB7 and MYB32 are activated by MYB46, functioning as potential repressors to downregulate SND1 expression (Jin, 2000; Ko et al., 2009; H., Wang et al., 2011). VND-INTERACTING2 (VNI2) directly binds to VND7 and repress the expression of vessel-specific genes (Yamaguchi et al., 2010). Overexpression of VNI2 results in an inhibition of the normal xylem vessel development in roots and aerial organs (Yamaguchi et al., 2010). A WRKY transcription factor, WRKY12, has been reported to be a negative regulator, regulating secondary cell wall synthesis in pith cells by suppressing the expression of NST2 and C3H zinc finger transcription factors (H., Wang et al., 2010). The negative regulators are important in maintaining the metabolic homeostasis with respect to metabolic commitment to secondary cell wall formation especially under undesirable growth conditions (Jin, 2000; Li et al., 2012). A series of feed-forward loops are also involved in the secondary cell wall regulatory network to ensure regulation of the whole process under abiotic stress (Taylor-Teeples et al., 2015). By testing protein-DNA interactions, E2Fc was identified as a key upstream repressor as well as activator to regulate VND7 expression, and it can bound to numerous gene promoters including those of VND6, VND7, MYB46 and cellulose-, hemicellulose- and lignin-associated genes (Taylor-Teeples et al., 2015).   1.5 KNOTTED-like homeobox (KNOX) family proteins As KNAT7 has a demonstrated role in negatively regulating interfascicular fiber secondary cell wall thickness, it is interesting to consider other potential roles in secondary cell wall biosynthesis for other KNOTTED-like homeobox transcription factors. The Homeobox class of proteins is defined by the homeodomain (HD), a 60 amino acid long DNA-binding domain.  In the TALE (Three Amino acid Loop Extension) homeobox genes, three additional 12   residues are observed between helix 1 and 2 of the HD, and the TALE homeobox genes can be found in all eukaryotic lineages (Bürglin, 1997; Derelle et al., 2007). Based on the sequence similarity, the TALE homeobox genes are classified into two different subfamilies, KNOTTED-like homeobox (KNOX) and BELL-like (BELL) (Mukherjee et al., 2009). Based on gene duplication and diversification events in a common ancestor of land plants, KNOX genes are grouped into two classes, class I (KNOX1) and class II (KNOX2; Kerstetter et al., 1994; Mukherjee et al., 2009) (Figure 1.1a).	A new class of KNOX genes, KNATM, which does not have a homeodomain, has been shown to regulate the leaf proximal-distal patterning and has only been identified in dicot species (Kimura et al., 2008; Magnani and Hake, 2008).   KNOTTED1, a maize KNOX1 gene, was the first plant homeobox gene identified, and gain-of-function KNOTTED1 plants have altered leaf phenotypes (Vollbrecht et al., 1991). In Arabidopsis, KNOX1 genes include SHOOTMERISTEMLESS (STM), BREVIPEDICELLUS (BP)/KNAT1, KNAT2 and KNAT6, which are involved in cell proliferation (Sakakibara et al., 2008) and required for development of shoot apical meristem (SAM; Long et al., 1996; Vollbrecht et al., 2000; Groover et al., 2006; Tsuda et al., 2011). The expression pattern of STM in Arabidopsis is very similar to that of KNOTTED1 in maize, with the presence in all four types of SAMs: vegetative, axillary, inflorescence and floral (Long et al., 1996). The stm mutants induced by ethylmethane sulphonate (EMS) fail to develop a SAM during embryogenesis (Long et al., 1996). The bp mutants exhibit reduced internode and pedicel length, downward-pointing flowers and a compact inflorescence architecture, which indicates that BP acts in the differentiation of the inflorescence stem, pedicel, and style in Arabidopsis (Venglat et al., 2002; Douglas et al., 2002). KNAT2 and KNAT6 are required for floral organ abscission downstream of BP/KNAT1, as loss-of-function KNAT2 and KNAT6 rescues the bp floral abscission phenotype (Shi et al., 2011).   Northern blot analysis in maize showed that KNOX1 genes are highly expressed in meristem-enriched tissues, while KNOX2 genes are broadly expressed in all tissues except the 13   meristematic regions (Kerstetter et al., 1994). In Arabidopsis, there are four KNOX2 genes, KNAT3, KNAT4, KNAT5 and KNAT7, which have very similar broad expression profiles as maize KNOX2 genes (Serikawa, 1997; Truernit et al., 2006; Li et al., 2012; Furumizu et al., 2015). As mentioned previously, KNAT7 expression is highly associated with the secondary cell wall formation and knat7 mutants exhibit an irregular xylem vessel phenotype and increased secondary cell wall thickness in interfascicular fibers (Li et al., 2012). Potential roles for KNAT3, KNAT4, and/or KNAT5 in secondary cell wall synthesis have not been explored.  Mutant analyses in Arabidopsis have demonstrated some functions of these additional members of the KNOX2 clade (Furumizu et al., 2015). KNAT3 is reported to be not only involved in the embryo sac development, but also in ABA-mediated seed dormancy and early seedling development by interacting with BLH1 transcription factor (Pagnussat et al., 2007; D., Kim et al., 2013). KNAT3, KNAT4 and KNAT5 act redundantly in regulating plant development. The knat3 knat4 and knat3 knat5 double mutants have more deeply serrated leaves compared with knat4 knat5 (Furumizu et al., 2015). The knat3 knat4 and knat3 knat4/+ knat5 are female sterile with abnormal integument development while knat3/+ knat4 knat5 plants are phenotypically wild-type with normal amount of seeds (Furumizu et al., 2015). Interestingly, the loss-of-function KNOX2 phenotypes resemble the gain-of-function KNOX1 phenotypes, and visa versa (Hake et al., 2004; Hay and Tsiantis, 2010). Furumizu et al. (2015) suggested that the way that KNOX2 can suppress meristem activity, while KNOX1 activates meristem activity is by acting oppositely on common downstream elements.   1.6    Ovate Family Proteins (OFPs)   KNAT7 was found to interact with both Ovate Family Protein 1 (OFP1) and OFP4 using yeast two-hybrid assays and bimolecular fluorescence complementation tests (Li et al., 2011). The loss-of-function mutants are morphologically identical to Arabidopsis wild-type plants, but cross-sections taken from ofp4 stems revealed an irregular xylem (irx) phenotype 14   similar to that found in knat7 (Li et al., 2011). OFP1 and OFP4 function also depends on KNAT7 at least partially, since their pleiotropic over-expression phenotypes are repressed in a knat7 loss-of-function mutant (Li et al., 2011). As such the researchers proposed that a functional complex of KNAT7-OFP negatively regulates secondary cell wall formation.  Both OFP1 and OFP4 belong to the plant-specific OVATE family, which was named after the first OVATE gene cloned in the tomato (Liu et al., 2002). A premature stop codon of the OVATE gene in tomato caused a pear-shaped fruit phenotype (Liu et al., 2002). It revealed a new class of proteins found throughout land plants, that have a conserved 70-aa C-terminal domain, which was defined as an OVATE domain (Hackbusch et al., 2005; Wang et al., 2011; Liu et al., 2014). Through a large-scale yeast two-hybrid screen, nine members of OFPs were also found to interact with Arabidopsis TALE homeodomain proteins, and the subcellular localization of TALE proteins can be regulated by OFPs (Hackbusch et al., 2005). Phylogenetic analysis revealed 19 OFPs in Arabidopsis with three major clades and 11 sub-groups (Liu et al., 2014).  So far, there is little information regarding OFP function in Arabidopsis. In addition to participating in secondary cell wall development, OFP1 was reported to negatively regulate GA20ox1, inhibiting cell elongation through Gibberellic acid biosynthetic pathway in Arabidopsis (Wang et al., 2007). By interacting with the protein Ku70, OFP1 is involved in DNA repair through the non-homologous end-joining (NHEJ) pathway (Wang et al., 2010). Also, OFP1 interacts with BLH3 to regulate flowering time (Zhang et al., 2016). These data suggest that OFP1 has different functions when interacting with different partners. Alternately, OFP5 was shown to regulate Arabidopsis embryo sac development by interacting with BLH1 and KNAT3 (Pagnussat et al., 2007). According to OFP overexpression phenotypes, OFP genes in Arabidopsis were divided into three classes, and OFP1, OFP2, OFP4, OFP5 and OFP7 belong to class I, as their overexpression results in kidney-shaped cotyledons, round and curled leaves, small rosette size, later flowering, reduced fertilization and round seeds (Wang et al., 2011). OFP1 and OFP4 negatively 15   regulate secondary cell wall development by interacting with KNAT7 (Li et al., 2011), but it remains unclear if other OFPs are similarly involved in secondary wall development.  1.7 Research objectives and Significance of Findings The main goal of my PhD thesis is to investigate the functions of Arabidopsis KNOX2 and OFP genes focusing on those that may be involved in secondary wall development. To test the hypothesis that KNOX2 and OFP proteins are involved in the regulation of secondary wall formation through interaction with other partners, the following three objectives were addressed: 1. To determine the functions of KNOX2 genes, KNAT3, KNAT4 and KNAT5, in secondary cell wall development. 2. To identify the interaction partners of OFP4.  3. To explore the potential functions of OFP2, OFP3, and OFP5 in the same phylogenetic clade as OFP1 and OFP4.  Chapter 2 addresses the first research objective. By using combined molecular and genetic approaches, I concluded that other KNOX2 genes are also playing an important role in secondary cell wall formation, especially KNAT3. It is the first time to discover the transcription factors in KNOX2 class may function antagonistically for secondary cell wall formation in interfascicular fibers. Also, all the differentially expressed genes regulated by KNAT3 and KNAT7 are revealed by mRNA sequencing.  Chapter 3 addresses research objective 2 and 3. Previous studies have shown that OFP4 is involved in secondary cell wall development, but a role for OFP4 secondary cell wall formation was not supported by my studies. To test what other proteins interact with OFP4, I performed a yeast two-hybrid screen. In this chapter, Nucleosome Assembly Protein 1 (NAP1) was identified as a new interaction partner of OFP4, and new roles for both NAP1 and OFPs  in the regulation of cotyledon development in Arabidopsis were discovered.   16       Figure 1.1 Phylogenetic analysis of Arabidopsis KNOX (a) and OFP (b) gene families Bootstrap values are indicated above the branch. Figure is modified and reproduced from Figure 9 in Li et al. (2012) and Figure 1 in Liu et al. (2014). By permission from the publisher.   17     Figure 1.2 Transcriptional network of secondary cell wall development in Arabidopsis The red line with blunt ends indicates negative regulation, and the blue arrow indicates positive regulation. Figure is reproduced from Figure 2 in Yang and Wang (2016). By permission from the publisher.          18   Chapter 2: The Class II KNOX genes KNAT3 and KNAT7 work cooperatively to regulate secondary cell wall deposition and provide mechanical support to Arabidopsis stems  2.1     Introduction Secondary cell walls confer the main structural components of mature tracheary elements and fibers, providing additional mechanical strength and stiffness to plant tissues (Cosgrove and Jarvis, 2012). The formation of secondary cell walls requires a complex network of transcriptional regulation in land plants, which includes at least three layers of regulators (reviewed in Nakano et al., 2015;  Yang and Wang, 2016). The VASCULAR-RELATED NAC-DOMAIN gene family (VND1-7) have been identified as key regulators of secondary cell wall deposition in vessel elements (Kubo et al., 2005), while the SECONDARY WALL-ASSOCIATED NAC DOMAIN PROTEIN1 (SND1) and NAC SECONDARY WALL THICKENING PROMOTING FACTOR1 (NST1) gene families regulate secondary cell wall deposition in interfascicular fibers (Zhong et al., 2006; Mitsuda et al., 2007).  The second-layer master switches are MYB-type transcription factors, such as MYB46 and MYB83, which are directly regulated by the NAC transcription factors, and act as regulators of secondary wall formation in both xylem vessels and fibers (Zhong et al., 2007; McCarthy et al., 2009; Ohashi-Ito et al., 2010; Wang et al., 2011; Zhong and Ye, 2012). Many other transcription factors also function downstream of the NAC and MYB domain master switches, forming a feed forward regulatory network for secondary wall deposition (Zhong et al., 2008a; Ko et al., 2009; Taylor-Teeples et al., 2014a).  One of the Arabidopsis Knotted-like homeobox (KNOX) genes, KNOTTED ARABIDOPSIS THALIANA7 (KNAT7), has been shown as a direct target of both SND1 (Zhong et al., 2008a) and MYB46 (Ko et al., 2009). Unlike the NAC and MYB domain transcription factors that activate secondary cell wall deposition, KNAT7 was reported to act as a negative regulator 19   of secondary cell wall formation in interfascicular fibers, with thicker fiber walls observed in loss-of-function KNAT7 mutants (Li et al., 2012). However, Zhong et al. (2008) found that transgenic plants with dominant repression of KNAT7 had a reduction in secondary wall thickness. Paradoxically, knat7 mutants also had an irregular xylem (irx) phenotype (Li et al., 2012), as reported previously in the original irregular xylem studies (Brown et al., 2005). Based on the identification of KNAT7 interactors, such as OVATE FAMILY PROTEIN4 (OFP4), BELL-LIKE HOMEODOMAIN6 (BLH6), and MYB75 (Hackbusch et al., 2005; Bhargava et al., 2010; Li et al., 2011; Y., Liu et al., 2014), KNAT7 may have defined functions in secondary wall formation, depending on specific cell type and interaction partners, however, the specificities of the mechanism of KNAT7 regulation of secondary cell wall biosynthesis are still poorly understood.  Including KNAT7, there are in total nine KNOX genes in Arabidopsis, which belong to the plant-specific THREE AMINO ACID LOOP EXTENSION (TALE) homeodomain superfamily (Hake et al., 2004; Hay and Tsiantis, 2010). Based on gene duplication and diversification events in a common ancestor of land plants, KNOX genes were grouped into two classes, class I (KNOX1) and class II (KNOX2; Kerstetter et al., 1994; Mukherjee et al., 2009). A new class of KNOX genes, KNATM, which does not have a homeodomain, has been shown to regulates the leaf proximal-distal patterning and has only been identified in dicot species thus far (Kimura et al., 2008; Magnani and Hake, 2008). KNOX1 genes include SHOOTMERISTEMLESS (STM), BREVIPEDICELLUS (BP), KNAT2 and KNAT6, which are involved in cell proliferation (Sakakibara et al., 2008) and required for development of shoot apical meristem(Long et al., 1996; Vollbrecht et al., 2000; Groover et al., 2006; Tsuda et al., 2011). KNOX2 genes KNAT3, KNAT4, KNAT5 function redundantly to regulate lateral organ differentiation in Arabidopsis (Furumizu et al., 2015). KNAT3 was also reported to modulate abscisic acid (ABA) responses during germination and early seedling development (Kim et al., 2013). KNAT7 also belongs to the KNOX2 clade, and since it plays diverse roles in secondary wall formation, the question arises whether the other three KNOX2 genes, KNAT3, KNAT4 and KNAT5, may play a role in secondary wall formation. 20    In this chapter, I report the secondary cell wall characteristics of knat3, knat4 and knat5 mutants in Arabidopsis stems. I found that all KNOX2 genes are expressed in the inflorescence stems, and loss-of-function analysis show knat3knat7 double mutants exhibit an enhanced irregular xylem (irx) phenotype, typified by weak inflorescence stems. In addition, the interfascicular fiber wall thickness is reduced and cell wall compositions are modified in knat3knat7 double mutants. Constitutive overexpression of KNAT3 led to thicker interfascicular fiber walls. Whole-genome expression profiling of knat3knat7 double mutants showed differential gene expression associated with cell wall and secondary metabolism genes. My results suggest that KNAT3 plays a positive role in interfascicular fiber secondary wall formation, and may work cooperatively with KNAT7 to contribute to plant stem strength by affecting the cell wall deposition and the integrity of the cell-wall matrix. Together, the KNOX2 proteins appear to form a complex network of positive and negative regulators of secondary cell wall formation.      2.2     Materials and methods 2.2.1 Plant material and growth condition Arabidopsis thaliana ecotype Columbia was used as wild type in all experiments, and all the transgenic lines and mutants are also in the Columbia background. T-DNA insertion lines for knat3 (SALK_136464), knat4-1 (SALK_020216) and knat5 (SALK_000339C) were obtained from the Arabidopsis Biological Resource Center (ABRC). The knat7-1 allele described in Li et al. (2012) was used for all knat7 phenotypic analyses. Homozygous T-DNA insertion lines were screened by PCR using gene-specific primers (Table 2.1). knat3, knat4-1, knat5, and knat7 alleles were used to generate knat3knat7, knat4-1knat7, knat5knat7, knat3knat4-1knat7 and knat3knat5knat7 double and triple mutants. The other double and triple mutants, knat3knat4-2, knat3knat5, knat4-2knat5 and knat3knat4-2knat5 were kindly gifted from Dr. John Bowman (Furumizu et al., 2015). For the overexpression 21   analysis of KNAT3, homozygous lines for 35S:KNAT3 (Furumizu et al., 2015a) were employed for all experiments.    In all experiments, seeds were sterilized with 70% ethanol and sown on Murashige and Skoog (MS) medium with 1% sucrose, then cold-treated at 4°C for 48 hours in the dark, and grown at 20°C under a 16/8 h (light/dark) photoperiod at about 120 µmol m-2 s-1 light for 7 to 10 days. Seedlings were transferred to soil and grown under long-day conditions (16/8 h light/dark cycle) at ~100 µmol m-2 s-1 light, 20°C in growth chambers for further analysis.     2.2.2 Cloning and plant transformation To generate ProKNAT3:GUS, ProKNAT4:GUS, ProKNAT5:GUS, and ProKNAT7:GUS constructs, the fragment upstream of the ATG start codons of KNAT3 (3207 bp), KNAT4 (3290 bp), KNAT5 (1119 bp) and KNAT7 (2591 bp) respectively, were sub-cloned into the pCR8/GW/TOPO entry vectors, and then subsequently cloned into the binary vector PMDC163. For complementation experiments, the genomic fragments of KNAT3 and KNAT7 containing the 5’ promoter and 3’ untranslated regions were amplified from Col-0 genomic DNA, respectively. Following verifification of the nucleotide sequences of the amplified fragments, each fragment was cloned into the binary vector PMDC107 containing the hygromycin resistance genes to generate the ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP constructs. Gene-specific oligonucleotides used for cloning and construct generation are shown in Table 2.1.  All the constructs were introduced into Agrobacterium tumefaciens strain GV3101 for plant transformation. Wild-type Columbia was transformed for expression pattern analysis, while the complementation constructs were transformed into knat3knat7 double mutant lines via the floral dip method (Clough and Bent, 1998) to generate transgenic plants.   22   2.2.3 GUS expression assay GUS activity was assayed on hand-sections of 8-week-old inflorescence stems by incubating tissues in a solution containing 0.1M sodium phosphate buffer (PH 7.0), 1mM substrate 5-bromo-4-chloro-3-indolyl-β-D-glucuronide (X-Gluc), 0.5mM potassium ferricyanide and 0.01%(v/v) Triton X-100 at 37°C for 1 h to overnight. The resulting stained tissues were fixed with FAA (50 [v/v] ethanol, 5% [v/v] acetic acid and 10% [v/v] formaldehyde), and observed with an Olympus AX70 light microscope.   2.2.4 Microscopy Freshly harvested 8-week-old inflorescence stems were hand-sectioned, stained with aqueous 0.05% toluidine blue O for 1-2 min and mounted with water, or stained with 2% (v/v) phloroglucinol for 30 sec and mounted with concentrated HCl. Mäule staining was performed by treating hand sections for 5 min with 0.5% KMnO4, rinsing in water, and treating with 30% HCl until the brown colour disappeared. Samples were mounted in concentrated NH4OH and viewed using an Olympus AX70 light microscope. ImageJ was used to measure the thickness of interfascicular fibers. For each cell, three measurements were taken, and the average used as the value of that cell. Statistical analysis was performed by Student’s T-test and one-way ANOVA followed by Tukey’s post hoc test.   Roots of ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP plants in the knat3knat7 background were counterstained with propidium iodide and observed on a Perkin-Elmer UltraView VoX spinning disk confocal mounted on a Leica DMI6000 inverted microscope with a Hamamatsu 9100-02 CCD camera. The microscope used for all live-cell imaging was fit with the following excitation/emission filters: GFP (488 and 525) and propidium iodide (620/720). ImageJ software was used for image processing.   23   2.2.5 Physical tests Tensile and three-point flexural tests were carried out using a 5565 model Instron universal testing machine according to the detailed methods describe by MacMillan et al. (2010). 60mm basal stems samples were harvested and used for all tensile tests. The 20mm basal segments were attached to paper tabs by moisture-activated cyanoacrylated Loctite 454 gel (Henkel, Dusseldorf, Germany). The instron testing machine was fit with a 5kN load cell, and employed a cross-head speed of 10 mm/min. Flexural bending strength was estimated on a specimen 15mm from the stem base, using the Instron fitted with a 10N load cell and a cross head speed of 10mm/min. A digital caliper was used to measure the diameter of each stem. BlueHill software was employed to capture and calculate the force and modulus of elasticity for each sample. Statistical analysis was performed by one-way ANOVA followed by Tukey’s post hoc test.   2.2.6 Cell-wall analysis All the cell-wall analyses were performed on dried stems harvested 0-11cm from the base.  Lignin content was determined by the modified Klason method described previously (Coleman et al., 2008), in which 100mg of ground, hot acetone-extracted Arabidopsis stem tissue was incubated with 72% H2SO4 for 2 hours with regular stirring and then diluted to 3% H2SO4 and autoclaved to completely hydrolyze cell wall polysaccharides. The monosaccharides were quantified by HPLC (DX-500; Dionex) equipped with a CarboPac PA1 column (Thermofisher) and a pulsed amperometric detector with a gold electrode. Acid-insoluble lignin was quantified gravimetrically, while the acid-soluble lignin content was measured spectrophotometrically at 205 nm.   α-cellulose content was estimated according to the method of Porth et al. (2013). 100 mg of ground, extract-free stem tissue was mixed with 3.5 ml buffer (60ml of glacial acetic acid, 1.3g NaOH/L) and 1.5 ml of 20% sodium chlorite, and then gently mixed at 50 °C for 16 hours. The reaction was repeated again and the residual sample was transferred to crucible to 24   determine holocellulose gravimetrically. α-cellulose content was then quantified by reacting 30 mg of holocellulose with 2.5ml of 17.5% NaOH for 30 min, followed by a second 8.75% NaOH for an additional 30 min. Then α-cellulose was finally determined gravimetrically.   Lignin monomer composition was estimated by thioacidolysis as described by Robinson and Mansfield (2009). Statistical analysis was performed by one-way ANOVA followed by Tukey’s post hoc test, using three biological replicates and two technical replicates.     2.2.7 Total RNA isolation and quantitative RT-PCR The basal 0-11cm of 8-week-old inflorescence stems were frozen and homogenized. Total RNA isolation and subsequent cDNA synthesis was completed as per Liu et al., (2014). For qRT-PCR analysis of the basal stem segments, PCR amplification was performed using a CFX ConnectTM real-time system (Bio-Rad), using 40 quantitative PCR cycles that were run under the following parameters: denaturation step, 95°C for 20 sec; annealing step, 55°C for 30 sec; elongation step, 72°C for 1 min. ACTIN2 was used as the reference housekeeping gene. All primers are listed in Table 2.1. The calculation of differences in gene expression was according to Bhargava et al. (2010). Three biological replicates consisting of three technical replicates were used for each analysis.  2.2.8 RNA-seq analysis Total RNA was extracted from wild type, knat3, knat7, and knat3knat7 inflorescence stems (1-15cm from the top) using TRIZOL reagent (Invitrogen, Life Technologies) following the manufacturer’s protocol. In total, 12 samples were sent to UBC-Djavad Mowafaghian Centre for Brain Health (DMCBH) Next Generation Sequencing Centre for library preparation and Ion ProtonTM semiconductor-based transcriptome sequencing (4 genotypes times 3 biological replicates).   25   The sequencing quality of the RNAseq reads was evaluated by FastQC software (http://www.bioinformatics.babraham.ac.uk/projects/fastqc), and clean reads were mapped to the A. thaliana TAIR 10 reference genome using STAR (https://github.com/alexdobin/STAR). FPKM (fragments per kilobase of exon per million fragments mapped) was used to estimate the gene transcript abundance. Cuffdiff (Trapnell et al., 2012) was used to calculate FPKM and identify differentially expressed genes across genotypes. GO functional enrichment analysis was performed using PANTHER (https://www.arabidopsis.org/tools/go_term_enrichment.jsp) with a significance level of P < 0.05 (Mi et al., 2013). The metabolism overview of DEGs between mutants and wild type was detected by MAPMAN software (http://mapman.gabipd.org/).   2.3     Results 2.3.1 KNOX2 genes are expressed in the inflorescence stems  In Arabidopsis, there are four genes encoding KNOX2 proteins, KNAT3, KNAT4, KNAT5 and KNAT7 (Furumizu et al., 2015a). To investigate if these KNOX2 genes are expressed in cells undergoing secondary wall deposition, and overlapping with that of KNAT7, lines containing ProKNAT3:GUS, ProKNAT4:GUS, ProKNAT5:GUS, and ProKNAT7:GUS were generated, and similar stem expression patterns were observed among all independent transformants (Figure 2.1). The KNOX2 promoter directed GUS expression in cortical cells and developing vascular bundles near the apex of the stems (Figure 2.1a-d). This expression pattern was retained in the more mature stem regions, where GUS staining was also detected in the differentiating interfascicular fibers (Figure 2.1e-h). Interestingly, the expression of all four KNOX2 genes was absent in fully differentiated interfascicular fibers and metaxylem vessels in the lower regions of mature inflorescence stems, but was still apparent in cortical cells, phloem, and protoxylem cells (Figure 2.1i-l). These findings suggest that KNAT3, KNAT4 and KNAT5 share similar expression patterns as KNAT7 in the different developmental stages of inflorescence stems, and they are expressed in differentiating interfascicular fibers and xylem cells where secondary cell walls are actively being deposited.   26    2.3.2    knat3knat7 double knock-out mutants have enhanced irregular xylem (irx) phenotype To test the potential role of KNOX2 genes in xylem development, their null alleles were obtained, which included SALK_136464 (knat3), SALK_020216 (knat4-1), N759461 (knat4-2), SALK_000339C (knat5), and SALK_002098 (knat7) (Kim et al., 2013; Furumizu et al., 2015a; Li et al., 2012). Stem cross-sections were examined at the base of mature inflorescence stems for each genotype to determine xylem morphology. knat3, knat4, and knat5 single mutants exhibited normal vessels morphology (Figure 2.2b-d). As previously reported (Brown et al., 2005; Li et al., 2012; Y., Liu et al., 2014a), knat7 single mutant exhibited an irregular xylem (irx) phenotype (Figure 2.2e). Due to their overlapping expression patterns, gene redundancy may potentially mask these observations, so different combinations of double and triple mutants were generated (knat3knat7, knat4-1knat7, knat5knat7, knat3knat4-1knat7, knat3knat5knat7) or generously obtained from Dr. John Bowman (knat3knat4-2, knat3knat5, knat4-2knat5, knat3knat4-2knat5).   Interestingly, the vessels of knat3knat7 double mutants were more frequently irregular and collapsed compared to wild-type plants and the associated single mutants (Figure 2.2f).  The enhanced irx phenotype of knat3knat7 was obvious in all vascular bundles of plants examined. knat4-1knat7 and knat5knat7 showed a mild irx phenotype (Figure 2.3d,e), which appeared to vary in severity between vascular bundles and plants. Their mild irx phenotype was similar to that commonly observed in knat7 single mutants (Y., Liu et al., 2014a).  The other double mutants did not show any obvious changes in xylem morphology compared to wild-type plants (Figure 2.3f-h). Cross-sections of triple mutant plants revealed that knat3knat4-1knat7 and knat3knat5knat7 had severe irx phenotypes, mimicking that of knat3 knat7 double mutants, while knat3knat4knat5 vessels appeared similar to that of wild-type plants (Figure 2.3i-k). These results indicate that in addition to KNAT7, KNAT3 may also play a role in the xylem vessel development.  27    2.3.3    Knockout of KNAT3 and KNAT7 affects stem mechanical properties and fiber wall thickness The knat3knat7 double mutants exhibited a pendent stem phenotype following 7 weeks growth, while the other single and double mutants showed no apparent stem morphological differences compared to wild-type plants (Figure 2.4a, 2.5). Often the double mutants of knat3knat7 displayed shorter inflorescence stems than wild type, but this was not consistent across all generations and experiments. The obvious weak stem phenotype implies that the mechanical properties of knat3knat7 inflorescence stems may have been impaired, so both uniaxial tensile tests and three-point flexural tests of stems were performed. As described previously (Turner, 1997; MacMillan et al., 2010), the stem strength is measured as the maximum stress at yield for breaking the sample, and the stem stiffness (as defined by modulus of elasticity) is a measure of the force required to deform the sample. The bases of knat3knat7 fresh stems were considerably weaker than wild-type plants, displaying an approximately 80% reduction in the tensile strength and 60% reduction in the tensile stiffness compared to wild type (Figure 2.4b,c). In comparison, knat3 and knat7 single mutants do not possess weaker tensile strength and stiffness than wild type (Figure 2.4b,c). However, the flexural strength of the base of knat7 fresh stems showed a significant increase compared to wild-type stems (Figure 2.4d).  Unlike tensile strength, knat3knat7 fresh stems had no change in flexural strength (Figure 2.4d). When it comes to the bases of dried stems, the trends in flexural strength were similar to tensile strength, with a pronounced decrease in knat3knat7 double mutants, and no considerable changes in the corresponding single mutants (Figure 2.4d). The differences in flexural strength between fresh and dried stems may be a function of the turgor pressure coming from stem water contents. Furthermore, the knat3knat7 stems had reduced flexural stiffness compared to wild type and single mutants, for both fresh and dried stems (Figure 2.4e). The reduction in flexural stiffness in knat3knat7 stem bases was particularly remarkable in dried stems (82%), whereas fresh stems showed a 56% reduction compared to wild-type plants (Figure 2.4e). 28    Arabidopsis interfascicular fibers are particularly important for providing the necessary mechanical strength to inflorescence stems, as demonstrated in a significant number of mutants, including rev and fra8 (Zhong, 1997; Zhong, 2005), to mention only a few. To determine whether the striking stem phenotype in knat3knat7 double mutants was caused by defects in interfascicular fibers, the fiber wall thickness was examined. Cross-sections of stem bases stained with phloroglucinol-HCL revealed that the interfascicular fiber wall thickness of knat3 knat7 double mutant was smaller compared (decreased by 37%; Figure 2.6e) with that of wild type (Figure 2.6a-d), as well as knat3, and knat7 mutant plants. In contrast, knat7 had a significant increase in interfascicular fiber wall thickness, as previously reported (Li et al., 2011; Li et al., 2012; Y., Liu et al., 2014a), while knat3 did not show differences compared to that of wild-type (Figure 2.6e). .   2.3.4  KNAT3 and KNAT7 expression affect secondary cell wall composition Secondary cell wall is the major constituent of xylem vessels and fiber cells in Arabidopsis stems. The enhanced irx and decreased fiber wall thickness phenotypes predict an alteration in secondary cell wall composition in knat3knat7 double mutants. Monosaccharide compositional analysis was performed on the basal segments of inflorescence stems. Single mutants had no significant or mild changes in the concentrations of sugars derived from cell wall carbohydrates, while the changes in the double mutants knat3knat7 were significant (Table 2.2). The glucose, xylose and mannose content in the double mutants was significantly reduced by 22%, 43% and 40% respectively, but considerable increases were also observed in arabinose, rhamnose and galactose of the double mutants, displaying 75%, 41% and 41% more than wild-type (Table 2.2). Given cellulose is a polymer of β(1,4)-linked glucose, the decreased glucose content maybe attributed to a reduction in cellulose deposition in stems, so cellulose content was further investigated. The α-cellulose content in knat3knat7 stem bases was approximately 20% lower than in wild type, and no significant difference were observed in knat3 and knat7 single mutants (Table 2.3). Lignin, another important composition of secondary cell wall, was also quantified. There was no discernible 29   change in the lignin content of knat3 single mutant relative to the wild-type control (Table 2.3). However, a different trend was observed between knat7 and the double mutants relative to the wild type (Table 2.3). knat7 had an approximate 11% increase in acid-insoluble lignin and 19% decrease in soluble lignin, while knat3knat7 displayed a 4% decrease in acid-insoluble lignin and 26% increase in soluble lignin (Table 2.3). It is quite interesting that knat7 single mutant had increased total lignin content as previously reported (Li et al., 2012; Liu et al., 2014), while the double knockout mutant knat3knat7 did not have a significant change in total lignin (Table 2.3).  The lignin monomer composition, as determined by thioacidolysis, was further characterized in single and double mutant stems (Figure 2.7), and revealed that knat3 and knat7 single mutants had lower S/G lignin ratio than the wild-type control plants. However, the ratio was much lower in the knat3knat7 double mutant stems (around 84% lower than the wild-type S/G lignin; Figure 2.7a). Mäule staining of stem cross-sections was concurrently conducted in attempts to selectively localize syringyl (S) lignin in single and double mutants (Figure 2.7b). The vascular elements and interfascicular fibers displayed the classical the red coloration in wild type and single mutants, which is indicative of syringyl-derived lignin (Figure 2.7b), while the same cells in the knat3knat7 double mutants were much lighter (Figure 2.7b). The decreased S lignin in the double mutant stems is consistent with the chemical determinatin of S/G monomer ratio.   2.3.5    knat3knat7 phenotypes are complemented by ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP  To investigate whether the knat3knat7 phenotypes were caused by loss of function of KNAT3 and KNAT7, complementation experiments were performed. knat3knat7 was transformed with two different constructs, ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP. Expression of KNAT3 under the control of its native promoter partially rescued the severe irx phenotype consistently observed in knat3knat7, manifesting in a phenotype similar to knat7 (Figure 30   2.8a). Plants transformed with ProKNAT7:KNAT7-GFP in the knat3knat7 background fully rescued the irx phenotypes of knat3knat7 (Figure 2.8a). The pendent phenotype of knat3knat7 was also complemented in both transformants (Figure 2.8b). The interfascicular fiber wall thickness was quantified in inflorescence stems of the complementation lines (Figure 2.8c,d), and shown to be restored to a level relative to the wild type by expressing KNAT3-GFP under the control of its native promoter in knat3knat7 background (Figure 2.8c), mimicking the knat7 single mutant. As with knat3 single mutants, the interfascicular fiber cell wall thickness was not different than wild type when KNAT7-GFP was expressed in knat3knat7 mutants (Figure 2.8c).   KNAT3 and KNAT7 were also confirmed to have overlapping expression patterns during secondary cell wall deposition, as reflected in the GFP signals using the complemented lines, since both lines were under the control of their native promoters (Figure 2.9). KNAT3 had a broader expression pattern compared with KNAT7, but both were detected in the protoxylem and metaxylem vessel precursors in the roots, and in the interfascicular fiber cells in inflorescence stems of the complemented lines (Figure 2.9).   2.3.6    Overexpression of KNAT3 increases interfascicular fiber wall thickness of stems The contrasting phenotypes of knat3knat7 double mutants having decreased interfascicular fiber wall thickness, while knat7 single mutants had increased fiber wall thickness, suggested that KNAT3, along with KNAT7, plays a regulatory role in the development of interfascicular fibers. In an attempt to enhance the levels of KNAT3, plant lines overexpressing KNAT3 under the control of 35S promoter were generated. Cross-sections of the basal inflorescence stems were examined by Toluidine Blue staining (Figure 2.10) and compared to wild type, and the overexpression lines appeared to have thicker interfascicular fibers secondary cell walls (Figure 2.10a). Measurements taken via high-magnification light microscopy images confirmed that constitutive expression of KNAT3 (Pro35S:KNAT3) significantly increased interfascicular fiber wall thickness (Figure 2.10b), suggesting that KNAT3 may play a role in the positive regulation of secondary wall formation in interfascicular fibers.  31   2.3.7    Genome-wide transcript profiling shows altered expression of secondary cell wall related genes in knat3knat7 plants In an attempt to uncover the mechanistic role of KNAT3 and KNAT7 in secondary cell wall development, the transcript abundance of known cellulose, hemicellulose and lignin biosynthetic genes was profiled by qRT-PCR in cDNA isolated from basal stems of knat3, knat7 and knat3knat7 mutants. It was clearly shown that the expression levels of secondary cell wall specific CELLULOSE SYNTHASE (CESA) genes (CESA4, CESA7, CESA8), and one of xylan biosynthetic genes (IRX9) were dramatically increased in the double mutants compared with wild-type and the individual single mutants (Figure 2.11a). Another xylan synthetic gene, IRX10, was significantly down regulated in knat7 and knat3knat7 plants, and knat3 showed a slightly increase in IRX10 expression (Figure 2.11a). Loss of KNAT3 and KNAT7 functions together also resulted in both up-regulation and down-regulation of several phenylpropanoid biosynthetic genes (Figure 2.11b). Specifically, the expression of C4H, C3H1, CCoAMT1 and CCR1, was elevated 1.5- to 2-fold in the basal segments of knat3knat7 inflorescence stems, while the abundance of FERULATE-5-HYDROXYLASE (F5H), a key enzyme for S lignin biosynthesis, was considerably decreased (about 50%) in knat3knat7 mutants (Figure 2.11b). This latter finding is consistent with the reduced S lignin deposition observed in our biochemical and histochemical assessment of the knat3knat7 stems.  To gain further insights into KNAT3 and KNAT7 function, mRNA sequencing was carried out to identify all the genes showing significant changes in expression among wild-type, knat3, knat7 and knat3knat7 stems. Compared with wild type, knat3knat7 had a total of 959 up-regulated and 1001 down-regulated genes, whereas knat3 and knat7 had 484 and 367 down-regulated, respectively (Figure 2.12a,c). Comparative analyses revealed that of all the mis-regulated genes, 32 and 117 genes were found in common in knat3, knat7 and knat3knat7 plants (Figure 2.12a,c). To determine the identities of these differentially expressed genes (DEGs), GO ontology classifications were performed based on their known functions. In knat7, genes involved in cell wall organizational processes were significantly 32   up-regulated and over-represented compared to wild-type plants (Figure 2.12b), which is consistent with a repressor-type function in secondary cell wall formation (Li et al., 2012). Compared with knat3 and knat7, a significant number of up-regulated genes in knat3knat7 were related to glucuronoxylan biosynthesis, plant-type secondary cell wall biogenesis, nucleotide-sugar biosynthetic processes, microtubule-based processes and pectin metabolic processes (Figure 2.12b). Additionally, a few biological processes were over-represented in the down-regulated DEGs found in knat3knat7 plants, such as aromatic amino acid family catabolic process, phloem transport, and response to red light, to mention only a few (Figure 2.12d). MapMan was further used to visualize the differentially expressed genes in various metabolic pathways (Figure 2.13, 2.14). Generally consistent with the qRT-PCR data, knat3knat7 exhibited both increased and decreased transcript levels of phenylpropanoid biosynthetic genes (Figure 2.13, 2.14, Table 2.4). The weak stem and enhanced irx phenotypes of knat3knat7 may be caused by the mis-regulation of a number of secondary cell wall related genes, representing all major biochemical pathways (lignin, cellulose and hemicellulose).   2.4    Discussion KNAT7, one of the four KNOX2 genes in Arabidopsis, has been proposed to function as a transcriptional repressor regulating secondary cell wall formation in interfascicular fibers (Li et al., 2011; Li et al., 2012). Three other Arabidopsis KNOX2 genes, KNAT3, KNAT4, and KNAT5, were found to act redundantly in regulating plant development, with serrated leaves apparent on knat3knat4, knat3knat5 and knat3knat4knat5 plants (Furumizu et al., 2015), however, little is known about their exact biological role(s), if any, in secondary cell wall formation. In this study, beyond the developmental roles of KNOX2 in leaf development, we focused on their putative roles in secondary cell wall formation. KNAT3 was identified as a potential transcriptional activator, working antagonistically with KNAT7 to regulate secondary wall formation in interfascicular fibers, while in xylem vessels, KNAT3 and KNAT7 may function together to activate secondary cell wall deposition.   33   Expression data shows that all four KNOX2 genes are co-expressed in the same cell types of inflorescence stems (Figure 2.1, 2.9), suggesting KNAT3, KNAT4 and KNAT5 may also play a role(s) in secondary cell wall formation (Li et al., 2012; Y., Liu et al., 2014a). Interestingly, ProKNAT3:KNAT3-GFP has an expanded expression pattern compared with ProKNAT7:KNAT7-GFP in the roots, with expression apparent throughout other root cell layers, other than protoxylem and metaxylem vessel precursors (Figures 2.9). This suggests that KNAT3 may play multiple roles in plant development, which is consistent with previous findings that KNAT3 can modulate ABA responses (Kim et al., 2013) and regulate lateral organ development (Furumizu et al., 2015a).   The collapsed xylem vessel phenotype, irx, has been widely used to isolate Arabidopsis mutants defective in the secondary cell wall biosynthesis (Turner, 1997; Jones et al., 2001; Brown et al., 2005). Analysis of knat4knat7, knat5knat7 mutants revealed an irx phenotype, which phenocopied knat7 (Li et al., 2012), while knat3knat7 afforded an enhanced irx phenotype (Figure 2.2, 2.3). The synergistic effect of the xylem vessel phenotype in double mutants indicates that KNAT3 and KNAT7 may function redundantly in modulating vessel secondary cell wall development, although KNAT7 likely plays the dominant role, as the single mutant knat7 displays the irx phenotype, while knat3 has no obvious xylem vessel phenotypes (Figure 2.2).   A striking effect of loss-of-function KNAT3 and KNAT7 was the weak inflorescence stem with reduced tensile and flexural strength and stiffness (Figure 2.4). Thinner interfascicular fiber cell walls in the knat3knat7 double mutants is undoubtedly the reason for the pendent stem phenotype, as interfascicular fibers are important for the mechanical strength of Arabidopsis inflorescence stems (Zhong, 2005; Zhong et al., 2007). Although the thicker secondary cell walls phenotype in interfascicular fibers in knat7 mutants was observed here, as well as in previous studies (Li et al., 2012), the double mutants knat3knat7 displayed an opposing trend with thinner interfascicular fiber wall thickness (Figure 2.6). Thus, the thicker secondary cell wall phenotype in knat7 appears to be associated with a functional KNAT3. 34   KNAT3 may be able to activate secondary cell wall formation in fibers, as ectopic expression of KNAT3 manifested in thicker interfascicular fiber secondary cell walls (Figure 2.10). However, knat3 single mutants do not show an obvious interfascicular fiber secondary cell wall development phenotype (Figure 2.6), which suggests KNAT3 has some redundancy with other genes in plants to regulate secondary wall formation. KNAT7 may have distinct roles by interacting with different partners, such as the helix-loop-helix leucine zipper domain (bHLHZip) transcription factor family proteins demonstrated in animals (reviewed in Lüscher, 2001; Amoutzias et al., 2008). Zhong et al. (2008) found dominant transcriptional repression of KNAT7 caused reduction in secondary cell wall thickening in interfascicular fibers, which is similar to knat3knat7 double mutants. Since the dominant repression approach may not only inhibit the function of targeted transcription factor, but also their homologs (Hiratsu et al., 2004), it is possible that the thinner fibers phenotype emanates from the repression of both KNAT3 and KNAT7.   The cellulose polymer is the major load bearing component affecting stem tensile and flexural strength, as demonstrated in Arabidopsis and rice mutants (Turner and Somerville, 1997; Li et al., 2003). The lower cellulose content of knat3knat7 stem cell walls (Table 2.3) can easily explain the decreasing tensile and flexural strength of knat3knat7 stems. In addition, the lignin composition was modified in the knat3knat7 double mutants, with significant reductions in both S lignin and S/G lignin ratio in inflorescence stems (Table 2.3). Secondary cell walls are the main structural component of xylem vessels and interfascicular fibers in Arabidopsis inflorescence stems, and therefore the altered cell wall chemical composition in knat3knat7 mutants may account for the irx and thinner interfascicular fiber wall phenotypes. The complementation tests in our study confirmed the involvement of KNAT3 and KNAT7 in the regulation of secondary cell wall development in Arabidopsis stems (Figure 2.8).   The impact of KNAT3 and KNAT7 on cell wall development was also assessed by qRT-PCR and messenger RNA sequencing. Both techniques clearly showed increasing expression of 35   the secondary cell wall-specific cellulose biosynthetic genes (CESA4, CESA7 and CESA8) in the basal stems of knat3knat7 mutants (Figure 2.11), which was unexpected as the knat3knat7 mutants had lower cellulose contents (Table 2.3). We reasoned that the observed up-regulation of cellulose synthase genes may reflect a feedback mechanism where that plant is attempting to over compensate for impaired secondary cell wall development in knat3knat7 stems. The key enzyme for syringyl monomer biosynthesis and integration into the lignin polymer, F5H, was significantly down regulated in the knat3knat7 tissue (Figure 2.11), which is consistent with the observed reduction in both S lignin and the associated S/G lignin ratio (Figure 2.7). RNA-seq experiments provided an overview of the effects of the various mutation on the global stem transcriptomes. The down-regulation of F5H was again apparent in the transcriptome analysis, and was one of the most highly down-regulated genes in knat3knat7 stems. In addition, a number of secondary cell wall-related transcription factors and biosynthetic genes were only differentially expressed in the knat3knat7 double mutants. Among the down-regulated cell wall-related genes were MAP70-5, which encodes a plant-specific microtubule-associated protein that is important for secondary cell wall patterning (Pesquet et al., 2010; Oda and Fukuda, 2012), VND-INTERACTING2 (VIN2), which is a transcriptional repressor regulating xylem cell specification (Yamaguchi et al., 2010), and PECTIN METHYLESTERASE35 (PME35), which has been shown to have a pendent stem phenotype and an increased deformation rate of stem in the loss-of-function mutant (Hongo et al., 2012). GO classification analysis clearly showed the cell wall organization process was overrepresented in knat7 DEGs (Figure 2.12b), which may explain the thicker interfascicular fibers and higher lignin phenotypes in knat7 mutants. The mRNAseq also highlighted a number of up-regulated genes in the knat3knat7 plants involved in secondary cell wall biogenesis process, which is consistent with qRT-PCR results.  This specific up-regulation in knat3knat7 may represent a negative feedback mechanism to fine-tune secondary cell wall development. MAPMAN analysis revealed a number of cell wall and secondary metabolism-related genes that were uniquely mis-regulated in the knat3knat7 mutants (Figure 2.13, 2.14), which may explain their weak stem and heighten irx phenotypes.   36   Liu et al. (2014) showed that the expression of REV was negatively regulated by KNAT7, resulting in thicker interfascicular fibers in knat7 mutants. However, the REV transcript level was not significantly changed in knat3knat7 mutants in my RNA-seq analysis, suggesting the thinner interfascicular fiber phenotype of knat3knat7 may not be caused by REV functions (Table 2.4). Taken together, we found that the knat3 mutation could enhance knat7 xylem vessel and S lignin secondary cell wall phenotypes (Figure 2.2,2.7), while concurrently repressing the knat7 interfascicular fiber phenotype (Figure 2.6). KNAT3 may be a potential activator of xylem vessel secondary cell wall formation, acting together with KNAT7, while it may act antagonistically with KNAT7 for secondary wall formation in interfascicular fiber wall biosynthesis. KNAT3 and KNAT7 appear to work cooperatively to activate S lignin biosynthesis and regulate secondary cell wall composition and integrity.                    37   Table 2.1 Oligonucleotides used in Chapter 2 Gene name Application Primer sequence (5' to 3') KNAT3 genotyping salk_136464-L: TCTCCTTCAATCATTTCACCG     salk_136464-R: ACATCTAATCCCCCATCGAAC        LBb1.3: ATTTTGCCGATTTCGGAAC KNAT4 genotyping salk_020216-L: AACTTTAGAAGCCGCTCAAGG     salk_020216-R: TGACAAGTTCTTGGTTGATTGG    N759461-L: GATCACCAAAAAGCTGGTACTC     N759461-R: CATGAAGTGGTCAAGCTCCTTGTC KNAT5 genotyping salk_000339-L: CTCTTCTCCGATCCCAAAAAC       salk_000339-R: AACGTGGTGTTGGAGTTGTTC   KNAT7 genotyping salk_002098-L: AAGTTTGGGCTTGGGCTTGAC     salk_002098-R: TTGCCTTGTCATCTTCCTGTTCA KNAT3 cloning ProKNAT3:GUS ProKNAT3-L: CAACATTTACGGGGGTTGTTACGT     ProKNAT3-R: CGCGAACCGCTCTCTTCCGCTATT KNAT4 cloning ProKNAT4:GUS ProKNAT4-L: CGCGGTGAACATGAAAAACTCT     ProKNAT4-R: CGTTTTCGTGTTGAATTTGTTTTTG  KNAT5 cloning ProKNAT5:GUS ProKNAT5-L: AAACTGGCCTATATGAAGAT      ProKNAT5-R: TGTTTTCCTGCGTTTTTGGG KNAT7 cloning ProKNAT7:GUS ProKNAT7-L: TCTTTTGTAAAAACGGTTTTAA      ProKNAT7-R: AACCTTGACACAAGACCGGA   38   Gene name Application Primer sequence (5' to 3') KNAT3 cloning  ProKNAT3: KNAT3-GFP  ProKNAT3-L: CAACATTTACGGGGGTTGTTACGT     KNAT3-R: CGCGAACCGCTCTCTTCCGCTATT KNAT7 cloning  ProKNAT7: KNAT7-GFP  ProKNAT7-L: TCTTTTGTAAAAACGGTTTTAA       KNAT7-R: GTGTTTGCGCTTGGACTTCAA ACTIN2 qRT-PCR ACTIN2-L: CCAGAAGGATGCATATGTTGGTGA     ACTIN2-R: GAGGAGCCTCGGTAAGAAGA CESA4 qRT-PCR CesA4-L: GGATCAGCTCCGATCAATTT     CesA4-R: ACCACAAAGGACAATGACGA CESA7 qRT-PCR CesA7-L: CAGGCGTACTCACAAATGCT     CesA7-R: TGTCAATGCCATCAAACCTT CESA8 qRT-PCR CesA8-L: ACGGAGAGTTCTTTGTGGCT     CesA8-R: GGTCTGTGTTGGAACAATGG IRX9 qRT-PCR IRX9-L: TTTGCGGGACTAAACAACAT     IRX9-R: ATCGGAGGCTTTGTCTCTGT IRX10 qRT-PCR IRX10-L:AATTGGCCTTATTGGAATCG     IRX10-R: TTCGTCCAAACAGACATGG PAL1 qRT-PCR PAL1-L: AAGATTGGAGCTTTCGAGGA     PAL1-R: TCTGTTCCAAGCTCTTCCCT PAL2 qRT-PCR PAL2-L: GAGGCAGCGTTAAGGTTGAG     PAL2-R: TTCTCGGTTAGCGATTCACC C4H qRT-PCR C4H-L: ACTGGCTTCAAGTCGGAGAT     C4H-R: ACACGACGTTTCTCGTTCTG 4CL1 qRT-PCR 4CL1-L: TCAACCCGGTGAGATTTGTA   4CL1-R: TCGTCATCGATCAATCCAAT  39   Gene name Application Primer sequence (5' to 3') C3H1 qRT-PCR C3H1-L: GTTGGACTTGACCGGATCTT     C3H1-R: ATTAGAGGCGTTGGAGGATG HCT qRT-PCR HCT-L: GCCTGCACCAAGTATGAAGA     HCT-R: GACAGTGTTCCCATCCTCCT CCoAOMT1 qRT-PCR CCoAMOMT1-L: CTCAGGGAAGTGACAGCAAA     CCoAMOMT1-R: GTGGCGAGAAGAGAGTAGCC CCR1 qRT-PCR CCR1-L: GTGCAAAGCAGATCTTCAGG     CCR1-R: GCCGCAGCATTAATTACAAA F5H1 qRT-PCR F5H-L: CTTCAACGTAGCGGATTTCA     F5H-R: AGATCATTACGGGCCTTCAC COMT1 qRT-PCR COMT1-L: TTCCATTGCTGCTCTTTGTC     COMT1-R: CATGGTGATTGTGGAATGGT CAD5 qRT-PCR CAD5-L: TTGGCTGATTCGTTGGATTA     CDA5-R: ATCACTTTCCTCCCAAGCAT   40   	Figure 2.1 The expression patterns of KNOX2 genes in Arabidopsis stems.  Histochemical localization of ProKNAT:β-GLUCURONIDASE (GUS) activity in 8 week-old inflorescence stems of ProKNAT3:GUS (a, e, i), ProKNAT4:GUS (b, f, j), ProKNAT5:GUS (c, g, k), and ProKNAT7:GUS (d, h, l) transgenic plants. Results shown are representative of more than 3 independent lines. (a-d) hand cross-sections from young upper inflorescence stems, and GUS signals were found in the cortex and vascular bundles. (e-h) hand cross-sections from the middle of inflorescence stems, showing GUS expression in cortex, interfascicular fibers and vascular bundles. (i-l) hand cross-sections from basal inflorescence stems. The promoter activities at the basal stems were detected in cortex, phloem and protoxylem, but not in the fiber cells. co, cortex; if, interfascicular fibers; p, phloem; x, xylem. px, protoxylem; mx, metaxylem. Scale bars = 30µm. 41   	Figure 2.2 Xylem vessel morphology of wild-type (WT) and KNOX2 mutant plants. Cross-sections of basal stem vascular bundles stained with phloroglucinol from 8-week-old WT (a), knat3 (b), knat4 (c), knat5 (d), knat7 (e), and knat3knat7 (f) plants; Arrows indicate collapsed xylem vessels. Scale bars = 30µm.    42    43   Figure 2.3 Cross-sections of stem vascular bundles in wild-type (WT) and different combinations of knox2 mutants. Stem sections stained with toluidine blue. A single representative vascular bundle is shown from each mutant. (a) WT; (b) knat7; (c) knat3knat7; (d) knat4-1knat7; (e) knat5knat7; (f) knat3knat4-2; (g) knat3knat5; (h) knat4-2knat5; (i) knat3knat4-2knat5; (j) knat3knat4-1knat7; (k) knat3knat5knat7. Scale bars = 30µm.                       44   Figure 2.4 Plant morphology and stem biomechanical properties of WT, knat3, knat7 and knat3knat7.   45   Figure 2.4 Plant morphology and stem biomechanical properties of WT, knat3, knat7 and knat3knat7.  	(a) Fifty-day-old knat3knat7 plants had a pendant stem phenotype compared with WT, knat3 and knat7 single mutants. Bar = 10cm. Tensile strength (b) and tensile stiffness (c) tests show that the bases of fresh stems (0-60mm from the base) of knat3knat7 had reduced tensile strength and tensile stiffness compared to WT or single mutants. 3-point flexural tests measured the stress at yield (d) and the modulus of elasticity (e) of both fresh and dried stems at 15mm from the base. White bars, WT; light-grey bars, knat3; dark-grey bars, knat7; black bars, knat3knat7. The error bars reperesent means ± SD. Statistical differences among the samples are labeled with different letters (n=5-10; P<0.05, one-way ANOVA followed by Turkey’s post hoc test).  						46   	Figure 2.5 Plant morphology of WT, knat3knat7, knat4knat7, knat5knat7.  Forty-day-old plants of knat3knat7 have a pendent stem phenotype compared with WT, knat4knat7 and knat5knat7. Bar = 10cm. 47   		Figure 2.6 Interfascicular fibers of WT, knat3, knat7 and knat3knat7 inflorescence stems.  Cross-sections from the bases of stems of 8-week-old WT (a), knat3 (b), knat7 (c), and knat3knat7 (d) plants stained with phloroglucinol. Bars = 20µm. (e) Quantification of fiber wall thickness in WT, single and double mutants. knat7 had a significant increase in interfascicular fiber wall thickness, while the double mutants knat3knat7 had a dramatic decrease in interfascicular fiber wall thickness. Statistical differences among the samples are labeled with different letters (P<0.01, one-way ANOVA followed by Turkey’s post hoc test). Error bars = 2 × SD. n= 30-50.    48    Table 2.2 Cell-wall monosaccharide content of WT, knat3, knat7 and knat3knat7 stems Sample Glucose Xylose Mannose Arabinose Rhamnose Galactose WT 352.4±3.2a 120.6±1.2a 17.9±0.9a 9.6±0.3a 7.2±0.2a 15.8±0.3a knat3 330.2±7.2b 112.9±2.9b 15.4±0.8a 8.8±0.2b 6.7±0.5a 14.1±0.8a knat7 324.8±2.9b 121.2±0.9a 11.3±0.7b 9.5±0.2ab 7.2±0.2a 14.1±0.1a knat3knat7 274.4±2.4c 69.0±2.4c 10.8±0.4b 16.8±0.1c 10.2±0.4b 22.3±0.8b  Cell-wall monosaccharide contents were determined by HPLC following secondary acid hydrolysis and are represented as µg per mg basal stem dry weight (0-11 cm from the base). The knat3knat7 double mutants had significantly reduced glucose, xylose and mannose, but considerably increased arabinose, rhamnose and galactose contents compared with WT. Data are means ± SD values from three technical replicates for a single experiment, and were able to be repeated by three times with three different biological experiments. Statistical differences among the samples are labeled with different letters (P<0.01, one-way ANOVA followed by Turkey’s post hoc test).             49    Table 2.3 Cellulose and lignin contents of WT, knat3, knat7 and knat3knat7 stem bases Sample Cellulose content  (mg/100 mg ± SD) Acid-insoluble lignin (mg/100 mg ± SD) Acid-soluble lignin (mg/100 mg ± SD) Total lignin  (mg/100 mg ± SD) WT 27.16±0.26a 18.59±0.38 2.3±0.01 20.89±0.38a knat3 27.58±1.07a 18.45±0.40 2.22±0.06 20.67±0.35a knat7 26.47±1.20a 20.61±0.20 1.86±0.04 22.48±0.23b knat3knat7 21.84±0.65b 17.89±0.09 2.89±0.05 20.79±0.06a  Cellulose and lignin contents were measured from the bases of inflorescence stems (0-11cm from the base). knat3knat7 had lower cellulose content and acid-insoluble lignin, but more acid-soluble lignin compared with wild-type plants. No change in total lignin was detected in double mutant stems. Statistical differences among the samples are labeled with different letters (P<0.01, one-way ANOVA followed by Turkey’s post hoc test). Values are means ± SD for three technical replicates. The data was repeated by three times with three different biological experiments.     50   		Figure 2.7 Lignin composition of WT, knat3, knat7 and knat3knat7 stem cell walls.  (a) Syringyl to guaiacyl (S/G) lignin monomer ratio of dried stem bases (0-11cm from the base), as determined by thioacidolysis and histochemical staining with Mäules reagent. knat3knat7 stem had significantly lower S/G lignin ratio. Error bars represent the standard deviations; Statistical differences among the samples are labeled with different letters (P<0.01, one-way ANOVA followed by Turkey’s post hoc test). n = 3.  (b) Mäule stain produces red colour in syringyl lignin-rich cell walls.  Cross-sections from the bases of stems of 8-week-old WT, knat3, knat7, and knat3knat7 plants. The double mutant knat3knat7 had less S lignin deposition. Scale bars = 30 µm.  	   51   	Figure 2.8 Complementation of knat3knat7 phenotypes with ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP. (a) Xylem vessel morphology of WT, knat3knat7 and complemented lines. Cross-sections of basal stem vascular bundles with Mäule staining from 8-week-old plants show that the irx phenotype of knat3knat7 was complemented. Scale bars = 30µm.  52   Figure 2.8 Complementation of knat3knat7 phenotypes with ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP. (b) Fifty-day-old plants showing the pendent phenotype of knat3knat7 was complemented in both transformants. Bar = 10cm. (c) and (d) showing the interfascicular fibers in WT, knat3knat7 and complementation lines. The reduced wall thickness in knat3knat7 was restored to an increased level relative to the wild-type by expressing KNAT3 under the control of its native promoter at knat3knat7 background, while the thickness was fully recovered to wild-type by expressing KNAT7 back to knat3knat7 mutants. Error bars represent the standard deviations. Statistical differences among the samples are labeled with different letters (P<0.01, one-way ANOVA followed by Turkey’s post hoc test). Scale bars = 20µm. n= 30-50.              		53   	54   Figure 2.9 Expression patterns of KNAT3 and KNAT7 in the root and stem. Expression of ProKNAT3:KNAT3-GFP and ProKNAT7:KNAT7-GFP were detected in the protoxylem and metaxylem, and in the interfascicular fiber cells in 6-week-old inflorescence stems. Confocal laser scanning microscopy images of roots counterstained with propidium iodile. px, protoxylem; mx, metaxylem. Scale bars = 30µm.                 55   		Figure 2.10 Interfascicular fibers have thicker secondary cell walls in 35S:KNAT3 basal stems. (a) Cross-sections of the bases of 8-week-old inflorescence stems of WT and 35S:KNAT3 plants stained with Toluidine Blue. Bars = 20µm.  (b) Quantification of interfascicular fiber wall thickness in WT and KNAT3 overexpression lines. The results shown are representative of two independent overexpression lines.  Compared to wild-type, the overexpression lines appeared to have thicker secondary cell walls of interfascicular fibers. Asterisk (*) indicates the sample is significantly different from WT at P <0.01 determined by Student’s T-test. Error bars represent the standard deviations. n= 30-50.   		 56    Figure 2.11 Expression of secondary cell wall-related genes in lower stems of WT, knat3, knat7 and knat3knat7 double mutants.  RNA was extracted from the basal segment of inflorescence stems (0-11cm) of 8-week-old WT, knat3, knat7 and knat3knat7 plants for qRT-PCR analysis. Error bars represent the standard deviations of three biological replicates.  57    Figure 2.11 Expression of secondary cell wall-related genes in lower stems of WT, knat3, knat7 and knat3knat7 double mutants.  (a) Expression of CELLULOSE SYNTHASE (CESA) genes (CESA4, CESA7, CESA8) and hemicellulose (IRREGULAR XYLEM 9, IRX9; IRX10) biosynthetic genes. CESA4, CESA7, CESA8, and IRX9 were dramatically up-regulated in knat3knat7 double mutants, while IRX10 was significantly down-regulated compared with WT as determined by Student’s T-test (P<0.01).  (b) Expression of lignin biosynthetic genes. PHENYLALANINE AMMONIA LYASE 1 (PAL1), PAL2; CINNAMATE-4-HYDROXYLASE (C4H); 4-COUMARATE-COA LIGASE 1 (4CL1); COUMARATE 3-HYDROXYLASE 1 (C3H1); HYDROXYCINNAMOYL-COA:SHIKIMATE HYDROXYCINNAMOYL TRANSFERASE (HCT); CAEEROYL-COA 3-O-METHYLTRANSFERASE 1 (CCoAOMT1); CINNAMOYL-COA REDUCTASE 1 (CCR1); FERULATE-5-HYDROXYLASE (F5H); CAFFEIC ACID O-METHYLTRANSFERASE 1 (COMT1); CINNAMYL ALCOHOL REDUCTASE 5 (CAD5). C4H, C3H1, CCoAMT1, and CCR1, were elevated 1.5- to 2-fold in the basal segments of knat3knat7 inflorescence stems, but the transcript abundance of F5H, a key enzyme for S lignin biosynthesis, was considerably decreased in knat3knat7 mutants compared with WT at P<0.01 as determined by Student’s T-test.          58    Figure 2.12 Venn diagram and GO term enrichment analysis of differentially expressed genes (DEGs) of knat3 versus WT, knat7 versus WT, and knat3knat7 versus WT.  59   Figure 2.12 Venn diagram and GO term enrichment analysis of differentially expressed genes (DEGs) of knat3 versus WT, knat7 versus WT, and knat3knat7 versus WT. (a and b) Venn diagram (a) and GO biological processes (b) of genes with increased expression in knat3 versus WT, knat7 versus WT and knat3knat7 versus WT.  (c and d) Venn diagram (c) and GO biological processes (b) of genes with decreased expression in knat3 versus WT, knat7 versus WT and knat3knat7 versus WT Go terms were identified using PANTHER (https://www.arabidopsis.org/tools/go_term_enrichment.jsp) with default significant level (P<0.05).         60   		Figure 2.13 MAPMAN schematic providing a metabolic overview of the differential gene expression between WT and knat3knat7 stems.  The different colors represent the log2 (FC) values of the gene expression levels in knat3knat7 versus wild-type stems: red, down-regulation; white, no change; blue, up-regulation. 	61   			Figure 2.14 MAPMAN schematic providing a metabolic overview of the differential gene expression levels in knat3 and knat7 stems compared with WT.  The different colors represent the log2 (FC) values of the gene expression levels in knat3 and knat7 versus wild-type stems: red, down-regulation; white, no change; blue, up-regulation.       62   Table 2.4 Expression changes of secondary cell wall-related genes in knat3knat7 stems  Gene Name Gene ID knat3knat7 VS WT RNA-seq Sig. knat3knat7 VS WT RNA-seq FPKM log2(fold_change) qRT-PCR log2(fold_change) CESA4 AT5G44030 0.13 no 2.25 CESA7 AT5G17420 1.39 yes 2.66 CESA8 AT4G18780 1.51 yes 3.03 IRX9 AG2G37090 2.32 yes 3.15 IRX10 AT1G27440 -2.93 yes -1.65 REV AT5G60690 0.56 no 0.66 PAL1 AT2G37040 -0.76 yes -0.21 PAL2 AT3G53260 -0.56 yes 0.02 C4H AT2G30490 -0.04 no 1.00 4CL1 AT1G51680 -0.84 yes 0.23 C3H1 AT2G40890 0.12 no 0.62 HCT AT5G48930 -0.44 no 0.40 CCoAOMT1 AT4G34050 0.61 yes 1.00 CCR1 AT1G15950 0.00 no 0.89 F5H1 AT4G36220 -2.65 yes -1.09 COMT1 AT5G54160 0.15 no 0.21 CAD5 AT4G34230 0.19 no 0.34  RNA-seq experiments were carried out using the top 1-15cm of inflorescence stems of knat3knat7. qRT-PCR analysis was done using the basal stems of knat3knat7 double mutants. FPKM (fragments per kilobase of exon per million fragments mapped) was used to estimate the abundance of gene transcripts. The significance (Signi.) for RNA-Seq was determined by the q value (q<0.01).  63   Chapter 3: Ovate Family Proteins are associated with Brassinosteroid homeostasis and function in cotyledon development by interacting with Nucleosome Assembly Protein 1 in Arabidopsis  3.1    Introduction The ovate family of transcription factors is named after a tomato ovate mutant that has elongated fruit (Liu et al., 2002). In Arabidopsis, Ovate Family Proteins (OFP) interact with a number of TALE homeodomain proteins and regulate their subcellular localization in plant cells to control plant development (Hackbusch et al., 2005). More specifically, nine ovate family proteins were found to be involved in the interaction network of TALE proteins and OFP1 could change the nucleus localization of KNAT1 or BLH1 to the cytoplasm by protein-protein interactions (Hackbusch et al., 2005). OFPs were defined by their conserved C-terminal 60-70 amino acid ovate domain, and they are only found in land plants so far (Hackbusch et al., 2005; Wang et al., 2007, 2011, 2016;  Liu et al., 2014).  In Arabidopsis, there are 19 OFP genes, which were classified into 3 clades and 8 sub-groups according to the phylogenetic analysis (Liu et al., 2014). Among them, OFP1, OFP2, OFP3, OFP4 and OFP5 are close homologs. OFP1 has been reported to have multiple functions in plants, such as inhibiting cell elongation through repressing GA20ox (Wang et al., 2007), facilitating DNA repair by interacting with the Ku protein (Wang et al., 2010) and forming a complex with BLH3 to regulate the flowering time (Zhang et al., 2016). OFP1 and OFP4 were also shown to interact with KNAT7, negatively regulating secondary cell wall formation, and ofp4 mutants phenocopied knat7, with irx and increased fiber cell wall thickness (Li et al., 2011; Liu and Douglas, 2015). OFP5 was found to regulate the embryo development, by interacting with KNAT3 and BLH1 (Pagnussat et al., 2007). The previous chapter has demonstrated the roles of KNAT3 and KNAT7 in secondary cell wall development, and the goal of this chapter was to test if the uncharacterized OFP family 64   members, OFP5 and its homologs, OFP2 and OFP3, also play a role in secondary wall formation.  In this study, I took the reverse genetics approach to identify the secondary cell wall phenotypes in ofp single mutants. However, all of the ofp single mutants did not show any distinguishable phenotype from wild type, even for the ofp4, which was previously described as a secondary cell wall mutant (Li et al., 2011). Instead, I confirmed OFP overexpression phenotypes with kidney-shaped cotyledons (Wang et al., 2011). I also found the hypocotyls of OFP1, OFP2, OFP4 and OFP5 overexpression plants had cell swelling, disordered microtubules, and dark-grown de-etiolated phenotypes, resembling brassinosteroid (BR) deficient mutants. Exogenous BR treatment partially rescued the hypocotyl phenotypes of OFP overexpressing plants, which provided new insights into OFP functions in maintaining BR homeostasis in Arabidopsis.   To identify novel OFP4 interaction partners, I used yeast two-hybrid and bimolecular fluorescence complementation assays to identify Nucleosome Assembly Protein 1 (NAP1) family proteins. NAP1;1 and NAP1;2, were shown to interact with OFP4 in vitro and in vivo. Both YFP-NAP1;1 and YFP-NAP1;2 fusion proteins were localized abundantly in the cytoplasm, associated with the ER. By phenotypic analysis of higher-order loss-of-function mutants, I found that nap1;1nap1;2nap1;3 triple mutant; the ofp4nap1;1nap1;2nap1;3 quadruple mutant; and the ofp1ofp2ofp3ofp4ofp5 quintuple mutant exhibited similar phenotypes, with altered cotyledon shapes and reduced cotyledon width/length ratio. In addition, the kidney-shaped cotyledon phenotype of OFP4 overexpressing plants was suppressed in the nap1;1nap1;2nap1;3 triple mutant background, suggesting OFP4 functions in cotyledon development at least partially depend on NAP1 proteins. All together, these data indicate that the complex of OFP and NAP1 plays a significant role in the cotyledon development in Arabidopsis seedlings.   65   3.2    Materials and methods 3.2.1   Plant materials and growth condition Arabidopsis thaliana ecotype Columbia was used as wild type in all experiments, and all the transgenic lines and mutants are also in the Columbia background. T-DNA insertion lines for ofp2 (SALK_122550), ofp3 (GABI_167F01) and ofp5 (SALK_203823) were obtained from the Arabidopsis Biological Resource Center (ABRC). The ofp1 (SM_3_21689), ofp4 (SALK_022396) described in (Wang et al., 2007; Li et al., 2011) were used for ofp1, ofp4 phenotypic analyses. Homozygous T-DNA insertion lines were screened by PCR using gene-specific primers (Table 3.1). The nap1;1 (SALK_013610), nap1;2 (SAIL_84_B01), nap1;3 (SALK_131746) and nap1;1nap1;2nap1;3 were kindly obtained from Dr. Aiwu Dong’s group (Liu et al., 2009) and the homozygous transgenic plants pUBQ1: mRFP-TUB6 were kindly obtained from Dr. Chris Ambrose (2011). Quadruple and quintuple mutants of ofp4nap1;1nap1;2nap1;3 and ofp1ofp2ofp3ofp4ofp5 were generated by genetic crossing, and the genotypes confirmed with PCR.   In all experiments, Arabidopsis seeds were sterilized with 70% ethanol and sown on half Murashige and Skoog (MS) medium with 1% sucrose, then cold-treated at 4°C for 48 hours in the dark. For normal growth conditions, after cold treatment, seeds were grown at 20°C under a 16/8 h (light/dark) photoperiod at about 120 µmol m-2 s-1 light for 7 to 10 days. For dark growth conditions, plates were wrapped with aluminum foil and placed vertically in the growth chamber for 7 days. For whole plant growth, seedlings were transferred to soil and grown under long-day conditions (16/8 h light/dark cycle) at ~100 µmol m-2 s-1 light, 20°C in growth chambers for further analysis.  3.2.2   RNA isolation and quantitative RT-PCR Total RNA was extracted from tissues using Qiagen RNeasy columns following the manufacturer’s instructions, and treated with DNase I (Qiagen). 2 µg RNA was transcribed into cDNA using the Omniscript RT kit (Qiagen) according to the manufacturer’s 66   instructions. For qRT-PCR of the basal stem segments, PCR amplification was performed using a CFX ConnectTM real-time system (Bio-Rad), using 40 quantitative PCR cycles that were run under the following parameters: denaturation step, 95°C for 20 sec; annealing step, 55°C for 30 sec; elongation step, 72°C for 1 min. ACTIN2 was used as the reference housekeeping gene. All primers are listed in Table 3.1. The calculation of differences in gene expression was used as described by Bhargava et al. (2010). Three biological replicates were performed and each measurement consisted of three technical replicates.  3.2.3   Cloning and plant transformation To generate the overexpression constructs, the full-length open-reading frames of OFP1, OFP2, OFP3, OFP4, OFP5, NAP1;1, and NAP1;2 were amplified by PCR from the cDNA of Col-0 ecotype prepared as previously described, and sub-cloned into the pCR8/GW/TOPO entry vectors, respectively. After verifying the nucleotide sequences of the amplified fragments, they were cloned into the binary vector pEarlyGate 101, pEarlyGate 104 and PMDC43 for generating the 35S: OFP1-YFP, 35S: OFP2-YFP, 35S: OFP3-YFP, 35S: OFP4-YFP, 35S: OFP5-YFP, 35S: YFP-NAP1;1, 35S: YFP-NAP1;2 and 2 x 35S: GFP-OFP4 constructs, respectively. For generating BiFC constructs, the clone of OFP4 was transferred to the Gateway destination vector pSAT4-DEST-nYFP-C1 and NAP1;1/NAP1;2 were cloned into pSAT5-DEST-cYFP-C1 (Citovsky et al., 2006), to generate OFP4 fusions to the N-terminal half of YFP, and NAP1;1/NAP1;2 fusions to the C-terminal half of YFP. To generate ProOFP2:OFP2-GUS, ProOFP3: GUS, and ProOFP5: GUS constructs, the full genome sequence including the promoter region of OFP2 (2462bp), or only the promoter region of OFP3 (2143bp), and OFP5 (2156bp) were cloned into the binary vector PMDC163. Gene-specific oligonucleotides used for cloning and construct generation are shown in Table 3.1.  All the binary constructs were introduced into Agrobacterium tumefaciens strain GV3101 for plant transformation. The wild-type Columbia was transformed for expression pattern and overexpression analysis. The 2 x 35S: GFP-OFP4 construct was transformed to Col-0 and 67   nap1;1nap1;2nap1;3 mutants. The 35S:OFP-YFP constructs were also transformed to pUBQ1: mRFP-TUB6 for microtubule analysis. The floral dip method (Clough and Bent, 1998) was used for generating transgenic plants. T1 plants for 35S:OFP-YFP were selected with Basta resistance while the other transformants were selected with hygromycin resistance.   3.2.4   GUS expression assay The GUS activity was assayed using 10-day-old seedlings, which were incubated with 90% acetone by vacuum infiltration, then moved to an ice bath for 30 min, followed by washing with 0.1M sodium phosphate buffer three times and incubating tissues in a solution containing 0.1M sodium phosphate buffer (PH 7.0), 1mM substrate 5-bromo-4-chloro-3-indolyl-β-D-glucuronide (X-Gluc), 0.5mM potassium ferricyanide and 0.01%(v/v) Triton X-100 at 37°C for 1 h to overnight. The resulting stained tissues were fixed with FAA (50 [v/v] ethanol, 5% [v/v] acetic acid and 10% [v/v] formaldehyde), and observed with an Olympus AX70 light microscope.   3.2.5   Transient expression in N. benthamiana and BiFC assay To determine the subcellular localization, the leaves of 4-week-old Nicotiana benthamiana plants were agroinfiltrated with OFP and NAP1 overexpression constructs respectively using a syringe (1mL) without needle, and then the infiltrated tobacco plants were placed in the growth chamber for 72 to 96 h. For co-expression and BiFC assays, the Agrobacterium cultures with different expression constructs were mixed in the infiltration medium to a final OD600 of 0.05, and the infiltration in tobacco leaves was carried out as described previously (Velasquez et al., 2011).     3.2.6   Light and confocal microscopy For observing secondary cell wall phenotypes, the base of freshly harvested 8-week-old inflorescence stems were hand-sectioned, stained with aqueous 0.05% toluidine blue O for 1-68   2 min, mounted with water and viewed using an Olympus AX70 light microscope. To observe the cotyledon vascular pattern, one-week-old seedlings were cleared with 90% acetone, washed with phosphate buffer and mounted with chloral hydrate solution. All the cotyledon phenotypes were viewed using the dissecting microscope (Zeiss), and the one-week-old hypocotyls grown in the dark were viewed using the Olympus AX70 light microscope. Image J was used to measure the dark-grown hypocotyl length, and the cotyledon width and length. Statistical analysis was performed by one-way ANOVA followed by Tukey’s post hoc test and Student’s T-test.  YFP-fusion, GFP-fusion and mRFP-marker proteins were observed on a Perkin-Elmer UltraView VoX spinning disk confocal mounted on a Leica DMI6000 inverted microscope with a Hamamatsu 9100-02 CCD camera. The microscope used for all live-cell imaging was fit with the following excitation filters: YFP (514nm), GFP (488nm), and RFP (561nm). Image J software was used for image processing.  3.2.7   BR treatment To test the responses of OFP OX seedlings to exogenous BR, plants were germinated and grown on one-half-strength MS medium with different concentrations of epibrassinolide (epiBL, Sigma). 2 mg epiBL was dissolved in 1ml 80% EtOH as a stock. And the plates were placed vertically in the growth chamber equipped with light for one week before photographing. Image J software was used for hypocotyl length measurements. Statistical analysis was performed by Student’s T-test and one-way ANOVA followed by Tukey’s post hoc test.  3.2.8   Yeast two-hybrid assays The yeast two-hybrid screen was performed using the DUALhybrid Kit (Dualsystems Biotech) following the manufacturer’s instructions. The full-length of OFP4 cDNA was cloned into pLexA-N vector as the bait and the prey used an Arabidopsis cDNA library in the 69   pGAD-HA vector obtained from the Dualsystems Biotech. The yeast NMY51 strains were co-transformed with DBD-OFP4 and the empty cDNA library, and spread on selection media lacking Trptophan (Trp), Leucine (Leu), and Histidine (His) and supplemented with different concentrations of 3-Amino-1,2,4-triazole (3-AT) (SD/-Trp/-Leu/-His/3-AT), to determine the conditions of the screen according to the self-activation level of the bait. Then the cDNA library (Dualsystems Biotech) was transformed into NMY51 expressing the DBD-OFP4 bait and interactors were selected on plates that contained the concentration of 3-AT determined in the pilot screen. Plasmids from cells growing under selection and expressing both the reporter genes were isolated and all positive interactors were confirmed by the bait-dependency test.  To confirm the interactions, the ProQuest yeast two-hybrid system  (Invitrogen) was performed as described previously (Guo et al., 2009). OFP4 was cloned into the pDEST32 as a bait vector and NAP1;1/NAP1;2 were cloned into pDEST22 as the prey vector. The positive control used the known interactors MYB75 and TT8 (Zimmermann et al., 2004) and the negative control was the interaction between OFP4 and the empty prey vector. Positive interaction was determined by the yeast growth on the triple selective SD medium lacking leucine, tryptophan and histidine but supplemented with 40mM 3-AT, or the selective SD medium lacking leucine, tryptophan and uracil.   3.3    Results 3.3.1  OFP loss-of-function mutants have no secondary cell wall defects OFP1, OFP2, OFP3, OFP4, and OFP5 are close homologs in the same class in Arabidopsis. To investigate the potential roles of OFP2, OFP3 and OFP5 in regulating secondary cell wall formation, and to confirm the secondary cell wall functions of OFP1 and OFP4, the T-DNA insertion alleles of ofp2 (salk_122550), ofp3 (GABI_167F01) and ofp5 (salk_203823) were obtained from the Arabidopsis Biological Resources Centre (ABRC) and the null alleles of ofp1 and ofp4 were obtained from Liu and Douglas (2015). PCR-based genotyping 70   was carried out to identify homozygous mutants and loss-of-function mutant alleles were confirmed by RT-PCR. Stem cross-sections were examined at the bases of mature inflorescence stems for each genotype to determine the xylem or interfascicular fiber morphology.   Previous studies have shown that OFP4 loss-of-function mutants phenocopy that of knat7, with irregular xylem (irx) and thicker fiber cell wall phenotypes (Li et al., 2011; Liu and Douglas, 2015). However, in my study, there were no phenotypic differences in the vascular bundles detected between ofp4 mutants and wild type (Figure 3.1a, e) after testing at least three trials. ofp1, ofp2, ofp3 and ofp5 single mutants were similar to ofp4, exhibiting normal xylem vessels and no obvious changes in interfascicular fibers compared with that in wild type (Figure 3.1). Also, I did not observe other significant morphological defects in the stems of OFP loss-of function mutants.   3.3.2     OFP genes are expressed in Arabidopsis seedlings As there were no secondary cell wall phenotypes in loss-of-function mutants, I turned my attention to OFP gain-of-function phenotypes, focusing on OFP1, OFP2, OFP3, OFP4 and OFP5. The overexpression plants of each OFP were generated respectively, by transforming Col-0 plants with a binary vector containing OFP-YFP transgene driven by the strong CaMV 35S promoter. Plants overexpressing OFP1, OFP2, OFP4 and OFP5 presented kidney shaped cotyledons, while OFP3 overexpression plants did not display any significant difference compared with wild type (Figure 3.2). The OFP overexpression phenotypes are consistent with previous findings (Wang et al., 2011; Li et al., 2011).   Since the transgenic plants of 35S:OFP have cotyledon phenotypes, and both OFP1 and OFP4 are expressed in the veins and other tissues in cotyledons (Wang et al., 2007; Li et al., 2011), the expression pattern of the other three OFP genes, OFP2, OFP3 and OFP5, in cotyledons were tested. By generating the transgenic lines containing ProOFP2: OFP2-GUS, ProOFP3: GUS and ProOFP5: GUS, respectively, I found that OFP2, OFP3 and OFP5 had 71   similar expression patterns in 10-day-old seedlings, and they were detected in the cotyledon tips, veins, leaf trichomes and the shoot apex (Figure 3.3a-f). They were also expressed in the vascular cylinder in the seedling roots (Figure 3.3g-i).  3.3.3     OFP proteins are localized to the nucleus and cytoplasm  To determine the localization of OFP proteins in plant cells, 35S:OFP-YFP constructs were generated and expressed in the leaf lamina of tobacco plants transiently. The YFP fluorescence of OFP1 and OFP5 accumulated in the nucleus and labeled structures reminiscent of the cortical cytoskeleton in cytoplasm (Figure 3.4a, e), which is consistent with previous findings (Hackbusch et al., 2005). OFP2-YFP and OFP4-YFP proteins were localized in the nucleus and distributed throughout the cytoplasm (Figure 3.4b, d). Interestingly, OFP3-YFP was exclusively located in the nucleus (Figure 3.4c). All the YFP signals of OFP proteins were highly concentrated in the nucleolus (Figure 3.4).   3.3.4     OFP overexpression plants have cell swelling and de-etiolated phenotypes in hypocotyls In the four-day-old seedlings, overexpressing OFP proteins did not only affect the cotyledon leaf phenotype, but also the cell elongation and expansion in hypocotyl epidermal cells (Figure 3.5). Overexpression of OFP1, OFP2, OFP4 and OFP5 caused epidermal cells to swell in the hypocotyls (Figure 3.5), which led to the hypothesis that microtubule function could be impaired in these cells. To check if the microtubule organization has been changed in the epidermal cells, the binary vector of 35S:OFP-YFP was transformed into the RFP-βTubulin6 (pUBQ1: mRFP-TUB6) homozygous transgenic plants (Ambrose et al., 2011). Confocal observations showed that most cortical microtubules in wild type exhibited transverse and oblique orientations in the hypocotyl epidermal cells (Figure3.5c), while after overexpressing OFP1, OFP2, OFP4 and OFP5, the microtubule orientation was changed to predominantly longitudinal and oblique (Figure3.5f, i, l, o). The changes in the cortical 72   microtubule orientation may explain the hypocotyl cell swelling phenotype in Arabidopsis seedlings.   In dark growth conditions, the overexpression seedlings of OFP1, OFP2, OFP4 and OFP5 exhibited de-etiolated hypocotyl phenotypes (Figure 3.6a, b). The hypocotyl length was considerably shortened in the OFP overexpression plants compared with that of wild-type plants in one-week-old dark-grown seedlings (Figure 3.6b). Taken together, the hypocotyl elongation and epidermal cell expansion were significantly affected by overexpressing OFP proteins in Arabidopsis seedlings.    3.3.5     OFP overexpression plants show responses to exogenous BR treatment in hypocotyls  In the OFP overexpression plants, the microtubule disorganization and de-etiolation phenotypes in the dark mimic brassinosteroid (BR)-deficient or BR–signalling mutants. To test if these phenotypes are the result of BR-deficiency or BR-signalling, exogenous BR was added to overexpressing OFP plants. Without any synthetic BR (epiBL) treatment, the hypocotyl lengths of light-grown OFP2 overexpression seedlings were significantly shorter than those of wild type (Figure3.6c, d). When germinated on ½ MS medium containing epiBL, both wild type and OFP2 overexpression (OFP2 OX) seedlings showed a considerable increase in hypocotyl lengths with the increasing concentrations of epiBL (Figure 3.6c, d). Compared with the hypocotyl length of wild type seedlings growing on normal ½ MS medium, hypocotyl lengths of OFP2 OX were significantly shorter at 0.2 µM epiBL, but there was a sign of the rescue when epiBL concentration was increased to 0.5 µM (Figure 3.6d). The hypocotyl lengths of OFP2 OX at 0.5 µM epiBL were comparable with those of the wild type without any treatment (Figure 3.6d). The same rescue responses were found in the other OFP1, OFP4 and OFP5 overexpression plants upon treating with various concentrations of epiBL (Figure 3.7). However, the kidney-shaped cotyledon phenotypes of OFP overexpression plants were not changed by exogenously applied epiBL. These data 73   indicate that the BR signaling pathway was not blocked in the hypocotyls of OFP overexpression plants, and their hypocotyl phenotypes may be associated with BR deficiency.   3.3.6     OFP4 interacts with Nucleosome Assembly Protein 1 OFP4 has been reported to form different complexes with KNAT7 and BLH6, to regulate secondary cell wall formation (Li et al., 2011; Liu et al., 2014). In this study, OFP4 overexpressing plants exhibited pleiotropic phenotypes (Figure 3.2, 3.5, 3.6). To identify other possible components in complexes including OFP4 in regulating different aspects of plant development, a yeast two-hybrid screen was performed. Arabidopsis OFP4 cDNA was cloned into pLexA-N vector as the bait and an cDNA library made from leaves, inflorescence stems and roots of Arabidopsis, in the pGAD-HA vector as the prey (obtained from Dualsystems Biotech). The DBD-bait and AD-prey were co-transformed into yeast NMY51 strains and spread on selection media lacking Trp, Leu, and His and supplemented with 1mM 3-Amino-1,2,4-triazole (3-AT) (SD/-Trp/-Leu/-His/3-AT). Approximately 3 x 106 transformants were screened and 95 positive clones were isolated and sequenced, which could activate the reporter genes upon co-expression with the DBD-OFP4 bait but not with the negative controls. Sequence alignments were performed and five different genes, Nucleosome Assembly Protein 1;1 (NAP1;1), Nucleosome Assembly Protein 1;2 (NAP1;2), TON1 Recruiting Motif 20 (TRM20), Bromodomain and Extraterminal Domain Protein 9 (BET9) and Spiral 2 (SPR2) were identified (Table 3.2). Among them, NAP1;2 was found the most times with 87 positive clones (Table 3.2). NAP1;1 and NAP1;2 genes encode proteins that belong to the same Nucleosome Assembly Protein 1 (NAP1) family in Arabidopsis, although NAP1;1 was only identified once in my experiment (Table 3.2). So I subsequently focused my work on interactions between OFP4 and NAP1 proteins.  To confirm the interactions between OFP4 and full-length NAP1 proteins encoded by NAP1;1 or NAP1;2 cDNA, a yeast two-hybrid assay was carried out using another yeast strain MaV203. An OFP4 cDNA was cloned into a bait vector and the full-length NAP1;1 74   and NAP1;2 cDNAs were cloned into the prey vector. Figure 3.8(a) shows that the yeast cells expressing an OFP4-DBD fusion and a NAP1;1-AD fusion or a NAP1;2-AD fusion interacted well in the system, as judged by growth on both His- and Ura- selective media, which confirms the yeast two-hybrid screen result using the strain NMY51.   To test OFP4-NAP1 protein-protein interactions in plant cells, the Bimolecular Fluorescence Complementation (BiFC) was used (Hu et al., 2002). OFP4 and NAP1 were fused to N- and C-terminal fragments of enhanced Yellow Fluorescent Protein (OFP4-nYFP and NAP1;1/NAP1;2-cYFP), respectively. Different combinations of fusion constructs were transformed into Nicotiana benthamiana leaves and the complete Yellow Fluorescent protein (YFP) would be generated only when two proteins are able to interact with each other. The fluorescence was detected in the cytoplasm when OFP4-nYFP and NAP1;1 or NAP1;2-cYFP were co-transformed into tobacco leaves, while no fluorescence was observed when OFP4-nYFP was co-expressed with empty-cYFP that served as a negative control (Figure 3.8b). These results indicated that OFP4 interacted with NAP1;1 and NAP1;2 in vitro and in vivo.    3.3.7     NAP1;1 and NAP1;2 proteins are localized to the ER membrane in epidermal cells of N. benthamiana leaves The BiFC analysis indicates that OFP4 interacted with NAP1;1 and NAP1;2 in cytoplasm, with a structure reminiscent of the endoplasmic reticulum (ER) membrane adjacent to the cell nucleus (Figure 3.8b). To test the localization of NAP1;1 and NAP1;2 proteins in plant cells, and to determine if NAP1;1 and NAP1;2 are associated with the ER membrane, 35S:YFP-NAP1;1 or 35S:YFP-NAP1;2 constructs were generated, and co-transformed with the ER-marker HDEL-RFP (Napier et al., 1992) into epidermal cells of tobacco leaves. YFP-NAP1;1 and YFP-NAP1;2 fused proteins gave strong fluorescence labeling a reticulate structure in the epidermal cells (Figure 3.9 a,d), and the yellow fluorescence colocalized with the red fluorescence signals coming from the ER maker, which indicated NAP1;1 and 75   NAP1;2 were localized to the ER membrane in epidermal cells of tobacco leaves (Figure 3.9c,f).   3.3.8     NAP1 and OFP4 loss-of-function mutants have altered cotyledon shapes To investigate the roles of NAP1 in plant development, all the homozygous NAP1 loss-of-function mutants were obtained from Dr. Aiwu Dong’s group, which have been confirmed to be null alleles (Liu et al., 2009). As OFP4 had been reported to interact with KNAT7 to regulate secondary cell wall development, and knat7 mutants displayed wall phenotypes (Li et al., 2011; Li et al., 2012), I tested the potential secondary cell wall roles of our identified OFP4 interactors, NAP1;1 and NAP1;2. Stem cross-sections were taken from the bases of nap1;1 and nap1;2 mutants, and no secondary cell wall defects were observed in their stems compared with that of wild-type plants (Figure 3.10a-c). There are four NAP1 genes in Arabidopsis, and NAP1;1, NAP1;2 and NAP1;3 genes have previously been shown to be expressed ubiquitously, while the NAP1;4 gene was tissue-specifically expressed only in root segments and pollen grains (Zhu et al., 2006; Liu et al., 2009). Because of the gene redundancy, we focused on investigating the phenotypes of obtained triple mutants nap1;1nap1;2nap1;3. The xylem or interfascicular fiber morphology in triple mutants was still indistinguishable from wild type (Figure 3.10d), which was similar to ofp4 mutants. These data suggest that OFP4 and NAP1 may not be involved in secondary cell wall development.    Interestingly, we observed some changes in cotyledons of the triple mutants, nap1;1nap1;2nap1;3, which presented the oblong shape compared with the round ones in wild type seedlings (Figure 3.11a). All the other nap1 single and double mutants did not show any significant phenotypes compared with wild type in our experiments, although nap1;1 was previously reported to have enlarged size in early development (Galichet and Gruissem, 2006). The distal part (leaf tip) of cotyledons in nap1 triple mutants was narrower than that of wild type, and the cotyledon width to length ratio was significantly reduced in the triple mutants nap1;1nap1;2nap1;3 (Figure 3.11).  76    To further test the potential roles of the complex of NAP1 and OFP4 in cotyledon development, I generated the quadruple mutants ofp4nap1;1nap1;2nap1;3 by crossing ofp4 with the nap1 triple mutants. The cotyledons in the quadruple mutants exhibited similar phenotypes to the nap1 triple mutants, with narrower distal part, reduced width/length ratio and oblong shapes (Figure 3.11), which indicates that NAP1 and OFP4 may function in the same pathway. However, I did not detect any obvious cotyledon phenotype in ofp4 single mutants. Because of the functional redundancy of OFP proteins (Wang et al., 2011), the quintuple mutants ofp1ofp2ofp3ofp4ofp5 were generated. The cotyledon phenotypes of the ofp quintuple mutants had a more oblong shaped cotyledon and a higher degree of reduction in cotyledon width/length ratio compared to nap1;1nap1;2nap1;3 and ofp4nap1;1nap1;2nap1;3 mutants (Figure 3.11). Among these mutants, except the cotyledon phenotypes, I did not observe other morphological differences compared with wild type. All the findings confirmed that the complex of NAP1 and OFP proteins function in cotyledon development in Arabidopsis seedlings.   A number of cotyledon mutants, such as cvp1, cvp2, wox2 and stpl, to mention a few, presented disordered patterns of vascularization (Carland et al., 1999; Lie et al., 2012). Therefore I further tested the cotyledon vasculature development in the triple mutants nap1;1nap1;2nap1;3, quadruple mutants ofp4nap1;1nap1;2nap1;3, and quintuple mutants ofp1ofp2ofp3ofp4ofp5, to better understand the causes of observed changes in their cotyledons. However, in seedlings of wild type and all the mutants, there were no significant differences in the cotyledon vasculature forms, and all of them had a similar pattern with two to four closed loops around a single main vein (Figure 3.12). This suggests that the cotyledon vascular patterning may not be the reason for their shape changes in the nap1 and ofp mutants.   77   3.3.9     Genetic interactions between OFP4 and NAP1 To investigate the genetic interactions between OFP4 and NAP1 and to see whether the functions of OFP4 in cotyledon development depend on NAP1 proteins, the OFP4 constitutive construct 2 x 35S:GFP-OFP4 was generated and transformed to wild type and the triple mutants nap1;1 nap1;2 nap1;3. By screening T1 transformants on MS medium with hygromycin resistance, the majority of surviving seedlings in the wild type background were found to have kidney-shaped cotyledons (Figure 3.13b), which represented the OFP4 gain-of-function phenotypes, while all the surviving seedlings at the nap1 triple mutant background presented normal round shape cotyledons (Figure 3.13c). I further investigated the GFP-OFP4 expression in the survival of T1 seedlings in the triple mutant background by confocal microscopy. GFP-OFP4 fluorescence was detected in the cytoplasm in most of cotyledon epidermal cells (Figure 3.14c), and its nucleus localization was only detected in a few cells in most of transgenic seedlings (Figure 3.14b), which suggested that the normal wild-type look of cotyledons in transgenic plants at nap1 triple mutant background was not caused by silencing. All together, these data indicate that the functions of OFP4 in cotyledon development may require NAP1 proteins.   3.4     Discussion A previous study suggested, based on qualitative evidence, that ofp4 mutants exhibited irx and thicker fiber cell wall phenotypes, and KNAT7 could form complexes with OFP1 and OFP4, regulating secondary wall development (Li et al., 2011). However, I did not find the ofp4 phenotypes in secondary cell wall development (Figure 3.1). Liu and Douglas (2015) found similar phenotypes as mine, although they described the ofp4 mutants as having a mild irx phenotype, from the quantification of irx and fiber cell wall thickness in Liu and Douglas (2015), the ratio of ofp4 xylem bundles with irx to total xylem bundles, the number of irx per bundle, and the fiber wall thickness in ofp4 stems were indistinguishable from wild type, suggesting ofp4 mutants had no secondary cell wall defects. Statistically ‘mild’ irx phenotype was not significantly different than wild type. The reason that Li et al. (2011) 78   detected the irx phenotype, but it was not found in Liu and Douglas (2015) or my study, may be that their plants were stressed in some uncontrolled manner. I have observed that when plants are stressed, even wild-type plants may display some mild irregular xylem vessels. In addition, Li et al. (2011) presented that OFP1 and OFP4 functions partially depend on KNAT7, as their pleiotropic overexpression phenotypes were suppressed in the knat7 mutants. In my studies, by transforming the OFP1 and OFP4 overexpression constructs into knat7 mutants, the typical OFP overexpression phenotypes were still observed (data not shown). Dr. John Bowman’s group found the same results when they were studying the genetic interactions between OFP and KNOX genes (unpublished data). Therefore, it appears that OFP4 does not function in secondary cell wall development, and that the earlier claims must be reassessed in light of this new data.  The importance of critically evaluating previously reported phenotypes is also highlighted by my work on ofp5 mutants. The ofp5 mutants was reported that the T-DNA insertion line (SALK_010386) of OFP5 had collapsed ovules during embryo development (Pagnussat et al., 2007), but the T-DNA in this line was actually inserted in the 3’-UTR as described in the T-DNA Express database (Signal.salk.edu., 2018). I ordered a true OFP5 loss-of-function mutant (SALK_203823) with the T-DNA insertion in the exon, and found that the embryo development in the new line was normal and the homozygous plants were successfully obtained. All the ofp single mutants that I tested did not show any obvious defects, which suggested that OFP proteins have high functional redundancy for plant growth and development.   By OFP overexpression analysis, I confirmed the previous finding that overexpressing OFP1, OFP2, OFP4 and OFP5 resulted in kidney-shaped cotyledons (Wang et al., 2011), but overexpressing OFP3 had no obvious phenotypes (Figure 3.2). Interestingly, in the subcellular localization analysis, only OFP3-YFP was expressed in the nucleus, the other OFPs-YFP fluorescence was found in both nucleus and cytoplasm (Figure 3.4). OFP proteins were reported previously to be able to regulate the subcellular localization of TALE proteins 79   from nucleus to cytoplasm (Hackbusch et al., 2005). My data suggests the hypothesis that the cytoplasmic localization of OFPs may be contributing to their kidney-shaped cotyledon phenotypes in overexpression plants.   Recently, a couple of ovate family proteins in rice have been well studied. OsOFP8 can be phosphorylated by the signaling kinase OsGSK2 and then transported to the cytoplasm, playing a positive role in the brassinosteroid (BR) signaling pathway (Yang et al., 2016). OsOFP1 protein was localized to both nucleus and cytoplasm and the expression of OsOFP1 was highly induced by BR (Xiao et al., 2017). OsOFP19 was also involved in BR signaling pathway (Yang et al., 2018). In Arabidopsis, I found OFP overexpression plants demonstrated some phenotypes resembling the BR-deficient mutants, such as dwarfism described previously (Wang et al., 2011), disorganized microtubules in hypocotyl epidermal cells (Figure 3.5), and de-etiolated phenotypes grown in the dark (Figure 3.6). However, in the present study, exogenous BR treatment rescued the short hypocotyl phenotype in OFP overexpression seedlings (Figure 3.6), indicating the BR signaling pathway was not affected in Arabidopsis overexpression plants, which is different from the OsOFP8 findings in rice. An area of future work could test if the hypocotyl phenotypes of OFP overexpression plants in Arabidopsis may be associated with BR biosynthesis instead of signaling.   In Arabidopsis, OFP proteins have been reported to interact with TALE proteins regulating different aspects of plant development (Hackbusch et al., 2005; Pagnussat et al., 2007; Li et al., 2011; Liu et al., 2014; Zhang et al., 2016). Interestingly, by performing the yeast two-hybrid screen using OFP4 as bait, we did not identify any TALE transcription factors, which is consistent with the previous finding of yeast two-hybrid screen using Tomato OVATE gene as the bait (van der Knaap et al., 2014). More than 90% of interacting clones were identified as NAP1;1 and NAP1;2, which belong to the same Nucleosome Assembly Protein 1 family (Table 3.2). By the BiFC assay, the interactions between OFP4 and NAP1 were found to be in the cytoplasm in the epidermal cells of tobacco leaves (Figure 3.8). Subcellular localization of NAP1 revealed that NAP1;1 and NAP1;2 proteins were abundant in the 80   cytoplasm, associated with ER membrane (Figure 3.9). OFP4 was previously shown to localize to both the cytoplasm and nucleus (Figure 3.4). The complete YFP signal adjacent to the cell nucleus detected in the BiFC assay (Figure 3.8) indicated that OFP4 and NAP1 protein may interact with each other on the surface of the ER membrane in epidermal cells of tobacco leaves. Previous findings suggested that NAP1 proteins are primarily localized in the cytoplasm in Arabidopsis, and both NAP1;1 and NAP1;2 are expressed ubiquitously in plants (Liu et al., 2009), which is consistent with the model that OFP4 may form functional complexes with NAP1 proteins in Arabidopsis.   By checking the secondary cell wall phenotypes of NAP1 loss-of-function mutants, I found similar results as ofp4 mutants, with no distinguishable cell wall defects, which confirmed my previous work that OFP does not function in secondary cell wall development. In contrast, I found the triple mutants of nap1;1 nap1;2 nap1;3 had altered cotyledon shapes, which are consistent with the OFP overexpression lines. The phenotype in the triple mutants was mild but it was significantly different from wild type, according to the quantification of cotyledon width/length ratios (Figure 3.11). Because of functional redundancy of OFP proteins, the higher order OFP loss-of-function mutants were generated. We also observed the altered cotyledon shape in the ofp1 ofp2 ofp3 ofp4 ofp5 quintuple mutants, which was even more severe than the nap1;1 nap1;2 nap1;3 triple mutants (Figure 3.11). OFP overexpression plants had kidney-shaped cotyledons (Figure 3.2), but loss-of-function mutants had oblong shaped cotyledons (Figure 3.11), and all OFP genes were expressed in the seedlings (Figure 3.3). All these revealed that OFP and NAP1 proteins are functioning in the cotyledon development. In addition, the ofp4 nap1;1 nap1;2 nap1;3 cotyledon phenotypes were similar to that of nap1;1 nap1;2 nap1;3 triple mutants, rather than showing an additive phenotype, which suggested that OFP4 and NAP1 proteins may function in a common pathway. Overexpressing OFP4 in the nap1;1 nap1;2 nap1;3 background failed to show the kidney-shaped cotyledons, indicating OFP4 requires NAP1 proteins to function in the cotyledon development. All the evidence are consistent with the model that OFP and NAP1 regulate the cotyledon development by forming functional complexes. 81    Although the cotyledon shapes were changed in the OFP and NAP1 high-ordered loss-of-function mutants, their vascular patterning did not differ from wild type, suggesting the patterns of vascularization in cotyledons may not contribute to their shape modification. NAP1;1 and NAP1;2 proteins have been reported to have responses to BR treatment, with a significant decrease in NAP1;1 and a considerable increase in NAP1;2 in BL-treated cells (Shigeta et al., 2011). Our present work also suggests that OFP proteins may be associated with BR biosynthesis. It would be interesting to further investigate if the altered cotyledon phenotypes were caused by BR induced cell elongation and cell expansion in the future.   Overall, my data suggest that OFP proteins may be involved in maintaining BR homeostasis, and OFP4 interacts with NAP1;1 and NAP1;2, playing significant roles in the cotyledon development in Arabidopsis seedlings. The mechanism of OFP-NAP1 regulatory model in the cotyledon development requires further investigations.                82   Table 3.1 Oligonucleotides used in Chapter 3 Gene name Application Primer sequence (5' to 3') ofp1 genotyping SM_3_21689-L: ATGGGTAATAACTATCGGTTTAAG   SM_3_21689-R:  TTATTTGGAATGGGGTGGTGGAA Tran-element: TACGAATAAGAGCGTCCATTTTAGAGTGA ofp2 genotyping SALK_122550-L:   ACCAAATTCAAAGAAGCATCG   SALK_122550-R:  TGGTGAGTTATGGTGAGGAGG LBb1.3:   ATTTTGCCGATTTCGGAAC ofp3 genotyping GABI_167F01-L:  CAGAAAATGGGGACTCACAAG   GABI_167F01-R:  TGACTTTGAGAAAGAGGACGG GK T-DNA:  ATATTGACCATCATACTCATTGC ofp4 genotyping SALK_022396-L:  ATGAGGAACTATAAGTTAAGATTG     SALK_022396-R:  CTACTTCGATGCAAATGTAGAG ofp5 genotyping SALK_203823-L:  GACAACATCTTCATCTCCCTCC     SALK_203823-R:  ATTATGCACCTGCTGGAACAC nap1;1 genotyping SALK_013610-L:  TCTGTAAACTGTCCCGTGAGC     SALK_013610-R:  CATAGCCTTCTCAAGCAGTGG nap1;2 genotyping SAIL_84_B01-L:  GTCATCTGCCTCAACAGCTTC   SAIL_84_B01-R: TTTCTGCATTCGTGATTG     LB2 SAIL: GCTTCCTATTATATCTTCCCAAATTACCAATACA    83   Gene name Application Primer sequence (5' to 3') nap1;3 genotyping SALK_131746-L: TAATTGGCTTGGCATTCTTTG     SALK_131746-R: TTCAGGTTTGGGAAAACTTCC OFP2 cloning  ProOFP2-L:  GAAGCTTTTTTGGTGATGATG ProOFP2: OFP2-GUS  OFP2-R:  CTTTGTTTTTGTAAGTTGAAGC OFP3 cloning  ProOFP3-L:  TGAGAGGCGGCGAGAGAATTAG ProOFP3: OFP3-GUS  OFP3-R:  CTCTCAAATATTTTATGAGCTC OFP5 cloning  ProOFP5-L:  CTAACGTACTAACTCTATAA ProOFP5: OFP5-GUS OFP5-R:  TCCAAGAATCTGAAGAAGTT  OFP1 cloning  35S: OFP1-YFP OFP1-L:  ATGGGTAATAACTATCGGTTTA  OFP1-R:  TTTGGAATGGGGTGGTGGAAGA OFP2 cloning  35S: OFP2-YFP OFP2-L:  ATGGGGAATTACAAGTTCAGAA OFP2-R:  CTTTGTTTTTGTAAGTTGAAGC  OFP3 cloning  35S: OFP3-YFP OFP3-L:  ATGAAACAGAAAATGGGGAC  OFP3-R:  GAGAGAGATAGAGAGTCCTTGA  OFP4 cloning  35S: OFP4-YFP OFP4-L:  ATGAGGAACTATAAGTTAAGA OFP4-R:  CTTCGATGCAAATGTAGAGT OFP5 cloning  35S: OFP5-YFP OFP5-L:  ATGATGAGATGGGGAAGAAAGA OFP5-R:  ATGAAAATTAAAATCATTATGC NAP1;1 cloning  35S: YFP-NAP1;1 NAP1;1-L:  ATGAGCAACGACAAGGATAGCT NAP1;1-R:  TTACTGTTGCTTGCATTCGGGT   84   Gene name Application Primer sequence (5' to 3') NAP1;2 cloning  35S: YFP-NAP1;2 NAP1;2-L:  ATGAGCAACGACAAGGACAGCA   NAP1;2-R:  TCACTGCTGCTTACATTCCGGT  OFP4 cloning  OFP4-L:  ATGAGGAACTATAAGTTAAGA   2 x 35S: GFP-OFP4 OFP4-R:  CTACTTCGATGCAAATGTAGA OFP1 qRT-PCR OFP1-L:  ATGGGTAATAACTATCGGTTTA   OFP1-R: GCTATTTGGTTGGCTCTGAAGATTCT OFP2 qRT-PCR OFP2-L:  AGAGCAAACAAGATGTTCTA   OFP2-R:  TTTGTAAGTTGAAGCCAGAT  OFP3 qRT-PCR OFP3-L:  ATGAAACAGAAAATGGGGAC    OFP3-R:  TTGGGAGAAGAAAGATGGTG  OFP4 qRT-PCR OFP4-L:  ATGAGGAACTATAAGTTAAGA   OFP4-R:  TATGGAGTAAAGAGGAAGAGA  OFP5 qRT-PCR OFP5-L:  GATGGAGGAATGGAGAACGA    OFP5-R:  TTATGCACCTGCTGGAACAC  ACTIN2 qRT-PCR ACTIN2-L: CCAGAAGGATGCATATGTTGGTGA      ACTIN2-R:  GAGGAGCCTCGGTAAGAAGA     85    Figure 3.1 Cross-sections of stem vascular bundles in wild type (WT) and ofp single mutants. Stem sections from the base of 8-week old Arabidopsis plants, stained with toluidine blue. A single representative vascular bundle is shown from each mutant. (a) WT; (b) ofp1; (c) ofp2; (d) ofp3; (e) ofp4; (f) ofp5. No obvious differences are shown in xylem or interfascicular fiber morphology between ofp single mutants and wild type. Say how many batches of plants you tested. Scale bars = 20µm.       86        Figure 3.2 Phenotypes of wild type (WT) and OFP overexpression (OX) seedlings. Ten-day-old wild type (WT) (a), 35S:OFP1-YFP (OFP1 OX) (b), 35S:OFP2-YFP (OFP2 OX) (c), 35S:OFP3-YFP (OFP3 OX) (d), 35S:OFP4-YFP (OFP4 OX) (e), and 35S:OFP5-YFP (OFP5 OX) (f) seedlings  Plants overexpressing OFP1, OFP2, OFP4 and OFP5 presented kidney shaped cotyledons, while OFP3 overexpression plants did not display any significant difference compared with wild type. Scale bars = 5mm. 87     Figure 3.3 The expression patterns of OFP2, OFP3 and OFP5 in Arabidopsis seedlings. Histochemical localization of ProOFP:GUS activity in 10-day-old seedlings of ProOFP2:GUS (a, d, g), ProOFP3:GUS (b, e, h), and ProOFP5:GUS (c, f, i) transgenic plants. Results shown are representative of more than 3 independent lines. Bars: 500µm for a, b, c; 300µm for d, e, f; 100µm for g, h, i.  88      Figure 3.4 Subcellular localization of OFP-YFP fusion proteins in epidermal cells of N. benthamiana leaves. Confocal scanning microscopy observations of OFP1-YFP (a), OFP2-YFP (b), OFP3-YFP (c), OFP4-YFP (d) and OFP5-YFP (e) transiently expressed in tobacco leaf epidermal cells.  OFP1-YFP, OFP2-YFP, OFP4-YFP and OFP5-YFP proteins were localized in the nucleus and cytoplasm (a, b, d, e), while OFP3-YFP was exclusively located in the nucleus (c). Scale bars = 20µm.   89     90   Figure 3.5 Hypocotyl phenotypes of wild type (WT) and OFP overexpression (OX) plants.  Hypocotyls and the hypocotyl epidermal cells of four-day-old wild type (WT) (a, b), 35S:OFP1-YFP (OFP1 OX) (d, e), 35S:OFP2-YFP (OFP2 OX) (g, h), 35S:OFP4-YFP (OFP4 OX) (j, k), and 35S:OFP5-YFP (OFP5 OX) (m, n) seedlings; Overexpression of OFP1, OFP2, OFP4 and OFP5 caused epidermal cells to swell in their hypocotyls.  Cortical microtubules were observed in the hypocotyl epidermal cells of WT (c), OFP1 OX (f), OFP2 OX (i), OFP4 OX (l), OFP5 OX (o) plants by transforming OFP overexpression constructs into the RFP-βTubulin6 (RFP-TUB6) homozygous transgenic plants. After overexpressing OFP1, OFP2, OFP4 and OFP5, the microtubule orientation was changed to predominantly longitudinal and oblique (f, i, l, o).  Images of c, f, i were taken using different epidermal cells from b, e, h, respectively; Images of l, o were taken using same epidermal cells as k, n, respectively. Bars: 200 µm for a, d, g, j, m; 20 µm for b, c, e, h, i, k, l, n, o; 5 µm for f.           91     Figure 3.6 BR-deficient related morphological phenotypes of OFP overexpression (OX) mutants. (a) Hypocotyl elongation of one-week-old wild type (WT) and OFP OX plants grown in the dark. Bars, 5mm.  (b) Statistical analysis of hypocotyl length measurements of dark-grown plants as shown in a. Asterisk (*) indicates the sample is significantly different from WT at P <0.01 determined by Student’s T-test. Error bars represent the standard deviations. n= 5-15. (c) Hypocotyl elongation of one-week-old WT and OFP2 OX plants germinated on ½ MS plates containing 0, 0.2, 0.5 or 1 µM epiBL grown in the light. Bars, 10 mm.   92   Figure 3.6 BR-deficient related morphological phenotypes of OFP overexpression (OX) mutants. (d) Hypocotyl length measurements of WT and OFP2 OX seedlings germinated on ½ MS plates containing 0, 0.2, 0.5 or 1 µM epiBL in the light as shown in c. Statistical differences among different concentrations of epiBL treatments in the same genotype are labeled with different letters (P<0.01, one-way ANOVA followed by Turkey’s post hoc test). Asterisk (*) indicates the sample in OFP2 OX is significantly different from WT hypocotyl length at 0 µM epiBL treatment at P <0.01 determined by Student’s T-test. Error bars represent the standard deviations. n= 5-15.          93    Figure 3.7 Quantification of WT and OFP OX hypocotyl length under exogenous BR treatment. Hypocotyl length measurements of one-week-old WT, OFP1 OX, OFP4 OX, and OFP5 OX seedlings germinated on ½ MS plates containing 0, 0.2, 0.5 or 1 µM epiBL in the light condition. Error bars represent the standard deviations. n= 5-15.         94   Table 3.2 Identification of the cDNA encoding interactors of OFP4 Gene locus Protein name Accession no. Number of clone        Annotation AT4G26110 NUCLEOSOME ASSEMBLY PROTEIN 1;1 (NAP1:1)  2120785 1 Nucleotide excision repair (Liu et al., 2009);  Cell proliferation and cell expansion (Galichet and Gruissem, 2006)  AT2G19480 NUCLEOSOME ASSEMBLY PROTEIN 1;2 (NAP1;2)  2050424 87 Nucleotide excision repair (Liu et al., 2009) AT4G28760 TON1 RECRUITING MOTIF 20 (TRM20)  2117823 4 Cytoplasmic-localized protein (Drevensek et al., 2012); unknown function AT5G14270 BROMODOMAIN AND EXTRATERMINAL DOMAIN PROTEIN 9 (BET9)  2145673 2 Protein phosphorylation; Arabidopsis protein kinase (Nemoto et al., 2011) AT4G27060 Spiral2 (SPR2) 2136467 1 Microtubule-associated protein (Shoji et al., 2004)       95    96   Figure 3.8 OFP4 interacts with Nucleosome Assembly Protein 1;1 (NAP1;1) and NAP1;2 in vitro and in vivo. (a) Yeast two-hybrid assay of OFP4-NAP1 interactions. Top: diagram of used constructs in yeast two-hybrid assays. Bottom: assay of DNA-binding domain (DBD)-OFP4 interaction with activation domain (AD)-NAP1;1 and AD-NAP1;2 using two reporter genes, HIS3 (assayed on Histidine- +3AT 40mM medium) and URA3 (assayed on Uracil- medium), with growth controls in Trptophan-  Leucine- medium alone. DBD-OFP4- empty vector interaction was used as a negative control, and MYB75-TT8 interaction was served as a positive control. (b) Bimolecular Fluorescence Assay of OFP4-NAP1 interactions. Left: negative control, image of the representative tobacco leaf epidermal cell co-expressed with OFP4-nYFP and empty-cYFP; Middle and right: images of epidermal cells co-transformed by OFP4-nYFP and NAP1;1/NAP1;2-cYFP together. Scale bars = 50µm. nYFP: N-terminal enhanced YFP protein; cYFP: C-terminal enhanced YFP protein.           97     Figure 3.9 NAP1;1 and NAP1;2 are localized to the ER in epidermal cells of N. benthamiana leaves. 35S:YFP-NAP1;1 (a) and 35S:YFP-NAP1;2 (d) were co-transformed with ER-marker HDEL-RFP (b, e) into epidermal cells of N. benthamiana leaves. The colocalization of NAP1;1 and NAP1;2 with ER-marker protein appears as yellow in the merged images (c, f). Scale bars = 20µm.        98       Figure 3.10 Cross-sections of stem vascular bundles in WT, nap1;1, nap1;2 and nap1;1 nap1;2 nap1;3 triple mutants. Stem sections stained with toluidine blue. A single representative vascular bundle is shown from each mutant. (a) WT; (b) nap1;1; (c) nap1;2; (d) nap1;1 nap1;2 nap1;3. No obvious differences are shown in xylem or interfascicular fiber morphology between wild type and mutants. Scale bars = 30µm.    99    Figure 3.11 Cotyledon phenotypes of NAP1 and OFP loss-of-function mutants (a) one-week-old wild type (WT), nap1;1 nap1;2 nap1;3, ofp4 nap1;1 nap1;2 nap1;3, ofp1 ofp2 ofp3 ofp4 ofp5 cotyledons; Scale bars = 500 µm.  (b) Statistical analysis of the cotyledon width to length ratio of WT, nap1;1 nap1;2 nap1;3, ofp4 nap1;1 nap1;2 nap1;3, ofp1 ofp2 ofp3 ofp4 ofp5 seedlings as shown in a. Error bars represent the standard deviations; Statistical differences among the samples are labeled with different letters (P<0.01, one-way ANOVA followed by Turkey’s post hoc test). n = 20-30.  100     Figure 3.12 Cotyledon vasculature development in NAP1 and OFP loss-of-function mutants. In one-week-old WT (a), nap1;1 nap1;2 nap1;3 (b), ofp4 nap1;1 nap1;2 nap1;3 (c), ofp1 ofp2 ofp3 ofp4 ofp5 (d) seedlings, no significant differences were found in their cotyledon vasculature forms, and all of them had a similar pattern with two to four closed loops around a single main vein. Scale bars = 2mm.    101        Figure 3.13 Phenotypes of 2 x 35S: GFP-OFP4 transgenic seedlings in WT and nap1;1 nap1;2 nap1;3 background.  Ten-day-old wild type (WT) (a), 2 x 35S: GFP-OFP4 transgenic seedlings in the WT background (b), and in the nap1;1 nap1;2 nap1;3 background (c); Scale bars = 500 µm.            102      Figure 3.14 GFP-OFP4 expression in nap1;1 nap1;2 nap1;3 triple mutants.  WT has no fluorescence that serves as a negative control (a). GFP-OFP4 nucleus localization was only detected in a few epidermal cells of nap1;1 nap1;2 nap1;3 mutants (b), and GFP-OFP4 fluorescence was found in cytoplasm in most of epidermal cells of triple mutants (c). Scale bars = 15 µm.               103   Chapter 4: Conclusions  4.1    Major findings of the thesis Owing to the existence of secondary cell walls, large trees and other terrestrial plants are the foundations of ecosystems on earth. Secondary cell wall development requires a complex network of transcriptional regulation, controlling the biosynthetic genes to produce secondary cell walls spatially and temporally. In my thesis, I studied two different family proteins with the potential roles involved in secondary cell wall formation. And I discovered KNAT3, a transcription factor belonging to the KNOX family, is working cooperatively with KNAT7 to regulate secondary cell wall development and provide mechanical support to Arabidopsis stems (Chapter 2). Although expressed in the appropriate temporal and spatial window, I showed OFPs, another family of proteins, instead of participating in secondary cell wall formation as originally hypothesized, actually function to maintain plant hormone homeostasis and regulate the cotyledon development in Arabidopsis (Chapter 3).   4.1.1    KNAT3 and KNAT7 function together to activate xylem vessel secondary wall formation knat7 mutants, one of four KNOX2 family transcription factors in Arabidopsis, have been reported to have secondary cell wall defects, including irregular xylem (irx) and increased fiber wall thickness (Li et al., 2012). However, the functions of other KNOX2 genes, KNAT3, KNAT4, and KNAT5, involved in plant growth and development and secondary wall formation were unclear. In Chapter 2, the four Arabidopsis KNOX2 genes were shown to share the same expression patterns, largely being expressed in the cells where secondary cell walls are actively being deposited in inflorescence stems. knat3, knat4, knat5 single mutants did not show any obvious phenotypes, while the double mutants of knat3knat7 manifested an enhanced irx phenotype, compared to the other double and single mutants. ProKNAT3:KNAT3-GFP partially rescued the enhanced irx phenotype of the double mutants, and resulted in a 104   phenotype similar to the knat7 single mutants. This partial complementation suggests that KNAT3 plays a positive role in xylem vessel formation. When knat3knat7 mutants were transformed with ProKNAT7:KNAT7-GFP, the irx phenotype was completely recovered, mimicking the lack of phenotype of knat3 single mutants. Thus, KNAT3 and KNAT7 may function redundantly to activate secondary cell wall formation in xylem vessels, where KNAT7 plays a dominant role. My results confirmed previous observation concerning the knat7 irx phenotype and extend our understanding of the role of KNAT3 in xylem vessel secondary wall formation.   4.1.2    KNAT3 acts antagonistically with KNAT7 during secondary cell wall formation in interfascicular fibers In Chapter 2, we confirmed that knat7 single mutants had thicker interfascicular fiber wall thickness, and increased total lignin contents in inflorescence stems. Also, via RNA-seq analysis, we show an elaborate network of genes involved in the cell wall organizational processes are affected in knat7 mutants compared with wild-type plants, and conclusively show that KNAT7 is a transcriptional repressor in secondary cell wall formation in interfascicular fibers, as proposed before in Li et al. (2012).   Interestingly, Chapter 2 also showed that knat3knat7 displayed a pendent stem phenotype, culminating a significant reduction in tensile and flexural strength and stiffness in their inflorescence stems. It is very possible that this weak stem phenotype was caused by the extremely thin interfascicular fiber cell walls apparent in knat3knat7 double mutants. By expressing ProKNAT3::KNAT3-GFP in knat3knat7 double mutants, the thinner fiber secondary wall was restored to a level consistent with the knat7 single mutants, which suggested that the thicker fiber walls in knat7 require a functional copy of KNAT3. In addition, by overexpressing KNAT3, the secondary cell wall thickness of interfascicular fibers was significantly increased. These findings suggest that KNAT3 plays a positive role in regulating interfascicular fiber secondary wall formation, acting antagonistically with KNAT7. In 105   support of this conclusion, we observed altered cell wall chemical compositions and significant mis-regulation of associated cell wall related genes in knat3knat7 double mutants. Previously, it was reported that KNOX2 and KNOX1 had antagonistic roles in plant aerial organ development (Furumizu et al., 2015), and my results provided an additional dimension, implicating KNAT3 and KNAT7 as antagonistic interactors regulating secondary wall development in interfascicular fibers.   4.1.3    OFP genes are not associated with Brassinosteroid signaling pathway in Arabidopsis Previous data suggested that OFP1 is able to repress the expression of GA20ox, an enzyme involved in Gibberellic acid (GA) biosynthesis that is known to contribute to cell elongation. Contrary, in rice, a number of OsOFP genes have been reported to be involved in the BR signalling pathway (Yang et al., 2016; Xiao et al., 2017; Yang et al., 2018). However, to date, there is no evidence in Arabidopsis linking BR-related functions to OFPs. In Chapter 3, by examining the OFP overexpression phenotypes in detail, I observed that OFP mis-regulation results in dwarfism, disorganized microtubules, and dark-grown de-etiolated phenotypes mimic the BR-deficient or –signaling mutants. By treating the OFP overexpression plants with synthetic BR to distinguish BR-deficient from BR-signalling mutants, I found the short hypocotyl phenotypes in OFP overexpression plants were rescued, indicating that the BR signaling pathway is normal in these plants. Further work could be done to test if the OFP overexpressor phenotypes are associated with BR biosynthesis, to give new insights about OFP’s role in maintaining hormone homeostasis.   4.1.4    OFP proteins regulate cotyledon development by interacting with NAP1 To date, OFP overexpression phenotypes have been well studied (Hackbusch et al., 2005; Wang et al., 2007; Li et al., 2011; Wang et al., 2011), but there are only a few reports examining OFP loss-of-function mutant phenotypes in Arabidopsis (Li et al., 2011; Wang et al., 2010; Pagnussat et al., 2007). Among them, the secondary cell wall phenotypes of ofp4 106   and the embryo sac development phenotypes of ofp5 were not replicated in our work. The other phenotypes common to ofp mutants were apparent when the plants were subject to different stresses (Wang et al., 2010). Since overexpressing OFP would cause pleotropic phenotypes, we hypothesized that the functions of OFP are related to plant growth and development even in the normal conditions.   In Chapter 3, by generating the high-ordered loss-of-function mutants, I was able to uncover new phenotypes for ofp mutants. The quintuple mutants ofp1ofp2ofp3ofp4ofp5 displayed abnormal cotyledons with oblong shapes, and the cotyledon width to length ratio was significantly reduced compared with that in wild-type plants. In contrast, OFP overexpression plants diplay kidney-shaped cotyledons. Consistent with these phenotypic observations, an analysis of the expression pattern revealed that OFP1, OFP2, OFP3, OFP4 and OFP5 were expressed in the cotyledons, providing further support that OFPs function in the cotyledon development in Arabidopsis seedlings.    Previously, it has been shown that the Ovate Family Proteins interact with TALE homeodomain transcription factors regulating different aspects of plant development (Hackbusch et al., 2005; Pagnussat et al., 2007; Li et al., 2011; Zhang et al., 2016). In Chapter 3, in a yeast two-hybrid screen using OFP4 as the bait, I did not observe any TALE protein. Instead, I found that Nucleosome Assembly Protein1, histone chaperones, interacted with OFP4 in vitro and in vivo. In addition, they were found to interact with one another in the cytoplasm of tobacco leaf epidermal cells. In support of this finding, NAP1 loss-of-function mutants were shown to display an abnormal cotyledon phenotype, and the kidney-shaped cotyledon phenotype of OFP overexpression plants were suppressed when transforming the constitutive overexpressing OFP constructs into the nap1;1nap1;2nap1;3 triple mutants. In addition, by crossing ofp4 with the nap1 triple mutants, we did not detect any additive cotyledon phenotypes, and the cotyledon phenotypes of quadruple mutants were similar to the nap1 triple mutants. All the evidences we found suggest that OFP and NAP proteins can form a complex in Arabidopsis seedlings to regulate cotyledon development. 107   My findings provide a significant step forward in disclosing the OFP functions in plant growth and development.                                                                                    4.2    Future directions Throughout this thesis, I have investigated the functions of Arabidopsis KNOX2 and OFP genes, focusing specifically on their association to plant secondary cell wall development. However, this work has illustrated that KNOX2 and OFP genes also contribute and function as regulators in plant growth and development. As such, several questions have arisen regarding the mechanisms behind these phenotypes, and therefore, I propose future studies should focus specifically on understanding their unique contribution to plant growth and development in Arabidopsis, as outlined below.   4.2.1    Identifying direct target genes of KNAT3 and KNAT7 in different cell tissues My data (Chapter 2) suggested that KNAT3 may function as a transcriptional activator, regulating secondary cell wall formation in both fibers and xylem vessels, while KNAT7 may have opposing activities in interfascicular fibers and xylem vessels. RNA-seq analyses identified a number of genes differentially expressed between wild-type, knat3, knat7 and knat3knat7 stems. However, of the mis-regulated genes, we do not know which ones are directly regulated by KNAT3 and KNAT7, respectively, in different cell types. To address this, I would propose to employ chromatin immunoprecipitation sequencing (ChIP-seq). Since these transcription factors likely display different functions in the distinct cell tissues, we could use cell type-specific promoters to overexpress them to study their functions in different cell types, and couple this to ChIP-seq. More specifically, the fiber-specific promoter SND1 (Zhong et al., 2006) and xylem vessel-specific promoter VND7 (Kubo et al., 2005) could be used, and different transgenic plants could be generated by expressing ProSND1::KNAT3-GFP, ProSND1::KNAT7-GFP, ProVND7::KNAT3-GFP and ProVND7::KNAT7-GFP constructs in knat3knat7 mutants, respectively. The phenotypic analysis of these transgenic plants in secondary cell wall development should confirm our hypothesis about 108   their functions in different cell types. ChIP-seq using these complemented lines could help identify all the putative binding sites of KNAT3 and KNAT7, which would disclose their direct target genes in each cell type.   4.2.2    Identifying KNAT3 and KNAT7 cell-specific interacting partners  Many studies have shown that protein interactions among TALE homeodomain transcription factors play significant roles in varying aspects of plant development (Hackbusch et al., 2005; Pagnussat et al., 2007; Kim et al., 2013; Liu et al., 2014). KNAT7 has been reported to interact with BLH6, and the interaction appears to enhance KNAT7 repressor activity (Liu et al., 2014). Compared with single mutants, the knat7blh6 double mutants displayed an enhanced irx phenotype, but the fiber wall thickness was not enhanced (Liu et al., 2014). The additive irx phenotypes suggests that KNAT7 and BLH6 may function in different pathways to regulate xylem vessel secondary wall formation, while the phenotype in interfascicular fibers implies that KNAT7 forms a complex with BLH6 to impact fiber secondary cell wall development. In addition, KNAT3 was shown to interact with BLH1 in controlling embryo sac development and ABA signalling pathway (Pagnussat et al., 2007; Kim et al., 2013). However, the KNAT3 interacting partners involved in secondary cell wall development remain unclear. Moreover, according to my findings with KNAT7, I hypothesize that the same transcription factor may have different interacting partners in multiple tissues to regulate secondary cell wall formation.  To identify these putative protein complexes in Arabidopsis, the metabolic stable isotope labelling immuno-precipitation mass spectrometry (mSILIP) could be performed. And the transgenic lines mentioned above with KNAT3 or KNAT7 expression driven by tissue-specific promoters in the knat3knat7 background could be used to detect tissue-specific interactors of KNAT3 or KNAT7, respectively. Quantitative immuno-precipitation and mass spectrometry should allow us to distinguish specific from non-specific proteins revealed by co-immunoprecipitation, and employing this strategy, we may be able to identify novel interactors of KNAT3 and KNAT7. 109    4.2.3    Investigating upstream signals of KNAT3 and KNAT7 involved in secondary cell wall development Previous research has shown that SND1 and MYB46 are the upstream regulators of KNAT7 (Zhong et al., 2008; Ko et al., 2009). However, KNAT3 is a newly identified transcription factor involved in secondary cell wall formation. It would be interesting to see if KNAT3 and KNAT7 share the same upstream regulators, and if there are other “alternative” regulators controlling KNAT3 and KNAT7 functions in secondary cell wall development. It was reported that the xylem transcriptional network could be manipulated by environmental stresses (Taylor-Teeples et al., 2014a). However, there is little information about the role of KNOX2 in regulating secondary cell wall formation in response to different abiotic stresses. To address these questions, we could use a yeast one-hybrid (Y1H) protein-DNA interaction assays elucidate the direct upstream regulators of KNAT3 and KNAT7. This could be followed by electrophoretic mobility shift assays (EMSA) to confirm the interactions between identified regulators and the promoters of KNAT3 and KNAT7. By treating plants with different abiotic stresses, we could investigate the influence of stresses on the number of KNAT3 and KNAT7 upstream regulators in Arabidopsis.   4.2.4    Elucidating the roles of OFP in cotyledon development  In Chapter 3, the ofp1ofp2ofp3ofp4ofp5 quintuple mutants showed altered cotyledon shapes, but it is unclear how this phenotype manifests. RNA-seq analysis could be conducted on the quintuple mutants to reveal the genes responsible for this unique phenotype. Since overexpressing OFP1/2/4/5 resulted in kidney-shaped cotyledons, we could generate these same transgenic plants expressing OFPs under their native promoters, and again perform mSILIP to identify OFP interactors in planta. These approaches may help us improve our understanding of how OFPs contribute to the cotyledon development.   110   4.2.5    Investigating the relationships between OFP and plant hormones In addition to the cotyledon phenotype, OFP overexpression plants also exhibited shorter hypocotyl and swollen hypocotyl epidermal cells. The data in Chapter 3 highlighted a relationship between OFPs and BR, and we hypothesized that OFP may be involved in BR biosynthesis. Previous research has shown that OFP1 functions as a transcriptional repressor, by inhibiting GA biosynthesis to regulate cell elongation (Wang et al., 2007). Another hormone, Auxin, is also involved in cell elongation. Since crosstalk among Auxin, BR and GA signalling pathways exists (Jung et al., 2010; Bernardo-García et al., 2014; Bai et al., 2012), it would be interesting to see if OFP overexpression plants and loss-of-function (quintuple) mutants have any response to different hormones, and if the three hormones can induce or inhibit OFP gene expressions.  4.2.6    Characterizing the functions of OFP4-TRM20  van der Knaap et al. (2014) carried out a yeast two-hybrid screen using tomato OVATE gene as the bait, and they did not detect any TALE transcription factors in their results, which is consistent with my results. However, they identified many candidates from the TONNEAU1 Recruiting Motif (TRM) superfamily, including the otholog of Arabidopsis TRM17/20 (van der Knaap et al., 2014). Interestingly, in my experiment, TRM 20 was also identified as one of the top OFP4 interactors. These findings provide additional confidence and impetus to further study the potential complex of OFP4-TRM20. To verify their interaction in planta, BiFC or co-immunoprecipitation could be employed. Moreover, using a reverse genetics approach, we could explore the function(s) of a putative OFP4-TRM20 complex in plant growth and development, using the suite of similar techniques employed in this thesis.       111   References Albersheim, P., Darvill, A., Roberts, K., Sederoff, R., and Staehelin, A. (2010). Plant cell walls. Garlandscience. 18-19, 74. Ambrose, C., Allard, J.F., Cytrynbaum, E.N. and Wasteneys, G.O. (2011) A CLASP-modulated cell edge barrier mechanism drives cell-wide cortical microtubule organization in Arabidopsis. Nat. Commun., 2, 412–430.  Amoutzias, G.D., Robertson, D.L., Peer, Y. Van de and Oliver, S.G. (2008) Choose your partners: dimerization in eukaryotic transcription factors. Trends Biochem. Sci., 33, 220–229. Bai, M.-Y., Shang, J.-X., Oh, E., Fan, M., Bai, Y., Zentella, R., Sun, T. and Wang, Z.-Y. (2012) Brassinosteroid, gibberellin and phytochrome impinge on a common transcription module in Arabidopsis. Nat. Cell Biol., 14, 810–817.  Bernardo-García, S., Lucas, M. de, Martínez, C., Espinosa-Ruiz, A., Davière, J.M. and Prat, S. (2014) BR-dependent phosphorylation modulates PIF4 transcriptional activity and shapes diurnal hypocotyl growth. Genes Dev., 28, 1681–1694. Bhargava, A., Mansfield, S.D., Hall, H.C., Douglas, C.J. and Ellis, B.E. (2010) MYB75 Functions in Regulation of Secondary Cell Wall Formation in the Arabidopsis Inflorescence Stem. Plant Physiol., 154, 1428–1438.  Bonawitz, N.D. and Chapple, C. (2010) The Genetics of Lignin Biosynthesis: Connecting Genotype to Phenotype. Annu. Rev. Genet., 44, 337–363.  Brown, D.M., Zeef, L.A.H., Ellis, J., Goodacre, R. and Turner, S.R. (2005) Identification of Novel Genes in Arabidopsis Involved in Secondary Cell Wall Formation Using Expression Profiling and Reverse Genetics. Plant Cell, 17, 2281–2295. Brown, D.M., Zhang, Z., Stephens, E., Dupree, P. and Turner, S.R. (2009) 112   Characterization of IRX10 and IRX10-like reveals an essential role in glucuronoxylan biosynthesis in Arabidopsis. Plant J., 57, 732–746. Bürglin, T.R. (1997) Analysis of TALE superclass homeobox genes (MEIS, PBC, KNOX, Iroquois, TGIF) reveals a novel domain conserved between plants and animals. Nucleic Acids Res., 25, 4173–4180. Carland, F.M., Berg, B.L., Fitzgerald, J.N., Jinamornphongs, S., Nelson, T. and Keith, B. (1999) Genetic Regulation of Vascular Tissue Patterning in Arabidopsis. Plant Cell, 11, 2123–2137. Chen, F., Tobimatsu, Y., Havkin-Frenkel, D., Dixon, R.A. and Ralph, J. (2012) A polymer of caffeyl alcohol in plant seeds. Proc. Natl. Acad. Sci., 109, 1772–1777.  Citovsky, V., Lee, L.Y., Vyas, S., Glick, E., Chen, M.H., Vainstein, A., Gafni, Y., Gelvin, S.B. and Tzfira, T. (2006) Subcellular Localization of Interacting Proteins by Bimolecular Fluorescence Complementation in Planta. J. Mol. Biol., 362, 1120–1131. Clough, S.J. and Bent, A.F. (1998) Floral dip: A simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J., 16, 735–743. Coleman, H.D., Park, J.-Y., 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., 105, 4501–4506.  Cosgrove, D. (2005) Growth of the plant cell wall. Nat. Rev. Mol. Cell Biol., 6, 850–61.  Cosgrove, D.J. and Jarvis, M.C. (2012) Comparative structure and biomechanics of plant primary and secondary cell walls. Front. Plant Sci., 3, 1–7. Courtois-Moreau, C.L., Pesquet, E., Sjödin, A., Muñiz, L., Bollhöner, B., Kaneda, M., Samuels, L., Jansson, S. and Tuominen, H. (2009) A unique program for cell death in 113   xylem fibers of Populus stem. Plant J., 58, 260–274. Derelle, R., Lopez, P., Guyader, H. Le and Manuel, M. (2007) Homeodomain proteins belong to the ancestral molecular toolkit of eukaryotes. Evol. Dev., 9, 212–219. Dixon, R.A., Srinivasa Reddy, M.S. and Gallego-Giraldo, L. (2014) Monolignol Biosynthesis and its Genetic Manipulation: The Good, the Bad, and the Ugly. Recent Adv. Polyphen. Res., 4, 1–38. Donaldson, L.A. (2001) Lignification and lignin topochemistry - An ultrastructural view. Phytochemistry, 57, 859–873. Donner, T.J., Sherr, I. and Scarpella, E. (2009) Regulation of preprocambial cell state acquisition by auxin signaling in Arabidopsis leaves. Development, 136, 3235–3246.  Douglas, C.J. (1996) Phenylpropanoid metabolism and lignin biosynthesis: From weeds to trees. Trends Plant Sci., 1, 171–178. Douglas, S.J., Chuck, G., Dengler, R.E., Pelecanda, L. and Riggs, C.D. (2002) KNAT1 and ERECTA Regulate Inflorescence Architecture in Arabidopsis. Society, 14, 547–558. Drevensek, S., Goussot, M., Duroc, Y., Christodoulidou, A., Steyaert, S., Schaefer, E., Duvernois, E., Grandjean, O., Vantard, M., Bouchez, D. and Pastuglia, M. (2012) The Arabidopsis TRM1-TON1 Interaction Reveals a Recruitment Network Common to Plant Cortical Microtubule Arrays and Eukaryotic Centrosomes. Plant Cell, 24, 178–191.  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., Ellie, B.E., Bohlmannn, 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, 618–640. 114   Etchells, J.P., Provost, C.M., Mishra, L. and Turner, S.R. (2013) WOX4 and WOX14 act downstream of the PXY receptor kinase to regulate plant vascular proliferation independently of any role in vascular organisation. Development, 140, 2224–2234.  Etchells, J.P. and Turner, S.R. (2010) The PXY-CLE41 receptor ligand pair defines a multifunctional pathway that controls the rate and orientation of vascular cell division. Development, 137, 767–774.  Fisher, K. and Turner, S. (2007) PXY, a Receptor-like Kinase Essential for Maintaining Polarity during Plant Vascular-Tissue Development. Curr. Biol., 17, 1061–1066. Fukuda, H. (1996) XYLOGENESIS : INITIATION , PROGRESSION , AND CELL DEATH. , 299–325. Funk, V., Kositsup, B., Zhao, C. and Beers, E.P. (2002) The Arabidopsis xylem peptidase XCP1 is a tracheary element vacuolar protein that may be a papain ortholog. Plant Physiol., 128, 84–94.  Furumizu, C., Alvarez, J.P. aul, Sakakibara, K. and Bowman, J.L. (2015a) Antagonistic roles for KNOX1 and KNOX2 genes in patterning the land plant body plan following an ancient gene duplication. PLoS Genet., 11, e1004980.  Galichet, A. and Gruissem, W. (2006) Developmentally Controlled Farnesylation Modulates AtNAP1;1 Function in Cell Proliferation and Cell Expansion during Arabidopsis Leaf Development. Plant Physiol., 142, 1412–1426.  Greco, M., Chiappetta, A., Bruno, L. and Bitonti, M.B. (2012) Xylem tissue specification, patterning, and differentiation mechanisms. J. Exp. Bot., 63, 695–709. Groover, A.T., Mansfield, S.D., DiFazio, S.P., Dupper, G., Fontana, J.R., Millar, R. and Wang, Y. (2006) The Populus homeobox gene ARBORKNOX1 reveals overlapping 115   mechanisms regulating the shoot apical meristem and the vascular cambium. Plant Mol. Biol., 61, 917–932. Guo, D., Chen, F., Inoue, K., Blount, J.W. and Dixon, R. (2001) Downregulation of caffeic acid 3-O-methyltransferase and caffeoyl CoA 3-O-methyltransferase in transgenic alfalfa. impacts on lignin structure and implications for the biosynthesis of G and S lignin. Plant Cell, 13, 73–88. Guo, J., Wang, S., Wang, J., Huang, W.D., Liang, J. and Chen, J.G. (2009) Dissection of the relationship between RACK1 and heterotrimeric G-proteins in Arabidopsis. Plant Cell Physiol., 50, 1681–1694. Hackbusch, J., Richter, K., Müller, J., Salamini, F. and Uhrig, J.F. (2005) A central role of Arabidopsis thaliana ovate family proteins in networking and subcellular localization of 3-aa loop extension homeodomain proteins. P Natl Acad Sci USA, 102, 4908–4912. Hake, S., Smith, H.M.S., Holtan, H., Magnani, E., Mele, G. and Ramirez, J. (2004) the Role of Knox Genes in Plant Development. Annu. Rev. Cell Dev. Biol, 20, 125–51. Hao, Z. and Mohnen, D. (2014) A review of xylan and lignin biosynthesis: Foundation for studying Arabidopsis irregular xylem mutants with pleiotropic phenotypes. Crit. Rev. Biochem. Mol. Biol., 49, 212–241. Hay, A. and Tsiantis, M. (2010) KNOX genes: versatile regulators of plant development and diversity. Development, 137, 3153–3165. Hirakawa, Y., Kondo, Y. and Fukuda, H. (2010) TDIF Peptide Signaling Regulates Vascular Stem Cell Proliferation via the WOX4 Homeobox Gene in Arabidopsis. Plant Cell, 22, 2618–2629.  Hirakawa, Y., Shinohara, H., Kondo, Y., Inoue, A., Nakanomyo, I., Ogawa, M., Sawa, 116   S., Ohashi-Ito, K., Mastsubayashi, Y. and Fukuda, H. (2008) Non-cell-autonomous control of vascular stem cell fate by a CLE peptide/receptor system. Proc. Natl. Acad. Sci., 105, 15208–15213.  Hiratsu, K., Mitsuda, N., Matsui, K. and Ohme-Takagi, M. (2004) Identification of the minimal repression domain of SUPERMAN shows that the DLELRL hexapeptide is both necessary and sufficient for repression of transcription in Arabidopsis. Biochem. Biophys. Res. Commun., 321, 172–178. Hongo, S., Sato, K., Yokoyama, R. and Nishitani, K. (2012) Demethylesterification of the Primary Wall by PECTIN METHYLESTERASE35 Provides Mechanical Support to the Arabidopsis Stem. Plant Cell, 24, 2624–2634.  Hu, C.D., Chinenov, Y. and Kerppola, T.K. (2002) Visualization of interactions among bZIP and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol. Cell, 9, 789–798. Ito, Y., Nakanomyo, I., Hiroyasu, M., Iwamoto, K., Sawa, S., Dohmae, N. and Fukuda, H. (2006) Dodeca-CLE Peptides as Suppressors. Science (80), 842–845. Jin, H. (2000) Transcriptional repression by AtMYB4 controls production of UV-protecting sunscreens in Arabidopsis. EMBO J., 19, 6150–6161.  Jones, L., Ennos,  a R. and Turner, S.R. (2001) Cloning and characterization of irregular xylem4 (irx4): a severely lignin-deficient mutant of Arabidopsis. Plant J., 26, 205–216.  Jouanin, L., Goujon, T., Nadaï, V. de, Martin, M.T., Mila, I., Vallet, C., Poller, B., Yoshinaga, A., Chabbert, B., Petit-Conil, M. and Lapierre, C. (2000) Lignification in transgenic poplars with extremely reduced caffeic acid O-methyltransferase activity. Plant Physiol., 123, 1363–74.  117   Jung, J.H., Lee, M. and Park, C.M. (2010) A transcriptional feedback loop modulating signaling crosstalks between auxin and brassinosteroid in Arabidopsis. Mol. Cells, 29, 449–456. Kerstetter, R., Vollbrecht, E., Lowe, B., Veit, B., Yamaguchi, J. and Hake, S. (1994) Sequence-Analysis and Expression Patterns Divide the Maize Knotted1-Like Homeobox Genes into 2 Classes. Plant Cell, 6, 1877–1887.  Kim, D., Cho, Y.H., Ryu, H., Kim, Y., Kim, T.H. and Hwang, I. (2013) BLH1 and KNAT3 modulate ABA responses during germination and early seedling development in Arabidopsis. Plant J., 75, 755–766. Kim, W.C., Ko, J.H., Kim, J.Y., Kim, J., Bae, H.J. and Han, K.H. (2013) MYB46 directly regulates the gene expression of secondary wall-associated cellulose synthases in Arabidopsis. Plant J., 73, 26–36. Kimura, S., Koenig, D., Kang, J., Yoong, F.Y. and Sinha, N. (2008) Natural Variation in Leaf Morphology Results from Mutation of a Novel KNOX Gene. Curr. Biol., 18, 672–677. Knaap, E. van der, Chakrabarti, M., Chu, Y.H., Clevenger, J.P., Illa-Berenguer, E., Huang, Z., Keyhaninejad, N., Mu, Q., Sun, L., Wang, Y. and Wu, S. (2014) What lies beyond the eye: the molecular mechanisms regulating tomato fruit weight and shape. Front. Plant Sci., 5, 1–13.  Ko, J.-H., Kim, W.C. and Han, K.H. (2009) Ectopic expression of MYB46 identifies transcriptional regulatory genes involved in secondary wall biosynthesis in Arabidopsis. Plant J., 60, 649–665.  Ko, J.H., Yang, S.H., Park, A.H., Lerouxel, O. and Han, K.H. (2007) ANAC012, a member of the plant-specific NAC transcription factor family, negatively regulates xylary fiber development in Arabidopsis thaliana. Plant J., 50, 1035–1048. 118   Kondo, Y., Fujita, T., Sugiyama, M. and Fukuda, H. (2015) A novel system for xylem cell differentiation in arabidopsis thaliana. Mol. Plant, 8, 612–621.  Kondo, Y., Ito, T., Nakagami, H., Hirakawa, Y., Saito, M., Tamaki, T., Shirasu, K. and Fukuda, H. (2014) Plant GSK3 proteins regulate xylem cell differentiation downstream of TDIF-TDR signalling. Nat. Commun., 5, 1–11.  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 Dev., 19, 1855–1860. Lee, C., Teng, Q., Huang, W., Zhong, R. and Ye, Z.-H. (2010) The Arabidopsis Family GT43 Glycosyltransferases Form Two Functionally Nonredundant Groups Essential for the Elongation of Glucuronoxylan Backbone. Plant Physiol., 153, 526–541.  Lee, D., Meyer, K., Chapple, C. and Douglasa, C.J. (1998) Antisense suppression of 4-Coumarate: Coenzyme A Ligase activity in Arabidopsis leads to altered lignin subunit composition. , 9, 1985–1998. Li, E., Bhargava, A., Qiang, W., Friedmann, M.C., Forneris, N., Savidge, R.A., Johnson, L.A., Mansfield, S.D., Ellis, B.E. and Douglas, C.J. (2012) The Class II KNOX gene KNAT7 negatively regulates secondary wall formation in Arabidopsis and is functionally conserved in Populus. New Phytol., 194, 102–115. Li, E., Wang, S., Liu, Y., Chen, J. and Douglas, C.J. (2011) OVATE FAMILY PROTEIN4 ( OFP4 ) interaction with KNAT7 regulates secondary cell wall formation in Arabidopsis thaliana. , 4, 328–341. Li, J. and Nam, K.H. (2002) Regulation of Brassinosteroid Signaling by a GSK3 / SHAGGY-Like Kinase. Science (80-. )., 295, 1299–1301.  119   Li, Y., Qian, Q., Zhou, Y., Yan, M., Sun, L., Zhang, M., Fu, Z., Wang, Y., Han, B., Pang, X., Chen, M. and Li, J. (2003) BRITTLE CULM1 , which encodes a COBRA-like protein , affects the mechanical properties of rice plants. Plant Cell, 15, 2020–2031. Lie, C., Kelsom, C. and Wu, X. (2012) WOX2 and STIMPY-LIKE/WOX8 promote cotyledon boundary formation in Arabidopsis. Plant J., 72, 674–682. Liu, D., Sun, W., Yuan, Y., Zhang, N., Hayward, A., Liu, Y. and Wang, Y. (2014) Phylogenetic analyses provide the first insights into the evolution of OVATE family proteins in land plants. Ann. Bot., 113, 1219–1233. Liu, J., Eck, J. Van, Cong, B. and Tanksley, S.D. (2002) A new class of regulatory genes underlying the cause of pear-shaped tomato fruit. Proc. Natl. Acad. Sci. U. S. A., 99, 13302–6.  Liu, Y. and Douglas, C.J. (2015) A role for OVATE FAMILY PROTEIN1 (OFP1) and OFP4 in a BLH6-KNAT7 multi-protein complex regulating secondary cell wall formation in Arabidopsis thaliana. Plant Signal. Behav., 10, e1033126.  Liu, Y., You, S., Taylor-Teeples, M., Li, W.L., Schuetz, M., Brady, S.M. and Douglas, C.J. (2014a) BEL1-LIKE HOMEODOMAIN6 and KNOTTED ARABIDOPSIS THALIANA7 Interact and Regulate Secondary Cell Wall Formation via Repression of REVOLUTA. Plant Cell Online, 26, 4843–4861.  Liu, Z., Schneider, R., Kesten, C., Zhang, Y., Somssich, M., Zhang, Y., Fernie, A.R. and Persson, S. (2016) Cellulose-Microtubule Uncoupling Proteins Prevent Lateral Displacement of Microtubules during Cellulose Synthesis in Arabidopsis. Dev. Cell, 38, 305–315.  Liu, Z., Zhu, Y., Gao, J., Yu, F., Dong, A. and Shen, W.H. (2009) Molecular and reverse genetic characterization of NUCLEOSOME ASSEMBLY PROTEIN1 (NAP1) genes 120   unravels their function in transcription and nucleotide excision repair in Arabidopsis thaliana. Plant J., 59, 27–38. Long, J. a, Moan, E.I., Medford, J.I. and Barton, M.K. (1996) A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature, 379, 66–69. Lüscher, B. (2001) Function and regulation of the transcription factors of the Myc/Max/Mad network. Gene, 277, 1–14. MacMillan, C.P., Mansfield, S.D., Stachurski, Z.H., Evans, R. and Southerton, S.G. (2010) Fasciclin-like arabinogalactan proteins: Specialization for stem biomechanics and cell wall architecture in Arabidopsis and Eucalyptus. Plant J., 62, 689–703. Magnani, E. and Hake, S. (2008) KNOX lost the OX: the Arabidopsis KNATM gene defines a novel class of KNOX transcriptional regulators missing the homeodomain. Plant Cell, 20, 875–87.  Marita, J.M., Ralph, J., Hatfield, R.D. and Chapple, C. (1999) NMR characterization of lignins in Arabidopsis altered in the activity of ferulate 5-hydroxylase. Proc. Natl. Acad. Sci., 96, 12328–12332.  Mazur, E., Kurczyńska, E.U. and Friml, J. (2014) Cellular events during interfascicular cambium ontogenesis in inflorescence stems of Arabidopsis. Protoplasma, 251, 1125–1139. McCarthy, R.L., Zhong, R. and Ye, Z.H. (2009) MYB83 is a direct target of SND1 and acts redundantly with MYB46 in the regulation of secondary cell wall biosynthesis in arabidopsis. Plant Cell Physiol., 50, 1950–1964. McFarlane, H.E., Döring, A. and Persson, S. (2014) The Cell Biology of Cellulose Synthesis. Annu. Rev. Plant Biol., 65, 69–94.  121   Meents, M.J., Watanabe, Y. and Samuels, A.L. (2018) The cell biology of secondary cell wall biosynthesis. , 1–19. Mellerowicz, E.J., Baucher, M., Sundberg, B. and Boerjan, W. (2001) Unravelling cell wall formation in the woody dicot stem. Plant Mol. Biol., 47, 239–274. Mi, H., Muruganujan, A. and Thomas, P.D. (2013) PANTHER in 2013: Modeling the evolution of gene function, and other gene attributes, in the context of phylogenetic trees. Nucleic Acids Res., 41, 377–386. 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 Online, 19, 270–280.  Mottiar, Y., Vanholme, R., Boerjan, W., Ralph, J. and Mansfield, S.D. (2016) Designer lignins: Harnessing the plasticity of lignification. Curr. Opin. Biotechnol., 37, 190–200.  Mueller, S.C. and Malcolm Brown, R. (1980) Evidence for an intramembrane component associated with a cellulose microfibril-synthesizing complex in higher plants. J. Cell Biol., 84, 315–326. Mukherjee, K., Brocchieri, L. and Bürglin, T.R. (2009) A comprehensive classification and evolutionary analysis of plant homeobox genes. Mol. Biol. Evol., 26, 2775–2794. Nakano, Y., Yamaguchi, M., Endo, H., Rejab, N.A. and Ohtani, M. (2015) NAC-MYB-based transcriptional regulation of secondary cell wall biosynthesis in land plants. Front. Plant Sci., 6, 1–18.  Napier, R.M., Fowke, L.C., Hawes, C., Lewis, M. and Pelham, H.R. (1992) Immunological evidence that plants use both HDEL and KDEL for targeting proteins to the 122   endoplasmic reticulum. J. Cell Sci., 102, 261–71.  Nemoto, K., Seto, T., Takahashi, H., Nozawa, A., Seki, M., Shinozaki, K., Endo, Y. and Sawasaki, T. (2011) Autophosphorylation profiling of Arabidopsis protein kinases using the cell-free system. Phytochemistry, 72, 1136–1144.  Nixon, B.T., Mansouri, K., Singh, A., Du, J., Davis, J.K., Lee, J.G., Slabaugh, E., Vandavasi, V.G., O'Neil, H., Roberts, E.M., Roberts, A.W., Yingling, Y.G. and Haigler, C.H. (2016) Comparative Structural and Computational Analysis Supports Eighteen Cellulose Synthases in the Plant Cellulose Synthesis Complex. Sci. Rep., 6, 1–14.  Oda, Y. and Fukuda, H. (2012) Secondary cell wall patterning during xylem differentiation. Curr. Opin. Plant Biol., 15, 38–44.  Oehme, D.P., Downton, M.T., Doblin, M.S., Wagner, J., Gidley, M.J. and Bacic, A. (2015) Unique Aspects of the Structure and Dynamics of Elementary I β Cellulose Microfibrils Revealed by Computational Simulations. Plant Physiol., 168, 3–17.  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, 3461–3473.  Ohman, D., Demedts, B., Kumar, M., Gerber, L., Gorzsas, A., Goeminne, G., Hedenstrom, M., Ellis, B., Boerjan, W. and Sundberg, B. (2013) MYB103 is required for FERULATE-5-HYDROXYLASE expression and syringyl lignin biosynthesis in Arabidopsis stems. Plant J., 73, 63–76. Pagnussat, G.C., Yu, H.J. and Sundaresan, V. (2007) Cell-Fate Switch of Synergid to Egg Cell in Arabidopsis eostre Mutant Embryo Sacs Arises from Misexpression of the BEL1-Like Homeodomain Gene BLH1. Plant Cell Online, 19, 3578–3592.  123   Pawar, P.M.-A., Koutaniemi, S., Tenkanen, M. and Mellerowicz, E.J. (2013) Acetylation of woody lignocellulose: significance and regulation. Front. Plant Sci., 4, 1–8.  Peña, M.J., Kulkarni, A.R., Backe, J., Boyd, M., O’Neill, M.A. and York, W.S. (2016) Structural diversity of xylans in the cell walls of monocots. Planta, 244, 589–606. 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, 744–749.  Porth, I., Klápště, J., Skyba, O., . (2013) Populus trichocarpa cell wall chemistry and ultrastructure trait variation, genetic control and genetic correlations. New Phytol., 197, 777–90.  Rennie, E.A. and Scheller, H.V. (2014) Xylan biosynthesis. Curr. Opin. Biotechnol., 26, 100–107.  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. Rodriguez-Gacio, M. del C., Iglesias-Fernandez, R., Carbonero, P. and Matilla, A.J. (2012) Softening-up mannan-rich cell walls. J. Exp. Bot., 63, 3975–3988. Sakakibara, K., Nishiyama, T., Deguchi, H. and Hasebe, M. (2008) Class 1 KNOX genes are not involved in shoot development in the moss Physcomitrella patens but do function in sporophyte development. Evol. Dev., 10, 555–566. Samuels, A.L., Kaneda, M. and Rensing, K.H. (2006) The cell biology of wood formation: from cambial divisions to mature secondary xylemThis review is one of a selection of papers published in the Special Issue on Plant Cell Biology. Can. J. Bot., 84, 631–639.  124   Scheller, H.V. and Ulvskov, P. (2010) Hemicelluloses. Annu. Rev. Plant Biol., 61, 263–289.  Schneider, R., Tang, L., Lampugnani, E.R., Barkwill, S., Lathe, R., Zhang., Y., McFarlane, H.E., Pesquet, E., Niittyla, T., Mansfield, S.D., Zhou, Y. and Persson, S. (2017) Two Complementary Mechanisms Underpin Cell Wall Patterning during Xylem Vessel Development. Plant Cell, 29, 2433-2449.  Serikawa, K., Martinez-Laborda, A., Kim, H. and Zambryski, P. C. (1997) Localization of expression of KNAT3, a class 2 knotted1-like gene. Plant J., 11(4), 853–861.  Shi, C.-L., Stenvik, G.-E., Vie, A.K., Bones, A.M., Pautot, V., Proveniers, M., Aalen, R.B. and Butenko, M.A. (2011) Arabidopsis Class I KNOTTED-Like Homeobox Proteins Act Downstream in the IDA-HAE/HSL2 Floral Abscission Signaling Pathway. Plant Cell, 23, 2553–2567.  Shigeta, T., Yasuda, D., Mori, T., Yoshimitsu, Y., Nakamura, Y., Yoshida, S., Asami, T., Okamoto, S. and Matsuo, T. (2011) Characterization of brassinosteroid-regulated proteins in a nuclear-enriched fraction of Arabidopsis suspension-cultured cells. Plant Physiol. Biochem., 49, 985–995.  Shoji, T., Narita, N.N., Hayashi, K., Asada, J., Hamada, T., Sonobe, S., Nakajima, K. and Hashimoto, T. (2004) Plant-specific microtubule-associated protein SPIRAL2 is required for anisotropic growth in Arabidopsis. Plant Physiol., 136, 3933–3944. Signal.salk.edu. (2018). Salk_010386, accessed 22 Feb. 2018, <http://signal.salk.edu/cgi-bin/tdnaexpress?JOB=TEXT&TYPE=DATA&QUERY=SALK_010386.38.25.x>. Taylor-Teeples, M., Lin, L., Lucas, M. de, Turco, G., Toal, T.W., Gaudinier, A., Young, N.F., Trabucco, G.M., Veling, M.T., Lamothe, R., Handakumbura, P.P., Xiong, G., Wang, C., Corwin, J., Tsoukalas, A., Zhang, L., Ware, D., Pauly, M., Kliebenstein, D.J., Dehesh, K., Tagkopoulos, I., Breton, G., Pruneda-Paz, J.L., Ahnert, S.E., Kay, S.A., 125   Hazen, S. P. and Brady, S.M. (2014a) An Arabidopsis gene regulatory network for secondary cell wall synthesis. Nature, 517, 571–575.  Taylor, N.G. (2000) Multiple Cellulose Synthase Catalytic Subunits Are Required for Cellulose Synthesis in Arabidopsis. Plant Cell Online, 12, 2529–2540.  Taylor, N.G., Howells, R.M., Huttly, A.K., Vickers, K. and Turner, S.R. (2003) Interactions among three distinct CesA proteins essential for cellulose synthesis. Proc. Natl. Acad. Sci., 100, 1450–1455.  Taylor, N.G., Laurie, S. and Turner, S.R. (2000) Multiple Cellulose Synthase Catalytic Subunits Are Required for Cellulose Synthesis in Arabidopsis. Plant Cell, 12, 2529–2539.  Taylor, N.G., Scheible, W.R., Cutler, S., Somerville, C.R. and Turner, S.R. (1999) The irregular xylem3 locus of Arabidopsis encodes a cellulose synthase required for secondary cell wall synthesis. Plant Cell, 11, 769–780. Tobimatsu, Y., Chen, F., Nakashima, J., Escamilla-Trevino, L.L., Jackson, L., Dixon, R.A. and Ralph, J. (2013) Coexistence but Independent Biosynthesis of Catechyl and Guaiacyl/Syringyl Lignin Polymers in Seed Coats. Plant Cell, 25, 2587–2600.  Trapnell, C., Roberts, A., Goff, L., Pertea, G., Kim, D., Kelley, D.R., Pimentel, H., Salzberg, S.L., Rinn, J.L. and Pachter, L. (2012) Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat. Protoc., 7, 562–578.  Truernit, E., Siemering, K.R., Hodge, S., Grbic, V. and Haseloff, J. (2006) A map of KNAT gene expression in the Arabidopsis root. Plant Mol. Biol., 60, 1–20. Tsuda, K., Ito, Y., Sato, Y. and Kurata, N. (2011) Positive autoregulation of a KNOX gene is essential for shoot apical meristem maintenance in rice. Plant Cell, 23, 4368–4381.  126   Turner, S.R. (1997) Collapsed Xylem Phenotype of Arabidopsis Identifies Mutants Deficient in Cellulose Deposition in the Secondary Cell Wall. Plant Cell Online, 9, 689–701.  Turnera, S.R. and Somerville, C.R. (1997) Collapsed Xylem Phenotype of Arabidopsis ldentifies Mutants Deficient in Cellulose Deposition in the Secondary Cell Wall. , 9, 689–701. Ursache, R., Miyashima, S., Chen, Q., Vaten, A., Nakajima, K., Carlsbecker, A., Zhao, Y., Helariutta, Y. and Dettmer, J. (2014) Tryptophan-dependent auxin biosynthesis is required for HD-ZIP III-mediated xylem patterning. Development, 141, 1250–1259.  Ursache, R., Nieminen, K. and Helariutta, Y. (2013) Genetic and hormonal regulation of cambial development. Physiol. Plant., 147, 36–45. Vanholme, R., Cesarino, I., Rataj, K., Xiao, Y., Sundin, L., Goeminne, G., Kim, H., Cross, J., Morreel K., Araujo, P., Welsh, L., Haustraete, J., McClellam, 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, 342, 1327–1329. Velasquez, S.M., Ricardi, M.M., Dorosz, J.G., Fernandez, P.V., Nadra, A.D., Pol-Fachin, L., Egelund, J., Gille, S., Harholt, J., Cianacia, M., Verli, H., Pauly, M., Bacic, A., Olsen, C.E., Ulvskov, P., Petersen, B.L., Somerville, C., Lusem, N.D. and Esteves, J.M. (2011) O-Glycosylated Cell Wall Proteins Are Essential in Root Hair Growth. Science, 332, 1401–1403.  Venglat, S.P., Dumonceaux, T., Rozwadowski, K., Parnell, L., Babic, V., Keller, W., Martienssen, R., Selvaraj, G. and Datla, R. (2002) The homeobox gene BREVIPEDICELLUS is a key regulator of inflorescence architecture in Arabidopsis. Proc. Natl. Acad. Sci., 99, 4730–4735.  127   Vollbrecht, E., Reiser, L. and Hake, S. (2000) Shoot meristem size is dependent on inbred background and presence of the maize homeobox gene, knotted1. Development, 127, 3161–3172. Vollbrecht, E., Veit, B., Sinha, N. and Hake, S. (1991) The developmental gene Knotted-1 is a member of a maize homeobox gene family. Nature, 350, 241–243.  Wang, H., Avci, U., Nakashima, J., Hahn, M.G., Chen, F. and Dixon, R.A. (2010) Mutation of WRKY transcription factors initiates pith secondary wall formation and increases stem biomass in dicotyledonous plants. Proc. Natl. Acad. Sci., 107, 22338–22343.  Wang, H., Zhao, Q., Chen, F., Wang, M. and Dixon, R.A. (2011) NAC domain function and transcriptional control of a secondary cell wall master switch. Plant J., 68, 1104–1114. Wang, S., Chang, Y. and Ellis, B. (2016) Overview of OVATE FAMILY PROTEINS, A Novel Class of Plant-Specific Growth Regulators. Front. Plant Sci., 7, 1–8.  Wang, S., Chang, Y., Guo, J. and Chen, J.-G. (2007) Arabidopsis Ovate Family Protein 1 is a transcriptional repressor that suppresses cell elongation. Plant J., 50, 858–872.  Wang, S., Chang, Y., Guo, J., Zeng, Q., Ellis, B.E. and Chen, J.G. (2011) Arabidopsis Ovate family proteins, a novel transcriptional repressor family, control multiple aspects of plant growth and development. PLoS One, 6. Wang, Y.K., Chang, W.C., Liu, P.F., Hsiao, M.K., Lin, C.T., Lin, S.M. and Pan, R.L. (2010) Ovate family protein 1 as a plant Ku70 interacting protein involving in DNA double-strand break repair. Plant Mol. Biol., 74, 453–466. Weng, J.K., Mo, H. and Chapple, C. (2010) Over-expression of F5H in COMT-deficient Arabidopsis leads to enrichment of an unusual lignin and disruption of pollen wall formation. Plant J., 64, 898–911. 128   Wu, A.-M., Hornblad, E., Voxeur, A., Gerber, L., Rihouey, C., Lerouge, P. and Marchant, A. (2010) Analysis of the Arabidopsis IRX9/IRX9-L and IRX14/IRX14-L Pairs of Glycosyltransferase Genes Reveals Critical Contributions to Biosynthesis of the Hemicellulose Glucuronoxylan. Plant Physiol., 153, 542–554.  Xiao, Y., Liu, D., Zhang, G., Tong, H. and Chu, C. (2017) Brassinosteroids Regulate OFP1, a DLT Interacting Protein, to Modulate Plant Architecture and Grain Morphology in Rice. Front. Plant Sci., 8, 1–13.  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. (2010) 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, 906–914.  Yamaguchi, M., Mitsuda, N., Ohtani, M., Ohme-takagi, M., Kato, K. and Demura, T. (2011) VASCULAR-RELATED NAC-DOMAIN 7 directly regulates the expression of a broad range of genes for xylem vessel formation. , 579–590. Yamaguchi, M., Ohtani, M., Mitsuda, N., Kubo, M., Ohme-Takagi, M., Fukuda, H. and Demura, T. (2010) VND-INTERACTING2, a NAC Domain Transcription Factor, Negatively Regulates Xylem Vessel Formation in Arabidopsis. Plant Cell, 22, 1249–1263.  Yang, C., Ma, Y., He, Y., Tian, Z. and Li, J. (2018) OsOFP19 modulates plant architecture by integrating the cell division pattern and brassinosteroid signaling. Plant J., 489–501.  Yang, C., Shen, W., He, Y., Tian, Z. and Li, J. (2016) OVATE Family Protein 8 Positively Mediates Brassinosteroid Signaling through Interacting with the GSK3-like Kinase in Rice. PLoS Genet., 12, 1–15. Yang, J.H. and Wang, H. (2016) Molecular Mechanisms for Vascular Development and 129   Secondary Cell Wall Formation. Front. Plant Sci., 7, 1–8.  Yin, Y., Wang, Z.Y., Mora-Garcia, S., Li, J., Yoshida, S., Asami, T. and Chory, J. (2002) BES1 accumulates in the nucleus in response to brassinosteroids to regulate gene expression and promote stem elongation. Cell, 109, 181–191. Zeng, W., Lampugnani, E.R., Picard, K.L., Song, L., Wu, A.M., Farion, I.M., Zhao, J., Ford, K., Doblin, M.S. and Bacic, A. (2016) Asparagus IRX9, IRX10, and IRX14A Are Components of an Active Xylan Backbone Synthase Complex that Forms in the Golgi Apparatus. Plant Physiol., 171, 93–109.  Zhang, J., Elo, A. and Helariutta, Y. (2011) Arabidopsis as a model for wood formation. Curr. Opin. Biotechnol., 22, 293–299. Zhang, L., Zhang, X., Ju, H., Chen, J., Wang, S., Wang, H., Zhao, Y. and Chang, Y. (2016) Ovate family protein1 interaction with BLH3 regulates transition timing from vegetative to reproductive phase in Arabidopsis. Biochem. Biophys. Res. Commun., 470, 492–497.  Zhao, C., Avci, U., Grant, E.H., Haigler, C.H. and Beers, E.P. (2008) XND1, a member of the NAC domain family in Arabidopsis thaliana, negatively regulates lignocellulose synthesis and programmed cell death in xylem. Plant J., 53, 425–436. Zhao, Q. (2016) Lignification: Flexibility, Biosynthesis and Regulation. Trends Plant Sci., 21, 713–721. Zhao, Q. and Dixon, R.A. (2011) Transcriptional networks for lignin biosynthesis: More complex than we thought? Trends Plant Sci., 16, 227–233.  Zhao, Q., Wang, H., Yin, Y., Xu, Y., Chen, F. and Dixon, R.A. (2010) Syringyl lignin biosynthesis is directly regulated by a secondary cell wall master switch. Proc. Natl. Acad. 130   Sci., 107, 14496–14501.  Zhong, R. (2005) Arabidopsis Fragile Fiber8, Which Encodes a Putative Glucuronyltransferase, Is Essential for Normal Secondary Wall Synthesis. Plant Cell Online, 17, 3390–3408. Zhong, R. (1997) Disruption of Interfascicular Fiber Differentiation in an Arabidopsis Mutant. Plant Cell Online, 9, 2159–2170.  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 Online, 18, 3158–3170.  Zhong, R., Lee, C., Zhou, J., McCarthy, R.L. and Ye, Z.H. (2008a) A Battery of Transcription Factors Involved in the Regulation of Secondary Cell Wall Biosynthesis in Arabidopsis. Plant Cell Online, 20, 2763–2782.  Zhong, R., Lee, C., Zhou, J., McCarthy, R.L. and Ye, Z.-H. (2008b) A Battery of Transcription Factors Involved in the Regulation of Secondary Cell Wall Biosynthesis in Arabidopsis. Plant Cell Online, 20, 2763–2782.  Zhong, R., Richardson, E.A. and Ye, Z.-H. (2007) The MYB46 Transcription Factor Is a Direct Target of SND1 and Regulates Secondary Wall Biosynthesis in Arabidopsis. Plant Cell Online, 19, 2776–2792.  Zhong, R., Richardson, E.A. and Ye, Z.H. (2007) Two NAC domain transcription factors, SND1 and NST1, function redundantly in regulation of secondary wall synthesis in fibers of Arabidopsis. Planta, 225, 1603–1611. Zhong, R. and Ye, Z.H. (2015) The Arabidopsis NAC transcription factor NST2 functions together with SND1 and NST1 to regulate secondary wall biosynthesis in fibers of 131   inflorescence stems. Plant Signal. Behav., 10, 1–7. Zhong, R. and Ye, Z.H. (2012) MYB46 and MYB83 bind to the SMRE sites and directly activate a suite of transcription factors and secondary wall biosynthetic genes. Plant Cell Physiol., 53, 368–380. Zhong, R. and Ye, Z.H. (2007) Regulation of cell wall biosynthesis. Curr. Opin. Plant Biol., 10, 564–572. 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, 248–266. Zhou, J., Zhong, R. and Ye, Z.H. (2014) Arabidopsis NAC domain proteins, VND1 to VND5, are transcriptional regulators of secondary wall biosynthesis in vessels. PLoS One, 9, e105726. Zhu, Y., Dong, A., Meyer, D., Pichon, O., Renou, J.P., Cao, K. and Shen, W.H. (2006) Arabidopsis NRP1 and NRP2 encode histone chaperones and are required for maintaining postembryonic root growth. Plant Cell, 18, 2879–2892.  Zimmermann, I.M., Heim, M.A., Weisshaar, B. and Uhrig, J.F. (2004) Comprehensive identification of Arabidopsis thaliana MYB transcription factors interacting with R/B-like BHLH proteins. Plant J., 40, 22–34.   

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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.24.1-0366123/manifest

Comment

Related Items