"Forestry, Faculty of"@en . "DSpace"@en . "UBCV"@en . "McDonnell, Lisa Marie"@en . "2011-01-05T19:53:32Z"@en . "2010"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "Cellulose synthases are the enzymes responsible for the production of cellulose in plant cell walls. Mutations in any one of the Arabidopsis cellulose synthase (CesA) AtCesA4, AtCesA7, and AtCesA8 genes cause plants to develop collapsed xylem as a result of reduced cellulose content, demonstrating their critical role in secondary cell wall biosynthesis. A thorough characterization of the growth, cell wall properties, and cellulose ultrastructure of the AtCesA4irx\u00E2\u0081\u00B5-\u00C2\u00B9, AtCesA7irx\u00C2\u00B3-\u00C2\u00B9, and AtCesA8irx\u00C2\u00B9-\u00C2\u00B9 mutants, presented herein, is the first report of the changes to cellulose microfibril angle, cell wall crystallinity, and cellulose degree of polymerization (DP) in these mutants. This study suggests that the non-redundant functions of individual CesAs may be related to CesA-specific thresholds required for the formation of a cellulose synthesizing complex (CSC), and CesA-specific roles in regulating crystallinity and DP. Additionally, the results illustrate the importance of a fully formed CSC in regulating cellulose microfibril angle. \nBy identifying and characterizing three new CesA genes from spruce (Picea glauca), PgCesA1, PgCesA2, and PgCesA3, which are homologous to the Arabidopsis AtCesA8, A4, and A7 and the Populus trichocarpa PtiCesA8-A, A4, and A7-A genes, respectively, the degree of functional conservation among AtCesA homologs was explored. Expression of PgCesA1 or the PtiCesAs in AtCesAirx plants rescued the collapsed xylem phenotype, thus demonstrating for the first time that orthologs of AtCesA4, A7, and A8 have conserved functions. \nLastly, in planta techniques were used to measure interactions between AtCesAs to investigate if specific and consistent interactions exist. The results suggest that CesA8 and A4 can form homodimers in planta, and that there might be weak or transient interactions between AtCesA7-A4 and AtCesA7-A8. \n\tCollectively, the results presented suggest, indirectly, an unequal ratio of CesA subunits (AtCesA4:A7:A8) is required for proper cellulose biosynthesis, and that each CesA likely has a unique function which ultimately affects cellulose properties such as cell wall crystallinity and DP. Our conclusions shed new light on the role of CesAs in cellulose biosynthesis in secondary cell walls and elicit questions about the current model of CSC form and function."@en . "https://circle.library.ubc.ca/rest/handle/2429/30534?expand=metadata"@en . " INVESTIGATING THE ROLE OF CELLULOSE SYNTHASES IN THE BIOSYNTHESIS AND PROPERTIES OF CELLULOSE IN SECONDARY CELL WALLS by Lisa Marie McDonnell B.Sc., Carleton University, 2002 M.Sc., Carleton University, 2004 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Forestry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) December 2010 \u00C2\u00A9 Lisa Marie McDonnell, 2010 ii Abstract Cellulose synthases are the enzymes responsible for the production of cellulose in plant cell walls. Mutations in any one of the Arabidopsis cellulose synthase (CesA) AtCesA4, AtCesA7, and AtCesA8 genes cause plants to develop collapsed xylem as a result of reduced cellulose content, demonstrating their critical role in secondary cell wall biosynthesis. A thorough characterization of the growth, cell wall properties, and cellulose ultrastructure of the AtCesA4irx5-1, AtCesA7irx3-1, and AtCesA8irx1-1 mutants, presented herein, is the first report of the changes to cellulose microfibril angle, cell wall crystallinity, and cellulose degree of polymerization (DP) in these mutants. This study suggests that the non-redundant functions of individual CesAs may be related to CesA- specific thresholds required for the formation of a cellulose synthesizing complex (CSC), and CesA-specific roles in regulating crystallinity and DP. Additionally, the results illustrate the importance of a fully formed CSC in regulating cellulose microfibril angle. By identifying and characterizing three new CesA genes from spruce (Picea glauca), PgCesA1, PgCesA2, and PgCesA3, which are homologous to the Arabidopsis AtCesA8, A4, and A7 and the Populus trichocarpa PtiCesA8-A, A4, and A7-A genes, respectively, the degree of functional conservation among AtCesA homologs was explored. Expression of PgCesA1 or the PtiCesAs in AtCesAirx plants rescued the collapsed xylem phenotype, thus demonstrating for the first time that orthologs of AtCesA4, A7, and A8 have conserved functions. Lastly, in planta techniques were used to measure interactions between AtCesAs to investigate if specific and consistent interactions exist. The results suggest that CesA8 and A4 can form homodimers in planta, and that there might be weak or transient interactions between AtCesA7-A4 and AtCesA7-A8. Collectively, the results presented suggest, indirectly, an unequal ratio of CesA subunits (AtCesA4:A7:A8) is required for proper cellulose biosynthesis, and that each CesA likely has a unique function which ultimately affects cellulose properties such as cell wall crystallinity and DP. Our conclusions shed new light on the role of CesAs in cellulose biosynthesis in secondary cell walls and elicit questions about the current model of CSC form and function. iii Preface This thesis contains three chapters written with the intent of publication in peer-reviewed journals. Chapter 2: Lisa McDonnell designed the research project, performed the research, conducted data analysis, and prepared the manuscript. Laura Haley and Ilga Porth assisted in performing research. Shawn Mansfield was involved with design of the research program, providing research opportunity, and editing the manuscript. Chapter 3: Lisa McDonnell designed the research project, performed the research, conducted data analysis, and prepared the manuscript. Rachel Liu, Tracy Kim, and Laura Haley assisted in performing research. Shawn Mansfield was with identification and design of the research program, providing research opportunity, and editing the manuscript. Chapter 4: Lisa McDonnell designed the research project, performed the research, conducted data analysis, and prepared the manuscript. Shawn Mansfield was involved with design of the research program, providing research opportunity, and editing the manuscript. iv Table of contents Abstract ............................................................................................................................ii Preface ............................................................................................................................ iii Table of contents .............................................................................................................iv List of tables .................................................................................................................... x List of figures ...................................................................................................................xi List of abbreviations ...................................................................................................... xiv Acknowledgements ........................................................................................................xv CHAPTER 1 Introduction ............................................................................................... 1 1.1 Overview ................................................................................................................ 1 1.2 The cell wall ........................................................................................................... 1 1.3 Cellulose synthase genes ...................................................................................... 6 1.3.1 Bacteria and non-vascular plants .................................................................... 6 1.3.2 Vascular plants ................................................................................................ 6 1.3.2.1 Arabidopsis thaliana ................................................................................. 7 1.3.2.2 Monocots .................................................................................................. 8 1.3.2.3 Trees ........................................................................................................ 9 1.3.2.3.1 Poplar ................................................................................................. 9 1.3.2.3.2 Eucalyptus ....................................................................................... 11 1.3.2.3.3 Pine .................................................................................................. 11 1.4 CesA protein structure ......................................................................................... 12 1.5 Cellulose synthase complex ................................................................................ 12 1.6 The link between CesAs, rosettes, and the CSC ................................................. 13 1.7 Composition of the CSC ...................................................................................... 14 1.8 Transport of CSCs ............................................................................................... 16 v 1.9 The cytoskeleton and cellulose microfibril orientation .......................................... 17 1.10 Goals and hypothesis ........................................................................................ 20 1.10.1 Chapter 2: Characterization of Arabidopsis CesA secondary cell wall- specific mutants ..................................................................................................... 20 1.10.2 Chapter 3: Identification of secondary cell wall-specific cellulose synthase genes from Picea glauca and conservation of function among CesA orthologs ..... 21 1.10.3 Chapter 4: Measuring CesA-CesA interactions in planta ........................... 21 CHAPTER 2 Characterization of Arabidopsis CesA secondary cell wall-specific mutants ...................................................................................................................................... 22 2.1 Introduction .......................................................................................................... 22 2.2 Materials and methods ........................................................................................ 28 2.2.1 AtCesA mutant nomenclature ....................................................................... 28 2.2.2 Plant growth .................................................................................................. 28 2.2.3 Xylem morphology ........................................................................................ 28 2.2.4 Structural carbohydrate and starch analysis ................................................. 29 2.2.5 Microfibril angle and crystallinity by X-ray diffraction ..................................... 30 2.2.6 Preparation of \u00EF\u0081\u00A1-cellulose ............................................................................. 30 2.2.7 Degree of polymerization of cellulose fibres .................................................. 31 2.2.8 RNA extraction and cDNA synthesis for gene expression analysis ............... 32 2.2.9 Real-time PCR analysis ................................................................................ 32 2.2.10 Plant transformation vectors ........................................................................ 34 2.2.11 Plant transformation .................................................................................... 35 2.2.12 Live-cell image acquisition of YFP-CesAs ................................................... 36 2.3 Results ................................................................................................................. 36 2.3.1 AtCesAirx phenotype ................................................................................... 36 2.3.2 Carbohydrate and lignin content ................................................................. 40 vi 2.3.2.1 Cellulose ................................................................................................. 40 2.3.2.2 Lignin ...................................................................................................... 44 2.3.2.3 Hemicellulose and pectin ........................................................................ 47 2.3.3 Cellulose characterization ........................................................................... 47 2.3.4 Expression of AtCesAs in AtCesAirx and wild-type plants ........................... 51 2.3.4.1 P35S::AtCesA expression in AtCesAirx plants .......................................... 52 2.3.4.2 PAtCesA::AtCesA expression in AtCesAirx plants .................................... 55 2.3.4.3 Double-transgenic wild-type plants .......................................................... 55 2.3.5 Visual profile of YFP-CesA7 ....................................................................... 58 2.3.6 Visualization of YFP-CesA8 in wild-type plants .......................................... 71 2.4 Discussion....................................................................................................... 73 2.4.1 Growth defects of AtCesAirx mutants manifest differently under varied day- length conditions .................................................................................................... 73 2.4.2 Increased lignin and hemicellulose content in AtCesAirx mutants suggest compensatory feedback in response to secondary cell wall perturbations ............. 76 2.4.3 Mutant-specific differences indicate the importance of a complete CSC and suggest CesA-specific functions in cellulose biosynthesis ..................................... 78 2.4.4 Reporter tags may effect CesA function ........................................................ 82 2.4.5 Exogenous expression of AtCesAs in wild-type plants causes a severe mutant phenotype .................................................................................................. 84 2.4.6 YFP-CesA7 expression patterns reveal a strong presence of AtCesA7 in fibres ...................................................................................................................... 84 CHAPTER 3 Identification of secondary cell wall-specific cellulose synthase genes from Picea glauca and conservation of function among CesA orthologs ............................... 88 3.1 Introduction .......................................................................................................... 88 3.2 Materials and methods ........................................................................................ 89 vii 3.2.1 Isolation of full-length CesA cDNA ................................................................ 89 3.2.2 RNA extraction and cDNA synthesis for gene expression ............................ 90 3.2.3 Real-time PCR analysis ................................................................................ 92 3.2.4 Phylogenetic analysis .................................................................................... 92 3.2.5 Plant transformation vectors.......................................................................... 95 3.2.6 Plant growth and transformations .................................................................. 95 3.2.7 Complementation assay ................................................................................ 96 3.3 Results ................................................................................................................. 97 3.3.2 Spruce CesA gene isolation and sequence analysis ..................................... 97 3.3.3 Phylogenetic analysis .................................................................................. 102 3.3.4 Expression profiling of PgCesA1, A2, and A3 ............................................. 102 3.3.5 Ability of spruce CesAs to functionally complement Arabidopsis CesAirx mutants ................................................................................................................ 105 3.3.5.1 PgCesA1 .............................................................................................. 108 3.3.5.2 PgCesA3 .............................................................................................. 112 3.3.5.3 PgCesA2 .............................................................................................. 114 3.3.6 Ability of Poplar CesAs to complement Arabidopsis CesAirx mutants .......... 114 3.4 Discussion ......................................................................................................... 116 3.4.1 Three unique cellulose synthase genes identified in spruce ....................... 116 3.4.2 Evolutionary relationship of PgCesA1, A2, and A3 with other known cellulose synthases ............................................................................................................. 116 3.4.3 PgCesA1, A2, and A3 expression suggests involvement in secondary cell wall formation .............................................................................................................. 119 3.4.4 Significant conservation of function exists between CesA orthologs ........... 122 3.4.5 PgCesA2 and PgCesA3 may be distant orthologs of either AtCesA4 or AtCesA7 ............................................................................................................... 124 viii CHAPTER 4 Measuring CesA-CesA interactions in planta ......................................... 127 4.1 Introduction ........................................................................................................ 127 4.2 Materials and methods ...................................................................................... 130 4.2.1 Plant transformation vectors........................................................................ 130 4.2.2 Plant growth and transformations ................................................................ 132 4.2.3 BRET methods ............................................................................................ 133 4.2.4 BiFC in transiently transformed tobacco leaf cells ....................................... 134 4.3 Results ............................................................................................................... 135 4.3.1 BRET in Arabidopsis seedlings and stem tissues ....................................... 135 4.3.1.1 Comparing controls and tissue types .................................................... 135 4.3.1.2 CesA7-CesA7 interactions .................................................................... 138 4.3.1.3 CesA7-CesA8 interactions .................................................................... 138 4.3.1.4 CesA7-CesA4 interactions .................................................................... 142 4.3.1.5 CesA8-CesA8 interactions .................................................................... 142 4.3.2 BiFC in transiently transformed tobacco epidermal cells ............................. 147 4.4 Discussion ......................................................................................................... 149 4.4.1 Putative CesA homodimers and heterodimers detected in planta using BRET and BiFC .............................................................................................................. 149 4.4.2 Modifications to the CSC and rosette models ............................................. 152 CHAPTER 5 Conclusion ............................................................................................ 155 5.1 Thesis summary ................................................................................................ 155 5.2 A complete CSC influences the fundamental properties of cellulose ................. 157 5.3 The function of AtCesA4, A7, and A8 orthologs is highly conserved ................. 158 5.4 What is the ratio of CesAs within the CSC? ....................................................... 160 5.5 Should the cell wall-specific classifications of CesAs be revisited? ................... 161 ix 5.6 Critical comments and future work recommendations ....................................... 163 5.6.1 Understanding the secondary effects of AtCesA4irx5-1, A7irx3-1, and A8irx1-1 mutations ............................................................................................................. 163 5.6.2 Investigate CSC composition and function more thoroughly ....................... 163 5.6.3 Broaden our understanding of conservation of function among CesAs ....... 165 5.7 Final conclusions and relevance ........................................................................ 165 REFERENCES ............................................................................................................ 167 Appendix A: AtCesA gene expression in mature stem tissue of Arabidopsis thaliana plants. .......................................................................................................................... 182 x List of tables Table 2.1. A summary of AtCesAirx phenotypes of AtCesA4, A7, and A8 mutants reported in the literature. ............................................................................................... 27 Table 2.2. PCR primers used for gene isolation, cloning, screening, and real-time (RT) PCR ............................................................................................................................... 33 Table 2.3. Maximum stem height and rosette-leaf diameter of wild-type and AtCesAirx plants grown under long-day and short-day conditions. .............................................. 39 Table 2.4. Cell wall carbohydrate content of wild-type (WT) and AtCesAirx Arabidopsis inflorescence stems from plants grown under long-day and short-day conditions. ....... 48 Table 2.5. Microfibril angle, cell wall crystallinity, and relative degree of polymerization (DP) of cellulose from stems of wild-type (WT) and AtCesAirx plants. ........................... 49 Table 3.1. Sequences of primers used for isolation, cloning, real-time PCR (RT), and screening. ...................................................................................................................... 91 Table 3.2. A list of CesA sequences used for phylogenetic analysis and associated GenBank accession numbers.. ..................................................................................... 93 Table 3.3. Amino acid residue shared identities of deduced PgCesA proteins with each other and the most similar CesAs from Arabidopsis (At) and Populus trichocarpa (Pti) 99 Table 3.4. Phenotype survey of A8irx1-1-LUC-PgCesA1 transformed lines. Eight AtCesA8irx1-1 lines expressing LUC-PgCesA1 were assessed for complementation of the AtCesAirx phenotype based on xylem morphology (collapsed or not) and growth (stature). ...................................................................................................................... 110 Table 3.5. Real time PCR analysis of LUC-PtiCesA transgene expression in complemented Arabidopsis AtCesAirx mutant lines.. ................................................... 115 Table 4.1. PCR primers used for cloning, screening, and real-time (RT) PCR. .......... 131 Table 4.2. YFP, LUC, BRET, and transcript abundance in whole seedlings carrying two AtCesA7-BRET constructs .......................................................................................... 139 Table 4.3. LUC and YFP emissions of LUC-A4 x YFP-A7 transgenic plants .............. 144 xi List of figures Figure 2.1. A schematic diagram illustrating the location of various mutations in Arabidopsis AtCesA4 (A4), AtCesA7 (A7), and AtCesA8 (A8) genes ........................... 26 Figure 2.2. Growth and xylem phenotype of wild-type and AtCesAirx plants ................. 38 Figure 2.3. Stem growth rate of wild-type and AtCesAirx plants grown under varied day- length conditions. .......................................................................................................... 41 Figure 2.4. Chronological progression of flower and silique formation of wild-type and AtCesAirx Arabidopsis plants. ........................................................................................ 42 Figure 2.5. Structural carbohydrate content of wild-type and AtCesAirx Arabidopsis stems... .......................................................................................................................... 43 Figure 2.6.\u00EF\u0080\u00A0\u00EF\u0080\u00A0\u00EF\u0081\u00A1-cellulose content of wild-type and AtCesAirx stems ............................... 45 Figure 2.7. A summary of growth and cell wall changes measured in AtCesA8irx1-1, AtCesA7irx3-1, and AtCesA4irx5-1 plants grown under long and short day conditions. ..... 46 Figure 2.8. AtCesAirx plants transformed with P35S::LUC-AtCesA constructs do not show recovery of a wild-type phenotype. ...................................................................... 53 Figure 2.9. Real-time PCR analysis of transcript abundance in RNA from stems of 15- day-old plants. ............................................................................................................... 54 Figure 2.10. Phenotype of AtCesAirx plants transformed with wild-type AtCesAs under the control of native AtCesA promoters.. ....................................................................... 56 Figure 2.11. Real-time PCR analysis of transcript abundance of AtCesAirx plants transformed with endogenous AtCesAs under native AtCesA promoters ..................... 57 Figure 2.12. Wild-type (ecotype Columbia) plants transformed with two AtCesA- expression constructs show a severe AtCesAirx-like phenotype. ................................... 59 Figure 2.13. Wild-type (ecotype Landsberg) plants transformed with two AtCesA8 expression constructs show a mostly wild-type like phenotype. .................................... 60 Figure 2.14. YFP-CesA7 expression patterns in young tissues of AtCesA7irx3-1- PAtCesA7::YFP-AtCesA7 plants. .................................................................................. 61 xii Figure 2.15. Cellular anatomy and YFP-CesA7 expression patterns in the upper stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants. ............................................................. 64 Figure 2.16. Cellular anatomy and YFP-CesA7 expression patterns in the upper middle stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants ................................................. 66 Figure 2.17. Cellular anatomy and YFP-CesA7 expression patterns in the lower middle stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants. ................................................ 68 Figure 2.18. Cellular anatomy and YFP-CesA7 expression patterns near the base of the stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants ........................................... 70 Figure 2.19. YFP-CesA8 expression patterns in xylem of wild-type plants transformed with PAtCesA8::YFP-AtCesA8 (PA8-YFP-A8). ............................................................. 72 Figure 2.20. Hypothetical effects of AtCesA mutations on CSC form and function ..... 75 Figure 3.2. A dendrogram assembled from the alignment of the deduced amino acid sequences of PgCesA1, A2, and A3 with 48 other confirmed and putative full-length cellulose synthases ..................................................................................................... 104 Figure 3.3. Real-time PCR analysis of PgCesA1, PgCesA2, and PgCesA3 transcript abundance in tissues of 5-year-old spruce trees. ........................................................ 106 Figure 3.4. PgCesA1 does not rescue the A8irx1-1 mutant phenotype under expressional control of the 35S promoter. ........................................................................................ 107 Figure 3.5. Complementation of the A8irx1-1 mutant phenotype by LUC-PgCesA1 under expressional control of the native AtCesA8 promoter ................................................. 109 Figure 3.6. Additional A8irx1-1 plants expressing LUC-PgCesA1 under the regulation of the AtCesA8 promoter show varying degrees of complementation ............................. 111 Figure 3.7. P35S::LUC-PgCesA3 is not sufficient to rescue the A4irx5-1 mutant phenotype. .................................................................................................................. 113 Figure 3.8. Complementation of Arabidopsis AtCesAirx mutants by PtiCesAs. ........... 117 Figure 4.1. BRET and BiFC assays ............................................................................ 129 Figure 4.2. YFP expression patterns in BRET control plants observed using confocal fluorescence microscopy. ............................................................................................ 136 xiii Figure 4.3. Varied expression and BRET of the LUC-YFP control. ............................ 137 Figure 4.4. BRET measurements and transcript abundance in PA8::LUC-CesA8 x P35S::YFP-CesA7 seedlings. ..................................................................................... 141 Figure 4.5. BRET measured in stem tissue and seedlings of double transgenic plants carrying PA4::LUC-CesA4 and PA7::YFP-CesA7. ...................................................... 145 Figure 4.6. BRET measurements in AtCesA8xA8 double transgenic plants. ............. 146 Figure 4.7. BiFC assays in leaf epidermal cells of Nicotiana tabacum ....................... 148 Figure 4.8. CSC composition revisited. ...................................................................... 153 xiv List of abbreviations AI acid insoluble lignin AS acid soluble lignin BiFC bimolecular fluorescence complementation BRET bioluminescent resonance energy transfer CesA cellulose synthase CSC cellulose synthesizing complex CSR class-specific region DCB 2, 6-dichlorobenzonitrile DP degree of polymerization GFP green fluorescent protein GPI glycosylphosphatidylinositol GX Glucuronoxylan HVR hyper-variable region IRX irregular xylem KOR Korrigan LUC Luciferase MAP microtubule associated protein MASC microtuble-associated cellulose synthase compartment MF Microfibril MFA microfibril angle MT Microtubule RT-PCR real-time PCR SmaCC small CesA compartment SUSY sucrose synthase UDP uridine diphosphate WT wild-type YFP yellow fluorescent protein xv Acknowledgements I sincerely thank Dr. Shawn Mansfield for allowing me to join his lab (and stay for so long), and giving me the chance to learn and grow as a scientist and teacher. Thank you for providing me with the opportunity to supervise students, and take on new challenges. Thank you to Dr. Lacey Samuels for providing helpful scientific insight, encouragement, and guidance as a supervisory committee member and teacher. Thank you to Dr. Geoff Wasteneys for providing helpful scientific insight as a supervisory committee member, and for providing me with access to your fluorescence microscopy lab. A very big thank you to Dr. Carol Pollock and members of the Biology teaching faculty: I am grateful for the opportunities you gave me to learn about teaching, which were true highlights of time as a PhD student. Thank you to past and present members of the Mansfield Lab for helpful scientific and technical support. Thank you to Dr. Heather Coleman for participating in very important scientific discussions on a daily basis. An extended thank you goes to Dr. Thomas Canam for patiently answering my countless questions. Also, to Faride Unda for her technical support and for always being a patient, friendly person in the lab. To Drs. Kyu-Young Kang and Ji-Young Park for their technical support and insight. Thank you to summer NSERC students Rachel Liu, Laura Haley, and Tracey Kim for being fun to work with, and providing help with my research projects. Thank you to Dr. Miki Fujita for many hours of help and technical support with confocal microscopy. Thank you to the staff of the Bioimaging Facility for your technical support. Thank you to Dr. Sharon Regan, who instilled in me a true interest in research, and provided me with my first opportunities in the lab. Thank you to the UBC Let\u00E2\u0080\u0099s Talk Science Partnership Program for invaluable experiences in teaching, learning, and fun. Thank you to Dr. Catherine Anderson for providing me with incredible work opportunities, and for great conversations. Thank you to my friends for allowing me to complain (too much), making me laugh, and providing words of encouragement. Especially to Heather: for being an amazing person, and a dear friend. To Martha Mullally: for lending a sympathetic scientific ear, and being a ray of sunshine on things personal and professional. To Ryan: for your continued encouragement. To Laura: for always being a fun and pleasant person to be around, for, listening, and all your help and support. Lastly, the biggest thank you goes to my family. Your unwavering support has made these experiences possible. To Jim, for making our home the happiest place in my world, and always providing support and understanding, even on an empty tank. To our son, Colin, for being the most amazing boy in the world, for making me laugh every day, and reminding me that life is a miracle. To my parents, Joanne and Rick McDonnell, for always believing in me and teaching me the value of hard work, and always visiting. To my sisters, Vicky and Karen: for being wonderful people, fun friends to spend time with, and being so supportive and understanding. To Karen for giving up eight months of her life to help us get where we are today. To my grandparents, Don and Helen Cofell, for always expressing interest in what I do and your continued encouragement. To Joe and Debbie Cooke, for your positivity, encouragement, and fun times together. To my extended family: for always having words of encouragement. xvi Dedication I dedicate this thesis to my family. 1 CHAPTER 1 Introduction 1.1 Overview The inherent properties of the secondary cell walls of woody tissue ultimately affect the final properties of wood. Thus, understanding the molecular and mechanical mechanisms governing cellulose biosynthesis is an important and intriguing area of research. The research described herein focuses on the process of cellulose production in the secondary cell wall, with an emphasis on the cellulose synthase enzymes responsible for its biosynthesis. Elucidating the roles of cellulose synthase enzymes in regulating cellulose properties not only offers a means to understand the fundamentals of plant and wood cell wall development, but also may offer a unique means to improve one of the most abundant industrially used natural polymers. 1.2 The cell wall The natural structure and chemistry of plant cell walls allow cells to expand, become rigid, provide immense structural support, allow water to be transported great distances, and protect plants from pathogens. The diversity of cell wall structures and functions immediately signifies the complex nature of cell wall biosynthesis. Of interest in the research presented in this thesis is the secondary cell wall formation of xylem and fibres. However, this does not undermine the importance of the work done on primary wall synthesis, which provides the scaffold for the process of secondary cell wall formation. Primary cell walls differ from secondary walls in both composition and function. Both primary and secondary walls contain a great deal of cellulose and polysaccharides. However, the hemicellulose found in primary walls of dicots is mostly xyloglucans, whereas xylans are the predominant hemicellulose in secondary walls (Mellerowicz and Sundberg, 2008). Additionally, primary walls contain significantly more pectins and proteins compared to secondary walls (Cosgrove, 2005). In general, primary walls are thinner and more flexible, allowing for cell expansion. In contrast, secondary walls are thick, rigid, and lignified. Hemicelluloses are heteropolymeric polysaccharides, and most commonly include xyloglucans, xylans, mannans, and glucomannans in vascular plants. 2 Xyloglucan is the dominant hemicellulose in primary walls of dicots (Scheller and Ulvskov, 2010), composed of 1,4-\u00EF\u0081\u00A2-linked glucan chains with frequent integrations of xylosyl residues and side chains composed of xylosyl, galactosyl, and fucose (Liepman et al., 2010). Glucuronoxylan (GX), a 1, 4-\u00EF\u0081\u00A2-linked xylan is an example of a hemicellulose found in Arabidopsis and woody plants (Liepman et al., 2010), and is the most common hemicellulose in secondary walls of angiosperms (Scheller and Ulvskov, 2010). Mannans (e.g. glucomannan) are also found in dicot secondary cell walls (Mellerowicz et al., 2001) including those of Arabidopsis xylem and fibres (Handford et al., 2003) and several functional mannan-synthases have been identified in Arabidopsis (Liepman et al., 2005). Hemicelluloses form a cross-linking network between cellulose microfibrils (Cosgrove, 2005), which can be modified during cell development (Cosgrove, 2005; Scheller and Ulvskov, 2010). In secondary cell walls, the most likely function of hemicellulose appears to be associated with load-bearing capacity and structural support. Some Arabidopsis irregular xylem mutants, with reduced xylan content and alterations to xylan backbone structure, exhibit thinner secondary cell walls and collapsed xylem, illustrating the role of xylans in providing structural integrity to the walls of fibres and vessels (Zhong et al., 2005; Brown et al., 2005; Brown et al., 2007). For example, the putative xylan glycosyltransferases FRA8, IRX8, and IRX9 were shown to be co-expressed with Arabidopsis secondary cell wall biosynthetic genes (Brown et al., 2005). Changes to GX properties in these mutants suggested that IRX9 might be involved in GX chain elongation, whereas FRA8 and IRX8 might be involved in the addition of glucuronic acid residues to the GX reducing ends (Zhong et al., 2005; Brown et al., 2007). Cross sections of stems of irx8 plants revealed an early onset of collapsed xylem, compared to the AtCesA8irx1-1 mutant, and a double irx8- AtCesA8irx1-1 mutant exhibited a more severe irx-like phenotype, emphasizing the importance of xylans in maintaining cell wall structural integrity especially in cellulose deficient mutants (Persson et al., 2007a). Recently, Wu et al. (2010) reported redundant and partially redundant functions of some glycosyltransferases involved in xylan chain elongation. With such a diversity of hemicelluloses found in plant cell walls, it is not surprising that the hemicellulose glycosyltransferase gene families are extremely large, and that the 3 functions of the coded enzymes are diverse, including (but likely not limited to) chain elongation, side-chain addition, and reducing end synthesis (reviewed by Scheller and Ulvskov, 2010). Lignins are another heteropolymer found in secondary cell walls, which are composed of phenylpropanoid subunits assembled in a random amorphous macromolecule. Cross-linkages between lignin and hemicelluloses, or lignin- carbohydrate complexes (Jeffries, 1990) likely add additional stiffness to plant cell walls. A correlation between lignin content, wood density, and stiffness has been reported along with reductions in stiffness in transgenic trees with reduced lignin content (Bjurhager et al., 2010). The hydrophobic nature of lignins likely enhances the capacity for vessels and tracheids to transport water. Lignin polymers are composed of alcohol monomers: p-coumaryl alcohol, coniferyl alcohol, and sinapyl alcohol, which produce \u00E2\u0080\u0095H\u00E2\u0080\u0096, \u00E2\u0080\u0095G\u00E2\u0080\u0096, and \u00E2\u0080\u0095S\u00E2\u0080\u0096 lignins, respectively (Boerjan et al., 2003). Analysis of mutants with defects in lignin biosynthesis has played an important role in elucidating some of the key steps in the lignin synthesis pathway (for a review see Boerjan et al., 2003). Similar to plants with defects in polysaccharide biosynthesis, lignin-deficient mutants often exhibit collapsed xylem phenotypes, highlighting the importance of lignin in cell wall integrity. For example, the irx4 mutant of Arabidopsis (a cinnamoyl-CoA reductase mutant), which contains up to 50% less lignin than wild-type plants, exhibits stunted growth, and collapsed xylem (Jones et al., 2001). The ref8 mutant of Arabidopsis (a p- coumarate 3-hydroxylase, C3\u00E2\u0080\u0099H, mutant) also exhibits reduced lignin content, and consequently collapsed xylem, but additionally it was found that the lignin produced in the mutant varied in composition compared to wild-type plants (Franke et al., 2002) illustrating how changes to lignin are very dependent on the step within the biosynthetic pathway that is altered. Cell wall defects as a result of reduced lignin are not limited to Arabidopsis. For example, lignin reductions in hybrid poplar, the result of RNAi- suppression of p-coumaryl-3\u00EF\u0082\u00A2-hydroxylase caused a collapsed xylem phenotype in tree stems and an increase in susceptibility to vessel cavitation (Coleman et al., 2008). Some lignin-deficient mutants also exhibit a concomitant increase in cellulose (Coleman et al., 2008a; Hu et al., 1999), suggesting potential compensatory feedback between the regulation of lignin and cellulose biosynthesis. 4 Cellulose is the most abundant polymer in secondary cell walls, as well as a large component of primary cell walls next to pectins (Mellerowicz et al., 2001). Cellulose is a homopolymer composed of 1, 4-\u00EF\u0081\u00A2-linked glucose molecules. Cellulose is polymerized from the UDP (uridine diphosphate)-glucose precursor, that can originate from the phosphorylation of free glucose, or by the action of sucrose synthase to produce UDP-glucose and fructose from UDP and sucrose. The long chains of cellulose can lie parallel to one another and form hydrogen bonds between chains, forming strong and often crystalline microfibrils (Delmer and Amor, 1995; Saxena and Brown, 2005). The rigidity and hydrophobicity of crystalline cellulose provides structural support to cell walls and aids in the ability of xylem vessels to transport water. Two crystalline forms of cellulose exist in nature, designated as allomorphs I and II. Cellulose chains are aligned in parallel or anti-parallel in cellulose I and II, respectively. Allomorph II has additional hydrogen bonding, resulting in a more thermostable form of cellulose (Brown, 2004). Cellulose II is found in some algae and bacteria, but allomorph I is the most common form and found in vascular plants. The parallel alignment allows for the microfibril structure to form (Brown 2004). Cellulose I has two additional sub-forms: I\u00EF\u0081\u00A1 and 1\u00EF\u0081\u00A2. I\u00EF\u0081\u00A1\u00EF\u0080\u00A0and I\u00EF\u0081\u00A2 cellulose have different states of crystal packing, such that the crystalline unit cell of I\u00EF\u0081\u00A1 is triclinic, whereas I\u00EF\u0081\u00A2 is monoclinic. Triclinic refers to a unit cell with three axes (in the x, y, and z plane of a three dimensional unit cell) that are of different lengths, whereas monoclinic unit cells have at least two axes that are the same length but differ from the third axis. The differences in crystal packing result in varied hydrogen bonding between chains in the I\u00EF\u0081\u00A1 and I\u00EF\u0081\u00A2 forms (Delmer, 1999). I\u00EF\u0081\u00A2\u00EF\u0080\u00A0is the most common form of cellulose found in plants, whereas I\u00EF\u0081\u00A1 can be found in bacteria and algae (Delmer, 1999; Saxena and Brown, 2005). Groups of chains come together, perhaps spontaneously (Guerriero et al., 2010), to form microfibrils. The number of individual chains in a microfibril varies between plants and cell wall types. Typical microfibrils that have been measured in a variety of plant types have a diameter between 2 nm and 4 nm (Donaldson, 2007; Frey-Wyssling, 1968). Aggregates of cellulose have been known to form large sheets and ribbons in bacteria and algae (Brown et al., 1976; Herth, 1983). Additionally, microfibrils can 5 aggregate even further to form larger macrofibrils, between 50 and 250 nm in diameter in primary cell walls of maize parenchyma (Ding and Himmel, 2006), and between 14 nm and 27 nm in secondary cell walls of poplar and pine (Donaldson, 2007). The number of glucose units in a cellulose chain is described by the degree of polymerization (DP). Cellulose chain DP may affect the strength of a microfibril, and therefore how a cell wall grows and expands (Wasteneys and Fujita, 2006), however this has not yet been determined experimentally. It is feasible to speculate that the strength of a microfibril would be affected if the constituting glucose chains were shorter (lower DP): such a microfibril would likely be more susceptible to breakage, bending, or separation during expansion (Wasteneys and Fujita, 2006). Expanding to the cell wall, lower inherent DP could negatively affect cell wall rigidity, or increase cell wall flexibility, and therefore change the way a cell expands or withstands physical stresses. The range of DP of cellulose chains is quite large, from near 8000 in some primary walls to greater than 15000 in secondary cell walls (Brown, 2004). The factors, physical, chemical or enzymatic that dictate DP are not completely clear. Wasteneys and Fujita (2006) postulated that the life span of cellulose synthesizing complexes in the membrane, which could be influenced by the organization of microtubules, could affect DP. The differences in DP between cell wall types within a single plant, or perhaps even within a single cell, suggest that there are indeed specific mechanisms that affect and perhaps regulate cellulose DP. The primary cell wall is composed of many layers of cellulose microfibrils (MF), of which the organization can be irregular, but generally transverse with respect to the axis of elongation. The highly cellulosic secondary cell wall is made up of three layers (S1 to S3), which are deposited successively inside of the primary cell wall. The longitudinal angle of cellulose MFs in the cell wall, with respect to the axis of elongation, is described as microfibril angle (MFA). Large MFAs provide flexibility in the cell wall required for growth and bending early in development, whereas small angles confer the strength and rigidity needed to withstand turgor pressure (Barnett and Bonham, 2004). The S2 layer is the thickest of the secondary cell wall layers and has a small MFA compared to the thinner S1 and S3 layers, which generally have a higher MFA (Barnett and Bonham, 2004). 6 1.3 Cellulose synthase genes 1.3.1 Bacteria and non-vascular plants An operon containing four genes linked to cellulose biosynthesis was the first genetic identification of what came to be known as the cellulose synthases. The operon, identified in Acetobacter xylinium, was determined to be essential for cellulose biosynthesis (Wong et al., 1990), and further mutant studies revealed changes in cellulose crystallinity upon mutation in the fourth gene, acsD (Saxena et al., 1994). More recently, Roberts and Bushoven (2007) genetically characterized the CesA and CesA-like (CSL) gene family of the moss, Physcomitrella patens. The full-length CesA genes identified contain typical motifs common amongst all other known CesAs: a zinc-binding domain, a plant-conserved region, a class specific region, and eight transmembrane domains. They also show that many of the introns of the moss CesAs are conserved with Arabidopsis CesAs. Phylogenetic analysis places the P. Patens CesAs in separate clades from those of the higher plants. It is interesting that there are many CesAs in this moss, even though it does not form vasculature with thick secondary cell walls. When comparing the P. patens CesAs to those of Arabidopsis, Roberts and Bushoven (2007) highlighted the fact that there were no P. patens orthologs of the Arabidopsis primary and secondary wall CesAs, suggesting that the specialized functions of AtCesAs in primary versus secondary cell wall biosynthesis is not present in P. patens, at least not at the genetic level. For this reason, the current model of CesA specialization of function (primary versus secondary cell wall) may have evolved with vascular plants. 1.3.2 Vascular plants Evolutionarily, CesA gene families appear to have emerged prior to the divergence of gymnosperms from angiosperms. The cotton (Gossypium hirsutum) cellulose synthase 1 (GhCesA1) and GhCesA2 were the first CesA genes to be identified in higher plants (Pear et al, 1996). Compared to the CesAs identified in bacteria, higher plant CesAs contain additional sequences termed the plant-conserved region, a hypervariable region, and a larger N-terminal sequence (Doblin et al., 2002; 7 Pear et al., 1996; Delmer et al., 1999). These additional motifs could be responsible for putative diversified functions of plant CesAs (Delmer et al., 1999). 1.3.2.1 Arabidopsis thaliana The CesA gene family of Arabidopsis is by far the most extensively studied, greatly advanced by the sequencing of the Arabidopsis genome in 2000. From genomic data available, 10 individual CesA genes and 31 cellulose-synthase-like genes were identified (Richmond and Somerville, 2000). The function of many of the AtCesAs has been under investigation. Studies often utilize mutant phenotypes and large scale gene expression analysis to further characterize the putative functions of each of the 10 CesAs. Tissue specific expression of the known CesA genes resulted in the classification of CesAs as either primary or secondary cell wall-specific (Hamann et al, 2004). In Arabidopsis, AtCesA4, A7, and A8 are often described as displaying reduced expression in leaves, roots, flowers, and seedlings but elevated expression in stem tissue (Hamann et al, 2004). Analysis of AtCesA4, A7, and A8 mutants revealed that these three CesA subunits are required for proper formation of the cellulose network in secondary cell walls. Screening of mutagenized plants for alterations in xylem structure identified the first two secondary cell wall-cellulose deficient CesA-mutants: irregular xylem 1 (irx1) and irx3 (Turner and Somerville, 1997), later identified as mutations in the AtCesA8 and AtCesA7 genes, respectively (Taylor et al, 2000). A decrease in cellulose contributed to the collapsed xylem phenotype. It has been suggested that some CesA mutants might not produce cellulose at all in the secondary cell wall (Taylor, 2008); however, there is not yet any evidence to prove this, and thus it remains only speculative. Instead, it is likely that mutants produce cellulose in the secondary cell wall, but at dramatically reduced levels and quality. Of the 10 CesA genes identified in Arabidopsis, only three have thus far been conclusively linked to secondary cell wall formation, whereas the remaining seven appear to be primary cell wall-specific. AtCesA1 was first identified as a primary cell wall-specific CesA in the rsw1-1 mutant, which showed defects in cell elongation and reduced levels of cellulose synthesized at a restrictive temperature (Arioli et al, 1998). 8 In addition to CesA1, CesA3 (Sheible et al, 2001) and CesA6 (Fagard et al, 2000) have also been implicated in cellulose production in the primary cell wall. Furthermore, is has been shown that expression of CesA6 was not sufficient to rescue the rsw1-1 mutant phenotype, suggesting non-redundant functions between these two CesA proteins (Robert et al, 2004). Examination of double- and triple-mutant phenotypes revealed partial redundancy between AtCesA2, A5, A9 and A6 depending on stage of growth (Desprez et al., 2007; Persson et al., 2007b). Phylogenetic analysis also suggests that their functions could be slightly redundant. Interestingly, Stork et al. (2010) recently suggested that AtCesA9 could have a unique role in the formation of secondary cell walls in seed coats. Also, it has been postulated that AtCesA7 might have some role in primary cell wall formation (Bosca et al., 2006; Zhong et al., 2007). These latter two observations advocate some overlap in function of a CesA in both primary and secondary cell walls, and these results also imply that CesA function can vary depending on cell type, or developmental stage. 1.3.2.2 Monocots As in Arabidopsis, there are ten identified CesAs in rice, Oryza sativa. An investigation of the brittle culm mutants lead to the characterization of three rice CesAs proposed to be involved in secondary cell wall cellulose biosynthesis (Tanaka et al., 2003): OsCesA4, OsCesA7, and OsCesA9. Phylogenetic analysis confirmed the similarity to Arabidopsis secondary cell wall-specific CesAs; however, the cell-wall deficient phenotype differed slightly from that of the Arabidopsis mutants. The rice brittle culm mutants exhibited thinner walls in fibre cells, but no collapsed xylem. This has been attributed to inherent differences in cell wall composition between the monocot rice and dicot Arabidopsis (Tanaka et al., 2003). Another OsCesA4 mutant, bc11, exhibited both primary (increased callose, pectin) and secondary cell wall alterations suggesting a link between biosynthesis of the two cell walls (Zhang et al., 2009), as was also suggested in the case of the Arabidopsis AtCesA7mur10 mutant (Bosca et al., 2006). At least eight CesA genes have been shown to exist in barley, Hordeum vulgare (Burton et al., 2004). Expression profiling of the HvCesA gene family suggested two co- 9 expressed groups: HvCesA1, A2, and A6 (group 1), and HvCesA4, A7, and A8 (group 2). These groups are most similar to the Arabidopsis primary- and secondary cell wall- specific CesAs, respectively. Two HvCesA4 mutants have reduced cellulose crystallinity, but also exhibited a slight up-regulation of HvCesA7 and HvCesA8 in the mutant internodes (Burton et al., 2010), further suggesting non-redundant functions between members of a putative CesA group, as the A7 and A8 up-regulation could not rescue the mutant phenotype. Twelve CesA gene family members have been identified in Zea mays (Appenzeller et al., 2004). Gene expression analysis suggests that ZmCesA10, 11, 12 are co-ordinately expressed in tissues undergoing secondary cell wall formation, similar to the CesA4, A7, and A8 group in Arabidopsis (Appenzeller et al., 2004). The available expression data, however, shows a great deal of overlap between putative CesA-group members (e.g.: primary versus secondary cell wall-specific) in both the elongating and maturing regions. The overlap of expression, and likely protein activity, is surely not limited to corn. Further characterization of CesAs in monocot plants may reveal functional orthology to the known CesAs in Arabidopsis, for example, but could also reveal varied cell wall-specificity, perhaps as a result of the divergence between monocot and dicot species. 1.3.2.3 Trees 1.3.2.3.1 Poplar An evaluation of the Populus trichocarpa genome suggests there are 17 CesA genes (Kumar et al., 2009). The CesA nomenclature proposed by Kumar et al. (2009) will be used throughout this thesis when referring to poplar CesAs, whereby the PtiCesA genes are numbered based on sequence homology to the Arabidopsis CesAs, such that PtiCesA8-A and A8-B are homologs of AtCesA8. The added complexity of a tree genome due to genome duplication is highlighted by putative multiple copies of individual PtiCesAs. For example, there are currently six copies of PtiCesA6 (Kumar et al., 1999), and multiple copies of all other PtiCesAs except for PtiCesA4. Interestingly, some duplicates of a CesA are both transcribed, such asPtiCesA8-A and -B (Djerbi et 10 al., 2005). Separation of some of the CesA duplicates will likely occur with thorough functional characterization of the gene products, or classification of some as non- functional. In poplar, most of the information regarding the role of CesAs is limited to expression profiles. For example, analysis of transcript abundance in both normal and tension wood of P. trichocarpa display an up-regulation of PtiCesA4, PtiCesA6, and PtiA8-A and A8\u00E2\u0080\u0093B in developing xylem and tension wood (Djerbi et al., 2004). Further, Suzuki et al. (2006) suggested that PtiCesA8-B is the most highly expressed in young developing xylem. However, the other highly expressed CesAs were PtiCesA3-C and PtiCesA1-A, which are orthologous to AtCesAs normally ascribed as primary cell-wall specific (Suzuki et al., 2006). In aspen, P. tremuloides, PtdCesA7-B and PtdCesA8-A were found to be are highly up-regulated in developing xylem (Samuga and Joshi, 2002; Joshi, 2003). Expression of secondary cell wall CesAs is altered in tension wood, however, the extent of change observed varies between studies. In hybrid aspen (Populus tremula x tremuloides), PtxtCesA8-A is the only gene confirmed to have increased expression using both PCR and microarray analyses (Djerbi et al, 2004 and Anderson-Gunneras et al., 2006, respectively). The tension wood gelatinous cell wall layer (G-layer) is highly cellulosic, with larger macrofibrils, more crystalline cellulose, and a very small MFA (Mellerowicz et al., 2001; Mellerowicz and Sundberg, 2008). The differences in cellulose ultrastructure between the S2 and G-layers suggest there are mechanisms in place that control these properties. Whether or not individual CesAs contribute to these properties is not known. Also, the degree to which various CesAs are tension wood-specific is not completely clear, nor is it apparent if particular CesAs have a distinct role in altering the cell wall properties to deposit a G-layer. Perhaps small changes in gene expression represent substantial alterations to the ratio of CesA subunits in a given cell, which could contribute to altered cellulose of tension wood. From just a few reports it is clear that a variety of CesAs are putatively involved in cellulose biosynthesis in xylem tissue of poplars. Further characterization of CesA specificity and functionality will be required to clearly elucidate any specificity among poplar CesAs. 11 1.3.2.3.2 Eucalyptus Eucalyptus, another economically important tree, has also been investigated to identify and classify the CesA genes (Lu et al., 2008; Ranik and Myburg, 2006). Six full- length EgCesA genes have been identified in Eucalyptus grandis (Ranik and Myburg, 2006), and were shown to be expressed at varying levels depending on tissue type. Gene expression profiling resulted in clustering of the identified CesAs into two groups: one that was most highly expressed in developing xylem (EgCesA1, EgCesA2, and EgCesA3) and the other group more highly expressed in primary-wall forming tissues of Eucalyptus (EgCesA4, EgCesA5, EgCesA6). It was further shown that the EgCesAs were less similar to each other than they were to the AtCesAs (at the nucleotide level) suggesting the six EgCesAs are distinct from one another. Additionally, GUS-EgCesA- promoter fusions expressed in Arabidopsis showed that EgCesA1 and A3 expression is limited to secondary cell wall-forming cells in the stem (Creux et al., 2008). Creux et al. (2008) also identified many cis-elements within the EgCesA promoters that are conserved with AtCesA promoters. In another independent study, (Lu et al., 2008), the results of Ranik and Myburg were supported by expression profiling of three putative secondary cell wall-specific EgCesA genes, homologous to AtCesA4, AtCesA8, or AtCesA3. It was found that only the EgCesAs homologous to AtCesA4 and AtCesA8 were up-regulated during xylem maturation. As with the characterization of CesAs in poplar, a great deal of functional characterization is required for the EgCesA proteins to identify if wall-specific functions exist. 1.3.2.3.3 Pine There are fewer reports of coniferous CesAs, despite the economic importance of many coniferous species worldwide. A query of Arabidopsis CesA gene sequences against publically available sequence data, returned several hits to pine and spruce sequences. However, most of the matches proved to be mRNA fragments and putative CesA sequences. To date only a handful of full-length CesAs have been identified in conifers; and those characterized are from Pinus taeda (Nairn and Haselkorn, 2005; Nairn et al., 2008) and Pinus radiata (Krauskopf et al., 2005). As in the other tree species, there has been an emphasis on isolating secondary cell wall-specific CesAs. 12 Gene expression analysis has revealed at least three CesAs in Pinus taeda that are up- regulated in wood-forming tissues (Nairn and Haselkorn, 2005: Nairn et al., 2008), and one gene (PrCesA10) in Pinus radiata that is homologous to AtCesA7, which is expressed in tracheids undergoing secondary cell wall formation (Krauskopf et al., 2005). 1.4 CesA protein structure CesA proteins have a conserved structure containing a zinc-binding domain, eight transmembrane domains (two located at the amino terminus and six at the carboxyl terminus), a plant-conserved region, a class-specific region (CSR), and two highly conserved domains, A and B (Richmond, 2000; Krauskopf et al., 2005). Within domains A and B there are four amino acid motifs that are characteristic of processive glycosyltransferases: three aspartic acid residues (D) and a QXXRW motif (Joshi and Mansfield, 2009). All components of domains A and B are necessary for the binding of UDP-glucose (Pear et al, 1996), suggesting that this region is part of the CesA catalytic domain (Pear et al., 1996; Doblin et al., 2002). In vitro experiments also indicated that the zinc-binding domains of cotton CesA1 and A2 are required for proper interactions (Kurek et al., 2002). Predicted CesA amino acid sequences from genes within a given plant are less similar than comparative CesA sequences from other plants (orthologs) (Ranik and Myburg, 2006; Joshi et al, 2004; Nairn and Haselkorn, 2004; Samuga and Joshi, 2002). Although the conserved regions are highly similar (90%) within and among plants, CSRs tend to have extremely low similarity within a plant (11 - 40%) and slightly higher similarity (70%) among orthologs (Ranik and Myburg, 2006). The CSR may dictate differences in function among CesAs within a plant, and has been used to aid in classification. 1.5 Cellulose synthase complex Hexameric complexes associated with cellulose production, described as rosettes or terminal complexes, were first visualized in bacteria (Brown et al, 1976) and have been described in some plants as scattered throughout the plasma membrane 13 (Mueller and Brown, 1980; Haigler and Brown, 1986; Kimura et al., 1999a, 1999b, Arioli et al., 1998). Some of the first images of rosettes were in the plasma membrane of algae (Brown and Montezinos, 1976). Rosettes and terminal complexes have been observed in a variety of arrangements within plasma membranes of diverse species (Tsekos, 1999). For example, in Oocystis, the terminal complexes have been shown to be arranged in linear arrays, and these organisms tend to produce large microfibrils (Tsekos 1999) and sheets of cellulose (Brown and Montezinos, 1976). Comparatively, large arrays consisting of dozens of six-lobed rosettes have been visualized in the plasma membrane of the algae Micraterias denticulate (Giddings et al., 1980) and Spirogyra (Herth, 1983), and according to Gidding et al. (1999) the size of cellulose microfibril bundles was proportional to the size of rosette arrays observed in M. denticulate. Furthermore, the arrangement of the rosettes in the plasma membrane can differ between primary and secondary cell walls in the same organism (Tsekos, 1999). It has been postulated (Tsekos, 1999) that differences in terminal complex arrangement may be involved in the production of I\u00EF\u0081\u00A1 cellulose versus I\u00EF\u0081\u00A2 cellulose. The diversity of rosette arrangements in various species and during different stages of cell wall biosynthesis might explain the inherent variability in cellulose microfibril characteristics, such as size and shape (Tsekos, 1999), and perhaps crystallinity, thus affecting cell wall properties. The active unit responsible for the production of cellulose is called the cellulose synthesizing complex (CSC). The CSC contains cellulose synthases (Kimura et al., 1999a), and approximately six CSCs associate to form a hexameric complex described as a rosette (Saxena and Brown, 2005). It is postulated that at least three unique CesAs are required for secondary cell wall CSCs to form in Arabidopsis: AtCesA4, A7, and A8 (Taylor et al., 2003). 1.6 The link between CesAs, rosettes, and the CSC The first link between CesA genes and the rosette structure was provided by analysis of the AtCesA1 mutant, rsw1-1 (Arioli et al., 1998). When grown at the restrictive temperature, the mutants had reduced cellulose content and subsequently cell wall defects that affected growth such as stunted growth in seedlings and oddly- 14 shaped epidermal cells. In addition, the plants were shown to have fewer rosettes in the plasma membrane. In particular, after a long exposure to the restrictive temperature there were no rosettes visible, just a dense scattering of single particles. This implies that in the absence of a functional CesA, the CSCs and rosette structures are altered. Mutant forms of AtCesA4 (irx5) have been used to examine the interactions among CesA proteins and confirm that three CesA subunits are required for CesA-CesA interactions, detected in vitro using immunoprecipitation (Taylor et al. 2003). Null mutations of AtCesA4 caused a reduction in AtCesA8 and A7 protein levels, and eliminated in vitro detection of interactions between AtCesA8 and AtCesA7. However, interactions were apparent in a non-null mutant of AtCesA8. These results suggest that the three CesAs interact to form the CSC, and all three are required for interactions to occur. In the CesA mutants, reduced cellulose production is believed to be the result of improper assembly and function of the CSC, which is supported by the results of Arioli et al. (1998). However, it is interesting to note that cellulose is still produced in the absence of one CesA. It has been postulated that in the absence of a functioning CesA, aberrant CSCs may form and produce \u00EF\u0081\u00A2-1,4-glucan chains that do not crystallize properly (Arioli et al, 1998). Labelling of AtCesA7 with GFP allowed Gardiner et al. (2003) to visualize the location of CesA-containing complexes or CesA-containing organelles within developing roots. They showed that the GFP-CesA signal was localized along the spiral cell wall thickenings of root xylem vessels, which was reduced, but not completely eliminated, in a null-AtCesA4 or null-AtCesA7 mutant background. This suggests that the remaining CesAs in a mutant background are likely present, and functioning, during cell wall biosynthesis. 1.7 Composition of the CSC Elucidating the components of the CSC and how they affect cellulose production and properties remains an intriguing area under investigation. Some in vitro, and a few in vivo studies, have shown that the CesA proteins form homodimers and heterodimers (Taylor et al, 2003; Atanassov et al, 2009; Timmers et al., 2009; Desprez et al., 2007), and putative CSCs have been isolated and characterized (Atanassov et al., 2009; Song 15 et al., 2010). In Arabidopsis, the use of immunoprecipitation methods lead to the conclusion that all secondary cell wall-specific CesAs interact (Taylor et al., 2000: Taylor et al., 2003) within a large complex (Atanassov et al., 2009), but yeast-two hybrid assays by Timmers et al. (2009) suggested that AtCesA7 does not form homodimers. Furthermore, in vivo experiments (using bimolecular fluorescence complementation, BiFC) suggested all CesA-CesA interaction combinations are possible except CesA7 homodimers (Timmers et al., 2009). Very recently, Gu et al. (2010) reported that a novel protein, cellulose synthase interactive-1 (CSI1) interacts with AtCesA1, A3, and A6 (interactions detected using yeast-two hybrid assays). The function of CSI1 is not yet determined, although mutant CSI1 plants have a reduced cellulose phenotype. In the developing xylem of poplar (Populus deltoides x trichocarpa) Song et al. (2010) reported on the use of immunoprecipitation to isolate protein complexes containing CesAs, and showed that two types of putative CSCs might exist during secondary cell wall formation: a type I CSC that contains PdxtCesA4, A7(A/B), A8(A/B), and a type II CSC that contains PdxtCesA1 (A/B), A3(C/D), and A6 (E/F). Additionally, they reported that sucrose synthase (SUSY), korrigan (KOR), and the GPI-anchored protein COBRA interacted with the CesA complexes during immunoprecipitation. Contrary to this, the interaction of KOR with Arabidopsis CesAs has not been found using immunoprecipitation (Desprez et al., 2009) and yeast-two hybrid assays (Maloney, 2010). Additionally, using the same technique (immunoprecipitation of large complexes), Atanassov et al. (2009) reported only CesA-CesA interactions and no detection of non-CesA proteins. However, it must be noted that detection of CesA-interacting proteins may differ between plants and cell types. Additionally, the techniques and conditions used to isolate interacting proteins or complexes could affect the ability to detect interactions. As evidenced by the recent reports of CSI1 (Gu et al., 2010), there may be additional proteins interacting with CesAs that have not previously been reported. The particularly large, globular, cytoplasmic domain of rosettes observed by Bowling and Brown (2008) also lends support to the idea that additional proteins could be components of the CSC. The size of the putative CSC isolated by Song et al. (2010), however, was not reported, thus it is difficult to interpret whether the size of isolated CSCs supports the idea of a 16 large, diverse, multi-protein complex. Clearly, further investigation is required to fully elucidate the composition of CSCs. 1.8 Transport of CSCs The abundance and organization of CSCs, and rosettes, within the plasma membrane during cellulose biosynthesis is believed to greatly affect cellulose properties, leading to an important question: how do CSCs ultimately end up in the plasma membrane at sites of cell wall thickenings? Haigler and Brown (1986) visualized rosette structures within Golgi vesicles (and in the plasma membrane) of Zinnia cells undergoing secondary cell wall thickening and it has been suggested that complete CSCs are formed, but inactive, within the Golgi. Control of CSC localization from the Golgi has been attributed to microtubules, pre-existing cellulose networks, and a variety of proteins. The mechanisms of delivery, and consequent distribution of CSCs in the plasma membrane may affect the final abundance and properties of cellulose produced, and is thus an intriguing area of research. The use of fluorescence confocal microscopy has facilitated the visualization of active CSC movement in live cells, and the trafficking of CSCs via Golgi vesicles and other transport vesicles (Paredez et al., 2006; Wightman and Turner, 2008; Crowell et al., 2009; Gutierrez et al., 2009; Wightman and Turner, 2010). Changes in the pattern of Golgi-CSCs after treatment with an actin-disrupting drug have been shown to alter the distribution of CSCs in the plasma membrane, suggesting that CSC organization is dependent on the actin cytoskeleton, perhaps by affecting the delivery of Golgi-CesA complexes to particular sites or regions of the plasma membrane (Wightman and Turner, 2008; Gutierrez et al. 2009; Crowell et al., 2009). Crowell et al. (2009) also observed that CesA-Golgi complexes move and pause at various sites along MTs, and that pause events are often associated with presumable CSC delivery into the plasma membrane, and that the distribution of CSCs was dependent on a functional MT array. Wightman and Turner (2008) have also reported on CesA-Golgi compartments which the authors postulated also pause to deliver CSCs to sites of secondary cell wall thickening. Furthermore, reports of smaller CesA-containing particles have been observed below the plasma membrane plane in the cytosol, and are described as 17 MASCs (microtubule-associated cellulose synthase compartments, Crowell et al., 2009) and SmaCCs (small CesA compartments, Gutierrez et al., 2009). Both MASCs and SmaCCs have been suggested to be another, mostly non-Golgi vesicular compartment involved in regulating the distribution of CSCs and therefore cellulose biosynthesis during primary cell wall formation (Crowell et al., 2009; Gutierrez et al., 2009). MASC abundance and distribution were increased in response to drug-induced disruptions to cellulose biosynthesis and osmotic stress, further suggesting that they are involved in regulating cellulose biosynthesis by perhaps internalizing CSCs from the plasma membrane (Crowell et al., 2009). The same microtubule-associated localization and drug-induced accumulation was observed for SmaCCs (Guiterrez et al., 2009), and a comparison of images presented by Crowell et al. (2009) and Guiterrez et al. (2009) suggest that MASCs and SmaCCs are similar in size: between 300 and 500 nm in diameter. Together, these results strongly suggest that SmaCCs and MASCs could be the same population of CSC-containing cortical compartments (Whightman and Turner, 2010). Occasional association of SmaCCs (and possibly MASCs) with the trans-Golgi network has also been observed (Guiterrez et al., 2009; Whightman and Turner, 2010). Recent increases in the use of confocal fluorescence microscopy to observe CSCs in live cells has greatly enhanced our understanding of the dynamic nature of CSC delivery (and perhaps removal) to the plasma membrane, and that a variety of CSC- containing compartments exist in plant cells. However, it is clear that further characterization of the composition and function of MASCs/SmaCCs is required to completely understand their role in CSC regulation, as well as determine if they are the same population of CSC-compartments. 1.9 The cytoskeleton and cellulose microfibril orientation The organization of cellulose microfibrils is crucial for proper cell formation, growth, anisotropy, and ultimately plant development. The mechanism, by which CSCs are localized within the plasma membrane, and the regulation of cellulose microfibril (MF) orientation within the cell wall, are not yet known. In addition, whether or not the localization of CSCs within the plasma membrane solely dictates cellulose MF orientation or which, if any, post-deposition remodeling occurs is still under question. 18 The role of microtubules (MTs) in cellulose MF organization has been the topic of much debate (Baskin, 2001; Wasteneys, 2004; Paredez et al., 2008; Crowell et al., 2010). It has been many years since microtubules were observed to lay parallel to cell wall components, such as cellulose microfibrils (Ledbetter and Porter, 1963), and since then various models have been proposed to explain the role of MTs in cellulose MF organization, such as the alignment hypothesis and template-incorporation hypothesis (reviewed by Baskin, 2001) and the microfibril-length-regulation hypothesis (Wasteneys, 2004; Wasteneys and Fujita, 2006). In combination with previous reports, the more recent use of fluorescently-tagged CesAs (often in combination with fluorescently labeled microtubules) provides more information about the relationship between MT and MF organization. Colocalization of CSCs and MTs has been observed in xylem vessels (Gardiner et al, 2003; Wightman and Turner, 2008) and during primary cell wall formation (Paredez et al, 2006). Disruption of the MT array (using MT-depolymerizing drugs) causes CSC trajectories and localization to become disorganized (Gardiner et al, 2003; Paredez et al, 2006; Crowell et al., 2009), suggesting that proper localization of the CSC is dependent on MTs. However, Himmelspach et al. (2003) demonstrated that in the mor1-1 mutant grown under restrictive temperature conditions, causing microtubule disorganization, and after the application of DCB (2, 6-dichlorobenzonitrile), resulting in disorganization of the cellulose microfibril array, newly formed cellulose microfibrils aligned in an organized, parallel fashion, indicating that neither the MT network nor a pre-existing cellulose network was required for organized microfibrils to form. Additionally, Paredez et al. (2006) showed that after several hours of MT disruption by oryzalin treatment, CSC patterns and trajectories become highly organized, despite the lack of an organized MT array, further supporting the idea that MT organization is not required for CSC organization (Paredez et al., 2006) and hence cellulose MF organization (Himmelspach et al., 2003). These results suggested that the organization of cellulose MFs was not dependent on organized MTs or pre-existing MFs, and that other factors such as a self- assembly mechanism, CSC density, or the rate of cellulose production may regulate cellulose MF organization (Himmelspach et al., 2003). However, it is hard to argue that 19 MTs are not, at least indirectly, involved in the regulation of cellulose deposition or cellulose MF properties at particular times in cell wall development. As detailed by Fujita (2008) MT disorganization in the mor1-1 mutant affects cellulose crystallinity, CSC density, and CSC velocity in primary cell walls. This clearly indicates a role for MTs in regulating aspects of cellulose microfibril properties, potentially by affecting CSC distribution and movement, but perhaps not microfibril orientation. It has been proposed that MTs could regulate the length of cellulose MFs, (Wasteneys 2004; Wasteneys and Fujita, 2006), which could affect cell expansion dynamics. Additionally, the alteration of the patterning and movement of non-Golgi CesA compartments (MASCs and SmaCCs, described above) in response to changes in the MT array further illustrates that some aspect of CSC organization is affected by the MT network (Crowell et al., 2009; Gutierrez et al., 2009), which could ultimately determine properties of the cellulose microfibrils produced. It should also be considered that MT-CesA associations (whether direct or indirect) could differ in primary and secondary cell walls, or between developmental stages. The existence of secondary mechanisms involved in MT-aided CSC organization is supported by the phenotype of the Arabidopsis fra1 mutant (Zhong et al., 2002). A mutation of a kinesin-like protein resulted in slight disorganization of the cellulose microfibrils in the secondary cell walls of fibers, resulting in reduced fiber strength. No change in cellulose content or MT organization was observed, and therefore it was suggested that the kinesin-like protein may guide CSCs along MTs, or may regulate the positioning of other barrier proteins along MTs that would restrict the localization of CSCs and therefore cellulose MFs (Zhong et al., 2002). An increase in the expression of a kinesin-like protein in tension wood of Eucalyptus (Paux et al., 2005) also suggests a role for such a protein in cellulose biosynthesis in secondary cell walls. Microtubule associated proteins (MAP) may act as an intermediate between MTs and CSCs. For example, it has been shown that treatment of plants with DCB inhibits cellulose production causing abnormal cell wall formation (Taylor et al., 1992), and more recently a MAP20 protein found in developing xylem of hybrid poplar was shown to bind DCB (Rajangam et al., 2008). Additionally, mis-regulation of an Arabidopsis MAP (AtMAP70-5) caused alterations to microtubules and secondary cell wall thickening in 20 xylem cells (Pesquet et al., 2010). Wightman et al. (2009) observed that the movement of YFP-CSCs in secondary cell wall forming cells was slower following treatment with DCB. Taken together, it is feasible to speculate that there may be a population of proteins that facilitate linkages between MTs and CSCs, potentially affecting CSC movement and therefore cellulose deposition and potentially MF organization. Disruption of KORRIGAN or CesA6 (AtCesA6prc mutant) in Arabidopsis was found to increase the sensitivity of plants to a microtubule destabilizing drug, oryzalin, and the mutant plants had disorganized MT arrays (Paredez et al., 2008). Previously, it had been shown that disruption of the cellulose MF network using DCB (Himmelspach et al,.2003) or isoxaben (Fisher and Cyr, 1998) causes MT-disorganization. Combined, these results strongly suggest there is feedback in response to perturbations of the cellulose network that ultimately affects the MT array. 1.10 Goals and hypothesis Although a great deal of research has been conducted around the fact that a group of conserved genes are required for cellulose biosynthesis in the secondary cell wall of Arabidopsis (CesA4, A7, and A8), there is little to no information available to identify if each CesA has a unique role in cellulose biosynthesis, and if their precise roles confer specific cellulosic properties. The main goal of my research was to contribute to our understanding of the role of secondary cell wall-specific cellulose synthases during cellulose production. These questions were investigated in three separate studies, detailed in Chapters 2, 3, and 4. The scope and aim of each study is briefly described here. 1.10.1 Chapter 2: Characterization of Arabidopsis CesA secondary cell wall- specific mutants The goal of this research was to characterize the growth and cell wall properties of AtCesA secondary cell wall-specific cellulose synthase mutants. The emphasis was on understanding how the individual CesAs contribute to cellulose biosynthesis and cellulose properties in plant secondary cell walls. The hypothesis was that each AtCesA has a unique role in cellulose production, such that a mutation in any one 21 protein will result in altered cellulose properties. To examine the effect of each mutation, I measured growth rates and cell wall properties of plants grown under both long and short-day lighting conditions. Additionally, I examined the distribution of AtCesAs in planta using YFP-CesA fusions. 1.10.2 Chapter 3: Identification of secondary cell wall-specific cellulose synthase genes from Picea glauca and conservation of function among CesA orthologs The goal of this research was to identify secondary cell-wall specific CesA genes in a gymnosperm, spruce (Picea glauca), that are orthologous to the Arabidopsis secondary cell wall-specific CesAs. It was hypothesized that the isolated genes would share similar conserved domains and would exhibit varied transcript abundance such that the highest expression would correspond to tissues undergoing significant secondary cell wall formation. Additionally, I tested the hypothesis that true orthologous CesAs would be functionally conserved. This was tested by transforming Arabidopsis AtCesAirx mutants with candidate CesA genes from spruce and poplar, to determine if presence of the spruce or poplar CesA could recover the mutant phenotype, thus demonstrating conservation of function among CesA orthologs. 1.10.3 Chapter 4: Measuring CesA-CesA interactions in planta The goal of this work was to examine if specific and consistent interactions exist among Arabidopsis CesA proteins in vivo during secondary cell wall formation. The hypothesis was that CesA subunits form homodimers and heterodimers within the CSC, and that these interactions are specific and consistent during a defined developmental stage. I utilized two methods, bioluminescence energy transfer (BRET) and bimolecular fluorescence complementation (BiFC) to study CesA-CesA interactions in vivo. 22 CHAPTER 2 Characterization of Arabidopsis CesA secondary cell wall-specific mutants 2.1 Introduction Although a variety of cellulose synthase (CesA) mutants have been studied and characterized, the precise mechanism by which the cellulose synthase proteins and the cellulose synthesizing complex (CSC) affect cellulose properties such as abundance, crystallinity, microfibril angle, and degree of polymerization are still relatively unknown. It has been shown that at least three CesAs are required for proper secondary cell wall formation in Arabidopsis: AtCesA4, AtCesA7, and AtCesA8 (Taylor et al., 2000; Taylor et al., 2003; Gardiner et al., 2003; Taylor et al., 2004). Mutations in any one of these genes results in reduced cellulose synthesis and altered secondary cell wall structure (for review see Taylor, 2008). However, a comparison of all three mutants has not been provided in a single, unified study. Instead, several, independent reports provide a general summary of the effects of each mutant, all of which generally lack a complete evaluation of the effects of each mutation on cellulose properties. In an effort to better understand the role of AtCesA4, A7, and A8 in determining cellulose properties, a complete and thorough characterization of three Arabidopsis secondary cell wall CesA- mutants was performed: AtCesA4irx5-1, AtCesA7irx3-1, and AtCesA8irx1-1. Growth phenotype, secondary cell wall composition, and cellulose ultrastructure were determined for each mutant, and a critical review of the information collected provides some new insight into the complex process of cellulose biosynthesis. Cellulose is a linear, unbranched, homopolymer composed of \u00EF\u0081\u00A2-1,4-linked glucan units, polymerized from the UDP (uridine-diphosphate)-glucose precursor (Delmer and Amor, 1995). Cellulose chains are aligned parallel to one another and form hydrogen bonds between chains, and among fibrils, ultimately generating a polymer with both crystalline and amorphous regions (Delmer and Amor, 1995; Saxena and Brown, 2005). Bundles of glucan chains, putatively 16 to 36 glucan chains in vascular plants (Saxena and Brown, 2005), form cellulose microfibrils. Microfibrils with diameters of 3 to 5 nm were observed in maize cell walls, a size which supports the idea that microfibrils could contain a range of glucan chains (Ding and Hemmel, 2006). 23 The orientation, and thus microfibril angle (MFA) of cellulose microfibrils in a cell wall, or layers of a cell wall, is thought to greatly affect cell wall structure and function. Cellulose MFA is quite diverse between plant species (Lichtenegger et al., 1999), tissue, cell type, and developmental stage (Donaldson, 2007; Barnett and Bonham, 2004; Mansfield et al., 2009), and cell wall layer (Chan et al., 2010; Barnett and Bonham, 2004; Donaldson, 2007). Such diversity in MFA illustrates the importance of cellulose MFA in cell wall structure and function. Variations in MFA are thought to affect cell wall stiffness and extensibility (Reiterer et al.,1999). For example, in secondary cell walls, the cellulose microfibrils of the dominating S2 layer have a small MFA to provide the structural integrity required for water transport (Barnett and Bonham, 2004; Mellerowicz and Sundberg, 2008; Fang et al., 2004). Comparatively, the S1 and S3 layers of secondary cell walls, as well as microfibrils in primary cell walls, have a larger MFA (Barnett and Bonham, 2004; Seagull, 1992) and alterations to cellulose microfibril orientation in primary walls affects the degree and direction of cell expansion (Himmelspach et al., 2003; Wasteneys and Fujita, 2006; MacKinnon et al., 2006). The present study aims to determine the MFA of secondary cell wall cellulose in whole Arabidopsis stems, providing novel information regarding the ultrastructure of cellulose in a very important model plant. Cellulose ultrastructure has a prominent effect on cell wall structure and function. For example, the degree of polymerization and extent of crystallization dictate the strength of the individual microfibril. Consider, for example, the differences in cellulose ultrastructure between primary and secondary cell walls. The primary cell wall cellulose of some plants is less crystalline than secondary cell walled tissues (Harris and DeBolt, 2008; M. Fujita, personal communication). Harris and DeBolt (2008) also found that cellulose crystallinity varied greatly among diverse species, and the percentage of crystalline cellulose in stems decreased when plants were exposed to wind stress. These results infer a relationship between cellulose crystallinity and the function of the cell wall (for example decreased stiffness under wind conditions). Cellulose microfibrils of primary walls may be composed of glucan chains with a lower degree of polymerization (DP) compared to those of secondary walls, although experimental evidence comparing cellulose DP in various tissues and cell walls is lacking. 24 Mechanically, it is thought that microfibrils composed of short glucan chains (low DP) would be more susceptible to breakage or separation, therefore affecting cell wall strength and expansion (Wasteneys and Fujita, 2006). In synthetic and bacterial produced cellulose films, a decrease in DP is correlated with reduced cellulose strength (Henriksson and Berglund, 2007). Variation in the properties of cellulose microfibrils, such as MFA and crystallinity, between cell wall layers evokes many questions pertaining to how, what, and when these qualities are determined, and to what extent the CesA complexes affect these properties. In CesA mutants, reduced cellulose production is believed to be the result of improper assembly and function of the CSC, or impaired function of a CesA subunit. In fact, the loss of hexagonal rosettes at the plasma face of plants with a mutant form of a primary cell wall CesA, AtCesA1, has been observed (Arioli et al., 1998). However, cellulose is still produced in the absence of one CesA. It has been postulated that in the absence of a functioning CesA, aberrant CSCs may form and produce \u00EF\u0081\u00A2-1,4-glucan chains that do not crystallize properly (Arioli et al., 1998). A number of secondary cell wall-specific CesA mutants have been described in the literature, and Figure 2.1 summarizes an overview of the types of mutants reported. Table 2.1 provides details of the mutant phenotype characterization for some of the mutants for which cell wall composition has been reported. Of the information currently available, the most critical observations revealed thus far are that all three CesAs (AtCes4, A7, and A8) are required at the same time for proper cellulose biosynthesis, such that a mutation in any one CesA results in a dramatic reduction in cellulose content. Due to the diversity of published information available, only a limited comparison of the mutants can be made, mostly limited to broad changes in phenotype. Also, discrepancies in reports on changes to cellulose content further limit the comparisons that can be made about each mutant. For example, the same AtCesA7irx3-1 mutant has been reported to have both a 70% decrease (Taylor et al., 1999) and a 92% decrease (Taylor et al., 2003) in cellulose compared to wild-type plants. Such differences could be due to many factors, including the age of plants harvested for analysis, the region of stem analyzed, and the growth conditions. In an effort to gain more insight into the role of each CesA in cellulose biosynthesis, it is 25 prudent to thoroughly investigate the CesA4, A7, and A8 mutants simultaneously. Additionally, cellulose properties such as MFA, crystallinity, and DP have not been measured in any of the mutants. Our study, therefore, provides the means for direct comparison of each mutant and novel information about cellulose structure in these mutants, and such a thorough comparison may shed new light on the individual role of each protein in cellulose biosynthesis. 26 Figure 2.1. A schematic diagram illustrating the location of various mutations in Arabidopsis AtCesA4 (A4), AtCesA7 (A7), and AtCesA8 (A8) genes (modified from Fujita, 2008; Zhong et al., 2003). Only mutants for which cell wall properties have been published are shown. Mutants have been given new labels (e.g.: A7fra5 instead of fra5), as shown in Table 2.1. Further details about the amino acid substitutions can be found in Table 2.1 HVR \u00E2\u0080\u0093 hypervariable region Zn \u00E2\u0080\u0093 zinc-binding domain PM \u00E2\u0080\u0093 plasma membrane TMD \u00E2\u0080\u0093 transmembrane domain CSR \u00E2\u0080\u0093 class-specific region D1, D2, D3, QXXRW \u00E2\u0080\u0093 conserved residues within the catalytic domain. 27 Table 2.1. A summary of AtCesAirx phenotypes of AtCesA4, A7, and A8 mutants reported in the literature. The summary is restricted to reports of cell wall composition. Values are a percentage of the levels reported for wild-type plants from within the same literature source. The new labels presented are a nomenclature suggestion (Taylor, 2008). Ds - dissociation insertion mutation; arrow \u00E2\u0080\u0093 substitution event; OE \u00E2\u0080\u0093 35S promoter-directed over-expression; *conserved residue in catalytic site. 28 2.2 Materials and methods 2.2.1 AtCesA mutant nomenclature In order to integrate the gene identity with previously assigned irx labels a modified nomenclature will be used throughout this chapter and thesis, as proposed by Taylor (2008). The revised nomenclature is reviewed in Figure 2.1 and Table 2.1. The following nomenclature will be used to describe the mutants that were used in the subsequent studies, irx1-1: AtCesA8irx1-1 (A8irx1-1); irx3-1: AtCesA7irx3-1 (A7i rx3-1); and irx5-1: AtCesA4irx5-1 (A4irx5-1). 2.2.2 Plant growth Wild-type Arabidopsis (Landsberg ecotype, WT), AtCesA8irx1-1 (Turner and Somerville, 1997), AtCesA7irx3-1 (Turner and Somerville, 1997), and AtCesA4irx5-1 (Taylor et al., 2003) were grown and used for these studies. Seeds were first surface sterilized by washing for 2 minutes in 70% (v/v) ethanol followed by an 8 minute wash in a 30% (v/v) bleach and 0.2% (v/v) Triton X-100 solution. Washed seeds were rinsed ten times with sterilized water, after which, seeds were stratified by storing in water, in the dark, at 4oC for at least 2 days. Stratified seeds were germinated on half- concentration MS medium (Mirashige and Skoog, 1962) with no sucrose under continuous light. Seedlings were transferred to soil approximately 7 days post germination, and grown in a growth chamber at 21oC under either a long-day light cycle (16-h light/8-h dark) or a short-day light cycle (8-h light/16-h dark). Plants under both long and short-day conditions were monitored and measured regularly for growth analysis. For growth, rosette diameter and stem height were measured weekly after the first rosette leaf exceeded 0.5 cm. The appearance of flower buds, open flowers, siliques, and mature siliques was also recorded. When plants were fully mature (yellowed siliques), they were harvested for cell wall analysis. For live-cell imaging, inflorescence stems of mature (15-day to 30-day old) plants were used. 2.2.3 Xylem morphology Toluidine blue staining was employed to assess xylem morphology. Stem bases (approximately 1cm from the soil level) from plants (21-30 days old) were hand- 29 sectioned using a double-edged razor blade. Sections were stained for 5 minutes in a 0.25% (w/v) Toluidine blue solution, and rinsed for 5 minutes in water. Hand-sections were viewed through a Leica Light Microscope, and pictures taken with a Q-imaging camera. 2.2.4 Structural carbohydrate and starch analysis Two and three-month old Arabidopsis stems were harvested from fully mature plants that had grown under long-day or short-day conditions. Dried stems were ground in a Wiley mill to pass a 0.4-mm mesh screen (40 mesh). Extractives were removed by subjecting the ground tissue to a hot acetone extraction for 16 hours using a Soxhlet apparatus. Lignin and carbohydrate content was determined using a modified Klason technique. Approximately 200 mg of dried, acetone-extracted tissue was macerated at 10 minute intervals in 3 mL of 72% sulphuric acid for a total of 2 hours. Samples were then diluted to 3% acid by the addition of 112 mL deionized water and autoclaved for 1 hour at 121oC. The acid-insoluble lignin fraction was isolated by filtration of the filtrate through a pre-weighed medium coarseness sintered-glass crucible, the retentate thoroughly rinsed with distilled water, dried at 105oC for 24 hours, and weighed to determine the amount of acid insoluble lignin. The acid soluble lignin was determined by measuring the absorbance at 205 nm according to TAPPI Useful Method UM-250. The filtrate was then used to determine the amount of structural monosaccharides using HPLC analysis. Specifically, samples were analysed using a DX-600 anion-exchange HPLC (Dionex) equipped with an ion exchange PA1 column (Dionex). Concentrations of glucose, xylose, mannose, galactose, arabinose, rhamnose, and fucose were determined using regression equations from calibration curves that were derived from external standards. Starch content of mature stems was also determined. To extract starch, approximately 20 to 50 mg of the acetone-extracted tissue described above was hydrolyzed in 5 mL of 4% sulphuric acid at 121oC for 3.5 minutes. Glucose content was determined by HPLC analysis and represents starch content. 30 2.2.5 Microfibril angle and crystallinity by X-ray diffraction Microfibril angle (MFA) was measured using an X-ray diffraction technique (Megraw et al., 1998). Measurements were taken from at least five individual mature stems (as described in 2.2.3) for each genotype. The first 3 cm of the stem base were used for X-ray diffraction. The 002 diffraction spectra from at least five individual mature (see section 2.2.3) stems for each genotype were screened for T value distribution and symmetry on a Bruker D8 discover X-ray diffraction unit. Wide-angle diffraction was used in the transmission mode, and measurements were made with CuK\u00EF\u0081\u00A11 radiation (\u00EF\u0081\u00AC\u00EF\u0080\u00A0represents 1.54 \u00C7\u00BA). The X-ray source was fit with a 0.5 mm collimator and the scattered photon was collected by a GADDS detector. The X-ray source and the detector were set to a theta angle of 0o. The diffraction data were integrated using GADDS software and further analyzed to estimate MFA values. Cellulose crystallinity was determined from the same stems used for MFA analysis. Crystallinity estimates were generated by analysis of X-ray diffraction. The X- ray parameters used were the same as those described for MFA determination with the exception of the source theta being set at 20o. Diffraction intensities were counted at 0.1o increments between 4 and 40o in the 2 theta angle range. The diffraction data were integrated using GADDS software and the output data further analyzed using a crystallinity calculation program based on the Vonk method (Vonk, 1973) to estimate the degree of crystallinity. 2.2.6 Preparation of \u00EF\u0081\u00A1-cellulose Ground, acetone-extracted stem tissue samples (as described in 2.2.3.2) were employed for quantitative assessment of \u00EF\u0081\u00A1-cellulose content. First, to remove lignin, approximately 50-200 mg of ground tissue was mixed with 3.5 mL of buffer solution (per 1L: 60 mL glacial acetic acid, 1.3 g NaOH) and 1.5 mL of 20% (w/v) sodium chlorite solution in a capped and parafilm-sealed glass tube and incubated with gentle shaking for 16 hours in a 50oC water bath. The overnight reaction was then quenched by placing the tubes in an ice bath for 8 hours. The reaction solution was then carefully removed, leaving the wood meal in the tube. A second identical reaction was set-up and left to react overnight. The wood meal was then washed twice with 50 mL of 1% (v/v) glacial acetic acid followed by washing with 10 mL of acetone and filtered through 31 a pre-weighed, coarse, sintered glass crucible. Samples were dried overnight in a 50 oC oven and weighed to determine the mass of hollocellulose. Approximately 25-100 mg of hollocellulose was used to isolate \u00EF\u0081\u00A1-cellulose. The hollocellulose was placed in a small beaker with 8 mL of 17.5% (v/v) NaOH. After a 30 minute incubation, 8 mL of deionized water was added and the sample stirred and left to react for an additional 29 minutes. The solution was then filtered through a pre- weighed, coarse, sintered glass crucible and washed three times (3 x 50 mL) with deionized water. The samples, in the crucible, were soaked in 1.0 M acetic acid for 5 minutes and then washed again (3 x 50 mL) with deionized water. The \u00EF\u0081\u00A1-cellulose was dried overnight in a 50oC oven and weighed. 2.2.7 Degree of polymerization of cellulose fibres The molecular weight distributions of \u00EF\u0081\u00A1-cellulose isolated from wild-type and the AtCesAirx mutant plant lines were measured using gel permeation chromatography (GPC) coupled to a multi angle light scattering detector (MALS; Dawn Helos-II, Wyatt Technologies). Between 8.5 mg to 40 mg of \u00EF\u0081\u00A1-cellulose was individually weighed into glass vials. Prior to dissolution, solvent exchange from water (Nanopure), to anhydrous ethanol to N,N-Dimethylacetamide (DMAC) was performed in series. In brief, the cellulose was stirred for three days in water, followed by two days in ethanol and, finally two days in DMAC. At each solvent change, the solvent was removed by centrifugation for 20 minutes at 6000 rpm. The final pellet of cellulose was resuspended in a 9% (w/v) lithium chloride in DMAC solution (LiCl/DMAC) at a ratio of 6 mg of cellulose per 1 mL of solvent. This suspension was then stirred for three days at room temperature, and stored at 4\u00EF\u0082\u00B0C prior to GPC analysis. The cellulose in LiCl/DMAC was diluted 1:20 with HPLC grade DMAC. Samples were filtered through a 0.45 \u00EF\u0081\u00ADm nylon filter and run on an Agilent 1100 series GPC instrument equipped with two Styragel columns in series (Waters; 5E, 4E) maintained at room temperature with an isocratic flow rate of 1mL/min of 0.9% LiCl/DMAC. Following separation, cellulose polymer size was estimated by multi-angle light scattering and processed using the ASTRA program (Wyatt Technologies) to determine the average molecular weights. 32 2.2.8 RNA extraction and cDNA synthesis for gene expression analysis Plant tissues (namely 21-30 day-old stems) were harvested and immediately frozen in liquid nitrogen. Total RNA was extracted using TRIzol (Invitrogen) according to the manufacturer\u00E2\u0080\u0099s instructions. RNA was treated with TURBO DNase (Ambion) to remove DNA. First-strand cDNA was synthesized from 1 \u00EF\u0081\u00ADg of DNAse-treated RNA using Superscript II Reverse Transcriptase (Invitrogen) with dT18 oligonucleotides. 2.2.9 Real-time PCR analysis Gene expression was measured using quantitative real-time PCR (RT-PCR). RT-PCR reactions were set-up in triplicate for each sample with Platinum SYBR Green qPCR Master Mix (Invitrogen) and run on an Mx3000p real-time PCR system (Stratagene). The primers used to detect AtCesA and the housekeeping control gene AtUBQ5 are listed in Table 2.2. 33 Table 2.2. PCR primers used for gene isolation, cloning, screening, and real-time (RT) PCR. Primer Name Sequence (5' to 3') Use A4HVRIIFW TGACATGTGATTGTTGGCCGTCGT RT A4HVRIIRV AATCGCCTCCGTCGATGATCGTTT RT A7HVRIIFW ACATGAATGGTGACGTAGCAGCCCTT RT A7HVRIIRV ACCGCAGCTTATGACATGGATTGCCT RT A8HVRIIFW GCAAAGCGAGAAGAACTTGATGCTGC RT A8HVRIIRV TTTACAGAGTCGGGAACACCGCCATT RT UBQ5FW ACACCAAGCCGAAGAAGATCAAGCAC RT UBQ5RV AAATGACTCGCCATGAAAGTCCCAGC RT LUC3'FW AGAGTGCTGAAGAACGAGCAGAGA screening/RT YFP3'FW ATCACATGGTCCTGCTGGAGTTCGT screening/RT A45'RV TAAACGCACACGTGACACGCCACAAA screening/RT A75'RV TCAGAGGCTTTGGCTCTTCATGGTTGTG screening/RT A85'RV TCACCACAAGTGTTGCAGATGGGA screening/RT A4GWFW CACCGAACCAAACACCATGGCCAGC isolation/pENTR cloning A4GWRV TTAACAGTCGACGCCACATTGCTTCA isolation/pENTR cloning A7GWFW CACCGAAGCTAGCGCCGGTCTTGTC isolation/pENTR cloning A7GWRV TCAGCAGTTGATGCCACACTTGGA isolation/pENTR cloning A8GWFW CACCGAGTCTAGGTCTCCCA isolation/pENTR cloning A8GWRV TTAGCAATCGATCAAAAGACA isolation/pENTR cloning PA7FWSDAI CCTGCAGGGGGTACAGAGTTTGGGGAGTGATGG promoter cloning PA4FWCLAI AGTCATCGATGGGTACAGAGTTTGGGGAGTGATGG promoter cloning PA4RVAVRII AGTCCCTAGGTCGGAAGCAGAGCAGAAGGTGGG promoter cloning PA7FWSDAI CCTGCAGGGCAGCAACAGCAGGAGAGGTACG promoter cloning PA7FWCLAI AGTCATCATGCAGCAACAGCAGGAGAGGTACG promoter cloning PA7RVAVRII AGTCCCTAGGAGGGACGGCCGGAGATTAGCAG promoter cloning PA8FWHINDIII AAGCTTAGCTCACAATCTTCTTCCTGGTCG promoter cloning PA8FWCLAI ATCGATAGCTCACAATCTTCTTCCTGGTCG promoter cloning PA8RVAVRII CCTAGGCCCTGTTTGGAGAAACAGAGAAATGAACCC promoter cloning LUCFWAVRII AGTCCCTAGGACAAACGAATCTCAAGCAATCAAGC LUC-YFP cloning Term-YFPRVKPN AGTCGGTACCTCCGGCTCGTATGTTGTGTGGAAT LUC-YFP cloning 34 Conditions for the RT-PCR reactions were 95oC for 10 minutes followed by 35 cycles of 95oC for 30 seconds, 59oC for 1 minute, and 72oC for 30 seconds. Relative gene expression was calculated using the equation described by Pfaffl (2001), where \u00EF\u0081\u0084ct represents 2-(ctCesA-ctEIF5\u00EF\u0081\u00A1), in which CesA gene expression is relative to that of the control gene, AtUBQ5. Melting curve analysis, agarose gel electrophoresis, and DNA sequencing of RT-PCR products was used to confirm that the RT-PCR amplicons represented the expected RT-PCR products. 2.2.10 Plant transformation vectors Arabidopsis thaliana cellulose synthases (AtCesA) 4, 7, and 8 cDNAs were isolated from cDNA generated from wild-type Arabidopsis (Colombia ecotype) total RNA. PCR was used to isolate the cDNAs lacking a start codon for N-tagging in the plant transformation vectors, and adding a CACC sequence to the N-terminus for ligation into the pENTR cloning vector. The primers used for AtCesA isolation by PCR are listed in Table 2.2. PCR products were cloned into the pENTR-D/TOPO cloning vector (Invitrogen) and sequenced using universal M13 FW and RV primers, as well as gene-specific internal primers (Table 2.2) to verify identity and confirm that there were no PCR-induced mutations. Confirmed pENTR-CesA plasmids were subsequently used for Gateway-Clonase insertion of the CesA cDNA into plant transformation vectors. Expression vectors containing Luciferase under the 35S promoter, pPZPhRLuc- attR (P35S::Luc), or enhanced Yellow Fluorescent Protein under the 35S promoter, pBin19YFP-attR (P35S::YFP) are described by Subramanian et al. (2006). LR Clonase II (Invitrogen) reactions between pENTR-CesA plasmids and the destination expression vectors were performed according to manufacturer\u00E2\u0080\u0099s instructions to create P35S::Luc- CesA and/or P35S::YFP-CesA expression vectors. Vectors were sequenced to ensure placement of the AtCesA cDNA in frame, downstream from the Luc or YFP tag. To create expression vectors containing AtCesAs under the control of native Arabidopsis cellulose synthase promoters, the 35S promoter from the above Luc or YFP vectors were removed and replaced with AtCesA promoters. For PAtCesA::LUC destination vectors the Arabidopsis CesA promoters were amplified from DNA (between 1 and 2 kb upstream of gene start codons) using PCR (primers in Table 2.1), adding 35 SdaI (PAtCesA4, PAtCesA7) or HindIII (PAtCesA8) to the 5\u00EF\u0082\u00A2 terminus and AvrII (PAtCesA4, 7, 8) to the 3\u00EF\u0082\u00A2 terminus of the promoter fragment. The promoter fragments were digested from cloning vectors with SdaI/AvrII or HindIII/AvrII and ligated into the P35S::Luc-CesA backbone which had the 35S promoter removed by digestion with SdaI/AvrII or HindII/AvrII. LR Clonase II reactions were performed between the promoter-modified destination vectors and pENTR-CesA vectors to create PAtCesA4::Luc-CesA4 and PAtCesA8::Luc-CesA8. PAtCesA::YFP-CesA vectors were created in a similar fashion, except that XmaJI and AvrII were used to replace the 35S promoter and replace with the CesA promoter, followed by an LR Clonase II reaction with pENTR-CesA vectors to create PAtCesA7::YFP-CesA7 and PAtCesA8::YFP-CesA8. The promoter-tag-gene fragments of all binary vectors were confirmed with sequencing. A binary construct containing a Luciferase-YFP fragment under the control of the AtCesA4 promoter was also created. The Luc-YFP fragment was amplified by PCR adding an AvrII site to the 5\u00EF\u0082\u00A2 terminus and a KpnI site to the 3\u00EF\u0082\u00A2 terminus and cloned into the pBLUNT cloning vector (Invitrogen). The YFP fragment was removed from the PAtCesA4::YFP binary vector using endogenous AvrII/KpnI sites. The Luc-YFP fragment was then cut and isolated from the cloning vector by a AvrII/KpnI double digest, and subsequently ligated into the previously linearized PAtCesA4::YFP vector, to create PAtCesA4::Luc-YFP. 2.2.11 Plant transformation Plants were transformed using a method modified from Clough and Bent (1998). Agrobacterium tumefaciens GV3101-pMP90 (Hellens et al., 2000) were transformed with a binary vector using a freeze-thaw method. Cultures were grown overnight at 28oC in Luria-Bertani medium containing 50 \u00EF\u0081\u00ADg L-1 kanamycin or 50 \u00EF\u0081\u00ADg L-1 spectinomycin, 25 \u00EF\u0081\u00ADg L-1 rifampicin, and 25 \u00EF\u0081\u00ADg L-1 gentamycin. The overnight culture was centrifuged to pellet the cells and resuspended in a 5% (w/v) sucrose solution to an OD600 of at least 0.400. Silwet L-77 (LEHLE Seeds) was added to each resuspended culture at a final concentration of 0.02% (v/v). Newly flowering plants (approximately 4 weeks old) were sprayed with the Agrobacterium solution using a fine mist spray nozzle. Sprayed plants were placed in a dark, humid environment for 16 to 24 hours and then returned to the light. A second spraying was often conducted 5 days after the first spray 36 to increase the transformation rate. After all spray treatments, plants were maintained as usual and seeds were harvested from mature, dried plants. Seeds were screened for putative transformants by germinating on half-concentration MS medium (no sucrose) with the addition of either 75 \u00EF\u0081\u00ADg L-1 kanamycin or 50 \u00EF\u0081\u00ADg L-1 glufosinate ammonium sulfate. Seedlings that successfully grew on antibiotics were grown in soil for further analysis. Transformed plants were confirmed by PCR on genomic DNA using gene specific oligonucleotides (Table 2.1) 2.2.12 Live-cell image acquisition of YFP-CesAs Both seedlings and stem sections were used for live cell imaging. For seedlings, whole, dark- or light-grown, five to ten-day-old seedlings were mounted in water between a 44 x 22 mm (#1.5) cover slip and a glass slide. For stem sections, a segment of stem from plants 15 to 21-days old was excised and longitudinal hand sections were made and immediately mounted in water as above. YFP fluorescence was detected via a Leica DMI6000 inverted microscope with a Quorum Wave FX system which had a modified Yokogawa CSU-10 spinning disk scan head (Yokogawa Electric Corporation). YFP was excited with a 491 nm laser and emissions passed through a 528/38 band filter (Chroma Technology). Images were acquired using a Hamamatsu 9100-13 EMCCD camera (Hamamatsu) controlled by Volocity software (Improvision). 2.3 Results 2.3.1 AtCesAirx phenotype The AtCesA8irx1-1, AtCesA7irx3-1, and AtCesA4irx5-1 mutants were used in this study (Figure 2.1 and Table 2.1). These mutants are of the Arabidopsis Landsberg ecotype. The AtCesA8irx1-1 mutant contains an amino acid substitution at a conserved residue in the catalytic domain, likely rendering the mutant CesA8 protein non-functional but still produced (Taylor et al., 2003). The AtCesA7irx3-1 and AtCesA4irx5-1 mutants are believed to be null-mutations (Taylor et al., 1999; Taylor et al., 2003) as a result of truncation of the protein just after the fourth transmembrane domain. These mutants were used for three reasons: 1) previous studies had reported some changes to cell wall content in these mutants, providing some background and context for the work 37 presented in this thesis; 2) homozygous mutant seed stocks were readily available; and 3) These mutants grew more easily than other AtCesAirx mutants which exhibited very weak and minimal growth and reproduction, which would have hindered some of the experiments conducted (namely genetic transformations and the potential for crossing). Homozygous A8irx1-1, A7irx3-1, and A4irx5-1 mutants have a distinctive stunted growth and rounded-leaf phenotype compared to wild-type plants grown under the same conditions (Figure 2.2). Compared to wild-type, the AtCesAirx plants are 20 to 42% shorter (Table 2.3). AtCesAirx plants grown under short-day conditions were more stunted than those grown under long-day conditions. Xylem vessels of all mutants were often collapsed and irregularly shaped (Figure 2.2, arrowheads). 38 Figure 2.2. Stature and xylem phenotype of wild-type and AtCesAirx plants. Mature, dried plants harvested after growth under short-day conditions (-S labels) or long-day conditions (-L labels). Leaf inset pictures are from 15-day-old plants grown under long- day conditions. Wild-type (WT) plants (A) looked nearly indistinguishable under long and short-day conditions, so only short-day plants are imaged here. B, E, I: AtCesA8irx1-1. C, F, J: AtCesA7irx3-1. D, G, K: AtCesA4irx5-1. Cross sections of the lower portion of inflorescence stems (within the bottom 4 cm) show normal xylem in wild-type (H) and collapsed xylem in the mutants (I-K). Arrowheads indicate xylem vessels. Scale bar represents 50 \u00EF\u0081\u00ADm. 39 Table 2.3. Maximum stem height and rosette-leaf diameter of wild-type and AtCesAirx plants grown under long-day and short-day conditions. Averages were calculated from measurements taken from 35-50 individual plants. Growth was measured for a total of 42 days (long-day, all lines), 98 days (short-day, wild-type), or 63 days (short-day, AtCesAirx lines) after which time growth ceased. Bold values represent averages that are statistically different from wild-type averages (t-test p<0.05). Long-day Short-day Rosette Diameter Stem height Rosette Diameter Stem height Line Day s Max. Avg. (cm) S.D. Days Max. Avg. (cm) S.D. Days Max. Avg. (cm) S.D. Days Max. Avg. (cm) S.D. Wild-type 28 7.3 1.1 42 37.3 4.6 49 8.8 1.1 98 34.3 8.7 A8irx1-1 28 4.0 0.8 42 15.7 2.3 49 4.7 0.8 63 7.2 2.6 A7irx3-1 28 3.8 0.8 42 13.6 2.1 42 5.1 0.8 56 12.5 1.3 A4irx5-1 28 4.0 0.8 42 14.1 1.3 49 5.3 0.9 63 10.4 3.1 40 Under long-days, stem heights increased steadily for 30 days, after which there was a brief growth plateau. Under short-day conditions AtCesAirx stems reached a maximum height by day 55, whereas the wild-type plants showed a steady increase in stem height after 60 days of growth and reached a maximum height after day 80. As seen in Figure 2.3, the stem growth rate of wild-type and AtCesAirx plants differ depending on day length growth conditions. Long-day AtCesAirx plants exhibit a reduced growth rate compared to wild-type plants. Contrastingly, short-day AtCesAirx plants exhibited an increased growth rate during the first 8 weeks of growth compared to wild-type plants. In particular, A7irx3-1 and A4irx5-1 plants experienced more rapid growth than wild-type and A8irx1-1plants. Day length conditions also affected reproductive stages of growth (Figure 2.4). Under long day conditions all lines formed buds and siliques at approximately the same times post-planting. However, under short-day conditions all the mutant lines formed buds and mature siliques significantly earlier than wild-type plants, and had fully matured nearly a month prior to wild-type plants. 2.3.2 Carbohydrate and lignin content Total lignin (soluble plus insoluble) and structural carbohydrate content of mature AtCesAirx stems were determined and compared to wild-type plants to identify and quantify any differences in cell wall composition (Figure 2.5 and 2.7). Additionally, stems from plants grown under both long and short-day light conditions were compared. 2.3.2.1 Cellulose Whole, mature, dried stems were evaluated for cellulose content. As seen in Figure 2.5, the AtCesAirx mutant lines had significantly lower levels of total glucose when compared to wild-type stems. Under both day lengths, A8irx1-1 plants had more glucose than A7irx3-1 and A4irx5-1 plants, as compared to wild-type. A8irx1-1 stems contained between 58% and 66% the amount of glucose found in wild-type plants, whereas A7irx3-1 and A4irx5-1 contained on average 65% less glucose than wild-type. 41 Figure 2.3. Stem growth rate of wild-type and AtCesAirx plants grown under varied day- length conditions. (A) Plants grown under long-day conditions (16 hours light). (B) Plants grown under short-day conditions (8 hours light). Plant height was measured weekly once the primary stem was at least 0.5 cm tall, until plants ceased growing. The average stem height calculated each week is from a population of approximately 50 plants. Error bars represent standard deviations. 3 4 5 6 0 5 10 15 20 25 30 35 40 45 A v e ra g e h e ig h t (c m ) Week WT IRX1 IRX3 IRX5 2 4 6 8 10 12 14 -10 -5 0 5 10 15 20 25 30 35 40 45 A v e ra g e h e ig h t (c m ) Week WT IRX1 IRX3 IRX5 A Long-day B Short-day 42 Figure 2.4. Chronological progression of flower and silique formation of wild-type and AtCesAirx Arabidopsis plants. Growth stages were monitored weekly for plants grown under long-day (A) and short-day (B) light conditions. Bars represent the time period during which 50% of the plants monitored were exhibiting that stage of growth. The red line indicates when silique maturation began (based on complete yellowing of siliques on 50% of the plants monitored). The legend in box A also applies to box B. 43 Figure 2.5. Structural carbohydrate content of wild-type and AtCesAirx Arabidopsis stems. Plants were grown under long-day conditions (A) and short-day conditions (B). Error bars represent standard deviation, n represents 3 (3 pools of at least 25 plants in each pool). *values are statistically significant compared to wild-type, **values are statistically significant compared to AtCesA8irx1-1, (p<0.05, ANOVA followed by Tukey's post-hoc test). WT IRX1 IRX3 IRX5 0 5 10 15 20 25 30 35 40 45 * ** ** *** * * * * * * * A v g . % d ry w e ig h t Line Glucose Hemicellulose Lignin WT IRX1 IRX3 IRX5 0 5 10 15 20 25 30 35 40 45 * ** ** * ** * * ** ** * ** * * * A v g . % d ry w e ig h t Line Glucose Hemicellulose Lignin B Short-day A Long-day WT A8irx1- A7irx3-1 A4irx -1 WT A8irx1-1 A7irx3-1 A4irx5-1 * ** 44 Interestingly, A8irx1-1 plants had increased starch levels under both long and short-day growth conditions, whereas A7irx3-1 and A4irx5-1 plants had reduced total glucose and starch under long-day conditions (Figure 2.7). To determine what proportion of the altered glucose content can be ascribed to the cellulose in AtCesAirx stems, the amount of \u00EF\u0081\u00A1-cellulose was determined from extractive-free stem samples (Figure 2.6). All the AtCesAirx lines, under both short and long-day conditions, showed significantly reduced levels of \u00EF\u0081\u00A1-cellulose compared to wild-type stems. The reductions were more significant in short-day plants, such that AtCesAirx stems contained on average 62% less \u00EF\u0081\u00A1-cellulose than wild-type stems. Under both day-length conditions A7irx3-1 showed the greatest reduction in \u00EF\u0081\u00A1-cellulose. Interestingly, the glucose levels in A8irx1-1 plants seemed least affected (compared to the other AtCesAirx lines), however, the \u00EF\u0081\u00A1-cellulose levels for all the AtCesAirx lines were similar suggesting an equivalent cellulose reduction in all mutant lines. 2.3.2.2 Lignin Acid-soluble (AS) and acid-insoluble (AI) lignin were determined for AtCesAirx and wild-type stems, and summed to represent total lignin content (Figure 2.5). Lignin content was elevated in all the AtCesAirx lines under both short and long day conditions The greatest changes in total lignin, above wild-type levels, were observed in A7irx3-1 and A4irx5-1 stems grown under long-day conditions (a 32% and 24% increase over wild- type, respectively). The A8irx1-1 stems also contained elevated lignin, but only significantly under short-day conditions, with a 15% increase over wild-type. Under long-day conditions the increases in lignin in AtCesAirx stems was due to an overall increase in both AS and AI lignin (Figure 2.7). A trend observed in all long-day mutant lines was a higher proportion of AI to AS lignin such that the ratio of AS:AI decreased from 8.1 in wild-type to between 5.5 and 7.3 in AtCesAirx lines. The observed changes in AI and AS proportions were only statistically significant in A4irx5-1 stems, and can be attributed more specifically to an excessive increase in AI lignin compared to all other lines (Figure 2.7). 45 Figure 2.6. \u00EF\u0080\u00A0\u00EF\u0081\u00A1-cellulose content of wild-type and AtCesAirx stems. Stems were harvested from mature, dried plants grown under long-day (white bars) and short-day (grey bars) conditions. Error bars represent standard deviation, n represents 3 (3 pools of at least 25 plants each). *values are statistically significant compared to wild-type (p<0.05, ANOVA followed by Tukey's post-hoc test). WT IRX1 IRX3 IRX5 0 5 10 15 20 25 30 35 40 45 * * ** * * \u00EF\u0081\u00A1 -c e llu lo s e ( A v g . % d ry w e ig h t) Line Longday Shortday WT A8irx1-1 A7irx3-1 A4irx5-1 46 Figure 2.7. A summary of growth and cell wall changes measured in AtCesA8irx1-1, AtCesA7irx3-1, and AtCesA4irx5-1 plants grown under long and short day conditions. Values represent the fraction percentage compared to wild-type plants (e.g.: 0.66 glucose corresponds to 66% of the wild-type level), such that values above and below 1.0 represent an increase and decrease compared to wild-type, respectively. Red colours indicate a large decrease, green colours a large increase. Orange and yellow colours represent a range of moderate decreases, no change, to moderate increases. MFA \u00E2\u0080\u0093 microfibril angle, DP \u00E2\u0080\u0093 degree of polymerization, AS \u00E2\u0080\u0093 acid soluble, AI \u00E2\u0080\u0093 acid insoluble 47 2.3.2.3 Hemicellulose and pectin Components of secondary cell wall hemicelluloses were estimated by HPLC determination by quantifying the amount of fucose, arabinose, galactose, xylose, and mannose in stem extractions (Table 2.4). All the AtCesAirx lines exhibited an overall increase in hemicellulose content compared to wild-type, under both long and short-day growth conditions (Figure 2.7). The greatest change was in stems of A7irx3-1and A4irx5-1 plants which had a 34% and 30% increase above wild-type, respectively. A8irx1-1 stems had a slight, but statistically significant, increase (8% above wild-type) in hemicelluloses, but only under long-day conditions. A more careful inspection of the monosaccharide composition (Table 2.4) indicates that the increase in hemicellulose content of AtCesAirx stems is due to broad increases in most of the monosaccharide components analyzed, with the greatest increase being in fucose, arabinose, and xylan, of which the latter two may represent an increase in arabinoxylan. The only exception was mannose, which generally decreased in all mutant lines, particularly under short-day conditions. Rhamnose values, which form part of the pectic components, remained similar to wild-type levels except in A4irx5-1 stems, which contained 17% to 22% greater levels of rhamnose under long and short-day conditions, respectively. 2.3.3 Cellulose characterization To broaden our understanding of how the AtCesA mutations affect cellulose ultrastructure, cellulose crystallinity, microfibril angle (MFA), and degree of polymerization (DP) were measured and compared to wild-type stems (Table 2.5). Cellulose MFAs in the A7irx3-1 and A4irx5-1 lines were significantly larger than the MFAs of wild-type and A8irx1-1, when grown under long-day conditions. Wild-type and A8irx1-1 MFA was approximately 14o, whereas A7irx3-1 and A4irx5-1 MFA was between 17o and 20o. Interestingly, there were no significant differences in MFA among wild-type and the AtCesAirx lines when grown under short-day conditions. In general, the MFAs were larger in all plants when grown under short-day conditions compared to long-day. 48 Table 2.4. Cell wall carbohydrate content of wild-type (WT) and AtCesAirx Arabidopsis inflorescence stems from plants grown under long-day and short-day conditions. Standard deviation is in parenthesis. Values in bold are statistically significant compared to wild-type values, *indicates a significant difference compared to A8irx1-1, \u00EF\u0081\u00B4indicates a difference compared to A4irx5-1 (p<0.05, ANOVA followed by a Tukey\u00E2\u0080\u0099s post-hoc test), n represents 3. Line Fucose Arabinose Rhamnose Galactose Glucose Xylose Mannose Long-day (\u00EF\u0081\u00ADg/mg dry weight) WT 3.02 (0.15) 10.71 (0.40) 8.04 (0.53) 12.36 (0.58) 433.06 (14.44) 144.28 (3.22) 17.41 (0.89) A8irx1-1 3.25 (0.07) 13.75 (0.12) 7.60 (0.38) 13.12 (0.27) 287.80 (5.76) 156.24 (3.76) 16.70 (0.45) A7irx3-1 3.51 (0.11) \u00EF\u0081\u00B4 13.21 (0.77)\u00EF\u0081\u00B4 7.94 (0.70)\u00EF\u0081\u00B4 13.47 (0.63) 163.78 (5.72)*\u00EF\u0081\u00B4 206.94 (7.88)* 13.42 (0.37)*\u00EF\u0081\u00B4 A4irx5-1 4.45 (0.50)* 15.19 (1.31)* 9.44 (1.07)* 13.67 (0.31) 138.35 (13.95)* 195.28 (6.42)* 15.29 (1.19) Short-day (\u00EF\u0081\u00ADg/mg dry weight) WT 3.58 (0.12) 15.41 (0.64) 8.48 (0.64) 15.69 (0.67) 439.79 (24.72) 141.61 (9.51) 20.66 (1.57) A8irx1-1 4.15 (0.16) 20.91 (0.71) 8.34 (0.35) 15.45 (0.63) 255.99 (11.39) 146.79 (5.11) 15.93 (1.18) A7irx3-1 4.41 (0.19) 15.78(1.29)*\u00EF\u0081\u00B4 8.64 (1.18)\u00EF\u0081\u00B4 14.42 (0.56) 152.66 (1.77)* 183.73 (3.02)* 12.43 (1.31)* A4irx5-1 4.61 (0.13)* 17.99 (0.25)* 10.31 (0.43)* 15.00 (0.35) 153.54 (1.97)* 186.68 (3.25)* 14.55 (0.52) 49 Table 2.5. Microfibril angle, cell wall crystallinity, and relative degree of polymerization (DP) of cellulose from stems of wild-type (WT) and AtCesAirx plants. The DP values presented are relative to WT (WT set to 100). Plants were grown under long-day and short-day light conditions. Standard deviation is in parenthesis. Values in bold are statistically significant compared to wild-type values, *indicates a significant difference compared to A8irx1-1, \u00EF\u0081\u00B4indicates a difference compared to A4irx5-1 (p<0.05, ANOVA followed by a Tukey\u00E2\u0080\u0099s post-hoc test), n represents 6 for MFA and crystallinity, n=2 or 3 for relative DP. a represents no day-length effect and b represents a day-length effect (p<0.05, two-way ANOVA followed by a Bonferroni post-hoc test). Line MFA % Crystalline Relative DP Long-Day WT 13.5 (0.8)b 42.6 (2.3) a 100.0 (17.5) A8irx1-1 13.9 (2.1) b 32.6 (4.3) a 117.4 (10.2) A7irx3-1 17.2 (1.6)* b 25.5 (5.5) a 72.7 (12.6)* \u00EF\u0081\u00B4 A4irx5-1 19.9 (1.0)* a 34.3 (8.5) a 38.6 (11.6)* Short-Day WT 20.8 (2.1) 46.1 (7.3) 100.0 (17.5) A8irx1-1 19.7 (1.4) 37.6 (9.3) 45.9 (16.6) A7irx3-1 20.8 (3.0) 35.8 (6.7) 44.5 (3.7) A4irx5-1 21.2 (2.1) 44.1 (8.6) 43.2 (16.8) 50 From long-day to short-day conditions, wild-type and A8irx1-1 plants exhibited an average increase in MFA of 6.5o, whereas A7irx3-1 and A4irx5-1 only had an average 2o increase in MFA when grown under short-day conditions. Two-way ANOVA analysis revealed a significant day-length effect on MFA in wild-type, A8irx1-1, and A7irx3-1 plants. Interestingly, no day-length effect was found for MFA of A4irx5-1 plants, suggesting that the A4irx5-1 mutation may mute the effect of day-length on MFA. Under long-day conditions, cell wall crystallinity was lower in the AtCesAirx mutants compared to wild-type plants, although only the decreases in A8irx1-1 and A7irx3-1 were statistically significant. Under short-day conditions, a similar trend was observed but none of the values were significantly different, and the decreases (compared to wild- type) were less than those observed for long-day grown plants. Although the general trend in cell wall crystallinity values was increased for all lines when grown under short- days, compared to long-days, there was no statistically significant effect of day-length effect on cell wall crystallinity. The degree of polymerization (DP) of the isolated \u00EF\u0081\u00A1-cellulose from the mutant lines was calculated relative to the wild-type samples. As seen in Table 2.5, under long- day conditions the \u00EF\u0081\u00A1-cellulose from the A7irx3-1 and A4irx5-1 mutants had a significantly lower DP compared to wild-type and the A8irx1-1 mutant plants. In particular, the relative DP of A4irx5-1 plants was approximately 60% lower than that of wild-type. Under short- day conditions, a more uniform reduction in DP was observed in the mutant lines, as all the mutants had relative DP values that were approximately 55% lower than that of the wild-type plants. Reduced \u00EF\u0081\u00A1-cellulose quantities limited the replications that could be performed for DP, and for this reason a two-way ANOVA was not performed to statistically determine the presence of a day-length effect on DP. However, comparison of the raw DP values (data not shown) suggests there may only be a negative day- length effect on DP of A8irx1-1 mutants grown under short-days. A summary of the changes to growth and cell wall properties in the AtCesAirx is presented in Figure 2.7. In general, the A7irx3-1 mutant plants exhibit a more severely altered cell wall phenotype compared to the A8irx1-1 and A4irx5-1 mutants. 51 2.3.4 Expression of AtCesAs in AtCesAirx and wild-type plants Various constructs were created to assess the expression of the wild-type AtCesA genes in both wild-type and AtCesAirx mutant backgrounds. Gene expression was controlled by either the constitutive cauliflower mosaic 35S promoter (P35S or the native AtCesA promoters (PAtCesA). For all constructs, the AtCesA cDNA was tagged with one of either an N-terminal yellow fluorescent protein (YFP) or Luciferase (LUC) tag (YFP-CesA or LUC-CesA). 52 2.3.4.1 P35S::AtCesA expression in AtCesAirx plants Expression of wild-type AtCesA in an AtCesAirx mutant background should restore mutant plants to a wild-type phenotype if the CesA transgene is functional and expressed at the appropriate temporal and spatial levels. Expression of P35S::LUC- AtCesA constructs in A4irx5-1 and A8irx1-1 plants were not sufficient to rescue the mutant phenotype (Figure 2.8). Three independent transgenic A4irx5-1 lines carrying the P35S::LUC-AtCesA4 construct showed no recovery of stunted growth or collapsed xylem vessels. Similarly, A8irx1-1 plants carrying the P35S::LUC-AtCesA8 construct showed no recovery from the mutant phenotype. Lack of complementation appears to be correlated to low transgene expression levels in the transgenic A4irx5-1 lines (Figure 2.9 A) compared to endogenous CesA expression in wild-type plants, and may be due to 35S-driven expression in the A8irx1-1. 53 Figure 2.8. AtCesAirx plants transformed with P35S::LUC-AtCesA constructs do not show recovery of a wild-type phenotype. P35S::LUC-AtCesA4 is expressed in the AtCesA4irx5-1 mutant background (C, G), and P35S::Luc-AtCesA8 in AtCesA8irx1-1 (D, H). Transgenic lines and non-transformed AtCesAirx (B, F) plants exhibit stunted growth and collapsed xylem compared to wild-type plants (A, E). Scale bar represents 50 \u00EF\u0081\u00ADm. 54 Figure 2.9. Real-time PCR analysis of transcript abundance in RNA from stems of 15- day-old plants. AtCesAirx plants were transformed with endogenous AtCesAs under the 35S promoter. (A) AtCesA4 expression was measured in wild-type (WT), AtCesA4irx5-1 (A4-irx), and A4irx5-1-P35S::LUC-AtCesA4 stems (#3, #4, and #5). (B) AtCesA8 expression was measured in wild-type (WT), AtCesA8irx1-1 (A8-irx), and A8irx1-1- P35S::LUC-AtCesA8 (35S-L-A8) stems. Expression levels are relative to the control gene, AtUBQ5. Error bars represent standard error of the mean, n represents 3 biological replicates. WT A4-irx #3 #4 #5 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 A R e la ti v e e x p re s s io n Line AtCesA4 WT A8-irx 35S-L-A8 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 B R e la ti v e e x p re s s io n Line AtCesA8 55 2.3.4.2 PAtCesA::AtCesA expression in AtCesAirx plants Proper temporal and spatial expression of the AtCesA transgenes should be achieved when expression is regulated by the native AtCesA promoters. Indeed, A4irx5-1 plants transformed with PAtCesA4::LUC-AtCesA4, A7irx3-1 plants transformed with PAtCesA7::YFP-AtCesA7, and A8irx1-1 plants transformed with PAtCesA8::LUC-AtCesA8 all showed varying levels of complementation. Transgenic A4irx5-1 and A7irx3-1 plants did not show the stunted growth phenotype and had open, normal-shaped xylem vessels (Figure 2.10). Transgenic A8irx1-1 plants exhibited only partial complementation as evidenced as growth remained stunted, but microscopically had fewer collapsed xylem (Figure 2.10). Varying degrees of complementation could be the result of transgene expression levels. Real-time PCR analysis revealed different levels of transgene expression in the transgenic lines (Figure 2.11). Levels of AtCesA7 in the A7irx3-1- YFP- AtCesA7 transgenic plants were 25% greater than endogenous AtCesA7 levels in wild- type plants. Comparatively, levels of AtCesA4 in A4irx5-1- LUC-AtCesA4 transgenic plants and levels of AtCesA8 in A8irx1-1- LUC-AtCesA8 transgenic plants were only 7% and 31% of endogenous gene expression levels, respectively. Two additional lines of A7irx3-1 expressing PAtCesA7::YFP-AtCesA7 had AtCesA7 transcript levels that were significantly lower than wild-type levels and showed little to no complementation of the mutant phenotype (data not shown). 2.3.4.3 Double-transgenic wild-type plants Wild-type plants of the Columbia (WT-C) or Landsberg (WT-L) ecotypes were transformed with two constructs to create double transgenic plants. The following transformations were recovered: WT-C transformed with both P35S::LUC-AtCesA7 and P35S::YFP-AtCesA7 (WTC-2xA735S), WT-C transformed with PAtCesA7::YFP-AtCesA7 and P35S::LUC-AtCesA7 (WTC-2xA735S/PA7), WT-C transformed with P35S::LUC- AtCesA4 and P35S::YFP-AtCesA4 (WTC-2xA435S), and WT-L transformed with PAtCesA8::LUC-AtCesA8 and PAtCesA8::YFP-AtCesA8 (WTL-2xA8PA8). The T1 generation plants were grown alongside untransformed WT and AtCesAirx plants under long-day growth conditions. 56 Figure 2.10. Phenotype of AtCesAirx plants transformed with wild-type AtCesAs under the control of native AtCesA promoters. Images are of 25-day-old plants. Wild-type (A and F), AtCesAirx (B and G), AtCesA4irx5-1 transformed with PAtCesA4::LUC-AtCesA4 (C and H), AtCesA7irx3-1 transformed with PAtCesA7::YFP-AtCesA7 (D and I), AtCesA8irx1-1 transformed with PAtCes8::LUC-AtCesA8 (E and J). Arrow heads indicate open xylem in F, H, I, and J and collapsed xylem in G. Scale bar represents 50 \u00EF\u0081\u00ADm. 57 Figure 2.11. Real-time PCR analysis of transcript abundance of AtCesAirx plants transformed with endogenous AtCesAs under native AtCesA promoters. A: AtCesA4 expression was measured in wild-type (WT), A4irx5-1 (IRX5), and A4irx5-1-PAtCesA4::Luc- AtCesA4 stems (PA4-L-A4). B: AtCesA7 expression was measured in wild-type (WT), A7irx3-1(IRX3), and A7irx3-1-PAtCesA7::YFP-AtCesA7 stems (PA7-Y-A7). C: AtCesA8 expression was measured in wild-type (WT), A8irx1-1 (IRX1), and A8irx1-1-PAtCesA8::Luc- AtCesA8 stems (PA8-L-A8). Expression levels are relative to the control gene, AtUBQ5. Error bars represent standard deviation, n represents 3 biological replicates. WT IRX5 PA4-L-A4 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 R e la ti v e e x p re s s io n AtCesA4 WT IRX3 PA7-Y-A7 0.0 0.5 1.0 1.5 2.0 2.5 R e la ti v e e x p re s s io n AtCesA7 WT IRX1 PA8-L-A8 0.0 0.2 0.4 0.6 0.8 1.0 1.2 R e la ti v e e x p re s s io n AtCesA8 A B C 58 All of the WT-C double transgenic plants (WTC-2xA735S, WTC-2xA735S/PA7, and WTC- 2xA435S) showed a severely stunted, AtCesAirx-like phenotype (Figure 2.12), whereas the WTL-2xA8PA8 double transgenic line exhibited only a mild reduction in stature (Figure 2.13). The WT-C double transgenic lines had a bushy appearance due to the growth of many axillary branches. Stems were extremely thin and weak, and the double WT-C transgenic plants produced fewer mature siliques and siliques devoid of any seeds. Additionally, these transgenic lines had collapsed xylem and all had reduced AtCesA transcript abundance compared to wild-type plants (data not shown). The stems of the WTL-2xA8PA8 transgenic line were slightly thinner compared to wild-type stems, and had a reduced ability to grow upright, but did not have collapsed xylem and had elevated AtCesA8 transcript abundance (data not shown). 2.3.5 Visual profile of YFP-CesA7 Expression of PAtCesA7::YFP-AtCesA7 in homozygous A7irx3-1 mutant plants restored the wild-type phenotype. This suggests that the YFP-AtCesA7 subunit is functional in the cellulose synthesizing complex. Using fluorescence spinning-disc confocal microscopy, the distribution patterns and movement of YFP-CesA7 was observed in a variety of cell types during different stages of plant growth (Figure 2.14- Figure 2.19). Expression was observed in many tissue types including seedlings, roots, leaf, and very young stem tissue (Figure 2.14). In the root tip, YFP-CesA7 particles appeared as small, dense puncta. It is interesting to note that all cells observed in the root tip contained YFP-CesA7 except the outer most layer of root cap columella cells (Figure 2.14). The YFP-CesA7 particles observed varied in size, shape, rate of movement, and abundance depending on tissue type and developmental stage. Compared to the puncta observed in root tips, the density of YFP-CesA7 particles in hypocotyl xylem was reduced and often a diffuse signal was observed (Figure 2.14). In the cotyledon and leaf tissue, large, fast moving (likely due to cytoplasmic streaming) puncta were observed in pavement cells and guard cells (Figure 2.14). Using a longitudinal section of very young stem tissue, where a great deal of cellular division and expansion is expected to be occurring, a high concentration of large, bright puncta were observed (Figure 2.14 D-E). 59 Figure 2.12. Wild-type (ecotype Columbia) plants transformed with two AtCesA- expression constructs show a severe AtCesAirx-like phenotype. Wild-type (A) and AtCesA4irx5-1 (B) plants were grown alongside the T1 generation double transgenic plants, (C) WTC-2xA735S, (D) WTC-2xA735S/PA7, and (E) WTC-2xA435S. Images are of mature, dried plants. Notice the reduced presence of full-sized siliques on the double transgenic plants. A B C D E 60 Figure 2.13. Wild-type (ecotype Landsberg) plants transformed with two AtCesA8 expression constructs show a mostly wild-type like phenotype. Wild-type (A, C) plants were grown alongside the T1 generation of double transgenic plants WTL-2xA8PA8 (B, D). Plants were 25-30 days-old. Arrow heads indicate. Scale bars represent 50 \u00EF\u0081\u00ADm. 61 Figure 2.14. YFP-CesA7 expression patterns in young tissues of AtCesA7irx3-1- PAtCesA7::YFP-AtCesA7 plants. (A) Root tip in a 5-day-old seedling. (B) Hypocotyl of a 7-day-old seedling. (C) Cells in cotyledons of a 5-day-old seedling. (D) A longitudinal section of the uppermost region of an inflorescence stem. The blue lines outline predicted cell edges based on the corresponding bright field image (not shown). (E) A magnification of the ring-shaped YFP-CesA7 particles observed in upper, developing stem tissue. White arrow heads indicate xylem vessels. Coloured arrow heads are positional markers to compare YFP and the corresponding brightfield image. Scale bars represent 20 \u00EF\u0081\u00ADm (A-D), 5 \u00EF\u0081\u00ADm (E). 62 The cells are believed to be a combination of pith parenchyma, immature xylem, and immature fibres. As seen in Figure 2.14-E, some of the YFP-CesA7 particles in this tissue were often observed to be contained in ring shapes. A profile of YFP-CesA7 in stems was created to document the varied expression in the various cell types of stem tissue, during different stages of development (Figure 2.15-2.18). From these profiles it is apparent that CesA7 is highly expressed in a variety of cell types from those in young developing stem tissue, to the mature stem base. Specifically, in stem tissue, it appears that CesA7 is expressed in cells undergoing cell expansion (based on the position of cells near the stem tip, Figure 2.15), young xylem vessels which exhibit spiral cell-wall thickenings (Figure 2.15 and 2.16), fibres (Figure 2.17), pith cells (Figure 2.15-2.18), and potentially phloem, epidermal tissue, cortex, and endodermal cells (Figure 2.15-2.18). In the upper stem there was a clear association of YFP-CesA7 with developing xylem (Figure 2.15) which decreases near the stem base (Figure 2.18). Additionally, it can be seen that as interfascicular fibre abundance increases from stem tip to base, there is a greater density of diffuse YFP-CesA7 signal visible in fibres (Figure 2.17). The mid-stem region, between 5 and 10 cm from the base of an approximately 20cm-tall plant, appeared to contain fibres with the strongest YFP-CesA7 signal. Focusing on the plasma membrane of these developing fibres, the YFP-CesA7 signal covered the plasma membrane, often appearing as a sheet of YFP signal (Figure 2.17-F). Only occasionally were small, distinct particles visible. The particles observed (both small and large) appeared to be moving at an extremely fast rate. However, measuring particle velocity or tracking the directionality of particle movement was not possible using traditional kymograph analysis due to the depth of fibre cells within the tissue, resulting in significant interference of YFP signal from surrounding cells. 63 64 Figure 2.15. Cellular anatomy and YFP-CesA7 expression patterns in the upper stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants. Toluidine Blue staining of a stem cross section (A) and longitudinal sections (B) was performed to observe cellular anatomy. (B) A collage of longitudinal sections was assembled to compare cell types. YFP- CesA7 signal in longitudinal sections of stem tissue (C-E). (C) A collage of longitudinal sections was assembled to compare expression in various cell types across the stem. (D) and (E) are magnified images of xylem elements. White arrow heads indicate xylem vessels. Coloured arrow heads are positional markers to compare YFP and the corresponding brightfield image. Letter markings indicate general regions containing the following cell types: ep-epidermis, c -cortex, en- endodermis, fi-fibres, x- xylem, p/c-phloem/cambium, pi-pith. Scale bars represent 100 \u00EF\u0081\u00ADm (A, B), 25 \u00EF\u0081\u00ADm (C), 5\u00EF\u0081\u00ADm (D, E). 65 66 Figure 2.16. Cellular anatomy and YFP-CesA7 expression patterns in the upper middle stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants. Toluidine Blue staining of a stem cross section (A) and longitudinal sections (B and C) were performed to observe cellular anatomy. YFP-CesA7 signal in longitudinal sections of stem tissue (D-F). (D) A collage of longitudinal sections was assembled to compare expression in various cell types across the stem (from one edge to just before the opposite outer edge). (E) and (F) are magnified images of a xylem element (E) and fibre (F). White arrow heads indicate xylem vessels. Coloured arrow heads are positional markers to compare YFP and the corresponding brightfield image. Letter markings indicate general regions containing the following cell types: ep-epidermis, c -cortex, en- endodermis, fi-fibres, x- xylem, p/c-phloem/cambium, pi-pith. Scale bars represent 100 \u00EF\u0081\u00ADm (A-C), 25 \u00EF\u0081\u00ADm (D), 10\u00EF\u0081\u00ADm (E), 5\u00EF\u0081\u00ADm (F). 67 68 Figure 2.17. Cellular anatomy and YFP-CesA7 expression patterns in the lower middle stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants. Toluidine Blue staining of a stem cross section (A) and longitudinal sections (B and C) were performed to observe cellular anatomy. YFP-CesA7 signal in longitudinal sections of stem tissue (D-F). (D) A collage of longitudinal sections was assembled to compare expression in various cell types across the stem (from the outer edge to the centre pith). (E) and (F) are magnified images of a xylem vessel (E) and two fibres (F). The red box highlights an area that contains a saturated signal due to the high abundance of YFP-CesA7 at the plasma membrane. The arrows indicate very small individual particles. White arrow heads indicate xylem vessels. Coloured arrow heads are positional markers to compare YFP and the corresponding brightfield image. Letter markings indicate general regions containing the following cell types: ep-epidermis, c -cortex, fi- fibres, x- xylem, pi-pith. Scale bars represent 100 \u00EF\u0081\u00ADm (A-C), 25 \u00EF\u0081\u00ADm (D), 10\u00EF\u0081\u00ADm (E), 5\u00EF\u0081\u00ADm (F). 69 70 Figure 2.18. Cellular anatomy and YFP-CesA7 expression patterns near the base of the stem of AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants. Toluidine Blue staining of a stem cross section (A) and longitudinal sections (B and C) were performed to observe cellular anatomy. YFP-CesA7 signal in longitudinal sections of stem tissue (D-F). (D) A collage of longitudinal sections was assembled to compare expression in various cell types across the stem (from the outer edge to the centre pith). (E) and (F) are magnified images of a xylem vessel (E) and fibres (F). (F) The blue lines highlight predicted cell edges based on the associated brightfield image (not shown). White arrow heads indicate xylem vessels. Coloured arrow heads are positional markers to compare YFP and the corresponding brightfield image. Letter markings indicate general regions containing the following cell types: ep-epidermis, c -cortex, en \u00E2\u0080\u0093 endodermis, fi-fibres, x- xylem, pi-pith. Scale bars represent 100 \u00EF\u0081\u00ADm (A-C), 25 \u00EF\u0081\u00ADm (D), 10\u00EF\u0081\u00ADm (E), 5\u00EF\u0081\u00ADm (F). 71 2.3.6 Visualization of YFP-CesA8 in wild-type plants Wild-type (Landsberg ecotype) transformed with both PAtCesA8::Luc-AtCesA8 and PAtCesA8::YFP-AtCesA8 (WTL-2xA8PA8) were used to visualize the distribution of CesA8 in stem tissue. Compared to YFP-CesA7-expressing plants, the YFP-CesA8 signal was weaker in intensity and less abundant in all cell types of the stem. In particular, it was not detected in developing fibres, compared to the high levels of YFP- CesA7 detected in fibres. As seen in Figure 2.19, it was possible to detect YFP-CesA8 in xylem cells. The observed patterns and abundance of YFP-CesA8 compared to YFP-CesA7 could be due to expression as a result of the transgene insertion, the expression of YFP-CesA8 in a wild-type background (compared to the irx background in the case of the YFP-CesA7 plants), the native differences in expression and function of CesA8 compared to CesA7, or a combination of these factors. 72 Figure 2.19. YFP-CesA8 expression patterns in xylem of wild-type plants transformed with PAtCesA8::YFP-AtCesA8 (PA8-YFP-A8). YFP-CesA8 signal in longitudinal sections of stem tissue from transformed, PA8-YFP-A8, plants (A) as compared to untransformed wild-type (B) and AtCesA7irx3-1-PAtCesA7::YFP-AtCesA7 plants (C). In (A) the yellow arrow head indicates a YFP-CesA8 ring structure. Coloured arrow heads are positional markers to compare YFP and the corresponding bright field image. Scale bars represent 10 \u00EF\u0081\u00ADm. 73 2.4 Discussion To date, AtCesA4, AtCesA7, and AtCesA8 are the most well-characterized secondary cell wall cellulose synthases in the plant kingdom, and mutants have been shown to possess altered cell wall composition (Turner and Somerville, 1997; Zhong et al., 2003; Ha et al., 2002; Taylor et al., 2003; Taylor et al., 1999; Brown et al., 2005; Bosca et al., 2006). However, to-date there has not been a comprehensive review of the AtCesA4, AtCesA7, and AtCes8 mutants in a single, independent study. Nor has there been a uniform report on the changes to cell wall composition (Table 2.1), and there is no information pertaining to cellulose ultrastructure such as MFA, DP, and crystallinity. In an attempt to increase our understanding of the potential role(s) of each CesA in cellulose biosynthesis we characterized how each mutant manifests differences in secondary cell wall composition and cellulose ultrastructure. Our findings provide evidence to suggest that AtCesA4 may have a unique effect on cellulose ultrastructure, and that the composition of the CSC has a very strong influence on cellulose MFA, crystallinity, and DP. Herein we provide the first report of three secondary cell wall AtCesA mutants analyzed together including novel information about the cellulose ultrastructure of these mutants, allowing for important comparisons among the mutants. 2.4.1 Growth defects of AtCesAirx mutants manifest differently under varied day- length conditions The reports on the growth characteristics of the AtCesA8irx1-1, AtCesA4irx5-1, and AtCesA7irx3-1 mutant plants are that they exhibit stunted growth and have weak stems, compared to wild-type plants (Turner and Somerville, 1997; Bosca et al., 2006). However, from the work presented herein it is apparent that when grown under short- day conditions the stunted AtCesAirx phenotype is more severe. Additionally, we reveal pleiotropic effects of the AtCesAirx mutations on growth rate and reproductive rates. The abundance of collapsed xylem in AtCesAirx mutants likely reduces water transport, which may in turn reduce growth potential resulting in overall stunting of both stem and leaves. Total glucose levels were significantly decreased in AtCesAirx plants compared to wild-type, which could signify a reduced photosynthetic capacity in the mutants, thus limiting the amount of growth possible. However, having not measured 74 other physical parameters such as soluble sugar content and rate of photosynthesis, this remains a putative explanation. The amount of \u00EF\u0081\u00A1-cellulose generally decreased under short-day conditions (in AtCesAirx stems, but not in wild-type), despite the fact that the total glucose content was equivalent between long and short-day plants, suggesting that glucose allocation may be influenced by the rate and extent of synthesis of the cell wall. Late-night cell wall biosynthesis has been suggested to occur in Arabidopsis seedlings (Harmer et al., 2000). The greater abundance of \u00EF\u0081\u00A1-cellulose produced in short-day wild-type plants corroborates night-time cellulose biosynthesis. Some diurnal regulation of secondary cell wall formation could also explain why the AtCesAirx plants exhibit a greater reduction in cellulose biosynthesis capacity under short day growth conditions. Day length conditions significantly affected cellulose microfibril angle (MFA) in AtCesAirx and wild-type plants. Under long-day conditions the AtCesA7irx3-1 and AtCesA4irx5-1 plants had significantly larger MFAs (20o) compared to wild-type and AtCesA8irx1-1 (14o). Under short-day conditions, however, all lines had a similar MFA (approximately 20o). Although it has been proposed that more rapid growth rates result in a larger MFA (Barnet and Bonham, 2004), this was not observed here as the opposite effect was observed: the wild-type and AtCesA8irx1-1 plants produced cellulose microfibrils aligned at a smaller angle to the longitudinal axis of the cell under long-day conditions, when they grow at a more rapid rate compared to short-day conditions. Correlations between MFA and growth rate have only been reported for various tree species, and thus the contrasting results of this study, compared to those reported in Barnett and Bonham (2004) for example, could be due to inherent differences in cellulose biosynthesis and cell wall properties between trees and Arabidopsis. Also, the differences in growth rate between long and short days may be very different from the growth rate differences suggested to affect MFA in trees. Further investigation to the variety of MFA in Arabidopsis, and the effect of growth rate on MFA in Arabidopsis, is required to fully understand why MFA increased from long to short days. It is interesting that the mutants and wild-type plants had similar MFA and crystallinity under short-days, suggesting that day-length muted, to some degree, the effects of the AtCesAirx mutations. Extending this idea, it may be inferred that rate of 75 cellulose biosynthesis has a very significant role on the cellulose properties, perhaps more significant than the AtCesA mutations in the case of MFA and crystallinity. Wasteneys and Fujita (2006) proposed that the rate of cellulose biosynthesis could affect DP and crystallinity, which would in turn affect microfibril function in cell expansion. Clearly there are some intriguing areas of research required to elucidate the role of day length and growth rate on cellulose biosynthesis and therefore cell wall structure and function. Cell wall crystallinity and cellulose DP in the AtCesAirx mutants also appeared to be affected by day length growth conditions as there was a general decreasing trend in DP for mutant plants. It could be that day length conditions effect the rate of cellulose biosynthesis, which manifests differences more strongly in the mutants due to reduced cellulose biosynthesizing capacity. In general, there was a trend of increased cell wall crystallinity in plants grown under short-day conditions. Under the assumption that cellulose biosynthesis occurs at a slower rate under short days, it could be that a slow rate of synthesis permits the formation of more crystalline structures, in particular cellulose. Cellulose DP was generally lower in all of the AtCesAirx mutants when grown under short-day conditions, compared to long-days. This suggests that CesA- composition of the CSC and the rate of cellulose biosynthesis could influence DP. Stunted growth may be the result of reduced longitudinal cell expansion, or fewer cell divisions, or both. The reasons for reduced primary growth as a secondary consequence of altered secondary cell wall biosynthesis are unknown. However, primary cell wall alterations in the CesA7mur10 mutant (Bosca et al., 2006) do suggest there is some developmental feedback occurring between primary cell walls and secondary cell walls. Further analysis revealed that short-day AtCesAirx plants have elevated growth rates compared to wild-type, such that they reproduce and mature when wild-type plants are still undergoing significant growth. This suggests that there could be secondary effects of the AtCesAirx mutations on developmental cues. To further elucidate the mechanism for reduced growth in AtCesAirx mutants it would be interesting to identify if fewer cell divisions occur or if the cells produced undergo reduced longitudinal expansion. 76 2.4.2 Increased lignin and hemicellulose content in AtCesAirx mutants suggest compensatory feedback in response to secondary cell wall perturbations Lignin values were calculated as a percentage of the total stem dry weight, and shown to be increased in all of the AtCesAirx plants. A portion of the increase observed could be a perceived increase due to reduced cellulose constituting the dry weight. Also, it could be the result of an increase in lignin biosynthesis as a compensatory mechanism to maintain structural integrity of the cell wall. In long-day AtCesA8irx1-1 plants there is very little change in lignin content compared to wild-type plants, despite the fact that there was a 37% decrease in \u00EF\u0081\u00A1-cellulose. In comparison, AtCesA4irx5-1 plants had a 25% reduction in \u00EF\u0081\u00A1-cellulose but a 22% increase in lignin content. This suggests that changes in lignin biosynthesis could be affected not only by a reduction in cellulose content (mass balance), but general perturbations to cell wall properties, as the AtCesA4irx5-1 lines exhibited more severe perturbations in the form of altered MFA. Although AtCesA8irx1-1 plants clearly have an \u00EF\u0081\u00A1-cellulose deficiency, the cellulose network seems to be more wild-type-like in terms of MFA, and these plants had a smaller increase in lignin compared to AtCesA7irx3-1 and AtCesA4irx5-1 plants. The differences between the mutant lines could be the result of stronger positive feedback for lignin biosynthesis in response to greatly perturbed secondary cell walls. Similar compensatory responses have been observed in other systems. For example, in lignin- deficient poplar it was suggested that there was an increase in cellulose (Hu et al., 1999). RNAi-suppression of korrigan in poplars caused a slight decrease in \u00EF\u0081\u00A1-cellulose (10% below wild-type) and a concomitant increase in lignin (Maloney and Mansfield, 2010). Interestingly, ectopic lignification in the eli mutants of AtCesA3 has been postulated as a defence response as a result of alterations to cell wall integrity (Cano- Delgado et al., 2003). However, a compensatory response is not always observed. In Arabidopsis irx4 mutant plants (mutation of a cinnamoyl-CoA reductase gene) that have perturbed lignin biosynthesis, there was no increase in cellulose or hemicellulose content of secondary cell walls (Jones et al., 2001). It could be that feedback in response to secondary cell wall alterations varies depending on 1) mutant phenotype 77 severity, 2) the gene(s) mutated and therefore biosynthetic pathways affected, and 3) the developmental stage when problems manifest. The hemicellulose network provides cross-linkages between cellulose microfibrils and lignin to create a strong network that imparts a great deal of structural integrity to the secondary cell wall (Cosgrove, 2005). A range of reports exist in the literature regarding the degree of change in hemicellulose content in AtCesA mutants including significant increases in the AtCesA7fra5 mutant (Zhong et al., 2003), whereas others report little to no change (Turner and Somerville, 1997; Brown et al., 2005). Herein we provide a more thorough report on the changes to hemicellulose content in stems from long- and short-day AtCesA8irx1-1, AtCesA7irx3-1, and AtCesA4irx5-1 plants. The proportion of total stem dry weight constituted by hemicelluloses did not significantly change between long and short-days. This is likely a reflection of reduced overall growth under short-days. The general increase in hemicellulose content in all AtCesAirx plants is believed to be a pleiotropic effect of the cellulose defects. Similar to the alterations in lignin content, an increase in hemicelluloses in response to decreased and altered cellulose could be a compensatory response. Given that xylans are very prominent hemicelluloses in plant secondary cell walls (Mellerowicz et al., 2001; Brown et al., 2009), it is not surprising that xylose levels showed some of the greatest increases. Secondary changes in cell wall composition, in response to cellulose deficiencies, have been reported for other mutants. The proscute1 mutant (mutation of AtCesA6), which also has dramatically reduced cellulose content, also exhibits a concomitant increase in hemicellulose content, and a slight change to pectin content (Fagard et al., 2000). Other non-CesA mutants with alterations to cellulose have exhibited changes to other cell wall components. For example, mutations of the Arabidopsis endo-1,4-\u00EF\u0081\u00A2-glucanse, korrigan, cause a decrease in cellulose content in primary cell walls (Sato et al., 2001; Peng et al., 2000) and crystalline cellulose in secondary cell walls (Szjanowicz et al., 2004) and secondary effects such as altered cell wall pectin content (Sato et al., 2001; Peng et al., 2000; His et al., 2001), minor changes to hemicellulose content (Lane et al., 2001; Peng et al., 2000), increased starch (Peng et al., 2000), and decreased structural integrity of the cellulose-xyloglucan network (Nicol et al., 1998). Changes to hemicelluloses and pectin content in korrigan mutants 78 have been attributed to a compensatory response as a result of altered cellulose (Peng et al., 2000). Widespread cell wall changes were also observed in AtCesA7mur10 mutant plants (Bosca et al., 2006). These plants had altered secondary cell walls, but also exhibited significant changes to the monosaccharide composition and xyloglucan fucosylation in primary walls of seedlings. The results presented in this chapter, and those from other reports, suggest that a compensatory increase in hemicellulose content is likely the result of a feedback response to cell wall perturbations to provide additional structural integrity to the cell walls of mutant plants. 2.4.3 Mutant-specific differences indicate the importance of a complete CSC and suggest CesA-specific functions in cellulose biosynthesis The non-redundant nature of AtCesA4, A7, and A8 (Taylor et al., 2008) could be because 1) the CSC will only form properly when three unique CesAs are present, 2) each CesA, as part of a CSC and rosette, contributes to the production of cellulose with unique properties, or both of these factors. Current literature, thus far, indirectly supports hypothesis one whereby it has been shown that CesA interactions are limited when one CesA is not functional (Taylor et al., 2003; Atanassov et al., 2009), which could lead to impaired CSC formation. Presented herein, the array of changes to cellulose content and ultrastructure among the AtCesAirx mutants also provide support for hypothesis one, and some support for hypothesis two. The differences between the mutant phenotypes may be explained, in part, by the type of mutation and how the mutations affect CSC form and function. The AtCesA7irx3-1 and AtCesA4irx5-1 mutants are thought to be null-mutants due to a truncation of the CesA protein, whereas AtCesA8irx1-1 is thought to be a reduced- function mutant due to an amino acid substitution at a conserved residue in the catalytic domain (Taylor et al., 2003; Figure 2.1, Table 2.1). The reduced function of AtCesA8irx1- 1 enzymes was indirectly confirmed by the low levels of \u00EF\u0081\u00A1-cellulose produced in AtCesA8irx1-1plants; levels that were similar to those of the null AtCesA7irx3-1 and AtCesA4irx5-1 mutants. Under the assumption that CesA4, A7, and A8 are the only CesAs contributing to secondary cell wall-specific cellulose biosynthesis, it stands to 79 reason that cellulose biosynthesis in AtCesA7irx3-1 plants is performed only by CesA4 and CesA8 subunits and only by CesA7 and CesA8 in AtCesA4irx5-1 plants. Therefore, it is hypothesized that aberrant and incomplete CSCs exist in the null mutant plants (Figure 2.20). In contrast, the CSC structure in AtCesA8irx1-plants may be more wild- type-like if the CesA8 subunit (a proportion of normal, or a modified form of the CesA8 subunit) is still synthesized and incorporated (Figure 2.20). As such, it was expected that there would be differences in the cell wall properties between the mutant lines, which could be attributed to the presence of a CSC or the unique functions of the remaining CesAs in each mutant background. Compared to the null mutants, the AtCesA8irx1-1 plants did not show a significant change in cellulose MFA, suggesting that a properly formed CSC could have a direct role in dictating MFA. The null mutants, AtCesA7irx3-1 and AtCesA4irx5-1, had similar alterations in MFA, however, the changes to other cellulose properties varied between these two mutant lines. In particular, the AtCesA7irx3-1 mutants had greatly reduced \u00EF\u0081\u00A1- cellulose content and the lowest crystallinity. In comparison, AtCesA4irx5-1 plants had near wild-type levels of crystallinity and more \u00EF\u0081\u00A1-cellulose than AtCesA7irx3-1 plants. Combined, these results suggest that the presence of a functional AtCesA7 subunit may have a greater influence on total cellulose production and crystallinity than AtCesA4. 80 Figure 2.20. Hypothetical effects of AtCesA mutations on CSC formation and cellulose properties. Based on the alterations to cellulose content, MFA, crystallinity, and DP, the effect of each CesA mutation on CSC form and function was inferred. Predicted CSC structures in (A) wild-type plants, (B) AtCesA8irx1-1 plants, (C) AtCesA7irx3-1 plants, and (D) AtCesA4irx5-1 plants are presented. It is proposed that a fully-formed CSC provides optimal conditions for potential interactions with other factors that regulate MFA (A, B). In null-mutant plants, it is assumed that the remaining functional CesAs form aberrant CSCs (C, D), and it is proposed that CesA-specific functions influence CSC function and ultimately manifest altered cellulose properties (C and D). For example, perhaps AtCesA7 facilitates better CSC formation and higher crystallinity (C), and AtCesA4 creates cellulose chains of greater length (D). A B C D Wild-type A8 irx1-1 : Less cellulose A7 irx3-1 : Less cellulose, less crystalline, larger MFA A4 irx5-1 : Less cellulose, lower DP, larger MFA 81 Perhaps the aberrant CSCs formed in the AtCesA7irx3-1 mutant plants do not provide a suitable environment for synthesized glucan chains to form highly crystalline bonds (Figure 2.20 C). Also, these results could suggest that AtCesA7 is required at higher levels to form CSCs, such that the presence of CesA7 and CesA8 in AtCesA4irx5- 1 plants permits the formation of more stable and more CesA-rich CSCs (compared to the CesA4-CesA8 CSCs in AtCesA7irx3-1 plants) which are able to synthesize greater amounts of cellulose (Figure 2.20 C, D). Although these are inferences about CSC form and function, the results presented herein suggest that CesA4, A7, and A8 subunits may be required at different levels in a CSC. To date, there have been no successful experiments to determine the ratio of CesA4:CesA7:CesA8 proteins within active CSCs. However, gene expression analysis suggests that AtCesA7 and AtCesA8 are more highly expressed than AtCesA4 (Appendix A). Perhaps CesA7 has a unique role compared to CesA4 and A8, such as acting as the initiation site for CSC formation. Zhong et al. (2003) found that the expression of a mutant form of AtCesA7 (fra5) caused a dominant negative effect, whereas expression of mutant form of CesA8 (fra6) did not, suggesting that CesA7 has a larger influence on cellulose biosynthesis. Anti-CesA antibody probing suggests near equal amounts of each CesA in stem total protein extracts (Atanassov et al., 2009). However, (what are presumed to be) the same antibodies used on stem tissue prints suggest that there are higher levels of AtCesA4 in the xylem compared to AtCesA8 and AtCesA7 (Taylor et al., 2003). Additionally, it appears as though more AtCesA4 and AtCesA8 are present in interfascicular cells compared to AtCesA7. These results must be interpreted with caution, as the visual differences could be the result of differences in antibody efficiencies, the region from which stem sections were taken, or potential masking of the epitope by CesA-protein interactions. As mentioned previously, another putative reason for the requirement of all three CesA proteins (AtCesA4, A7, and A8) for proper cellulose biosynthesis could be that each CesA has a unique function that influences the final cellulose properties. For example, one proposed possibility is that one (or two) CesAs have specific roles in using a cellulose biosynthesis primer, sitosterol-\u00C3\u009F-glucoside, such that these primer- utilizing CesAs are required for proper cellulose biosynthesis by the other CesA 82 subunits (Doblin et al., 2002). However, if the remaining CesA subunits do not readily form hetero-CesA complexes in the mutant plants (Taylor et al., 2003; Atanassov et al., 2009), and priming of biosynthesis is required by one of the CesA subunits in a fully formed CSC, we would expect to see no cellulose produced, but this is not the case. From the results obtained herein, it appears that AtCesA7 is required for a more crystalline cell wall to be produced. This could be the result of CesA7 being required for proper CSC formation (as discussed above), or it could be an indication of an AtCesA7- specific role in producing higher ordered crystalline cellulose. Under long-day conditions, the cellulose DP of AtCesA4irx5-1 mutants was the lowest of all the mutant lines. This may indicate that AtCesA4 has a unique role in producing long chains of cellulose, a role that may not be shared by AtCesA8 or AtCesA7. Combined, these results could indicate that 1) a fully formed CSC influences MFA; 2) AtCesA7 may be required at a greater level for proper CSC formation; 3) CSCs containing AtCesA7-AtCesA8 are stable than AtCesA8-AtCesA4, allowing for increased crystalline cellulose to be produced; 4) AtCesA4 may have a role in producing cellulose chains with a greater DP than those of AtCesA8 or AtCesA7. 2.4.4 Reporter tags may effect CesA function Complementation of the null AtCesA4irx5-1 and AtCesA7irx3-1 mutants was achieved by expression of the missing wild-type-CesA under expressional control of the native AtCesA promoter, but not by the 35S promoter. This emphasizes the requirement for appropriate spatial and temporal expression of AtCesA4, A7, and A8 for proper cellulose biosynthesis. In AtCesA4irx5-1 plants, very low expression of LUC- AtCesA4 (by PAtCesA4::LUC-AtCesA4) resulted in an almost fully-complemented phenotype. This result suggests that the Luciferase tag does not significantly interfere with AtCesA4 function in these plants, and it is interesting that a low threshold level of AtCesA4 transcript, and presumably protein, is required to facilitate normal cellulose biosynthesis and growth. In comparison, expression of YFP-AtCesA7 in A7irx3-1 plants resulted in a complemented phenotype, but only when expression levels were excessive compared to endogenous AtCesA7 transcript levels. In the non-null mutant AtCesA8irx1-1, expression of LUC-AtCesA8 under the native AtCesA8 promoter was not 83 sufficient to fully recover a wild-type phenotype as plants remained significantly stunted, but exhibited some recovery from the collapsed xylem phenotype. This is likely due to the continued presence of the mutant form of AtCesA8 in the AtCesA8irx1-1 background, and that the mutant AtCesA8 protein likely successfully out-competes the LUC-AtCesA8 protein for positions within the CSC. LUC-AtCesA8 transcript abundance was relatively low compared to wild-type. Full complementation of the mutant phenotype can likely only be achieved by excessive expression of the wild-type AtCesA8. Atanassov et al. (2009) reported that an N-terminal tag on AtCesA4 and AtCesA8 resulted in a non- functional protein, based on lack of mutant phenotype complementation by expression of a STREP-AtCesA4 or STREP-AtCesA8 fusion in the A4irx5-1 and A8irx1-1 mutants, respectively. Interestingly, there are multiple reports of successful complementation of the AtCesA7irx3-1 mutant tagged-CesA7, including STREP-AtCesA7 (Atanassov et al., 2009), GFP-AtCesA7 (Gardiner et al., 2003), and a small epitope-tagged CesA7 (Taylor et al., 2004), suggesting that tagged AtCesA7 subunits function normally. Furthermore, this highlights the fact that the AtCesA7irx3-1 mutant appears to be more easily complemented compared to the AtCesA4 and A8 mutants. If all three of AtCesA4, A7, and A8 are required in the CSC, then why do some tagged-CesA subunits seem more functional than others? Could these disparities in ease of complementation suggest that the CesA subunits have different roles within the CSC, such that tags may interfere with the function of AtCesA4 and A8 but not AtCesA7? The role of transgene expression levels on complementation cannot be ignored, however, the results presented herein, and those of others (Atanassov et al., 2009; Gardiner et al., 2003; Taylor et al., 2004), suggest that there may be some differences in the inherent role of each CesA in a CSC, such that complementation varies between CesAs when a tag is added to the N-terminus. These differences could be due to 1) CesA-specific functions, such that tags interfere more or less with CesA function; 2) CesA-CesA interactions, such that each CesA has a different ratio within a CSC, and these ratios are dependent on interactions which could be altered by an N-terminal tag; and 3) non-CesA interactions. Perhaps some of the CesAs within a CSC interact with other proteins during biosynthesis and a tag would interfere with these interactions thus making complementation more or less difficult. As the role of each CesA in CSC formation, 84 function, and the ratios of CesAs within a CSC are elucidated it may become clear why there are differences in complementation ability. 2.4.5 Exogenous expression of AtCesAs in wild-type plants causes a severe mutant phenotype The expression of two AtCesA transgenes in wild-type Columbia Arabidopsis plants resulted in a severe AtCesAirx phenotype (Figure 2.12). It is postulated that the mutant phenotype observed is the result of sense suppression, which was supported by reduced AtCesA transcript abundance in RNA from stem tissue of the mutant plants. Interestingly, expression of AtCesA8 (wild-type Columbia cDNA) in wild-type Landsberg Arabidopsis plants (WTL-2XA8) did not cause the same severe mutant phenotype. In fact, the WTL-2xA8 plants looked similar to wild-type plants, although stems were thinner and slightly less upright. Additionally, sense suppression did not appear to occur, as AtCesA8 transcript remained at near wild-type levels. It has been noted by others that AtCesA mutants in the Arabidopsis Columbia ecotype appear more severe than those in the Landsberg ecotype (Taylor et al., 2008). It was postulated that the reduced severity of AtCesAirx mutations in Landsberg plants is potentially the result of reduced signalling, and therefore feedback, in response to cell wall changes in Landsberg compared to Columbia (Taylor et al., 2008). 2.4.6 YFP-CesA7 expression patterns reveal a strong presence of AtCesA7 in fibres AtCesA7 is believed to be strongly associated with secondary cell wall development (Turner and Somerville, 1997; Taylor et al., 2003; Zhong et al., 2003; Brown et al., 2005), although it has also been thought to also affect (directly or indirectly) primary cell wall formation (Bosca et al., 2006). Additionally, phylogenetic comparisons of the secondary cell wall AtCesAs suggests that AtCesA7 is less divergent than AtCesA4 and AtCesA8 (Nairn and Haselkorn, 2005; Roberts and Bushoven, 2007; McDonnell, this study and Chapter 3). A visual profile of YFP-CesA7 expression in Arabidopsis plants, from seedlings to mature stem tissue, suggests that AtCesA7 is expressed in almost all tissue types, at a variety of developmental stages. 85 Additionally, our results support a potential involvement of AtCesA7 in primary growth, and that AtCesA7 (or entire AtCesA7-containing CSCs) are frequently compartmentalized. The presence of large, fast moving, CesA-containing compartments have been previously reported for YFP-CesA6 (Gutierrez et al., 2009; Paredez et al., 2006) and GFP-CesA3 (Crowell et al., 2009) (reviewed in Wightman and Turner, 2010) and YFP- CesA7 (Wightman and Turner, 2008). Co-localization of GFP-CesA3, YFP-CesA6, and YFP-CesA7 particles with Golgi-specific markers identified the ring structures (GFP- CesA3, YFP-CesA6) and large YFP-CesA7 puncta as Golgi-distinct compartments. Using drug treatments, it was determined that positioning of CesA-Golgi bodies near the plasma membrane is influenced by the microtubule array (Crowell et al., 2009), and their movement appears to be regulated somewhat by actin networks (Crowell et al., 2009; Wightman and Turner, 2008). As in these previous reports, the putative CesA- Golgi bodies observed herein (YFP-CesA7) were fast moving and often moved in non- linear trajectories (casual observation). As well, at times it would appear as though a compartment would pause within the cell (or at the periphery) and then continue moving, or disappear. Erratic movement and pausing of large CesA puncta was also observed by Wightman and Turner (2008) who postulated the pause events represented delivery of CSCs from Golgi compartments to the plasma membrane under sites of secondary cell wall thickening. The periphery-limited location of the YFP signal in the ring-shaped puncta suggests that the YFP-CesAs could be within the membrane of the compartment. Putative CesA-Golgi bodies observed herein were in most tissue types, although very rarely observed in the root tip. Large and small, with respect to the CesA-Golgi bodies observed, bright YFP-CesA7 particles were also observed in all cell types documented, but they did not have the same ring-like structure. Thus, the two groups of signal could represent different compartments, or different states of YFP- CesA7 within a compartment. The non-Golgi YFP-CesA7 compartments observed could be part of the MASC/SmaCC population of vesicles observed by Crowell et al. (2009) and Gutierrez et al. (2009). These putative CSC-containing vesicles were observed in the cytosol, are approximately 300 to 500 nm in diameter (personal estimation, based on images presented in the literature) and are believed to be involved 86 in regulating the distribution, and perhaps internalization of CSCs to and from the plasma membrane (Crowell et al., 2009; Gutierrez et al., 2009; Wightman and Turner, 2010). It is not clear if these compartments are completely distinct from the ring-shaped Golgi-CesA compartments, as some SmaCCs (and perhaps MASCs) have co-localized with a trans-Golgi network marker (Gutierrez et al., 2009). Some of the non-ring- shaped puncta observed in the research presented herein range in size from approximately 300 to 1000 nm in diameter, placing some of these puncta in a similar size range as SmaCCs and MASCs. Putative SmaCCs-MASCs observed herein were present in almost all tissues observed, therefore it cannot be speculated if they have a role in a particular developmental stage or tissue type. It is feasible that they are involved in regulating CSC delivery or internalization to recycle CSCs or reduce cellulose biosynthesis. However, further investigation is required to truly elucidate the role of SmaCCs and MASCs in cellulose biosynthesis. The most commonly observed puncta were: cytosolic, 300 to 1000 nm, bright, non-ring shaped puncta in all tissues (MASCs-SmaCCs); cytosolic, large, ring-shaped puncta in all tissues (CesA-Golgi bodies), and very small, bright puncta at the plasma membrane of fibres. The variety of YFP-CesA7 signal observed likely represents a dynamic population of active and non- active CSCs being transported to and from the plasma membrane. .A single ring-like structure was observed in YFP-CesA8 expressing plants (wild-type Landsberg expressing PAtCesA8::YFP-CesA8) alongside developing xylem. The reduced frequency of YFP-CesA8-Golgi bodies is likely due to the expression of YFP-CesA8 in a wild-type background. On the plasma membrane face of many fibres, presumed to be undergoing secondary cell wall thickening, the YFP-CesA7 signal was often observed as a layer of small, dense, diffuse particles that appeared to be moving very quickly (although the trajectories and speed could not be recorded). The particles were quite small and diffuse, and resembled the small puncta observed by others, which have been described as CSCs in the plasma membrane (Paredez et al., 2006; Crowell et al., 2009; Gutierrez et al., 2009; Wightman and Turner, 2008) . Fibres with dense, bright, diffuse YFP-CesA7 signal were most often found within the central region of a growing stem, believed to be where interfascicular fibres are maturing and developing thickened 87 secondary cell walls. To our knowledge, this is the first report of a secondary cell wall signal visualized in developing fibres, in planta. In contrast, at the base or tip of the stem, the plasma membrane of fibres was not found to be saturated with similar fine YFP-CesA7 puncta. The density and rapidly moving signal in developing fibres reiterates the fact that an extremely high abundance of CesA protein is likely to be involved in cellulose biosynthesis, that the production of secondary cell wall material is rapid, and that there is an extremely high turnover rate of CesA proteins. It is not known if CesA subunits behave similarly in xylem and fibres. Are the ratios of CesA subunits within a CSC similar in all cell types? Observing the significant abundance of YFP- CesA7 signal in fibres emphasizes that CesA7 has a prominent role in fibre secondary cell wall biosynthesis. The characterization of AtCesA8irx1-1, AtCesA7irx3-1, and AtCesA4irx5-1 growth and cell wall composition has revealed much new information with respect to cellulose biosynthesis. The results presented herein suggest that CSC formation and composition affect cellulose properties such as MFA, DP, and crystallinity, and that day- length further affects MFA in Arabidopsis. There appear to be CesA-specific roles in cellulose biosynthesis which can be detected by measuring changes to cell wall and cellulose ultrastructure. 88 CHAPTER 3 Identification of secondary cell wall-specific cellulose synthase genes from Picea glauca and conservation of function among CesA orthologs 3.1 Introduction Cellulose, a \u00EF\u0081\u00A2-1,4 linked homopolymer comprised of glucose, represents as much as 40% of the plant secondary cell walls (Barnett and Bonham, 2004; Mellerowicz et al, 2001). The genes coding for the enzymes responsible for cellulose biosynthesis in secondary cell walls, cellulose synthases (CesAs) have been studied in many plant species including Arabidopsis (Hamman et al., 2004), Populus (Djerbi et al., 2005; Kumar et al., 2009), Eucalyptus (Lu et al., 2008; Ranik and Myburg, 2006), rice (Tanaka et al., 2003), maize (Holland et al., 2000; Appenzeller et al., 2004), barley (Burton et al., 2004), potato (Oomen et al., 2004), and moss (Roberts and Bushoven, 2007) to name a few. However, comparatively little is known about CesAs from gymnosperm species. To date, there are only a few papers reporting the identification of CesAs in pine species (Nairn and Haselkorn, 2005; Krauskopf et al., 2005; Nairn et al., 2008). Thus, there is a pressing need to further identify CesAs, particularly those involved in secondary cell wall formation, in other economically important gymnosperms, such as spruce. CesA proteins have a conserved structure containing a zinc-binding domain, eight transmembrane domains (two located at the amino terminus and six at the carboxyl terminus), a plant-conserved region, a class-specific region (CSR), and two highly conserved domains, A and B (Richmond, 2000; Doblin et al., 2002), and conserved motifs believed to be involved in the catalytic activity of CesAs: three aspartic acid residues (D) and a QXXRW motif. Evolutionarily, CesA gene families appear to have emerged prior to the divergence of gymnosperms from angiosperms. The cotton (Gossypium hirsutum) cellulose synthase 1 (GhCesA1) and GhCesA2 were some of the first CesA genes to be identified in higher plants (Pear et al, 1996). With the advancement of genome and EST-library sequencing, large CesA families have emerged from other plants. For example, up to 10 CesA and 30 cellulose synthase-like (Csl) genes have been identified in Arabidopsis (Somerville et al., 2000), 17 CesA genes in poplar (Kumar et al., 2009), at least six in Eucalyptus (Ranik and Myburg, 89 2006), and 11 CesAs in moss (Roberts and Bushoven, 2007). Phylogenetic analysis of plant CesAs reveals a distinct separation of primary and secondary cell wall-specific CesAs (Roberts and Bushoven, 2007; Kumar et al., 2009). Additionally, for secondary cell wall cellulose biosynthesis, it has been postulated that three unique CesAs are required to produce cellulose, thus forming proper secondary cell walls. This has been clearly illustrated in Arabidopsis mutants (Turner and Somerville, 1997; Taylor et al., 2000; Taylor et al., 2003). Sequence homology suggests the same requirements exist in Populus (Djerbi et al., 2005), Eucalyptus (Ranik and Myburg, 2006), and Pinus taeda (Nairn and Haselkorn, 2005). Predicted CesA amino acid sequences, particularly the class-specific regions (CSRs) of CesA sequences, are highly similar among putative orthologs, but less similar between unique CesA family members (Ranik and Myburg, 2006; Joshi et al, 2004; Nairn and Haselkhorn, 2004; Samuga and Joshi, 2002). The CSR may dictate differences in function among CesAs within a plant, and has recently been used to aid in classification. Although orthology of CesAs has been inferred based on phylogenetic analysis, there is no evidence to suggest that CesAs from different plant species are functionally orthologous. Examining the conservation of function among CesA orthologs will provide information critical to understanding the evolution of cellulose synthases, and the specific functions of individual CesA family members in cell wall biosynthesis. The objectives of this study were three fold. The first was to identify the secondary cell wall-specific cellulose synthase genes from spruce, Picea glauca. Second, to characterize the identified genes based on phylogenetic analysis and gene expression profiling. Third, we wanted to examine the degree of conservation of function among CesA orthologs by assessing the ability of spruce and poplar CesAs to complement the Arabidopsis CesAirx mutants. 3.2 Materials and methods 3.2.1 Isolation of full-length CesA cDNA A spruce EST and contig database (Genome BC) provided short sequences from the 3\u00EF\u0082\u00A2 end of putative CesA cDNAs (based on sequence similarity to known CesAs). 90 These were identified as PgCesA1, PgCesA2, and PgCesA3. First-strand cDNA was synthesized from total RNA harvested from spruce xylem tissues. To amplify the full- length coding region of PgCesA1 5\u00EF\u0082\u00A2 RACE (Rapid Amplification of cDNA Ends, FirstChoice RLM RACE kit, Ambion) was used in combination with a primer specific for the known 3\u00EF\u0082\u00A2 end. To amplify the full-length coding region of PgCesA2 and PgCesA3, PCR was used with reverse primers specific for the known 3\u00EF\u0082\u00A2 end and forward primers specific for the 5\u00EF\u0082\u00A2 end of known pine cellulose synthase sequences (PitCesA2, AY89650.1; PitCesA3, AY789652.1). The three full-length cDNAs were analyzed by DNA sequence analysis. Primer oligonucleotides are shown in Table 3.1. Populus trichocarpa CesA gene sequences with homology to Arabidopsis AtCesA4, AtCesA7, and AtCesA8 described by Kumar et al. (2009) were retrieved from the JGI website (http://genome.jgi-psf.org/Poptr1_1.home.html). The following genes models were used as search terms to retrieve the sequences from JGI: PtiCesA4 \u00E2\u0080\u0093 eugene3.00002636; PtiCesA8-A \u00E2\u0080\u0093 gw1.XI.3218.1; and PtiCesA7-A \u00E2\u0080\u0093 estExt_Genewise1_v1.C_LG_V12188. The full-length PtiCesA cDNAs were amplified and cloned using PCR (primers listed in Table 3.1) and ligated into the Gateway compatible pENTR-D/TOPO donor cloning vector (Invitrogen) to create pE-PtiCesA4 (PtiA4), pE-PtiCesA7A, pE-PtiCesA7B, pE-PtCesiA8A, pE-PtCesiA8B. Clones were sequenced to confirm identity and accuracy. 3.2.2 RNA extraction and cDNA synthesis for gene expression Tissues from 5-year-old Picea glauca trees grown in Vancouver, British Columbia, in pots, outside, were harvested in early May. Specifically, young- unexpanded needles, young- expanded needles, phloem/cambium, xylem, and roots, were collected and immediately frozen in liquid nitrogen. Total RNA was extracted using methods described by Kolosova et al. (2004). RNA was treated with TURBO DNase (Ambion) to remove DNA. 91 Table 3.1. Sequences of primers used for isolation, cloning, real-time PCR (RT), and screening. Primer Name Sequence (5' - 3') Use PgA1RV3\u00E2\u0080\u0099 TAGCGATGGGAACCGACACTGA Isolation PgA2RV3\u00E2\u0080\u0099 TGGTAGCTTGGAGACGCCCTTGA Isolation PgA3RV3\u00E2\u0080\u0099 GTGGGCGTCGCTCCTTTTGTTAC Isolation PitCesA2 FW AGTCACTAGTATGGAGGCCAAGGCGGGACTTGTTGCA Isolation PitCesA3 FW TTGAACTTAACAATGGAAGCCAGCGC Isolation PgCesA1HVRIIFW TGCTGTTGCTGCTGTTGTTGTCCG RT PgCesA1HVRIIRV AACACCGAGGACTGGCCAAAGACT RT PgCesA2HVRIIFW TCAACATGAAGGGTCTGGATGGCA RT PgCesA2HVRIIRV AACCGCCGAAACAGCAACACCAT RT PgCesA3HVRIIFW TATGGGTATGGGCCTCCCAAAGGC RT PgCesA3HVRIIRV TCCAGATTGAAGGCAGGAGCTGGG RT PgeIF5a-1FW GTGCCATCTTCACACAACTGC RT PgeIF5a-1RV CAGATTCAGTCAGCAGGCTAAC RT PgCesA1FW CACCGCTTCCAACGGCAATATGAA Cloning PgCesA1RV GTACCGATGAAACATTTTATAGCACC Cloning PgCesA2FW CACCGAGGCCAAGGCGGGACTTGTT Cloning PgCesA2RV GGCGCGCCCACCCTTCTAGCAGTCAA Cloning PgCesA3FW CACCGAAGCCAGCGCCGGCTTGGT Cloning PgCesA3RV TTCTTTGATATTCAAGtTTCAaCCTT Cloning PtiCesA4FW CACCATGGCTGGCCTTGTCACGGGCAGT Cloning PtiCEsA4RV AATCTACCATGTTTGGTTGACTTT Cloning PtiCesA7AFW CACCATGGAAGCCAGTGCTGGACTTGTC Cloning PtiCEsA7ARV TTTAACAGTTGATTCCACATTGCTT Cloning PtiCesA8AFW CACCATGATGGAATCTGGGGCTCCTTTAT Cloning PtiCEsA8ARV GCAATCTATAGAAATGCAGGTTTCA Cloning A4HVRIIFW TGACATGTGATTGTTGGCCGTCGT RT A4HVRIIRV AATCGCCTCCGTCGATGATCGTTT RT A7HVRIIFW ACATGAATGGTGACGTAGCAGCCCTT RT A7HVRIIRV ACCGCAGCTTATGACATGGATTGCCT RT A8HVRIIFW GCAAAGCGAGAAGAACTTGATGCTGC RT A8HVRIIRV TTTACAGAGTCGGGAACACCGCCATT RT AtUBQ5FW ACACCAAGCCGAAGAAGATCAAGCAC RT AtUBQ5RV AAATGACTCGCCATGAAAGTCCCAGC RT 92 First-strand cDNA was synthesized from 1 \u00EF\u0081\u00ADg of DNAse-treated RNA using Superscript II Reverse Transcriptase (Invitrogen) with dT18 oligonucleotides. 3.2.3 Real-time PCR analysis PgCesA gene expression was measured using quantitative real-time PCR (RT- PCR). RT-PCR reactions were set-up in triplicate for each sample with Platinum SYBR Green qPCR Master Mix (Invitrogen) and run on a Mx3000p real-time PCR system (Stratagene). The primers used to detect PgCesA and housekeeping control genes are listed in Table 3.1. Conditions for the RT-PCR reactions were 95oC for 10 minutes followed by 35 cycles of 95oC for 30 seconds, 59oC for 1 minute, and 72oC for 30 seconds. Relative gene expression was calculated using the equation described by Pfaffl (2001), \u00EF\u0081\u0084ctrepresent2-(ctCesA-ctEIF5\u00EF\u0081\u00A1), in which CesA gene expression is relative to that of the housekeeping control, PgeIF5\u00EF\u0081\u00A1-1. Melting curve analysis, agarose gel electrophoresis, and DNA sequencing of the RT-PCR products were used to confirm that the RT-PCR amplicons represented the expected RT-PCR products. 3.2.4 Phylogenetic analysis Several cellulose synthase amino acid sequences were obtained from GenBank (NCBI) and compared to the predicted PgCesA amino acid sequences (Table 3.1). Amino acid alignments and subsequent dendrogram were created using the MEGA software (http://www.megasoftware.net/). For the box-shade figure, a sequence alignment file created using ClustalW (http://align.genome.jp/) was submitted to the BoxSHADE server (http://www.ch.embnet.org/software/BOX_form.html) to create the shaded alignment image. 93 Table 3.2. A list of CesA sequences used for phylogenetic analysis and associated GenBank accession numbers. When available, relevant literature sources are listed. Species CesA Accession # Reference Arabidopsis thaliana AtCesA1 O48946 Arioli et al., 1998 AtCesA2 O48947 Arioli et al., 1998 AtCesA3 NP_196136 Wang et al., 2006 AtCesA4 NP_199216 Taylor et al., 2003 AtCesA5 NM_121024 Scheible et al., 2004 AtCesA6 NM_125870 MacKinnon et al., 2006 AtCesA7 NP_197244 Taylor et al., 1999 AtCesA8 NP_567564 Turner and Somerville, 1997 AtCesA9 Q9SJ22 Beeckman et al., 2002 AtCesA10 Q9SKJ5 Beeckman et al., 2002 Gossipium hirsutum GhCesA1 ABG06122 Pear et al., 1996 GhCesA4 AAL37718 Kim and Triplett, 2001 Oryza sativa OsCesA1 AAU44296 OsCesA2 A2XN66 OsCesA3 BAD30574 OsCesA4 A2WV32 Tanaka et al., 2003 OsCesA5 A2XNT2 OsCesA6 Q6YVM4 OsCesA7 Q9AV71 Tanaka et al., 2003 OsCesA8 Q84ZN6 OsCesA9 A2Z1C8 Tanaka et al., 2003 Pinus radiata PirCesA1 AAT57672.1 For all: Krauskopf et al., 2005 PirCesA2 AAQ63936.1 PirCesA10 AAQ63935.1 Pinus taeda PitCesA1 AY789650.1 For all: Nairn and Haselkorn, 2005 PitCesA2 AY789651.1 PitCesA3 AY789652.1 Populus trichocarpa PtiCesA1-A For all: Djerbi et al., 2005, and Kumar et al., 2009 PtiCesA3-A PtiCesA3-C PtiCesA4 PtiCesA7-A PtiCesA7-B PtiCesA8-A PtiCesA8-B PtiCesA6-A PtiCesA6-D 94 Species CesA Accession # Reference Populus tremuloides PtdCesA1 Wu et al., 2000 PtdCesA2 Samuga and Joshi, 2002 PtdCesA3 Joshi, 2003 Eucalyptus grandis EgCesA1 AAY60843.1 For all: Ranik and Myburg, 2006 EgCesA2 AYY60844.1 EgCesA3 AAY60845.1 Physcomitrella patens PpCesA7 ABI78960 For all: Roberts and Bushoven, 2007 PpCesA6 ABI78959 PpCesA4 XP_001767133 PpCesA8 XP_001769255 95 3.2.5 Plant transformation vectors For expression of PgCesA cDNAs in various AtCesAirx mutants, the PgCesA cDNA fragments were cloned into the P35S::LUC-attR or the P35S::YFP-attR expression constructs (Subramanian et al., 2006) with either the 35S promoter (P35S) intact, or with the 35S promoter replaced by Arabidopsis AtCesA promoters (detailed below). The placement of PgCesAs into these vectors resulted in a fusion of Luciferase (LUC) or yellow fluorescent protein (YFP) at the N-terminus of the cloned gene. For AtCesA promoter driven expression, PgCesA cDNAs were cloned into one of the following binary vectors (described in Chapter 2): PAtCesA4::Luc-attR, PAtCesA7::YFP- attR, or PAtCesA8::LUC-attR. The full length PgCesA1, PgCesA2, PgCesA3, PtiCesA4, PtiCesA7-A, and PtiCesA8-A cDNAs were amplified by PCR (primers are listed in Table 3.1) and ligated into the pENTR-D cloning vector (Invitrogen). Using LR Clonase II (Invitrogen), PgCesA cDNAs were transferred to the desired destination expression vector to create the following binary vectors: P35S::LUC-PgCesA1, PAtCesA8::LUC- PgCesA1, P35S::LUC-PgCesA3, PAtCesA4::YFP-PgCesA3, PAtCesA7::YFP-PgCesA3, PAtCesA7::YFP-PgCesA2, PAtCesA8::LUC-PtiCesA8-A, PAtCesA4::LUC--PtiCesA4, and PAtCesA7::LUC-PtiCesA7-A. 3.2.6 Plant growth and transformations AtCesA8irx1-1 (Turner and Somerville, 1997), AtCesA7irx3-1 (Turner and Somerville, 1997), and AtCesA4irx5-1 (Taylor et al., 2003) Arabidopsis plants were germinated on half-concentration MS medium (Mirashige and Skoog, 1962) with no sucrose, and grown under a 16-hour light/8-hour dark cycle. Seedlings were transferred to soil approximately seven days post-germination, and grown in a growth chamber at 21oC under a 16-hour light/8-hour dark cycle. Plants were transformed using a method modified from Clough and Bent (1998). Agrobacterium tumefaciens GV3101-pMP90 (Hellens et al., 2000) was transformed with the binary vector using a freeze-thaw method. Cultures were grown overnight at 28oC in Luria-Bertani medium containing 50 \u00EF\u0081\u00ADg L-1 kanamycin (YFP vectors) or 50 \u00EF\u0081\u00ADg L-1 spectinomycin (Luc vectors), 25 \u00EF\u0081\u00ADg L-1 rifampicin and 25 \u00EF\u0081\u00ADg L-1 gentamycin. The overnight culture was centrifuged and resuspended in a 5% (w/v) sucrose solution to an OD600 of at least 0.8. Silwet L-77 96 (LEHLE Seeds) was added to each resuspended culture at a final concentration of 0.02% (v/v). Newly flowering plants (approximately 4 weeks old) were sprayed with the Agrobacterium solution using a fine mist spray nozzle. Sprayed plants were placed in a dark, humid environment for 16 to 24 hours and returned to light. A second spraying was done 5 days after the first spray in attempts to improve the transformation efficiency. After all spray treatments, plants were maintained as previously described, and seeds were harvested from mature, dried plants. Seeds were screened for transformations by germinating on half-concentration MS medium (no sucrose) with the addition of either 75 \u00EF\u0081\u00ADg L-1 kanamycin (YFP vectors), 50 \u00EF\u0081\u00ADg L-1 glufosinate ammonium sulfate (Luc vectors), or 25 \u00EF\u0081\u00ADg L-1 hygromycin (PAtCesA4::YFP vectors). Seedlings that successfully grew on antibiotics were grown in soil for further analysis. Transformed plants were confirmed by PCR of genomic DNA using gene specific primers (Table 3.1) 3.2.7 Complementation assay In an attempt to investigate conserved functional homology between spruce and Arabidopsis cellulose synthases and poplar and Arabidopsis cellulose synthases the ability of PgCesAs and PtiCesAs to rescue an Arabidopsis CesA mutant phenotype was assessed. Homozygous AtCesAirx mutant plants expressing a PgCesA or PtiCesA were grown in soil under long-day conditions (16-h light/8-h dark). Transgene expression was measured in RNA extracted from stem tissues of approximately 20-day-old plants. To assess xylem morphology, stem bases (approximately 1cm from soil) from 20-30 day old plants were hand-sectioned using a double-edged razor blade. Sections were stained for 5 minutes in a 0.25% (w/v) Toluidine Blue solution, and rinsed for 5 minutes in water. Hand-sections were viewed through a Leica Light Microscope, and pictures taken with a Q-imaging camera. 97 3.3 Results 3.3.2 Spruce CesA gene isolation and sequence analysis Three putative secondary cell wall-specific cellulose synthase cDNAs were isolated from reverse transcribed xylem RNA of white spruce, Picea glauca: PgCesA1, A2, and A3. For PgCesA1 and A3, approximately 1 kb from the 3\u00EF\u0082\u00A2 end of the final 3.2 kb cDNA was available from the Genome BC EST library. PgCesA1 was isolated using primers designed against the known 3\u00EF\u0082\u00A2 region and 5\u00EF\u0082\u00A2 RACE PCR. In contrast, several rounds of 5\u00EF\u0082\u00A2 RACE PCRs to isolate the PgCesA2 and PgCesA3 cDNA were unsuccessful. PCR performed with a forward primer designed to amplify the 5\u00EF\u0082\u00A2 region of a pine CesA (Pinus taeda, PitCesA3 or PitCesA2) in combination with a reverse primer designed against the known 3\u00EF\u0082\u00A2 end of PgCesA2 or PgCesA3 was employed to isolate the full length PgCesA2 and A3 cDNAs. All three genes were isolated from xylem RNA, whereas it was difficult to isolate the sequences from RNA extracted from leader tissue. This suggested that the three isolated CesAs are more highly expressed in xylem tissues. At the nucleotide level, the maximum identity shared between the three identified PgCesAs is between 59 and 72% (Table 3.3) which is comparable to the level of similarity between unique secondary cell-wall specific CesAs in other plants, for example AtCesA4-AtCesA8: 58%, AtCesA8-AtCesA7: 56%, and AtCesA4-AtCesA7: 65%. A comparison of the nucleotide sequences of PgCesA1, A2, and A3 cDNA sequences with those of other known CesAs by a standard BLASTN search revealed a strong similarity (68 to 95%) to secondary cell-wall specific CesAs from a variety of plants including Arabidopsis, Populus, and Pinus. Among the highly studied Arabidopsis secondary cell wall-specific CesAs the highest similarities were found between PgCesA1 and AtCesA8 (69%), PgCesA2 and AtCesA7 (69%), and PgCesA3 with AtCesA7 (74%). As expected, the shared identity between the spruce CesAs and those from another gymnosperm, Pinus taeda, was high: 93% between PgCesA1 and PitCesA1, 93% between PgCesA2 and PitCesA2, and 95% between PgCesA3 and PitCesA3. 98 The predicted amino acid sequences of the three putative PgCesAs were aligned to determine the level of consensus and to identify conserved domains (Figure 3.1). All three CesA genes contain domains known to be conserved among cellulose synthases of higher plants. This includes a putative zinc-binding domain (Kurek et al., 2002), a region of variability also referred to as a hypervariable region (HVR) and a class specific region (CSR), the conserved catalytic motif consisting of four aspartate residues and the QXXRW residue sequence, and eight trans-membrane domains (TMD) (two at the N- terminus and six at the C-terminus). The CSR regions of PgCesA1, A2, and A3 only share between 25 - 35% identity. 99 Table 3.3. Amino acid residue shared identities of deduced PgCesA proteins with each other and the most similar CesAs from Arabidopsis (At) and Populus trichocarpa (Pti). Three comparisons are presented: entire amino acid sequences, the hypervariable region sequences (HVR), and the class-specific region sequences (CSR). The number in brackets refers to the percentage identity of the nucleotide sequences. Values in bold represent the highest similarities within one group (e.g.: PgCesA1 within Arabidopsis CesAs has the highest similarity with AtCesA8). Grey shading is to differentiate the comparisons between PgCesAs, AtCesAs, and PtiCesAs. Entire sequence HVR CSR PgCesA1 PgCesA2 PgCesA3 PgCesA1 PgCesA2 PgCesA3 PgCesA1 PgCesA2 PgCesA3 PgCesA2 60 (70) 23 35 PgCesA3 66 (59) 63 (72) 11 19 25 40 AtCesA4 63 (58) 62 (67) 66 (56) 22 14 13 37 63 48 AtCesA7 64 (65) 62 (69) 77 (74) 6 13 34 12 22 36 AtCesA8 64 (69) 59 (57) 65 (56) 20 16 12 40 27 19 PtiCesA4 66 64 67 11 5 6 41 64 44 PtiCesA7-A 64 62 78 5 7 6 27 23 44 PtiCesA8-A 73 61 79 5 8 8 50 24 31 100 101 Figure 3.1. Alignment of the predicted amino acid sequences of spruce (Picea glauca) putative secondary cell wall-specific cellulose synthases, PgCesA1, PgCesA2, and PgCesA3. Black shading indicates identical conserved residues, grey shading indicates non-identical residues with similar properties. The known, essential, conserved catalytic motifs are indicated below the sequence with an asterisk (D, D, D, QXXRW motif). The predicted zinc-binding domain, class specific regions I and II, and trans-membrane domains (TMD) are indicated above the sequence. 102 3.3.3 Phylogenetic analysis The putative amino acid sequences of PgCesA1, A2, and A3 were compared to 44 other full-length amino acid sequences (Table 3.2). The multiple sequence alignment produced by ClustalW was used to create a dendrogram to view phylogenetic relationships (Figure 3.2). The dendrogram shows six major clades: three containing primary cell-wall specific cellulose synthases and three containing secondary cell-wall specific proteins. Each major clade contains CesAs from monocot and dicotyledonous plants, as well as those from gymnosperms. The three PgCesAs fall into unique clades, each clustered with one of the previously identified secondary cell wall-specific CesAs from Arabidopsis. Specifically, PgCesA1 groups with AtCesA8, PgCesA2 groups with AtCesA4, and PgCesA3 groups with AtCesA7. Conflicting results were obtained for PgCesA3: whole amino acid sequence similarity is highest with AtCesA7 (Table 3.3, Figure 3.2), but CSR similarity groups PgCesA3 with AtCesA4 (Table 3.3). Reviewing the alignment of the N-terminal regions only, the degree of similarity is highest between PgCesA3 and AtCesA7 (58%), followed by PgCesA2 with AtCesA7 (44%), and PgCesA1 with AtCesA8 (36%). Based on all the alignments, it is thought that PgCesA1 is orthologous to AtCesA8, PgCesA2 is likely an ortholog of AtCesA4, but PgCesA3 could be an ortholog of either AtCesA4 or AtCesA7. The class-specific regions of PgCesA1, A2, and A3 share an extremely low level of similarity with each other (Table 3.3). The degree of similarity was not much greater when compared to potential orthologs from Arabidopsis or Populus. However, when compared to the CSRs of the Pinus taeda orthologs the similarities were significantly higher, between 75 and 95%. This suggests that the CSR conservation is limited to more closely related orthologs. 3.3.4 Expression profiling of PgCesA1, A2, and A3 Using quantitative real-time PCR, the expression levels of PgCesA1, A2, and A3 were measured in tissues expected to be undergoing varying levels of secondary cell wall biosynthesis. Transcript abundance of the translation initiation factor eIF5\u00EF\u0081\u00A1 was measured as a control for data normalization. 103 104 Figure 3.2. A dendrogram assembled from the alignment of the deduced amino acid sequences of PgCesA1, A2, and A3 with 48 other confirmed and putative full-length cellulose synthases (Table 3.2). The AtCesAs known to be involved in secondary cell wall biosynthesis are marked with a coloured star, and the putative spruce and poplar homologues are marked with a similarly coloured star. Genus-species names were abbreviated as follows: At \u00E2\u0080\u0093 Arabidopsis thaliana; Eg \u00E2\u0080\u0093 Eucalyptus grandis; Gh \u00E2\u0080\u0093 Gossipium hirsutum; Os \u00E2\u0080\u0093 Oryza sativa; Pir \u00E2\u0080\u0093 Pinus radiata; Pit \u00E2\u0080\u0093 Pinus taeda; Ptd \u00E2\u0080\u0093 Populus tremuloides; Pti \u00E2\u0080\u0093 Populus trichocarpa; Pp \u00E2\u0080\u0093 Physcomitrella patens. 105 Expression was measured in six different tissues from three individual trees. Expression of all three PgCesA genes showed to be highest in xylem tissue (Figure 3.3). Expression was also relatively high in young needles (both expanded and unexpanded), as well as in root tissue. By contrast, low expression was found in old needles and phloem tissue. In all tissues the transcript abundance of PgCesA1 and A2 was consistently higher than that of PgCesA3. PgCesA1 and A2 were expressed at very similar levels within a particular tissue. 3.3.5 Ability of spruce CesAs to functionally complement Arabidopsis CesAirx mutants To test if the function of the spruce CesAs is conserved with the function of the Arabidopsis orthologs, PgCesA cDNAs were expressed in one of the Arabidopsis ortholog-mutant plants. The Arabidopsis CesAirx mutants AtCesA4irx5-1, AtCesA7irx3-1, and AtCesA8irx1-1 were used. These AtCesAirx mutants exhibit an obvious mutant phenotype consisting of stunted growth, rounded leaves, and weaker stems compared to wild-type plants (Figure 3.4 B). At the cellular level, the AtCesAirx plants have collapsed xylem vessels. The PgCesAs were expressed in homozygous Arabidopsis mutant plants under the expressional control of the 35S promoter (P35S::LUC-PgCesA), or an Arabidopsis CesA promoter (PA4::LUC-PgCesA, PA7::LUC-PgCesA, or PA8::LUC- PgCesA). All expression constructs resulted in an N-terminal Luciferase tag (LUC) or an N-terminal yellow fluorescent (YFP) tag on the PgCesA. Tagged versions of AtCesAs have been previously found to allow complementation of AtCesAirx mutants (McDonnell, unpublished), and were used because of 1) the ease with which PgCesAs could be cloned into the vector using Gateway technology, and 2) the LUC tag could provide the opportunity to monitor protein expression via detection of. 106 Figure 3.3. Real-time PCR analysis of PgCesA1, PgCesA2, and PgCesA3 transcript abundance in tissues of 5-year-old spruce trees. Average expression is relative to levels of a house-keeping gene, PgeIF5\u00EF\u0081\u00A1. Error bars represent standard error of the mean, n=3 (biological replicates). Tissue labels are as follows: ON, old needles; YN-C, young-unexpanded needles, YN-O, young-expanded needles; PH, phloem/cambium; XY, xylem; RT, roots. ON YN-C YN-O PH XY RT 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.35 0.40 0.45 0.50 0.55 0.60 R e la ti v e e x p re s s io n Tissue PgCesA1 PgCesA2 PgCesA3 107 Figure 3.4. LUC-PgCesA1 does not rescue the A8irx1-1 mutant phenotype under expressional control of the 35S promoter. A-G: Phenotype of plants. A and E: Wild- type (WT). B and F: AtCesA8irx1-1 mutant. C, D, G: AtCesA8irx1-1 plants transformed with the P35S::LUC-PgCesA1 construct, expressing LUC-PgCesA1 (A1 #4, A1 #7). White arrow heads mark regular xylem in the wild-type, and collapsed xylem in the mutant and transgenic plants. H: Real-time PCR analysis of transcript abundance. AtCesA8 (white bars) expression in WT and AtCesA8irx1-1(IRX1), and PgCesA1 expression (grey bar) in stem tissue of an AtCesA8irx1-1 -P35S::LUC-PgCesA1 transgenic plant (35S::PgA1). Expression is calculated relative to the control gene. Error bars represent standard error of the mean, n=3. R e la ti v e E x p re s s io n WT IRX1 35S::PgA1 0 2 4 6 8 10 12 14 16 18 Line H H R e la ti v e e x p re s s io n H 108 3.3.5.1 PgCesA1 Two sets of transformed plants expressing the LUC-PgCesA1 cDNA were recovered after multiple transformation attempts. The first set consisted of two lines transformed with P35S::LUC-PgCesA1, and the second set consisted of eight lines transformed with PA8::LUC-PgCesA1. None of the P35S::LUC-PgCesA1 lines showed complementation of the mutant phenotype (Figure 3.4 C, D, G), and only one was selected for further real-time PCR analysis. Expression of PgCesA1 measured with real-time PCR revealed very low transcript abundance, compared to the level of AtCesA8 in wild-type stem tissue (Figure 3.4 H), despite PgCesA1 being regulated by the 35S promoter. The transcript abundance of AtCesA8 in wild-type plants was nearly 15x higher than PgCesA1. In contrast, it appears as though the PA8::LUC-PgCesA1 construct complemented the AtCesA8irx1-1 mutant (Figure 3.5 A-F), and one line showed slightly greater height growth compared to wild-type plants after 21 days growth. The xylem vessels did not appear to be collapsed, suggesting that the expression of PgCesA1 had rescued the cellulose-deficient phenotype of the AtCesA8irx1-1 mutation. However, leaf size and morphology appeared similar to the mutant plants (smaller and more round than wild-type leaves). The complemented line also produced more axillary stems. Expression levels of PgCesA1 were very high (over 8x higher) compared to levels of AtCesA8 in wild-type plants (Figure 3.5 G). The remaining seven AtCesA8irx1-1- PA8::LUC-PgCesA1 lines exhibited varying degrees of complementation of the AtCesAirx phenotype (Figure 3.6) which appear to be related to varying levels of PgCesA1 expression (Table 3.4). Expression of PgCesA1 at levels exceeding those of AtCesA8 in wild-type plants seems sufficient to rescue the mutant phenotype, however, levels significantly lower than those of AtCesA8 (from 12% to 78%) resulted in partial or full complementation of the AtCesA8irx1-1 phenotype. 109 Figure 3.5. Complementation of the A8irx1-1 mutant phenotype by LUC-PgCesA1 under expressional control of the native AtCesA8 promoter. A-F: Phenotype of plants. Wild- type (A, D), A8irx1-1 mutant (B, E) and an A8irx1-1 plant transformed with the PA8::LUC- PgCesA1 construct (C, F). Black arrow heads mark regular xylem in stems of a wild- type plant (D) and transgenic plant (F), and collapsed xylem in an A8irx1-1 stem (E). Scale bar = 50 \u00EF\u0081\u00ADm. G: Real-time PCR analysis of AtCesA8 (white bars) expression in wild-type (WT) and A8irx1-1 (IRX1) stem tissue, and PgCesA1 expression (grey bar) in stem tissue of A8irx1-1-PA8::LUC-PgCesA1 transgenic plant #6. Expression is calculated relative to the control gene, AtUBQ5, Error bars represent standard error of the mean, n=3. G 110 Table 3.4. Phenotype survey of A8irx1-1-LUC-PgCesA1 transformed lines. Eight AtCesA8irx1-1 lines expressing LUC-PgCesA1 were assessed for complementation of the AtCesAirx phenotype based on xylem morphology (collapsed or not) and growth (stature). Using real-time PCR, AtCesA8 expression was measured in wild-type plants (Ler-WT) and AtCesA8irx1-1 plants, and PgCesA1 expression was measured in the transgenic AtCesA8irx1-1 plants (A1\u00E2\u0080\u00946, 2, 3, 4, 5, 7, 8, 9). AtCesA8 and PgCesA1 expression levels were calculated relative to that of the control gene, AtUBQ5. Expression was then calculated as a percentage of the AtCesA8 values in wild-type plants. Plant A1-6 is imaged in Figure 3.5, plants A1-2, -3, -4, -5, -7, -8, and -9 are imaged in Figure 3.6. Line Xylem Growth Gene Expression (% of endogenous AtCesA8) Phenotype Ler-WT normal normal 100 wild-type AtCesA8 irx1-1 collapsed stunted 19 mutant A1-6 normal normal 798 wild-type A1-2 normal normal 78 wild-type A1-3 normal stunted 23 intermediate A1-4 normal normal 164 wild-type A1-5 normal normal 191 wild-type A1-7 normal normal 12 wild-type A1-8 collapsed stunted 1 mutant A1-9 some collapsed normal 16 intermediate 111 Figure 3.6. Additional A8irx1-1 plants expressing LUC-PgCesA1 under the regulation of the AtCesA8 promoter show varying degrees of complementation. AtCesA8irx1-1 transgenic plants (A1-2, -3, -4, -5, -7, -8, and -9) transformed with the PA8::LUC- PgCesA1 construct are compared to a wild-type plant (WT-Ler) and an A8irx1-1 mutant plant. Stem cross sections were taken from the stem base (within 2 cm of the rosette) to compare xylem morphology. Scale bar = 150 \u00EF\u0081\u00ADm. 112 3.3.5.2 PgCesA3 Multiple attempts were made to transform AtCesA7irx3-1 plants with P35S::LUC- PgCesA3 or PA7::YFP-PgCesA3, and AtCesA4irx5-1 plants with P35S::LUC-PgCesA3 or PA4::YFP-PgCesA3. However, only three transgenic lines were recovered: AtCesA4irx5- 1 transformed with P35S::LUC-PgCesA3. None of the transgenic lines showed a complemented phenotype (Figure 3.7 C and F). Expression of PgCesA3 in the three transgenic lines varied (Figure 3.7 G), but was not as high as AtCesA4 expression in wild-type plants. Although AtCesA4irx5-1 - P35S::LUC-PgCesA3 line #4 expressed PgCesA3 at levels only slightly lower than AtCesA4 in wild-type plants, complementation of the mutant phenotype was not observed. This implies that the 35S promoter does not express PgCesA3 in the necessary cells and tissues at levels sufficiently high enough to complement the mutant phenotype, or that PgCesA3 is not able to functionally complement the AtCesA4irx5-1 mutation. These results could also suggest that PgCesA3 is not able to function properly with the LUC tag attached at the N-terminus of the PgCesA3 protein. 113 Figure 3.7. P35S::LUC-PgCesA3 is not sufficient to rescue the A4irx5-1 mutant phenotype. A-F: Phenotype of plants. Wild-type (A, D), A4irx5-1 mutant (B, E), and an A4irx5-1 plant transformed with the P35S::LUC-PgCesA3 construct (C, F). White arrow heads mark regular xylem in the wild-type (D), and collapsed xylem in the A4irx5-1 mutant (E) and LUC-PgCesA3 transgenic plant (F). Scale bar represents 100 \u00EF\u0081\u00ADm. G: Real-time PCR analysis. AtCesA4 expression (white bars) in wild-type (WT) and A4irx5-1 (IRX5), and PgCesA3 expression (grey bars) in stem tissue of A4irx5-1 - P35S::PgCesA3 transgenic plants (#1, #4, #3). Expression is calculated relative to the control gene, AtUBQ5. Error bars represent standard error of the mean, n=3. 114 3.3.5.3 PgCesA2 Multiple attempts to transform AtCesA7irx3-1 with P35S::YFP-PgCesA2 or PA7::YFP-PgCesA2, and AtCesA4irx5-1 with P35S::YFP-PgCesA2 were unsuccessful. Several plantlets were able to successfully grow on antibiotic selection medium, but the transgene could not be detected in genomic DNA or cDNA by PCR screening. For this reason, the ability of PgCesA2 to functionally complement these mutants is unknown. 3.3.6 Ability of Poplar CesAs to complement Arabidopsis CesAirx mutants The Arabidopsis AtCesA4, A7, and A8 amino acid sequences share a higher degree of similarity with the poplar CesA orthologs compared to the spruce CesA orthologs, as evidenced by phylogenetic analysis (Figure 3.2) and amino acid sequence alignments (data not shown). The alignment scores between whole predicted amino acid sequences were previously reported by Kumar et al. (2009) as 88% between AtCesA4 and PtiCesA4, 93% between AtCesA8 and PtiCesA8-A (93%), and 93% between AtCesA7 and PtiCesA7-A. An alignment of just the predicted class-specific region sequences (formerly hyper-variable region II) revealed that the AtCesA4, A8, and A7 CSRs are approximately 65%, 71% and 80% similar to those of PtiCesA4, A7-A, and A8-A, respectively. These similarities are approximately 20% higher than the similarity between the PgCesA and AtCesA CSRs. Based on sequence similarity, the following transformations were performed to conduct functional complementation assays: AtCesA4irx5-1plants transformed with PAtCesA4::LUC-PtiCesA4, AtCesA7irx3-1 plants transformed with PAtCesA7::LUC- PtiCesA7-A, and AtCesA8irx1-1 plants transformed with PAtCesA8::LUC-PtiCesA8-A. The Luciferase (LUC) tagged constructs were used for reasons described above (section 3.3.5). Real-time PCR analysis confirmed transgene expression at levels that varied compared to the abundance of endogenous AtCesAs in wild-type plants (Table 3.5). 115 Table 3.5. Real time PCR analysis of LUC-PtiCesA transgene expression in complemented Arabidopsis AtCesAirx mutant lines. Expression of endogenous AtCesA4, A7, or A8 in wild-type plants (grey highlighting) is compared to that of the LUC-PtiCesA transgene expression, which was measured in AtCesAirx plants transformed with PtiCesA7A (#4, #P3), A8A (#P1), or A4 (#1) Expression is calculated relative to the control gene, AtUBQ5, using the equation 2-(CTCesA-CTubq5) (delta CT). Line Gene Relative expression Standard deviation wild-type AtCesA7 2.18 0.20 AtCesA7irx3-1 -PtiCesA7A #4 PtiCesA7-A 0.24 0.12 AtCesA7irx3-1 -PtiCesA7A #P3 PtiCesA7-A 0.17 0.14 wild-type AtCesA8 5.24 1.61 AtCesA8irx1-1 -PtiCesA8A #P1 PtiCesA8-1 0.01 0.02 wild-type AtCesA4 0.48 0.50 AtCesA4irx5-1 -PtiCesA4 #1 PtiCesA4-A 1.22 0.85 AtCesA4irx5-1 -PtiCesA4 #2 PtiCesA4-A 7.32 0.94 116 As seen in Figure 3.8, the T1 generation transgenic plants exhibited varied complementation when compared to the wild-type growth phenotype, yet all lines had normal xylem. In particular, A7irx3-1 mutant lines expressing LUC-PtiCesA7-A seemed more complemented than the others. A4irx5-1 mutant plants expressing LUC-PtiCEsA4 had an intermediate complemented phenotype based on a somewhat stunted like stature, but rescuing of the collapsed xylem phenotype. A8irx1-1 mutant lines expressing the PAtCesA8::LUC-PtiCesA8-A construct, which had stunted growth similar to (but not quite as severe as) the AtCesirx mutants, did exhibit non-collapsed xylem. 3.4 Discussion 3.4.1 Three unique cellulose synthase genes identified in spruce At the nucleotide level, the three isolated PgCesA cDNAs are highly similar to cDNAs of CesAs from other gymnosperms, such as Pinus radiata and Pinus taeda. The predicted amino acid sequences of PgCesA1, A2, and A3 revealed motifs that are conserved among all other known cellulose synthases (Roberts and Roberts, 2007) including the defining D, D, D, QXXRW motif found in \u00EF\u0081\u00A2-glycosyltransferases (Pear et al., 1996), and believed to be important for catalytic function (Pear et al., 1996; Taylor et al., 2000; Doblin et al., 2002). The degree of similarity between the PgCesA CSRs was lower than alignments with CSR sequences of other CesA orthologs, therefore the isolated PgCesA1, A2, and A3 are likely to be uniquely different from one another. 3.4.2 Evolutionary relationship of PgCesA1, A2, and A3 with other known cellulose synthases A dendrogram of several plant cellulose synthases resulted in a clear separation of CesAs implicated in primary cell wall development from those implicated in secondary cell wall development. These findings are consistent with phylogenetic analysis previously reported (Ranik and Myburg, 2006; Kumar et al., 2009). 117 Figure 3.8. Complementation of Arabidopsis AtCesAirx mutants by LUC-PtiCesAs. Top: Plant morphology, bottom: stem cross sections to visualize xylem morphology. Left to right: Wild-type (WT) is compared to AtCesA7irx3-1 (IRX, representative of A4, A7, and A8 mutants), and the transgenic complemented lines AtCesA7irx3-1 \u00E2\u0080\u0093LUC- PtiCesA7-A (PA7::LUC-PtiCesA7-A), AtCesA8irx1-1\u00E2\u0080\u0093LUC-PtiCesA8-A (PA8::LUC- PtiCesA8-A), and AtCesA4irx5-1 \u00E2\u0080\u0093LUC-PtiCesA4 (PA4::LUC-PtiCesA4). White arrow heads mark regular xylem in the wild-type and transgenic plants, and collapsed xylem in A8irx1-1. Scale bar represents 100 \u00EF\u0081\u00ADm. 118 Putative orthologs of PgCesA1, A2, and A3 from Arabidopsis (Taylor et al., 2003), Populus (Joshi et al, 2004; Kumar et al, 2009), Pinus (Nairn and Haselkorn, 2005), Eucalyptus (Ranik and Myburg, 2006), and rice (Tanaka et al., 2003) are strongly associated with secondary cell wall development. The grouping of CesAs based on primary or secondary cell wall biosynthesis suggests a strong conservation of developmental specific functioning. Within both the AtCesA7 and AtCesA8-containing clades there are two CesAs from Populus trichocarpa (PtiCesA7-A and B, PtiCesA8-A and B, respectively). At the nucleotide level PtiCesA7-A and \u00E2\u0080\u0093B are 94% identical, and there close proximity within the dendrogram confirms that they are likely gene duplicates (Kumar et al., 2009). Contrastingly, the distinct positions of PgCesA1, A2, and A3 in separate clades further supports that they are unique cellulose synthases. Amino acid sequence similarity suggests that PgCesA1, A2, and A3 are orthologs of AtCesA8, AtCesA4, and AtCesA7, respectively. Other reports have indicated that alignment based phylogeny is consistent whether it is based on alignment of whole amino acid sequences or only the CSR sequences (Vergara and Carpita, 2001; Ranik and Myburg, 2006). While whole amino acid alignments placed PgCesA3 within the AtCesA7 clade, an alignment based on the CSR sequences places PgCesA3 with AtCesA4 (along with PgCesA2). The CSR sits within the putative catalytic domain, between the second and third conserved aspartate residues, and therefore homology of CSRs could implicate functional specificity. Nairn and Haselkorn (2001) suggested that the N-terminal region of AtCesA7-like CesAs was more conserved compared to N- terminal regions of other orthologous groups of CesAs, perhaps due to a specialized function or greater evolutionary divergence of AtCesA4 and A8 compared to AtCesA7. As evidenced by the amino acid alignment of PgCesAs with AtCesAs, the N-terminal regions are most similar between AtCesA7 and PgCesA3, followed by PgCesA2. It is interesting to note that in an un-rooted alignment of CesAs, including those of a moss (Physcomitrella patens), AtCesA7 is grouped with the moss CesAs, whereas AtCesA4 and AtCesA8 are in separate clades (Roberts and Bushoven, 2007). Again, this suggests that AtCesA7, and perhaps AtCesA7 orthologs, are less divergent compared to the other secondary cell wall-specific CesAs. Based on all alignment analysis, it seems most likely that PgCesA3 is an ortholog of AtCesA7. The grouping of PgCesA3 119 with AtCesA4 based on CSR alignments could be the result of less divergence between the PgCesA2 and A3 (compared to, for example, AtCesA7 divergence from AtCesA4). 3.4.3 PgCesA1, A2, and A3 expression suggests involvement in secondary cell wall formation The tissues harvested for gene expression analysis were collected in early May, during active growth and activity of the vascular cambium. The expression profiles of PgCesA1, A2, and A3 further support the notion that they are involved in secondary cell wall biosynthesis as expression was by far the highest in xylem tissue. The layers of tissue collected for xylem RNA extraction consisted of newly formed, immature xylem elements as well as xylem actively undergoing secondary cell wall formation and mature xylem tissue. Compared to the expression levels in xylem, only moderate expression was observed in young needles (both unexpanded and recently expanded). Expression of the three putative secondary cell wall CesAs (PitCesA1, A2, and A3) was also detected in young needles of pine (Nairn and Haselkorn, 2005). The synthesis of secondary cell walls in developing xylem of actively growing young needles could explain the expression of PgCesA1, A2, and A3 in these tissue samples. Moderately high levels of PgCesA1, A2, and A3 were also detected in root tissue, which can be attributed to developing xylem elements in the roots. Compared to mature needles there should be more xylem undergoing secondary cell wall formation, and therefore greater cellulose deposition in young needles and developing root tissue. Indeed, expression of the PgCesAs was the lowest in old needles, further supporting the need for PgCesA1, A2, and A3 in tissues that are actively developing xylem and therefore secondary cell walls. Although secondary cell walls are formed in phloem tissue the expression of PgCesA1, A2, and A3 was low. In Populus tremuloides, expression of PtrCesA1, A2, and A3 (orthologs of AtCesA8, A7, and A4, respectively) was found in developing phloem fibres based on in situ mRNA hybridization in stem sections (Joshi et al., 2004). In Eucalyptus, RT-PCR expression profiles of putative secondary cell wall EgCesAs revealed low expression in phloem tissue (Ranik and Myburg, 2006), consistent with our findings. Low, but detectable, expression of hybrid poplar PtxtCesA1, A2, and A3 (orthologs of AtCesA8, A7, and A4, respectively) in phloem was also reported by Djerbi et al. (2004). In both Eucalyptus (Ranik and Myburg, 2006) and 120 hybrid poplar (Djerbi et al., 2004) the CesAs more highly expressed in phloem had homology to the primary cell wall CesAs of Arabidopsis. It is postulated that PgCesA1, A2, and A3 might be involved in cellulose biosynthesis of secondary cell walls of phloem fibres, but that other yet to be examined PgCesAs are involved in primary cell wall biosynthesis in phloem which would be expressed at higher levels. RT-PCR analysis of PgCesA1, A2, and A3 transcript abundance shows that A1 and A2 are consistently more highly expressed than A3. Unequal expression between CesA family members, within a specific tissue, has been reported for other groups of cellulose synthases, including those from Arabidopsis (Brown et al., 2005), Populus (Samuga and Joshi, 2002), Pinus (Nairn and Haselkorn, 2005), and Eucalyptus (Ranik and Myburg, 2006). The biological implications for unequal gene expression, and if differences in transcript abundance are always conserved at the protein level, are not clear. Current reports of CesA protein abundance are ambiguous. For example, antibody labelling of Arabidopsis stem interfascicular fibres indicated higher abundance of AtCesA4 and A8, compared to those of AtCesA7 (Taylor et al., 2003). However, proteins from Arabidopsis stem tissue separated using SDS-PAGE and probed with CesA-specific antibodies showed higher levels of AtCesA8 monomers, followed closely by AtCesA7 and lower levels of AtCesA4 (Atanassov et al., 2009). Thus, it stands to reason that unequal gene expression levels likely represent some disparity in CesA protein ratios, but to what degree is not clear. Unequal expression, within a specific tissue or cell type, could be due to a few key factors, such as: 1) functional differences between individual CesAs dictate the need for unequal ratios; 2) the rate of protein turnover of individual CesAs varies; 3) cellulose synthesizing complexes require an unequal ratio of CesAs to function properly. Measuring protein levels within different tissues and cell types, and determining if individual CesAs have unique and specific functions, could explain the need for multiple CesAs and their unequal expression patterns. The comparative levels of expression between PgCesA1, A2, and A3 do not follow the expression pattern of their AtCesA orthologs. In inflorescence stems of 21 day-old Arabidopsis plants it was found that the AtCesA7 transcript (PgCesA3 ortholog) was most abundant followed closely by AtCesA8 (PgCesA1 ortholog), and lower expression of AtCesA4 (PgCesA2 ortholog) (Appendix A). A similar expression profile 121 was reported by Brown et al. (2005) in mature Arabidopsis stem tissue. However, Hamman et al., (2004) reported highest expression of AtCesA4 in stem tissue. In our analysis, PgCesA2, the AtCesA4 ortholog, showed the highest expression. In Eucalyptus xylem tissue, Ranik and Myburg (2006) reported the highest expression of EgCesA3 (the AtCesA7 ortholog) followed closely by EgCesA1 (AtCesA8 ortholog). In hybrid poplar, PtxtCesA1 (AtCesA8 ortholog) was the most abundant, followed by PtxtCesA2 (AtCesA7 ortholog) and PtxtCesA3 (AtCesA4 ortholog) (Djerbi et al., 2004). In Pinus taeda, Northern blot analysis (Nairn and Haselkorn, 2005) and RT-PCR analysis (Nairn et al., 2008) showed highest to expression in xylem of PitCesA2 (AtCesA7 ortholog) and slightly lower levels of PitCesA1 (AtCesA8 ortholog) and PitCesA3 (AtCesA4 ortholog). It is clear that unequal ratios exist amongst all secondary cell wall CesA gene families. However, there is some variability between different species as to which CesA is the most highly expressed and which CesA has the lowest expression in woody tissues. Different ratios between secondary cell wall CesAs from Arabidopsis, Eucalyptus, pine, and spruce could be due to experimental differences including tissue type, and time of harvest, both of which can have a dramatic effect on gene expression levels. Biologically, the differences could be due to the divergence of function between the angiosperm and gymnosperm CesAs. Interestingly, AtCesA8, EgCesA1, PtxtCesA1, PitCesA1, and PgCesA1 all show high expression (either the highest or nearly highest). This could suggest that the level of requirement of the AtCesA8 orthologs could be more conserved compared to the AtCesA7 and AtCesA4 orthologs. It is thought that cellulose biosynthesis in secondary cell walls requires the function of three distinct CesAs. In Arabidopsis this has been confirmed by mutant studies, revealing the need for AtCesA8, A7, and A4 to be functionally active for proper secondary cell wall formation. Based on sequence homology and gene expression profiling it is also thought that PgCesA1, A2, and A3 are the three subunits that would work in concert during secondary cell wall formation in spruce. However, the identification of additional PgCesAs, particularly primary cell wall-specific, is required to gain a complete understanding of the CesA gene family in spruce. 122 3.4.4 Significant conservation of function exists between CesA orthologs Wild-type Arabidopsis plants develop thick secondary cell walls in xylem vessels and interfascicular fibres which can be easily observed in hand sections of stem tissue. The cellulose synthase irregular xylem (irx) mutants have provided critical information about the role of AtCesA4, AtCesA7, and AtCesA8 in secondary cell wall formation in Arabidopsis stems (Turner and Somerville, 1997; Taylor et al., 2000; Taylor et al., 2003; Gardiner et al., 2003; Ha et al., 2002). Arabidopsis plants homozygous for the AtCesA8irx1-1 mutation contain a point mutation in the AtCesA8 gene, resulting in an amino acid substitution of one of the conserved aspartate residues within the catalytic domain, resulting in collapsed xylem with irregular cell wall deposition and reduced cellulose (Taylor et al., 2000), and mutant plants tend to be smaller in stature. The AtCesA8irx1-1 phenotype is easily observed, and provides a method to screen for functional complementation when transformed with other CesAs. In order to functionally complement AtCesA8irx1-1 it was expected that the orthologous gene product would be functionally equivalent to AtCesA8, and also expressed at temporally and spatially appropriate levels compared to the endogenous gene in wild-type plants. The cauliflower mosaic 35S promoter (P35S) is a constitutive promoter, however, in this study AtCesA8irx1-1 plants transformed with P35S::LUC-PgCesA1 expressed PgCesA1 at significantly lower levels than the native AtCesA8 in wild-type plants, and neither complete nor partial recovery of the AtCesA8irx1-1 phenotype was observed. Expression of PgCesA1 under the control of the endogenous AtCesA8 promoter (PA8) was hypothesized to provide proper spatial and temporal expression of a transgene, and therefore better conditions to examine functional orthology between CesAs. Eight transgenic lines were recovered, expressing PA8::LUC-PgCesA1 at levels from 1% to over 700% of the native AtCesA8 expression in wild-type plants. Such a range of transgene expression, despite all being driven by the AtCesA8 promoter may be explained by the position and copy number of the PA8::LUC-PgCesA1 fragment insertion within the genome of transgenic plants (Matzke and Matzke, 1998). Seven of these transgenic lines showed partial or complete recovery from the AtCesAirx phenotype based on overall plant growth and xylem morphology. Compared to the P35S::LUC-PgCesA1 transgenic lines it is clear that the AtCesA8 promoter is likely 123 providing critical temporal and spatial-specific expression that is required for LUC- PgCesA1 to complement the AtCesA8irx1-1 mutation. In addition to PgCesA1, we determined that the poplar CesAs (fused to Luciferase), PtiCesA4, PtiCesA7-A, and PtiCesA8-A, could at least partially complement the AtCesA4irx5-1, AtCesA7irx3-1, and AtCesA8irx1-1 mutant phenotypes (all under the expressional control of native AtCesA promoters), respectively. Although, the expression of LUC-PtiCesA8-A in AtCesA8irx1-1 and LUC-PtiCesA4 in AtCesA4irx5-1 only resulted in a partial complementation of the mutant phenotype, as transgenic plants exhibited AtCesAirx-like stunted growth but had normal xylem morphology. PtiCesA8-A transcript abundance was extremely low in these transgenic lines, which could have resulted in insufficient protein expression to fully complement the mutant phenotype. As well, the presence of the mutant A8irx1-1 protein in these mutants likely inhibits PtiCesA8- A from entering the CSC, thus maintaining some of the mutant phenotype. It should also be noted that the Luciferase tag attached to the 5\u00E2\u0080\u0099-terminus of PtiCesA8-A could interfere with CesA function or the ability of LUC-PtiCesAs to access the CSC, or both, which would cause only partial complementation to be achieved. However, the partial complementation suggests that PtiCesA8-A and PtiCesA4 likely has a high degree of functional conservation with AtCesA8 and AtCesA4, respectively. It would be relevant to determine if full complementation could be achieved with 1) higher expression of the LUC-PtiCesA fragment, or 2) expression of a non-tagged form of PtiCesA in the mutant background. The results presented herein are the first report of AtCesA orthologs capable of complementing an AtCesAirx mutation, and strongly support the idea that CesA-specific functions in secondary cell wall biosynthesis may have evolved prior to the divergence between angiosperms and gymnosperms. Recently, Maloney (2010) reported that a spruce KORRIGAN was capable of complementing the AtKORkor1-1 mutant. Combined with the results presented herein, this suggests that there could be a great deal of conservation of function among cell wall biosynthetic genes. It would be interesting to explore whether monocot CesA orthologs of AtCesA4, A7, and A8 share the same degree of functional conservation. The total abundance, crystallinity, microfibril angle, and degree of polymerization of cellulose in secondary cell walls of spruce tracheids, poplar vessels, and Arabidopsis vessels and fibres are likely very different. However, it 124 is clear that the functions of orthologous CesAs are sufficiently conserved. Thus, the ability of spruce and poplar CesAs to permit proper cellulose biosynthesis to occur in Arabidopsis suggests that the cellulose requirements of secondary cell walls are likely to be greatly influenced by factors other than the CesA- composition of CSCs. However, it must also be considered that the ratio of CesA subunits (e.g. CesA4:CesA7:CesA8) within a CSC may vary from species to species, and that such variety could dictate differences in cellulose properties. It is highly likely that the cellulose synthase complexes in the plasma membrane consist of multiple individual CesAs which interact within the complex. As such, in order to functionally complement the AtCesAirx1 mutants the orthologous spruce and poplar CesA subunits must be able to integrate into the complex and interact with neighbouring AtCesAs. Our results suggest that PgCesA1, PtiCesA8-A, PtiCesA7-A, and PtiCesA4 contain the domain(s) required for proper CesA-CesA interaction with AtCesA subunits. Kurek et al. (2002) implicated the zinc-binding domain in CesA-CesA interactions, and therefore complex assembly. Based on an amino acid alignment, the PgCesA1 and AtCesA8 zinc-finger region shows low similarity (approximately 48% identity and 67% similarity). Perhaps it is simply the presence of the zinc-finger region that allows integration into the complex and interaction with other CesAs. The central domain of CesA proteins (containing the catalytic site and CSR region) has been identified as necessary for proper CesA function based on mutant analysis (see Joshi and Mansfield, 2007 for review). Additionally, a chimeric protein consisting of a fusion between the AtCesA3 N-terminus with the central domain of AtCesA1 could functionally complement an AtCesA1 mutant phenotype (Wang et al., 2006), again illustrating the importance of the central domain in function. Our results suggest that the catalytic function of the AtCesA orthologs we tested are highly conserved with those of the AtCesA enzymes. 3.4.5 PgCesA2 and PgCesA3 may be distant orthologs of either AtCesA4 or AtCesA7 The inability of PgCesA3 to complement the AtCesA4irx5-1mutant could be due to several factors: 1) insufficient expression of PgCesA3 in the transgenic plants; 2) PgCesA3 is not a true ortholog of AtCesA4; 3) PgCesA3 is a distant ortholog of AtCesA4 or AtCesA7, 4) the 5\u00E2\u0080\u0099-terminal LUC tag on PgCesA3 results in poor function of 125 PgCes3, or 5) the 5\u00E2\u0080\u0099-terminal LUC tag on PgCesA3 restricts PgCesA3 from accessing the CSC. Expression of PgCesA3 in the recovered transgenic plants was lower compared to endogenous AtCesA expression in wild-type plants. As shown by the complementation of AtCesA8irx1-1 by LUC-PgCesA1 under expressional control of the native AtCesA8 promoter, it is likely that tissue and developmental-specific expression driven by an appropriate promoter is required to enable complementation. Hence, the plants recovered in this experiment, in which LUC-PgCesA3 expression is driven by the 35S promoter, do not provide enough evidence to conclude if PgCesA3 can functionally complement AtCesA4. A LUC-AtCesA4 fusion protein has been shown to be able to rescue the A4irx5-1 mutant phenotype (McDonnell, Chapter 2), illustrating that LUC- tagged CesAs can access the CSC, and function properly. However, it has not been overlooked that LUC-AtCesA4 may be a better fit in the CSC, with other AtCesAs, than LUC-PgCesA3. Thus it should be explored if a non-tagged form of PgCesA3 can complement the mutant phenotype. Additionally, testing the ability of PgCesA3 to complement the AtCesA7irx3-1 mutant phenotype may help resolve if PgCesA3 is a functional ortholog of AtCesA4 or AtCesA7. The lack of transgenic plants carrying the PgCesA2 construct is puzzling, and will require further examination in order to conclude if PgCesA2 is functionally orthologous to one of the Arabidopsis CesAs. In summary, herein we report on the identification of three unique cellulose synthase genes from spruce (Picea glauca). Sequence homology and expression profiles indicate that these PgCesAs are members of a secondary cell wall-specific CesA gene family. One gene, PgCesA1, was confirmed to be a true ortholog of AtCesA8 based its ability to complement the Arabidopsis AtCesA8irx1-1 mutant phenotype. Additionally, conservation of function between AtCesA4-PtiCesA4, AtCesA7-PtiCesA7-A, and AtCesA8-PtiCesA8-A was shown, suggesting that the catalytic function of CesA orthologs is highly conserved. This is the first example of cross-species CesA functional complementation. Despite extensive evolutionary divergence between spruce, poplar, and Arabidopsis there is clearly a very strong conservation of the primary functions among the AtCesA4, A7, and A8 orthologs. Together, the sequence homology results and functional complementation provide further evidence that CesA functions evolved prior to the divergence of angiosperms 126 from gymnosperms. Further investigation is required to identify additional members of the spruce CesA gene family, and to confirm if additional PgCesAs are able to functionally complement their AtCesA orthologs. 127 CHAPTER 4 Measuring CesA-CesA interactions in planta 4.1 Introduction The polysaccharide network in both primary and secondary cell walls is critical to the integrity, plasticity and morphology of plant cell walls, and hence cell wall function. Large particles, or groups of particles, within the plasma membrane of bacteria, algae, and plants have been observed and are believed to be a complex of cellulose synthase enzymes, responsible for producing cellulose. Rows of particles in the plasma membrane of bacteria, Acetobacter xylinium, that were found to be associated with ribbons of cellulose microfibrils were visualized in 1976 and believed to be a cellulose synthesizing complex (Brown et al., 1976). Similarly, linear arrays of particles were observed in the plasma membrane of the green alga, Oocystis (Brown and Montezinos, 1976). Later, groups of particles, sometimes forming a hexameric structure and called rosettes, were observed in the plasma membrane of plants (Mueller and Brown, 1980). The rosette structure has been postulated to be composed of six cellulose synthesizing complexes (CSCs), and current models hypothesize that six CesA subunits are bound together to form the CSC (Delmer, 1999; Doblin et al., 2002; Ding and Hemmel, 2006). Kimura et al. (1999a) used gold-labeled CesA-specific antibodies to reveal that rosette structures in the plasma membrane of bean plants, Vigna angularis, contain CesAs. CesA proteins have a conserved structure containing a zinc-binding domain at the N-terminus, eight transmembrane domains (two located at the amino terminus and six at the carboxyl terminus), a plant-conserved region, a class-specific region, and two highly conserved domains believed to be part of the catalytic site (Richmond, 2000; Krauskopf et al., 2005). The zinc-binding domains of cotton CesA1 and A2 have been shown with an in vitro assay to interact in a 1:1 ratio, in addition to forming homodimers, suggesting that the zinc binding domain could regulate CesA interactions in vivo (Kurek et al., 2002). Mutant forms of AtCesA4 (AtCesA4irx5) were used to examine the interactions among CesA proteins and confirm that three CesA subunits were required for proper cellulose biosynthesis (Taylor et al. 2003). Depending on the severity of the mutation, various levels of association between the three secondary cell wall-specific CesA proteins were observed using immunoprecipitation and CesA-antibody labelling (Taylor et al., 2003). For example, the authors found that null mutations of AtCesA4 128 eliminated interactions between AtCesA8 and A7. In addition, compared to wild-type plants, a reduced level of AtCesA8 and A7 protein was observed, possibly the result of degradation of improperly formed CSCs due to the lack of CesA4 (Taylor et al., 2003). Using an epitope-tagged AtCesA7, Atanassov et al. (2009) isolated a large protein complex containing CesA7, CesA4, and CesA8 proteins from Arabidopsis stem protein extracts, believed to be a CSC unit, or a portion of a CSC. The use of in vitro methods to study CesA-CesA interactions suggests that all the CesAs form homodimers and heterodimers (Taylor et al., 2000; Taylor et al, 2003; Atanassov et al., 2009). However, it is still unclear as to whether these interactions are representative of those that occur in planta, in tissues undergoing active cell wall formation. For this reason, we chose to adopt in vivo methods to study CesA-CesA interactions, specifically, the bioluminescence resonance energy transfer (BRET) assay and bimolecular fluorescence complementation (BiFC), reviewed in Figure 4.1. The BRET assay (reviewed by Subramanian et al., 2004 and 2006) relies on the excitation- emission of Luciferase (LUC) to excite yellow fluorescent protein (YFP). Interaction candidates are genetically fused to the LUC or YFP cDNA and simultaneously expressed in planta (by stable or transient plant transformations). If protein-protein interactions are occurring, the excitation-emission of LUC will excite YFP, resulting in a distinct emission spectrum that differs from non-interacting proteins. Interactions are therefore detected by measuring emission spectra from plant tissues transformed with the LUC and YFP expression constructs. Compared to FRET, which requires laser excitation of the donor molecule, BRET avoids interference from autofluorescence. BiFC is an in vivo assay that provides an opportunity to visually detect protein- protein interactions (reviewed by Citovsky et al., 2008; Kerppola, 2008). Briefly, the BiFC assay involves reconstitution of YFP when two proteins, one tagged with the N- terminal fragment of YFP and the other tagged with the C-terminal fragment of YFP, interact resulting in the two YFP fragments coming together to form a functional, fluorescent YFP signal that can be visualized using fluorescence microscopy. Moreover, we wanted to determine if it was possible to detect whether interactions were tissue-specific and consistent. 129 Figure 4.1. BRET and BiFC assays. (A) In BRET, candidate proteins are genetically fused to either the Luciferase protein (L, LUC) or the YFP protein (Y, YFP). (B) The Luciferase substrate, coelenterazine (X) is added to the tissue and resulting Luciferase and YFP emission intensities are recorded using specific filters to detect the LUC emission (480 nm) and YFP emission (530 nm). If the candidate proteins are interacting, the resulting emission spectrum consists of both LUC and YFP, such that the ratio of YFP/LUC is generally greater than one. (C) In BiFC, the YFP gene has been separated into an N- and C-fragment, which can be genetically fused to two candidate proteins (D). If the two proteins are interacting, the YFP fragments can fuse together to form a functional, fluorescent YFP, which can be detected by fluorescence microscopy (E). Diagrams are modified from Xu et al. (1999) (BRET) and Citovsky et al. (2008) (BiFC). 130 The use of in planta methods to study CesA interactions has the potential to reveal some critical and novel information that has not been provided by in vitro studies about how CesAs come together to form a CSC. In this study BRET was employed in Arabidopsis plants and BiFC in Nicotiana tabacum leaf tissue to determine if AtCesA4, A7, and A8 subunits form homodimers and heterodimers in planta. The results suggest that various combinations of interactions do occur in planta but that they are not consistent between seedlings and stem tissue. Additionally, interactions appear to be weak and potentially transient. Further optimization of the BiFC assay is required to fully elucidate the complexity of CesA-CesA interactions in the CSC. 4.2 Materials and methods 4.2.1 Plant transformation vectors Arabidopsis thaliana cellulose synthase (AtCesA) 4, 7, and 8 cDNAs were isolated using PCR and cloned into the pENTR (Invitrogen) vector for use in Gateway recombination reactions to produce BRET-CesA and BiFC-CesA expression constructs (cloning primers listed in Table 4.1). BRET destination expression vectors containing Luciferase (LUC), pPZPhRLUC-attR (P35S::LUC), and Yellow Fluorescent Protein (YFP), pBin19YFP-attR (P35S::YFP) are described by Subramanian et al. (2006). BiFC destination expression vectors pBIFP-2 (P35S::NYFP, contains the N-terminal fragment of YFP) and pBIFP-3 (P35S::CYFP, contains the C-terminal fragment of YFP) are described by Hu et al. (2002). LR Clonase II (Invitrogen) reactions between pENTR- CesA plasmids and BRET or BiFC destination vectors were performed according to manufacturer\u00E2\u0080\u0099s instructions to create BRET (P35S::LUC-CesA and P35S::YFP-CesA) and BiFC (P35S::NYFP-CesA and P35S::CYFP-CesA) expression vectors. The BRET expression vectors were also modified to replace the 35S promoter with AtCesA promoters (PAtCesA: PA4, PA7, and PA8). For PAtCesA::LUC destination vectors, the AtCesA promoters were amplified from DNA using PCR to add SdaI (PA4, PA7) or HindIII (PA8) to the 5\u00EF\u0082\u00A2-terminus and AvrII (PA4, PA7, PA8) to the 3\u00EF\u0082\u00A2-terminus. Primers are listed in Table 4.1. 131 Table 4.1. PCR primers used for cloning, screening, and real-time (RT) PCR. Primer Name Sequence Use A4GWFW CACCGAACCAAACACCATGGCCAGC pENTR cloning A7GWFW CACCGAAGCTAGCGCCGGTCTTGTC pENTR cloning A8GWFW CACCGAGTCTAGGTCTCCCA pENTR cloning A4GWRV TTAACAGTCGACGCCACATTGCTTCA pENTR cloning A7GWRV TCAGCAGTTGATGCCACACTTGGA pENTR cloning A8GWRV TTAGCAATCGATCAAAAGACA pENTR cloning PA7FWSDAI CCTGCAGGGGGTACAGAGTTTGGGGAGTGATGG promoter cloning PA4FWCLAI AGTCATCGATGGGTACAGAGTTTGGGGAGTGATGG promoter cloning PA4RVAVRII AGTCCCTAGGTCGGAAGCAGAGCAGAAGGTGGG promoter cloning PA7FWSDAI CCTGCAGGGCAGCAACAGCAGGAGAGGTACG promoter cloning PA7FWCLAI AGTCATCATGCAGCAACAGCAGGAGAGGTACG promoter cloning PA7RVAVRII AGTCCCTAGGAGGGACGGCCGGAGATTAGCAG promoter cloning PA8FWHINDIII AAGCTTAGCTCACAATCTTCTTCCTGGTCG promoter cloning PA8FWCLAI ATCGATAGCTCACAATCTTCTTCCTGGTCG promoter cloning PA8RVAVRII CCTAGGCCCTGTTTGGAGAAACAGAGAAATGAACCC promoter cloning LUCFWAVRII AGTCCCTAGGACAAACGAATCTCAAGCAATCAAGC LUC-YFP cloning Term-YFPRVKPN AGTCGGTACCTCCGGCTCGTATGTTGTGTGGAAT LUC-YFP cloning YFPNFW CTGACCCTGAAGTTCATCTG screening LUC3'FW AGAGTGCTGAAGAACGAGCAGAGA screening/RT YFP3'FW ATCACATGGTCCTGCTGGAGTTCGT screening/RT YFP5'RV AGATGAACTTCAGGGTCAGCTTGC screening/RT A45'RV TAAACGCACACGTGACACGCCACAAA screening/RT A75'RV TCAGAGGCTTTGGCTCTTCATGGTTGTG screening/RT A85'RV TCACCACAAGTGTTGCAGATGGGA screening/RT A4HVRIIFW TGACATGTGATTGTTGGCCGTCGT RT A4HVRIIRV AATCGCCTCCGTCGATGATCGTTT RT A7HVRIIFW ACATGAATGGTGACGTAGCAGCCCTT RT A7HVRIIRV ACCGCAGCTTATGACATGGATTGCCT RT A8HVRIIFW GCAAAGCGAGAAGAACTTGATGCTGC RT A8HVRIIRV TTTACAGAGTCGGGAACACCGCCATT RT UBQ5FW ACACCAAGCCGAAGAAGATCAAGCAC RT UBQ5RV AAATGACTCGCCATGAAAGTCCCAGC RT 132 Cloned promoter fragments were subsequently digested from cloning vectors with SdaI/AvrII or HindIII/AvrII and ligated into the P35S::LUC-CesA backbone, which had the 35S promoter removed by digestion with SdaI/AvrII or HindII/AvrII. LR Clonase II reactions were performed between the binary destination vectors and pENTR-CesA vectors to create PA4::LUC-CesA4, PA7::LUC-CesA7, PA8::LUC-CesA8. PCesA::YFP- CesA vectors were created in a similar fashion, however, using ClaI and AvrII to remove the 35S promoter and replaced with the CesA promoter, followed by an LR Clonase II reaction with pENTR-CesA vectors to create PA4::YFP-CesA4, PA7::YFP-CesA7, and PA8::YFP-CesA8. The promoter-tag-gene fragments of all binary vectors were verified by DNA sequencing. A binary construct containing the LUC-YFP fragment under the control of the AtCesA4 promoter (PA4) was also created as a comparative positive control to the P35S::LUC-YFP control. The LUC-YFP fragment was amplified by PCR adding an AvrII site to the 5\u00EF\u0082\u00A2 terminus and a KpnI site to the 3\u00EF\u0082\u00A2 terminus (primers in Table 4.1), and cloned into the pBLUNT cloning vector (Invitrogen). The YFP fragment was removed from the PAtCesA4::YFP binary vector by digestion of endogenous AvrII/KpnI sites. The LUC-YFP fragment was cut and isolated from the cloning vector by a AvrII/KpnI double digest, and subsequently ligated into the previously linearized PAtCesA4::YFP vector, to create PAtCesA4::LUC-YFP. 4.2.2 Plant growth and transformations Wild-type (WT, Landsberg ecotype), AtCesA8irx1-1 (Taylor et al., 2003), AtCesA7irx3-1 (Taylor et al., 1999), and AtCesA4irx5-1 (Taylor et al., 2003) plants were grown and used for transformations for BRET assays. Seeds were germinated on half- concentration MS medium (Mirashige and Skoog, 1962) with no sucrose under continuous light. Seedlings were transferred to soil approximately seven days post germination, and grown in a growth chamber at 21oC under a 16-hour light/8-hour dark cycle. Plants were transformed using a method modified from Clough and Bent (1998). Agrobacterium tumefaciens GV3101-pMP90 (Hellens et al., 2000) was transformed with individual binary vectors (listed above) using a freeze-thaw method. Cultures were grown overnight at 28oC in Luria-Bertani medium containing 50 \u00EF\u0081\u00ADg L-1 133 kanamycin (YFP vectors) or 50 \u00EF\u0081\u00ADg L-1 spectinomycin (LUC vectors), 25 \u00EF\u0081\u00ADg L-1 rifampicin and 25 \u00EF\u0081\u00ADg L-1 gentamycin. The overnight culture was centrifuged and resuspended in a 5% (w/v) sucrose solution to an OD600 of at least 0.8. Silwet L-77 (LEHLE Seeds) was added to each resuspended culture at a final concentration of 0.02% (v/v). Newly flowering plants (approximately 4 weeks old) were sprayed with the Agrobacterium solution using a fine mist spray nozzle. Sprayed plants were placed in a dark, humid environment for 16 to 24 hours and then returned to the light. A second spraying was done five days after the first spray to increase the transformation rate. After all spray treatments, plants were maintained as usual and seeds were harvested from mature, dried plants. Seeds were screened for transformations by germinating on half- concentration MS medium (no sucrose) with the addition of either 75 \u00EF\u0081\u00ADg L-1 kanamycin for YFP vectors, 50 \u00EF\u0081\u00ADg L-1 glufosinate ammonium sulfate for LUC vectors (or both for double transformations). Seedlings that successfully grew on antibiotics were grown in soil for further analysis. Transformed plants were confirmed by PCR screening of genomic DNA using primers listed in Table 4.1). 4.2.3 BRET methods BRET assays were performed as described by Subramanian et al. (2004 and 2006). Briefly, luminescence spectra were measured from seedlings (cotyledons alone, hypocotyls/roots alone, or whole seedlings) or stem cross sections. Seedlings were five to seven-days old, and individual plantlets were tested for BRET. For stem cross sections, stem segments spanning the lower 3cm to 5cm from the stem base from 21 to 30-day-old plants were excised and hand sectioned to produce transverse sections, which were then used as the tissue sample for BRET testing. All BRET assays were performed with tissue samples immersed in 180 \u00EF\u0081\u00ADl water in a well of a white 96-well Optiplate (Perkin Elmer), with native coelenterazine (Invitrogen) added to a final concentration of 2 \u00EF\u0081\u00ADM. Measurements were made using a PerkinElmer Victor3V microplate reader equipped with emission filters to capture LUC emissions (486/10 nm) and YFP emissions (530/10 nm). After the addition of coelenterazine, a protocol was initiated consisting of a 5-second shake, a 5-second pause, followed by sequential 0.5- second emission readings through the blue and yellow filters. Emission readings were 134 repeated two to eight times. BRET calculations were performed as described by Bacart et al. (2008). To calculate the BRET ratio, YFP emission values were divided by LUC emission values (BRET = YFPemission/LUCemission). To correct for background emission, the ratio in untransformed tissue was subtracted from all ratios calculated for experimental samples (BRETcorrected = BRETtest sample \u00E2\u0080\u0093 BRETuntransformed). The final ratios were multiplied by 1000 to be expressed as milliBRET units (milliBRET = BRETcorrected x 1000). To measure YFP abundance, YFP fluorescence intensity signals were recorded after a single excitation through a 490nm laser, from which background signal from untransformed plants was subtracted (YFP abundance = YFPtest sample-YFPuntransformed). To measure LUC abundance, the luminescence signal was recorded after the addition of coelenterazine, from which background signal from untransformed plants was subtracted (LUC abundance = LUCtest sample \u00E2\u0080\u0093 LUCuntransformed). 4.2.4 BiFC in transiently transformed tobacco leaf cells Tobacco leaf cells were transiently transformed with BiFC expression vectors (combinations of P35S::NYFP-CesA and P35S::CYFP-CesA) and control expression vectors (P35S::LUC-YFP, P35S::YFP-CesA7) following the methods of Sparkes et al. (2006). Briefly, leaves of 5-week-old tobacco (Nicotiana tabacum) were transformed by infiltration with Agrobacterium tumefaciens carrying the expression vectors. YFP fluorescence was detected 48 and 65-hours post-infiltration through a Leica DMI6000 inverted microscope with a Quorum Wave FX system, which has a modified Yokogawa CSU-10 spinning disk scan head (Yokogawa Electric Corporation). YFP was excited with a 491 nm laser and emissions passed through a 528/38 band filter (Chroma Technology). Images were acquired using a Hamamatsu 9100-13 EMCCD camera (Hamamatsu) controlled by Volocity software (Improvision). 135 4.3 Results 4.3.1 BRET in Arabidopsis seedlings and stem tissues The use of BRET to detect protein-protein interactions in vivo depends on a few key factors: 1) that interacting proteins are within 10 nm, and ideally closer; 2) that interactions are stable and abundant enough to be detected within the timeframe of a BRET assay (generally 20 minutes); and, 3) the stoichiometry of interacting proteins facilitates Luciferase to be within 10 nm of YFP. With these considerations in mind it was decided to explore the use of various tissue types from which to measure BRET, and to measure BRET multiple times over 20 minutes. 4.3.1.1 Comparing controls and tissue types A very strong positive control consists of the LUC protein directly fused to YFP, and constitutively expressed by the 35S promoter. Stably transformed Arabidopsis plants expressing such a control construct (P35S::LUC-YFP) were used as a positive control in the BRET experiments. To determine if the use of a P35S-driven positive control was appropriate for a comparison to potential CesA-CesA interactions, we compared the effectiveness of an AtCesA promoter-driven LUC-YFP fusion as a second positive control (PA4::LUC-YFP). To compare the two controls, the BRET signal was measured from various tissues of Arabidopsis plants, and the YFP expression pattern was observed using confocal fluorescence microscopy. As expected, there were varied YFP expression patterns between the controls observed by confocal fluorescence microscopy (Figure 4.2). P35S-driven expression produced abundant YFP signal, present in most tissues at all stages of development (Figure 4.2, A-C). In contrast, the PA4-driven YFP expression was surprisingly abundant in cotyledons but less so in young roots (compared to P35S::LUC-YFP). Additionally, in stem tissue, PA4-YFP was highly localized to xylem and fibres (Figure 4.2-G), whereas P35S-driven YFP was mostly found in pith parenchyma. LUC and YFP intensity, and BRET ratios agreed with the visual expression patterns observed: P35S::LUC-YFP had the highest intensities and BRET ratios in seedlings (Figure 4.3, A-B), but in stem cross sections the two controls exhibited similar BRET signals (Figure 4.3-C). 136 Figure 4.2. YFP expression patterns in BRET control plants observed using confocal fluorescence microscopy. YFP was visualized in cotyledons (A, D) and roots (B, E, and F) of 5-day-old seedlings and in longitudinal sections of maturing stem tissue (C, G). P35S::LUC-YFP transformed plants are imaged in A-C, and PA4::LUC-YFP in D-G. White arrow heads indicate xylem vessels. Scale bars represent 50 \u00EF\u0081\u00ADm (A, B, D) and 10 \u00EF\u0081\u00ADm (C, E, F, G). 137 Figure 4.3. Varied expression and BRET of the LUC-YFP control. Comparisons are made between two control constructs: P35S::LUC-YFP (P35S::L-Y) and the AtCesA4 promoter-driven control, PA4::LUC-YFP. (A) Relative LUC and YFP emission intensity in five-day-old seedlings. (B) Average milliBRET in seedlings compared to the LUC- only negative control (P35S::LUC-AtCesA4). (C) Average milliBRET in stem cross sections, compared to two negative controls (P35S::L-A4 and PA4::L-A4). (D) LUC-YFP transcript abundance in stem tissue of the control lines. Transcript abundance is calculated relative to AtUBQ5. Error bars are standard error of the mean. A and B, n=16 (P35S) or 32 (PA4); C, n=6 (all samples); D, n=3 (all samples). 138 Transcript abundance of the LUC-YFP fusion gene appeared to be lower in PA4::LUC- YFP stem tissue as measured by real-time PCR (Figure 4.3-D), and was reflected in lower overall LUC and YFP emission intensities in these plants compared to P35S::LUC-YFP. Despite the reduced transcript abundance, LUC luminescence, and YFP emissions detected in PA4::LUC-YFP stem tissue, the BRET ratios were similar to those of P35S::LUC-YFP stem sections, which suggests that there are saturation limits to the amount of LUC-YFP signal necessary to measure significant BRET interactions. From this assessment there were clear differences between the two control constructs in seedlings. Therefore, I chose to conduct BRET assays in both whole seedlings and stem cross sections to account for potential differences in spatial and temporal gene expression patterns. 4.3.1.2 CesA7-CesA7 interactions AtCesA7irx3-1 mutants transformed with two AtCesA7-BRET constructs (P35S::LUC-A7 x PA7::YFP-A7) were used to detect homodimerization of CesA7. Abundant YFP signal had previously been detected in seedling tissue, thus five-day-old plantlets were chosen as a suitable tissue for measuring BRET interactions. Although LUC and YFP signal was detectable during BRET assays, the signals were very low, and BRET ratios were always equivalent to, or lower than, ratios of the LUC-only negative control (Table 4.2). Gene expression analysis suggested elevated A7 transcript abundance in the transgenic lines. However, this failed to translate into elevated active protein levels as evidenced by the low LUC and YFP emissions. 4.3.1.3 CesA7-CesA8 interactions AtCesA8irx1-1 mutant plants were transformed with two BRET constructs (PA8::LUC-A8 x P35S::YFP-A7) to measure the interaction between AtCesA8 and AtCesA7. LUC and YFP emissions were measured, and BRET ratios subsequently calculated, in whole seedlings, cotyledons only, and roots only. In cotyledons, there was a slight but not always statistically significant increase in BRET ratios of LUC-A8 x YFP-A7 plants compared to the negative control (Figure 4.4, B-C). In whole seedlings the BRET ratio of LUC-A8 x YFP-A7 samples was significantly higher than that of the negative control. The ratio values, however, were extremely low. 139 Table 4.2. YFP, LUC, BRET, and transcript abundance in whole seedlings carrying two AtCesA7-BRET constructs. AtCesA7irx3-1 plants transformed with P35S::AtCesA7-LUC and PA7::YFP-AtCesA7 (irx3-P35S::A7-L x PA7::Y-A7) are compared to the BRET positive control (WT x P35s::L-Y) and a negative control expressing a LUC-construct only (irx3-P35S::A7-L). LUC and YFP abundance is average relative emission intensity units. MilliBRET is the average ratio of YFP to Luciferase emissions (x1000) after the addition of the Luciferase substrate, coelenterazine (note that milliBRET values are not calculated form the LUC and YFP abundance values, abundance is a separate test from BRET). AtCesA7 (A7) transcript abundance is relative to AtUBQ5. Standard error of the mean is in parenthesis, n=8 except for transcript abundance where n=3. Line LUC Abundance YFP abundance MilliBRET A7 transcript abundance WT x P35S::L-Y 1091.2 (279.5) 110708.2 (16960.3) 1.18 (0.15) 0.15 (0.04) irx3-P35S::A7-L 40.0 (7.0) 1832.6(1100.2) 0.01 (0.03) na irx3-P35S::A7-L x PA7::Y-A7 51.2 (8.7) 6587.8 (1343.7) 0.01 (0.03) 0.40 (0.00) 140 P35S::L-Y P35S::L-A8 A8xA7 0 50 100 150 200 250 300 A v e ra g e m il li B R E T milliBRET roots P35S::L-Y P35S::L-A8 A8xA7 0 100 200 300 400 1500 1600 1700 1800 1900 2000 A ve ra g e m ill iB R E T milliBRET cotyledons P35S::L-Y P35S::L-A8 A8xA7 -200 -150 -100 -50 0 50 400 450 500 550 A v e ra g e m ill iB R E T whole seedlings WT IRX1 A8XA7 0.0 0.2 0.4 0.6 0.8 20 40 60 80 100 120 140 160 180 200 Line R e la ti v e t ra n s c ri p t a b u n d a n c e AtCesA8 AtCesA7 141 Figure 4.4. BRET measurements and transcript abundance in PA8::LUC-CesA8 x P35S::YFP-CesA7 seedlings. BRET measurements were made after the addition of the Luciferase substrate, coelenterazine, to seedling roots (A), cotyledons (B), and whole seedlings (C). Tissues of AtA8irx1-1plants transformed with PA8::LUC-CesA8 and P35S::YFP-CesA7 (PA8::L-A8 x P35s::Y-A7) are compared to the positive control (P35S::L-Y) and the negative control (P35S::L-A8) plants and tissues. (D) Relative transcript abundance of AtCesA7 and AtCesA8 transcripts in PA8::L-A8 x P35s::Y-A7 seedlings compared to wild-type (WT) and AtA8irx1-1 seedlings. Transcript abundance is relative to that of AtUBQ5. Seedlings for all tests were seven days old. Error bars are standard error of the mean, n=8 (A-C) or n=3 (D). Stars indicate statistical significance compared to the negative control (p<0.05, t-test). 142 AtCesA7 and AtCesA8 transcript levels were found to be significantly higher In whole seedlings of the A8irx1-1-PA8::L-A8 x P35s::Y-A7 transgenic line, compared to the levels measured in untransformed wild-type and AtCesA8irx1-1 seedlings., suggesting that there was sufficient expression of the BRET constructs in the transgenic plants. Combined, these results suggest that a very weak or transient interaction between CesA7 and CesA8 might occur in seedlings. 4.3.1.4 CesA7-CesA4 interactions Crossing of AtCesA7irx3-1-PA7::YFP-A7 plants with AtCesA4irx5-1-PA4::LUC-A4 plants produced double transgenic plants carrying both PA7::YFP-A7 and PA4::LUC-A4 constructs. As seen in Table 4.3, LUC and YFP emission intensities varied greatly among the transgenic lines. In particular, the LUC signal is quite low in all crosses but this is likely the result of low expression in the AtCesA4irx5-1-PA4::LUC-A4 parental line. YFP-CesA7 signal was visible in the crossed lines, and resembled that of the original parent line (data not shown). Although there were no statistically significant differences between the BRET ratios of crossed plants and the negative control, the crossed lines showed a general trend of higher BRET in stem tissue compared to the negative control (Figure 4.5-A). Additionally, weak BRET interactions were also detected in seedling tissue. The variation is quite large in all samples evaluated, illustrating the difficulty in accurately measuring live protein-protein interactions in vivo. Large variation could also be an indication of transient or unstable interactions. It must be noted that the double transgenic plants recovered from the AtCesA7irx3-1-PA7::YFP-A7 X AtCesA4irx5-1- PA4::LUC-A4 cross were not selected for A7irx3-1 and A4irx5-1 homozygosity. Therefore, the transgenic plants could express a combination of endogenous wild-type, transgenic, and irx CesA4 and A7 proteins, which could impact BRET interactions and intensity. 4.3.1.5 CesA8-CesA8 interactions Several lines of transgenic wild-type (Landsberg) plants were recovered from a double transformation with two CesA8-BRET constructs: PA8::LUC-AtCesA8 and PA8::YFP-AtCesA8. Although LUC and YFP signals were detected in stem tissues of these transgenic lines, there was no BRET signal to indicate interaction between the 143 tagged CesA8 subunits (Figure 4.6-A). However, when BRET ratios were calculated over time in seedlings of one of the double transgenic lines a weak BRET ratio was detected that was significantly greater than the negative control at more than 50% of the time points measured (Figure 4.6-B). These results imply a transient interaction between CesA8 subunits, one that might be stronger in seedlings than in stem tissue. 144 Table 4.3. LUC and YFP emissions of LUC-A4 x YFP-A7 transgenic plants. Transgenic plants are compared to the P35S::LUC-YFP positive control ((P35S::L-Y) and the parental lines used to create the double transgenic plants (PA4::L-A4 and PA7::Y-A7). Double transgenic plants carry both the PA4::LUC-AtCesA4 and PA7::YFP-AtCesA7 constructs, and offspring from reciprocal crosses are shown (AtCesA4xA7 compared to AtCesA7xA4). Values represent relative intensity units, standard error of the mean in parentheses. Colouration is to highlight the different groups of plants tested (white \u00E2\u0080\u0093 controls, light grey \u00E2\u0080\u0093 crosses, dark grey-reciprocal crosses). Line LUC emission YFP emission P35S::L-Y 14728 (4096) 41502 (4903) PA4::L-A4 17 (12) .-5000 (1880) PA7::Y-A7 .-3 (12) 4023 (393) A4 x A7 #1 43 (11) 3506 (388) A4 x A7 #4 19 (3) 1912 (689) A4 x A7 #12 27 (8) 1882 (244) A4 x A7 #14 76 (15) 2797 (635) A7 x A4 #1 23 (5) 1753 (288) A7 x A4 #3 51 (9) 2218 (419) A7 x A4 #5 20 (6) 2828 (671) 145 Figure 4.5. BRET measured in stem tissue and seedlings of double transgenic plants carrying PA4::LUC-CesA4 and PA7::YFP-CesA7. MilliBRET values of the P35S::LUC- YFP positive control (L-Y) the PA4::LUC-A4 negative control (L-A4), and the crossed lines expressing LUC-A4 and YFP-A7 are presented. (A) BRET measured in stem cross sections. \u00E2\u0080\u0095C\u00E2\u0080\u0096 and \u00E2\u0080\u0095A\u00E2\u0080\u0096 lines are LUC-A4 x YFP-A7 plants. (B) Measurements from seedlings of the C14 line (LUC-A4 x YFP-A7). Error bars are standard error of the mean, n=8 (A-C) or n=3 (D). LY L-A4 C1 C4 C12 C14 A1 A3 A5 -- 0 50 100 150 200 250 300 350 400 A v e ra g e m ill iB R E T Line Stem LY L-A4 A4 x A7 -200 -150 -100 -50 0 50 100 1300 1400 1500 1600 1700 A v e ra g e m ill iB R E T Line Seedlings 146 Figure 4.6. BRET measurements in AtCesA8xA8 double transgenic plants. Plants expressing PA8::LUC-A8 and PA8::YFP-A8 (L-A8 x Y-A8) were compared to the P35S::LUC-YFP positive control (P35S::L-Y or LY) and the PA8::LUC-A8 negative control (PA8::L-A8 or L8). (A) BRET in maturing stem tissue from six individual L-A8 x Y-A8 lines (#21-27). (B) BRET measured over time (after substrate addition) in seven-day- old seedlings. Error bars are standard error of the mean, n=16 except for PA8::L-A8 (n=8). Stars in B indicate statistical significance compared to PA8::L-A8 (p<0.05, ANOVA followed by a Tukey\u00E2\u0080\u0099s post-hoc test ). LY L8 #21 #22 #23 #24 #25 #27 -300 -200 -100 0 100 200 300 A v e ra g e m ill iB R E T 0 2 4 6 8 10 12 14 16 18 20 -200 0 200 800 1000 1200 1400 * * ** A ve ra g e m ill iB R E T Time (min) P35S::L-Y PA8::L-A8 L-A8 x Y-A8 W T I R 147 4.3.2 BiFC in transiently transformed tobacco epidermal cells The BiFC assay was employed in attempts to visually detect CesA-CesA interactions in planta. Leaves from five-week-old Nicotiana tabacum were transiently transformed with combinations of N-YFP-CesA and C-YFP-CesA constructs. If the two tagged proteins interact it would be possible for the N-YFP and C-YFP fragments to reconstitute a functional YFP protein, which can then be visualized using fluorescence microscopy. To verify that YFP-tagged AtCesAs could be visualized in transiently transformed leaf tissue, leaves were infiltrated with YFP-CesA7 and YFP-CesA4 constructs. It was determined that YFP-CesA signal could not be detected after 24 hours, but that 48 to 65 hours post-infiltration was an optimal time to visualize signal (Figure 4.7). In transformed leaf cells, the YFP-CesA signal appeared as bright, large, puncta that moved along the edges of the cell. Much of the signal at cell edges is thought to be due to vacuole displacement. Additionally, most of the movement was thought to be due to cytoplasmic streaming. This pattern of fluorescence is similar to that of YFP-CesA7 in leaf tissue of Arabidopsis (data not shown). As expected with localized infiltration of the tissue, fluorescently labelled cells were found in clusters surrounding the site of infiltration. After infiltration with BiFC-tagged combinations of AtCesAs, leaf tissue was scanned for fluorescence as an indication of CesA-CesA interactions. In general, fluorescence observed in the BiFC tests was not greater than the background signal observed in negative controls (Figure 4.7 E-L). However, some weak fluorescence potentially indicating interaction was observed in the CesA4-homodimerization test (Figure 4.7, I-J), and the CesA8-homodimerization test (Figure 4.7, K-L). No signal above background was observed in the CesA7-homodimerization test or any of the AtCesA-heterodimer tests (e.g.: A4xA8, A4xA7). 148 Figure 4.7. BiFC assays in leaf epidermal cells of Nicotiana tabacum. YFP-fluorescence images are paired with the corresponding bright field image. Images were taken at 10x magnification (A, C, E, G, I, K) and 63x magnification (B, D, F, H, J, L). YFP-CesA7 (A-B) and YFP-CesA4 (C-D) were used as positive controls. Leaves infiltrated with the two BiFC constructs (P35S::N-YFP x P35S::C-YFP) were used as a negative control (E-F). CesA homodimer combinations are presented in panels G-L: CesA7xCesA7 (G-H), CesA4xCesA4 (I-J), and CesA8xCesA8 (K-L). Scale bars represent 50 \u00EF\u0081\u00ADm (A, C, E, G, I, K) and 50 um (B, D, F, H, J, L). 149 4.4 Discussion 4.4.1 Putative CesA homodimers and heterodimers detected in planta using BRET and BiFC The BRET results presented herein suggest there could be interactions between CesA8-A8 in seedlings, and potential interactions between CesA7 and CesA4 (strongest in stem), and CesA7 and CesA8 (seedlings). BiFC results suggested homodimerization of CesA8, and potentially CesA4, but not CesA7, although all positive BiFC results were weak. Together these results suggest there may be interactions between some of the CesAs in vivo, but they are difficult to detect. One limitation of the BRET and BiFC assays performed in this study is the potential interference of endogenous CesAs with the BRET- and BiFC-tagged CesAs. Previous experiments demonstrated that the YFP-AtCesA7, LUC-AtCesA4, and LUC- AtCesA8 fusions were functional in planta, based on full or partial complementation of AtCesAirx mutant phenotypes by expression of these fusion proteins (Chapter 2). When YFP-AtCesA7 was expressed in AtCesA7irx3-1 mutants, under the expressional control of the native AtCesA7 promoter, the mutant A7irx3-1 phenotype was completely rescued, based on plant stature and the absence of collapsed xylem (Figure 2.1, Chapter 2). These results suggested that the YFP-AtCesA7 fusion protein was functional and thus likely accessing the CSCs and rosettes similar to wild-type AtCesA7 subunits. When LUC-AtCesA4 and LUC-AtCesA8 were expressed (under expressional control of the native AtCesA4 and AtCesA8 promoters) in the A4irx5-1 and A8irx1-1mutant plants, respectively, only partial complementation of the mutant phenotype was achieved (Figure 2.1, Chapter 2). Partial complementation is likely due to low transgene expression in the case of LUC-A4 in A4irx5-1 plants. Partial complementation of the A8irx1-1mutant phenotype by LUC-AtCesA8 is likely the result of interference of mutant AtCesA8 subunits with LUC-AtCesA8 subunits. As discussed in Chapter 2, the non-null nature of the A8irx1-1 mutation likely results in production of mutant AtCesA8 subunit, which would therefore compete with LUC-AtCesA8 for positions within the CSC and rosette. For this reason, despite partial complementation, it is believed that the LUC-AtCesA4 and LUC-AtCesA8 fusions are functional in planta. It has not been 150 overlooked, however, that partial complementation could also signify that tagged-CesA4 and CesA8 do not function at wild-type levels. Additionally, the interference that may result from the expression of two tagged-CesAs in one plant may also hinder function. Some double-transgenic plants were created for BRET, but they were not always in a double-CesA mutant background. For example, the AtCesA8irx1-1 mutants expressing LUC-A8 and YFP-A7 would contain endogenous CesA7 proteins, as well as mutant CesA8 proteins (Taylor et al., 2003), which could have competed with the LUC- and YFP-fused CesAs for positions within the CSC. Such interference would reduce any potential BRET signal. The same interference could occur in tobacco leaf cells, if AtCesAs are capable of interacting with NtCesAs. Also, expression of two AtCesAs in tobacco leaf could result in interactions that are not normally present in Arabidopsis due to the fact that only two of the three AtCesAs required for CSC formation are present in the transiently transformed leaf cells. These limitations must be considered if future BRET or BiFC experiments are to be performed to study CesA-CesA interactions. In vitro assays performed by others have suggested that interactions occur between all of the secondary cell wall AtCesAs (A4, A7, and A8), although there are some conflicting results. Homodimerization of CesA4, A7, and A8, and heterodimerization between A4-A7, A7-A8, and A8-A4 has been reported using epitope- tagging purification and anti-CesA antibody labelling (Atanassov et al., 2009) and immunoprecipitation followed by anti-CesA antibody labelling (Taylor et al., 2000; Taylor et al., 2003). The proposed formation of AtCesA7 homodimers (Atanassov et al., 2009) does not agree with the yeast-two hybrid results of Timmers et al. (2009), and BRET and BiFC results presented here that suggested a lack of homodimerization of AtCesA7. Compared to the BiFC results of Timmers et al. (2009), our results suggest only very weak interactions, and fewer interactions. There were three key differences between the BiFC assay performed in this work, compared to that of Timmers et al. (2009): 1) here, 5-week-old leaves were infiltrated, whereas they used younger, 3-week- old tissue; 2) here, BiFC signal was measured 48 and 65 hours post-infiltration, whereas they looked at tissue 72-hours post-infiltration; 3) they included the start codon within the cloned and subsequently tagged CesAs, whereas herein we eliminated the start codon for the N-tagged fusions. These differences appear subtle, but could 151 explain some of the varied results. For example, the younger tissue used by Timmers et al. (2009) may have facilitated higher protein expression, thus enabling greater detection of interactions after infiltration. Conflicting results regarding CesA-CesA interactions could also be attributed to the system employed to study interactions. In vitro systems are clearly limited by a lack of simulation of the natural environment in which proteins would interact. Perhaps there are other mechanisms in planta that regulate when, where, and if CesAs interact and that these are limited in in vitro systems. An interaction observed in vitro suggests that two proteins can physically interact but not necessarily that they always do interact in planta. As identified here there were clear differences between expression patterns among tissues. It is highly possible that CesA interactions vary temporally and spatially, which adds to the complexity of studying interactions using an in vivo system. Transient or unstable interactions could also be difficult to measure in vivo. Interactions could also be affected by post-translational modifications. For example, the phosphorylation of serine residues within the N-terminal variable region appears to target CesA7 protein for degradation (Taylor et al., 2007). In fact, excess CesA7 produced in a transformed AtCesA7irx3-1 mutant line (over-expressing a tagged CesA7) was found to result in phosphorylation and subsequent degradation of the excess CesA7, likely as a mechanism to regulate the levels of active CesA7 or remove CesAs that are not part of a CSC (Taylor et al., 2007). Post-translational modifications may hamper the ability to measure CesA-CesA interactions by BRET in stably transformed Arabidopsis plants. If a tagged form of a CesA protein is not optimal for CSC formation and cellulose production, it is possible that the tagged-CesA will be degraded or prevented access to the CSC, thus reducing the potential BRET signal. Atanassov et al. (2009) noted that tagging of AtCesA4 and AtCesA8 did not result in fully functional CesAs (whereas tagging of AtCesA7 produced a functional enzyme). Some of the weak interactions detected and reported here could be the result of reduced function of tagged CesAs, and reduced production and placement of tagged- CesAs within functional CSCs. 152 4.4.2 Modifications to the CSC and rosette models Current models propose that cellulose microfibrils are produced by a rosette terminal complex (or rosette) which may be composed of six lobes (six cellulose synthesizing complexes, CSCs), each of which may contain six individual CesA proteins (Scheible et al., 2001; Doblin et al., 2002). It has been suggested that that each CSC contains three types of CesA: \u00EF\u0081\u00A2, \u00EF\u0081\u00A11, and \u00EF\u0081\u00A12, and that there are unequal ratios of each CesA (Doblin et al., 2002; Ding and Hemmel, 2006), as depicted in Figure 4.8A. Based on the proposed six-CesA composition of CSCs, and the idea that CSC-CSC interactions occur to form a rosette, Doblin et al. (2002) postulates that a CSC could contain three \u00EF\u0081\u00A2-CesA subunits, two \u00EF\u0081\u00A12-CesA subunits, and one \u00EF\u0081\u00A11-CesA subunit, and that CesA-interactions within the CSC could consist of \u00EF\u0081\u00A2-\u00EF\u0081\u00A11, \u00EF\u0081\u00A2-\u00EF\u0081\u00A12, and \u00EF\u0081\u00A12-\u00EF\u0081\u00A12. Ding and Hemmel (2006) propose fewer interactions within a CSC, only \u00EF\u0081\u00A11-\u00EF\u0081\u00A2 and \u00EF\u0081\u00A12-\u00EF\u0081\u00A2, but that \u00EF\u0081\u00A2-\u00EF\u0081\u00A2 interactions may exist between CSCs to form a rosette. Both models, then, only account for three CesA-CesA interactions: two heterodimers and one homodimer. Interaction studies (presented here and in the literature) do not agree with these models if it is assumed that each CesA is one of either \u00EF\u0081\u00A11, \u00EF\u0081\u00A12, or \u00EF\u0081\u00A2\u00EF\u0080\u00A0subunits, and that all three are required in a CSC. The in vivo results from Timmers et al. (2009) and presented herein suggest five possible interactions: A4- A4, A8-A8, A4-A7, A4-A8, and A7-A8. Thus, if the rosette models hold true, then more than one CesA must be able to fill the proposed \u00EF\u0081\u00A2, \u00EF\u0081\u00A11, and \u00EF\u0081\u00A12 positions (Figure 4.8 B). Recently, a new model proposed by Carpita (2010) suggests that CesAs form dimers to produce a single glucan chain and dimer-dimer interactions also occur for CSCs and rosettes to form. This model may account for the large variety of interactions currently reported, using both in vitro and in vivo methods. 153 Figure 4.8. CSC composition revisited. Models of the CesA-composition of CSCs have been proposed (Ding and Hemmel, 2006; Doblin et al., 2002) suggesting combinations of CesA-CesA interactions within a CSC and between CSCs in a rosette (A). Here, it is proposed that CSC composition may be somewhat flexible; that a pool of CesA proteins could somewhat randomly associate to form complete CSCs during secondary cell wall formation (B). As such, it would be expected that a mixed population of CSCs may exist, and therefore CesA-CesA interactions will vary, perhaps depending on tissue type and developmental stage. Models in A are based on those of Ding and Hemmel (2006), and Doblin et al. (2002). 154 AtCesA4 and A8 seem like suitable candidates to fill the \u00EF\u0081\u00A2 position, as they appear to form homodimers and interact with all other CesAs. Perhaps A7 could fill the \u00EF\u0081\u00A11 position which does not form homodimers in the current models. It is possible that the designations change depending on cell type and developmental stage (Ding and Hemmel, 2006). However, it still remains to be resolved if CesA-CesA interactions are specific and, if so, if the specificity is required for proper cellulose biosynthesis. Perhaps the composition of the CSC is somewhat flexible, such that they can form by different combinations of CesA-CesA interactions. This would mean that CesA-CesA interactions are required for CSC formation (and possibly rosette formation), and function, but that the specificity of interactions is not a limiting factor for CSC formation. The CesA interactions within a CSC may therefore be more random than specific, and the composition of CSCs could vary greatly spatially and temporally within a plant. The use of BRET and BiFC has provided some insight into CesA-CesA interactions in vivo, but has also generated more questions about the composition of CSCs. There is still a great deal of work required to clearly elucidate what interactions occur, if they are tissue- and development-specific, and the nature of the interactions. Does CesA8 always form homodimers? Do these homodimers then interact with CesA4 and CesA7? Optimization of the BiFC assay utilizing Arabidopsis double CesA mutants to avoid interference from endogenous CesAs will likely provide a great deal more information than has been achieved using BiFC in tobacco, or BRET in Arabidopsis. 155 CHAPTER 5 Conclusion 5.1 Thesis summary In recent years, a substantial amount of information regarding the genes and proteins involved in cellulose biosynthesis of both the primary and secondary cell walls has become available in a variety of plants including trees, particularly poplar. From mutant analysis, it appears that AtCesA1, A3, and A6 are key CesA subunits involved in primary cell wall cellulose biosynthesis (Arioli et al., 1998; Scheible et al., 2001; Fagard et al., 2000). Additionally, AtCesA2 and A5 may be active, partially redundant forms of AtCesA6 (Desprez et al., 2007; Persson et al., 2007b). The roles of AtCesA10 and AtCesA9 are less well characterized, although AtCesA9 may play a role in secondary cell wall formation in seed coats (Stork et al., 2010). We know that a mutation in any one of AtCesA4, A7, or A8 causes aberrant cellulose production and secondary cell wall structure (Taylor et al., 2000; Taylor et al., 2003; Gardiner et al., 2003; Taylor et al., 2004), resulting in cell wall defects that presumably prevent xylem vessels from withstanding the pressure of water transport resulting in collapse, and potentially weakening the walls of supportive fibre cells in the stem (Zhong et al., 2003). However, the precise mechanism by which the cellulose synthase proteins and the cellulose synthesizing complex (CSC) affect cellulose properties such as abundance, crystallinity, microfibril angle, and degree of polymerization are still unknown. From broadening our understanding of how plants regulate the inherent properties of cellulose in secondary cell walls, to the application for transgenic modifications of cellulose for downstream industrial applications, it is of interest to understand how the individual CesA protein subunits interact to form a fully functioning CSC and rosette, and ultimately how the CSC composition influences cellulose properties. This was the focus of the research presented in Chapter 2. The results obtained provided some support for our hypothesis, that each AtCesA has a unique role in cellulose production that could be manifested in altered cellulose properties, and additionally that the presence of a CSC influences cellulose ultrastructure independently of the abundance of cellulose biosynthesis. 156 The diversification of CesA genes into primary cell wall-specific or secondary cell wall-specific may have emerged prior to the divergence of gymnosperms from angiosperms. If this were true, it might be expected that CesA orthologs would have conserved functions. In planta evidence to suggest conservation of function between CesA orthologs has, however, not been determined. Elucidation of functional conservation may also lend information to understanding the composition of CSCs. Identifying CesAs from spruce and poplar, and determining if they are truly orthologous and have conserved functions with Arabidopsis was precisely the goal of the research presented in Chapter 3, and the results obtained supported our hypothesis: isolated spruce genes share similar conserved domains, exhibit the highest gene expression in tissues undergoing significant secondary cell wall formation, and both spruce and poplar CesAs are functionally orthologous to the Arabidopsis counterparts. Based on phylogenetic analysis, there appears to be a great deal of conservation between Arabidopsis CesAs and those from trees such as poplar, Eucalyptus, and even gymnosperms such as pine and spruce. Thus, information obtained from studying Arabidopsis secondary cell wall formation may be useful for application in a tree model. Current models of CSC form and function do not address how or why multiple CesAs are required to synthesize normal cellulose. The capability of CesAs to interact as homodimers or heterodimers has been suggested by in vitro and in vivo experiments but further investigation using in vivo methods are required to accurately ascertain if specific CesA-CesA interactions occur in order for CSCs to form and functionally give rise to cellulose polymers. Studying such interactions (Chapter 4), however, only partially supported our hypothesis. We did observe that CesA homodimers and heterodimers may form, but were unable to determine confidently that the interactions are specific and consistent in particular tissues or during a defined developmental stage. However, the development of a system to visualize YFP-CesAs in live fibre cells suggested to us that revisiting the use of BiFC in Arabidopsis CesA mutant plants may provide a better model to study CesA-CesA interactions in vivo. Combined, the results of the research presented in this thesis contribute to the current model of CSC composition and function. 157 5.2 A complete CSC influences the fundamental properties of cellulose As presented in Chapter 2, the effects of the AtCesA4irx5-1, AtCesA7irx3-1, and AtCesA8irx1-1 mutations go beyond decreased cellulose production. In addition to changes to cellulose properties (MFA, DP, crystallinity) the pleiotropy observed included changes to total growth, growth rate, and lignin and hemicellulose content. The secondary effects of altered cellulose, potentially on various developmental cues, appeared to vary depending on mutant phenotype severity and could be linked more specifically to the combined changes in cellulose crystallinity, DP, and MFA. Ultimately, the presence of a complete CSC appeared to dictate these fundamental cellulosic properties under long-day conditions. Our results suggest that a properly formed CSC does not have to contain all functional CesAs in order to produce cellulose with wild- type like properties. The phenotype of AtCesA8irx1-1, then, can be solely attributed to reduced cellulose in secondary cell walls, and not aberrant cellulose structure. To that end, the more severe pleiotropy observed in the AtCesA4irx5-1 and AtCesA7irx3-1 mutants are likely a result of stronger feedback by other developmental processes in response to changes in cellulose properties, or simply in response to more deficient cell walls. When a CSC is able to form even with a mutant CesA subunit, the resulting cellulose polymers are mostly of normal length (DP), the secondary cell wall is highly crystalline, and cellulose microfibrils have a normal MFA (under long day conditions). This strongly suggests that the purpose of the CSC, in addition to cellulose biosynthesis, is to facilitate a stable localized environment where newly synthesized glucan chains can crystallize. However, under long-day conditions, it does not appear that a high degree of crystallinity is correlated with a large DP. In contrast, it appears that the presence of AtCesA4 has a larger influence on DP than the presence of a fully formed CSC. The link between cellulose MFA and other cellulose properties, however, is not quite clear. There are some suggestions that rate of movement of CSCs affects microfibril organization (Himmelspach et al., 2003; Wasteneys and Fujita, 2006) and thus influences MFA. It could also be that a fully formed CSC is required to interact with another protein or complex that dictates MFA. Disorganized orientation of microfibrils has been found in response to mutations in various proteins, all of which may affect MFA. These include kinesins (Zhong et al., 2002), COBRA (Roudier et al., 2005), and a 158 CesA-interacting protein (CSI-1) with unidentified function (Gu et al., 2010). In transgenic poplar with RNAi-suppressed KORRIGAN, cell wall crystallinity increased, and MFA was smaller, but a link between the two cell wall characteristics was not clearly defined (Maloney and Mansfield, 2010). Greater crystallinity (as found in wild- type and AtCesA8irx1-1) may facilitate proper microfibril organization, and therefore a smaller MFA (as was found in wild-type plants grown under long day conditions). It is clear that the mechanisms that regulate cellulose MFA and DP require further investigation. 5.3 The function of AtCesA4, A7, and A8 orthologs is highly conserved The ability of spruce and poplar AtCesA orthologs to complement the AtCesAirx mutant phenotype suggests a very strong conservation of function among CesAs from evolutionarily divergent species. It is postulated that complementation of the AtCesAirx lines is the result of meeting a threshold of functional CesA, which then permits CSC formation and sufficient cellulose biosynthesis. This is the first indication of inter- species CesAs capable of integrating with AtCesA subunits to form a CSC, permits normal cell wall formation as exhibited by normal growth phenotype and a non- collapsed xylem of the complemented lines. Additionally, this suggests that the CesA- specific functions are maintained, despite as low as 64% amino acid sequence similarity (PgCesA1 with AtCesA8). There were subtle differences in the level of complementation observed, which may be attributed to slight differences in function among the orthologs. It was observed that even a low level of LUC-PgCesA1 transcript resulted in a complemented phenotype of AtCesA8irx1-1 plants. Conversely, low expression of the wild-type form of LUC- AtCesA8 or LUC-PtiCesA8-A in AtCesA8irx1-1 plants only partially complemented the mutant phenotype. This suggests that the LUC-PgCesA1 protein can readily access the CSC, or that the function of PgCesA1 is less affected by the LUC tag compared to LUC-AtCesA8 and LUC-PtiCesA8-A. Also, observing degrees of complementation could be a function of the level of CesA protein produced in the transgenic lines. Transcript abundance suggests some lines had elevated transgene expression, whereas others had low levels of gene expression, but how do these levels of transcript translate into active protein units? Post-translational modifications to CesAs, such as 159 phosphorylation (Taylor, 2007), may be at play, which could limit the amount of active AtCesA8 and PtiCesA8 in AtCesA8irx1-1 transgenic lines. A combination of the mutant genotype and post-translational modifications targeting transgenic CesAs for degradation may explain why PtiCesA8-A only partially complemented AtCesA8irx1-1 but PtiCesA7-A and PtiCesA4 more easily rescued the AtCesA7irx3-1 or AtCesA4irx5-1 mutants, respectively (based on the observed number of complemented lines). As discussed previously, the AtCesA8irx1-1 mutant is believed to continue to synthesize mutant AtCesA8 proteins, whereas little to no AtCesA7 protein is produced in AtCesA7irx3-1 mutants. Therefore, if there is a required threshold of CesAs in order for CSCs to form, and beyond that threshold CesA proteins are targeted for degradation, then the threshold will be more easily met in AtCesA8irx1-1 plants, whereas more transgenic CesA protein will be produced to meet the necessary CesA thresholds in AtCesA7irx3-1. Based on this threshold hypothesis, it is thus postulated that various populations of CSCs exist in the PgCesA1- and PtiCesA8-A- AtCesA8irx1-1 transgenic plants. The combinations of CSCs could be (using the PgCesA1 transgenic line as an example): CSC1: AtCesA4, AtCesA7, and non-functional AtCesA8. CSC2: AtCesA4, AtCesA7, PgCesA1. CSC3: AtCesA4, AtCesA7, non-functional AtCesA8, and PgCesA1. Under the assumption that PgCesA1, PtiCEsA8-A, PtiCEsA7-A, and PtiCEsA4 were properly integrated into CSCs within the transgenic plants, there are two generalized conclusions that can be made by these findings: The first is the structure of these AtCesA orthologs is sufficiently conserved to be integrated into the CSC. Although Kurek et al. (2002) determined that the zinc-binding domain of cotton CesA1 and A2 could form homodimers and heterodimers, Timmers et al. (2009) showed that mutations in conserved motifs of the zinc-binding domain did not eliminate CesA-CesA interactions in vitro. Therefore, the zinc-binding domain may be sufficient but not essential for CesA proteins to putatively interact and become part of a CSC. This may be why PgCesA1, which shares about 40% similarity with the AtCesA8 N-terminal region zinc binding domain, is able to become part of the CSC. It was demonstrated with AtCesA1-AtCesA3 chimeric proteins that the catalytic domain was sufficient to permit entry into the CSC (Wang et al., 2006) suggesting that this region could be involved in CesA-CesA interactions required for CSC formation. Secondly, our results 160 indicate that the catalytic function and role within the CSC of these AtCesA orthologs is highly similar to that of AtCesA8, thus allowing for normal cell wall formation to occur. This is the first report of functional conservation between CesA orthologs, one which suggests that CesA form and function has not diverged significantly since putative CesA-specific functions evolved, which likely occurred prior to the evolutionary split of angiosperms and gymnosperms. 5.4 What is the ratio of CesAs within the CSC? The inability of CesA4, A7, and A8 to complement one another when one is mutated suggests that each CesA protein subunit has a unique function. It could be that a ratio of each CesA is required for proper CSC form and function, or that each CesA has a unique function in cellulose biosynthesis or dictating cellulose properties, or both. The alterations to cellulose structure in AtCesA4irx5-1 and AtCesA7irx3-1plants were not always similar, despite both being null-mutant lines (Chapter 2). The present results suggest that the AtCesA7 enzyme may influence cellulose crystallinity more so than the other CesAs, and that AtCesA4 could have a unique role in producing long cellulose chains. The ability of a spruce and poplar CesA to take the place of AtCesA8, and poplar CesAs to substitute AtCesA7 and AtCesA4 suggests that CesA-specific functions for the AtCesA8, A7, and A4 orthologs are highly conserved. Thus, it seems possible that the following scenarios could be correct: a CSC must contain all three CesAs, the ratio of which is somewhat flexible, and that the uniqueness of each CesA is due to both interactions they facilitate and the cellulose properties they dictate. Interestingly, Song et al. (2010) recently reported that in poplar xylem there are CSCs which putatively contain five different CesA subunits: PttCesA7-A, PttCesA7-B, PttCesA8-A, PttCesA8-B, and PttCesA4. This contradicts the previous dogma that implied a three-CesA-CSC, given that the A and B isoforms are so similar. However, these results could also support the idea that a flexible ratio of CesAs is required for CSC formation and function, one which can be met by three or more CesAs. Various CSC models have been proposed, classifying the CesAs within the CSC depending on their interactions (Doblin et al., 2002; Ding and Hemmel, 2006). Interaction studies suggest more possibilities than the current CSC model, and therefore if these models hold true, then redundancy in the CesA must exist (to fill the 161 \u00EF\u0081\u00A11, \u00EF\u0081\u00A12, and \u00EF\u0081\u00A2\u00EF\u0080\u00A0positions, see Chapter 4, and Figure 4.8). Based on the CesA-CesA interactions putatively detected using the BRET and BiFC assays (Chapter 4), as well as those reported by Timmers et al. (2009), it is believed that AtCesA4 and A8 are suitable candidates to occupy the\u00EF\u0080\u00A0\u00EF\u0081\u00A2 position, as they appear to form homodimers and interact with all other CesAs. Perhaps CesA7 could fill the \u00EF\u0081\u00A11 position, which does not appear to form homodimers in the current models. However, this contradicts the idea that AtCesA7 is required at higher levels in the CSC than the other CesAs (Chapter 2), compared to the models that indicate lower levels of \u00EF\u0081\u00A11 compared to \u00EF\u0081\u00A12 and \u00EF\u0081\u00A2. Is it possible that CesAs can exist in the CSC without interacting with other CesAs? Clearly, there is a need to further investigate the composition of the CSC, and how CesA-CesA interactions are related to CSC composition. 5.5 Should the cell wall-specific classifications of CesAs be revisited? It is clear from mutant studies that CesA4, A7, and A8 affect cellulose biosynthesis of secondary cell walls. Additionally, mutant studies have determined that AtCesA1, A3, and A6 (and the potential A6 redundancies A2, A5, and A9) have a clear role in primary cell wall cellulose biosynthesis. However, results from some of the work presented in this thesis, and observations by other groups suggest that there could be some dual functionality, suggesting that our current model of primary or secondary wall- specificity are not fully representative of the roles of CesAs in cellulose biosynthesis. Dual functionality may be subtle, in some cases, but it begs the question: do the roles of CesAs need to be re-evaluated? From this thesis, observations of YFP-AtCesA7 expression in vivo revealed a diverse (in terms of tissue type) expression pattern, including expression in primary cell- walled cells (Chapter 2). The YFP-AtCesA7 fusion was under expressional control of the native AtCesA7 promoter, thus expression should be temporally and spatially precise with respect to native developmental cues and events. Additionally, the fusion protein was expressed in the AtCesA7irx3-1 mutant background, so there would be no competition with endogenous, functional CesA7 proteins. Therefore, the expression patterns observed are believed to be a true representation of where the CesA7 protein is normally present. In support of our observations, Bosca et al. (2006) also reported a potential role for CesA7 in primary cell walls. The authors identified an AtCesA7 mutant 162 plant, mur10, based on the changes to primary cell wall carbohydrates in leaves. This questioned whether the primary wall changes were a secondary effect of the AtCesA7 mutation, or if AtCesA7 has an active role in cellulose biosynthesis of primary cell walls. As reported in this thesis, the YFP-AtCesA7 was found in primary-walled cells in roots, cotyledons, mature leaves, and stems. Compared to AtCesA4 and A8, AtCesA7 shares a higher degree of similarity to moss (Physcomitrella patens) CesAs, (Roberts and Bushoven, 2007). This may be an indirect indication that AtCesA7 is less divergent compared to the other secondary cell wall-specific CesAs, perhaps retaining some function in primary cell walls. Dual functionality in both primary and secondary cell walls may also be true for AtCesA4 and AtCesA9. Under the expressional control of the native AtCesA4 promoter, a LUC-YFP fusion protein was observed to be highly expressed in cotyledons (non-vein tissue), mature leaves, and primary walled tissues in the stem, although to a lesser degree compared to the YFP-AtCesA7 fusion (Chapter 4). Also, AtCesA9, previously classified as primary wall-specific (Persson et al., 2007b; Desprez et al., 2007) was recently identified as a CesA required for secondary cell wall biosynthesis in seed coats (Stork et al., 2010). Is AtCesA9 functioning with other primary wall-specific CesAs to produce secondary cell walls in seed coats? Persson et al. (2007b) hypothesized that CesA9, A1, and A3 form the functional CSCs in floral organs and pollen. In hybrid poplar, Song et al. (2010) isolated CSCs from developing xylem and found that both primary and secondary cell wall-specific CesAs were expressed. CesA- interactions reported in large complexes isolated by immunoprecipitation do not support the conclusion that combinations of primary and secondary cell wall CesAs interact (Atanassov et al., 2009; Wang et al., 2008; Song et al., 2010). If CesAs do have dual functionality in both primary and secondary walls, it could be that combinations of primary and secondary CSCs are present, as suggested by Song et al. (2010). Further investigation into the expression patterns of all CesAs (particularly using fluorescent tagging), and perhaps more in depth analysis of CesA-mutant phenotypes, may shed light on whether or not CesAs have overlapping functions in both primary and secondary cell wall cellulose biosynthesis. 163 5.6 Critical comments and future work recommendations 5.6.1 Understanding the secondary effects of AtCesA4irx5-1, A7irx3-1, and A8irx1-1 mutations The reduced stature phenotype of the AtCesAirx mutants, identified in Chapter 2, strongly suggests that there are secondary effects of these mutations on primary growth. In an effort to more clearly elucidate the role of CesAs in primary growth, or to clarify if the effects measured are purely a secondary response, it would be interesting to identify the physiological reasons for reduced growth in the mutants. Using light microscopy could provide images of the cellular structure in these mutants and, as such, may offer a means to precisely measure the cellular dimensions in elongating stem tissues, to clarify if reduced stature is the result of fewer cell divisions of if the cells produced undergo less longitudinal expansion. 5.6.2 Investigate CSC composition and function more thoroughly It has not been overlooked that there are additional AtCesA4, A7, and A8 mutants that could be studied to determine if the changes in cell wall composition are affected by the type of genetic mutation. In particular, it would be interesting to study the effects of a null-AtCesA8 mutation, such as AtCesA8lew2-1 (Chen et al., 2005) or AtCesA8irx1-5 (Brown et al., 2005) and determine how it compares to the non-null mutant presented herein and additionally with the null AtCesA4irx5-1 and AtCesA7irx3-1 mutants. In addition, I am curious to know if non-null mutations in AtCesA4 and A7 would produce a similar phenotype as the AtCesA8irx1-1 mutant. Under the assumption that CSCs form in AtCesA8irx1-1, but have reduced cellulose biosynthesizing capacity, one might assume that an incomplete knockout of either AtCesA4 or AtCesA7 should have a similar phenotype as the AtCesA8irx1-1 mutants for the characteristics that are greatly affected by CSC structure, such as MFA. However, if the role of each CesA is different in the CSC, or if unequal ratios of each CesA are required for CSCs to form and function, the severity of the mutant phenotype in non-null A4 and A7 mutants might vary. 164 To better understand the effect of AtCesA4, A7, and A8 mutations on CSC form and function in secondary cell wall biosynthesis it would be interesting to compare the abundance and movement of CSCs in the various mutant backgrounds by utilizing YFP- CesA fusions. However, directly tracking the movement of CSCs in secondary cell wall forming tissues such as xylem or fibres, in planta, has not been possible due to interference from surrounding cells. Thus, it would be advantageous to utilize a system that could permit direct detection, such as that described by Yamaguchi et al. (2010). This system allows for cells in whole Arabidopsis plants to be induced to trans- differentiate into xylem vessels, including leaf and root epidermal cells, which would help eliminate interference. If this system could be applied to null-CesA mutant plants then it would be possible to study the effects of a CesA mutation on CSC patterning and movement. For example, the AtCesA7irx3-1-YFP-CesA7 line crossed with AtCesA4irx5-1 to produce a double mutant, but complemented with YFP-CesA7 so only one CesA is non-functional. These double-mutant-YFP-CesA7 plants could then be used to create the xylem trans-differentiation system, and subsequently used to determine CSC movement and patterns. Do CSCs move more slowly when one CesA is not- functional? Are they less abundant in the plasma membrane? Answers to these questions may provide information about the reasons for aberrant cellulose produced in CesA-mutants. The establishment of a xylem-inducible system would also be useful for documenting CesA interactions. Although BiFC results presented herein and by Timmers et al. (2009) have been informative, it cannot be overlooked that the interactions observed could be confounded by 1) expression of AtCesAs in tobacco that contain functional, endogenous CesAs, which could interfere with AtCesA-AtCesA interactions; and 2) expression of secondary cell wall-specific AtCesAs in primary cell walled cells. Therefore, we cannot confidently assume that the interactions observed in vivo thus far are representative of those that occur in secondary cell wall synthesizing cells such as xylem vessels and fibres. Alternatively, BiFC could be conducted and observed in Arabidopsis stem tissue. Based on the observation of YFP-CesA7 in longitudinal sections of mature and developing stem (Chapter 2) I suggest that a CesA-BiFC system be set up utilizing AtCesAirx mutant backgrounds. This would allow for interactions between CesAs to be 165 visualized in planta, in appropriate tissues, over developmental stages, and without interference from endogenous wild-type proteins. Additionally, such a system could be expanded to visualize the interactions between CesAs and other candidate proteins such as CSI-1, SUSY, and COBRA, for example. 5.6.3 Broaden our understanding of conservation of function among CesAs From the results presented in Chapter 3 it is clear that PgCesA1 and poplar CesAs are functional orthologs of Arabidopsis CesA proteins. In order to determine the extent of functional conservation it would be worthwhile determining the cell wall properties of the PgCesA- and PtiCesA-complemented plants, and comparing them to the properties of AtCesA-mutant lines (Chapter 2). If the PgCesA and PtiCesA subunits are functioning identically to the orthologous AtCesA subunits, then wild-type like cell wall properties would be expected in complemented lines. The complementation results presented herein also elicit questions about how easily AtCesAirx mutants can be complemented with other orthologous, and even non-orthologous CesAs. If the ability to complement is tightly linked to meeting required thresholds of CesA protein, then it could be hypothesized that orthologous and even non-orthologous CesAs could complement the mutant phenotype if they were temporally and spatially expressed at appropriate levels. It was not resolved if PgCesA2 and A3 are functional orthologs of the secondary cell wall CesAs of Arabidopsis. Further attempts to transform the AtCesAirx plants with PgCesA2 and A3 would be recommended. Also, it would be worthwhile to identify additional members of the PgCesA genes family to enable a more accurate comparison of homologous sequences. 5.7 Final conclusions and relevance Understanding the cellular mechanisms of cellulose biosynthesis has been a challenging area of research since the first identification of rosettes, CSCs, and CesA genes. This is particularly true of the CesAs involved in the synthesis of cellulose in the secondary walls, which have the added complexity and limitations of visualization. 166 The work presented in this thesis further supports the complexity of the relationship between CesAs, CSCs, and the cellulose product. The key findings of this thesis are as follows: 1. AtCesAirx mutants provide an invaluable system to study the role of each CesA in cellulose biosynthesis (Chapter 2). 2. A fully-formed CSC influences cellulose properties such as MFA (Chapter 2). 3. Day-length affects MFA in Arabidopsis 4. Longitudinal stem sections are a good system for visualizing the expression pattern of YFP-fused CesAs and measuring CesA-CesA interactions (Chapter 2, 4). 5. The newly isolated spruce CesA genes PgCesA1, A2, and A3 are likely involved in secondary cell wall cellulose biosynthesis (Chapter 3). 6. PgCesA1 and poplar PtiCesA8-A, PtiCesA7-A, and PtiCesA4 are functionally orthologous to AtCesA8, A7, and A4, respectively (Chapter 3). 7. CesA-CesA interactions occurring in vivo may include homodimers of CesA8 and CesA4, but not CesA7, CesA4-CesA7 heterodimers in stem tissue, and CesA8- CesA7 heterodimers in seedlings (Chapter 4). From these findings, it is speculated that: 8. AtCesA7 has a greater presence in CSCs, affecting cellulose crystallinity and abundance (Chapter 2, 3) 9. AtCesA4 may influence cellulose degree of polymerization. 10. The composition of the CSC may vary depending on tissue and developmental stage, such that the CesA-CesA interactions vary (Chapter 2, 3, and 4). The conclusions drawn from this thesis have enhanced our understanding of cellulose biosynthesis and the information herein can be used to guide new research questions in the pursuit of elucidating how CesAs and CSCs are intricately involved in cellulose biosynthesis. 167 REFERENCES Anderson, C. T., Carroll, A., Akhmetova, L. and Somerville, C. (2010). Real-Time Imaging of Cellulose Reorientation during Cell Wall Expansion in Arabidopsis Roots. Plant Physiol 152, 787-796. Andersson-Gunneras, S., Mellerowicz, E. J., Love, J., Segerman, B., Ohmiya, Y., Coutinho, P. M., Nilsson, P., Henrissat, B., Moritz, T., and Sundberg, B. (2006). Biosynthesis of cellulose-enriched tension wood in Populus: global analysis of transcripts and metabolites identifies biochemical and developmental regulators in secondary wall biosynthesis. The Plant Journal 45, 144-165. Appenzeller, L., Doblin, M., Barreiro, R., Wang, H., Niu, X., Kollipara, K., Carrigan, L., Tomes, D., Chapman, M., and Dhugga, K. S. (2004). Cellulose synthesis in maize: isolation and expression analysis of the cellulose synthase (CesA) gene family. Cellulose 11, 287-299. Arioli, T., Peng, L., Betzner, A., Joanne, B., Wittke, W., Herth, W., Camilleri, C., Hofte, H., Plazinski, J., Birch, R., et al. (1998). Molecular Analysis of Cellulose Biosynthesis in Arabidopsis. Science 279, 717-720. Atanassov, I. I., Pittman, J. K., and Turner, S. R. (2009). Elucidating the Mechanisms of Assembly and Subunit Interaction of the Cellulose Synthase Complex of Arabidopsis Secondary Cell Walls. Journal of Biological Chemistry 284, 3833-384. Bacart, J., Corbel, C., Jockers, R., Bach, S., Couturier, C. (2008). The BRET technology and its application to screening assays. Biotechnology Journal 3, 311-324. Barnett, J. R., and Bonham, V.A. (2004). Cellulose microfibril angle in the cell wall of wood fibres. Biological Reviews 79, 461-472. Baskin, T. I. (2001). On the alignment of cellulose microfibrils by cortical microtubules: A review and a model. Protoplasma 215, 150-171. Beeckman, T., Przemeck, G. K. H., Stamatiou, G., Lau, R., Terryn, N., De Rycke, R., Inze, D., and Berleth, T. (2002). Genetic Complexity of Cellulose Synthase A Gene Function in Arabidopsis Embryogenesis. Plant Physiol 130, 1883-1893. Bjurhager, I., Olsson, A.-M., Zhang, B., Gerber, L., Kumar, M., Berglund, L. A., Burgert, I., Sundberg, B. r. and Salme\u00C3\u008C \u00C2\u0081n, L. (2010). Ultrastructure and Mechanical Properties of Populus Wood with Reduced Lignin Content Caused by Transgenic Down-Regulation of Cinnamate 4-Hydroxylase. Biomacromolecules 11, 2359-2365. Boerjan, W., Ralph, J., and Baucher, M. (2003). Lignin biosynthesis. Annual Review Of Plant Biology 54, 519-546. Bosca, S., Barton, C. J., Taylor, N. G., Ryden, P., Neumetzler, L., Pauly, M., Roberts, K., and Seifert, G. J. (2006). Interactions between MUR10/CesA7 Dependent 168 Secondary Cellulose Biosynthesis and Primary Cell Wall Structure. Plant Physiol 142, 1353-1363. Bowling, A. J. and R. M. Brown (2008). The cytoplasmic domain of the cellulose- synthesizing complex in vascular plants. Protoplasma 233, 115-127. 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\u00E2\u0080\u00932295. Brown, D. M., Goubet, F., Wong, V. W., Goodacre, R., Stephens, E., Dupree, P., and Turner, S. R. (2007). Comparison of five xylan synthesis mutants reveals new insight into the mechanisms of xylan synthesis. The Plant Journal 52, 1154-1168. Brown, D. M., Zhang, Z., Stephens, E., Dupree, P., and Turner, S. R. (2009). Characterization of IRX10 and IRX10-like reveals an essential role in glucuronoxylan biosynthesis in Arabidopsis. The Plant Journal 57, 732-746. Brown, J., M.R., and Montezinos, D. (1976). Cellulose microfibrils: Visualization of biosynthetic and orienting complexes in association with the plasma membrane. PNAS 73, 143-147. Brown, R. M. (2004). Cellulose structure and biosynthesis: What is in store for the 21st century? Journal of Polymer Science Part A: Polymer Chemistry 42, 487-495. Brown, R. M., Willison, J. H. M., and Richardson, C. L. (1976). Cellulose Biosynthesis in Acetobacter xylinum: Visualization of the Site of Synthesis and Direct Measurement of the in vivo Process. PNAS 73, 4565-4569. Burk, D. H., and Ye, Z.-H. (2002). Alteration of Oriented Deposition of Cellulose Microfibrils by Mutation of a Katanin-Like Microtubule-Severing Protein. Plant Cell 14, 2145-2160. Burton, R., Shiley, N., King, B., Harvey, A., and Fincher, G. (2004). The CesA Gene Family of Barley. Quantitative Analysis of Transcripts Reveals Two Goups of Co- Expressed Genes. Plant Physiology 134, 224-236. Burton, R. A., Ma, G., Baumann, U., Harvey, A. J., Shirley, N. J., Taylor, J., Pettolino, F., Bacic, A., Beatty, M., Simmons, C. R., et al. (2010). A Customized Gene Expression Microarray Reveals That the Brittle Stem Phenotype fs2 of Barley Is Attributable to a Retroelement in the HvCesA4 Cellulose Synthase Gene. Plant Physiol 153, 1716-1728. Camilo Hern\u00C3\u00A1ndez-Blanco, D. X. F., Jian Hu, Andrea S\u00C3\u00A1nchez-Vallet, Laurent Deslandes, Francisco Llorente, Marta Berrocal-Lobo, Harald Keller, Xavier Barlet, Clara S\u00C3\u00A1nchez-Rodr\u00C3\u00ADguez, Lisa K. Anderson, Shauna Somerville, Yves Marco, and Antonio Molina (2007). Impairment of Cellulose Synthases Required for Arabidopsis Secondary Cell Wall Formation Enhances Disease Resistance. Plant Cell 19, 890-903. 169 Cano-Delgado, A., Penfield, S., Smith, C., Catley, M. and Bevan, M. (2003). Reduced cellulose synthesis invokes lignification and defense responses in Arabidopsis thaliana. The Plant Journal 34, 351-362. Carpita, N. C. (2010). Update on Mechanisms of Plant Cell Wall Biosynthesis: How plants make cellulose and other (1->4)-\u00EF\u0081\u00A2 -D-glycans. Plant Physiol, 110. Chan, J., Crowell, E., Eder, M., Calder, G., Bunnewell, S., Findlay, K., Vernhettes, S., Hofte, H. and Lloyd, C. (2010). The rotation of cellulose synthase trajectories is microtubule dependent and influences the texture of epidermal cell walls in Arabidopsis hypocotyls. J Cell Sci. 123, 3490-3495. Chen, Z., Hong, X., Zhang, H., Wang, Y., Li, X., Zhu, J., and Gong, Z. (2005). Disruption of the cellulose synthase gene, AtCesA8/IRX1, enhances drought and osmotic stress tolerance in Arabidopsis. The Plant Journal 43, 273-283. Citovsky, V., Gafni, Y., and Tzfira, T. (2008). Localizing protein-protein interactions by bimolecular fluorescence complementation in planta. Methods 45, 196-206. Clough, S. J., and Bent, A. F. (1998). Floral dip: a simplified method forAgrobacterium- mediated transformation of Arabidopsis thaliana. The Plant Journal 16, 735-743. Coleman, H., Beamish, L., Reid, A., Park, J.-Y., and Mansfield, S. (2010). Altered sucrose metabolism impacts plant biomass production and flower development. In Transgenic Research (Springer Netherlands) 19, 269-283. Coleman, H. D., Canam, T., Kang, K.-Y., Ellis, D. D., and Mansfield, S. D. (2007). Over- expression of UDP-glucose pyrophosphorylase in hybrid poplar affects carbon allocation. J Exp Bot 58, 4257-4268. Coleman, H. D., Ellis, D. D., Gilbert, M., and Mansfield, S. D. (2006). Up-regulation of sucrose synthase and UDP-glucose pyrophosphorylase impacts plant growth and metabolism. Plant Biotechnology Journal 4, 87-101. Coleman, H. D., Park, J.-Y., Nair, R., Chapple, C., and Mansfield, S. D. (2008a). RNAi- mediated suppression of p-coumaroyl-CoA 3\u00E2\u0080\u0099-hydroxylase in hybrid poplar impacts lignin deposition and soluble secondary metabolism. PNAS 105, 4501-4506. Coleman, H. D., Samuels, A. L., Guy, R. D., and Mansfield, S. D. (2008). Perturbed Lignification Impacts Tree Growth in Hybrid Poplar--A Function of Sink Strength, Vascular Integrity, and Photosynthetic Assimilation. Plant Physiol 148, 1229-1237. Coleman, H. D., Yan, J., and Mansfield, S. D. (2009). Sucrose synthase affects carbon partitioning to increase cellulose production and altered cell wall ultrastructure. PNAS 106, 13118-13123. Cosgrove, D. J. (2005). Growth of the plant cell wall. Nature Reviews Molecular Cell Biology 6, 850-861. 170 Crowell, E. F., Bischoff, V., Desprez, T., Rolland, A., Stierhof, Y.-D., Schumacher, K., Gonneau, M., Hofte, H., and Vernhettes, S. (2009). Pausing of Golgi Bodies on Microtubules Regulates Secretion of Cellulose Synthase Complexes in Arabidopsis. Plant Cell 21, 1141-1154. Crowell, E. F., Gonneau, M., Vernhettes, S., and H\u00C3\u00B6fte, H. (2010). Regulation of anisotropic cell expansion in higher plants. Comptes Rendus Biologies D\u00C3\u00A9veloppement v\u00C3\u00A9g\u00C3\u00A9tatif des plantes, Georges Pelletier, Jean-Fran\u00C3\u00A7ois Morot-Gaudy 333, 320-324. Delmer, D., and Amor, Y. (1995). Cellulose biosynthesis. The Plant Cell 7, 987-1000. Delmer, D. P. (1999). Cellulose Biosynthesis: Exciting Times for A Difficult Field of Study. Annual Review of Plant Physiology and Plant Molecular Biology 50, 245-276. Desprez, T., Juraniec, M., Crowell, E. F., Jouy, H., Pochylova, Z., Parcy, F., Hofte, H., Gonneau, M., and Vernhettes, S. (2007). Organization of cellulose synthase complexes involved in primary cell wall synthesis in Arabidopsis thaliana. PNAS 104, 15572-15577. Ding, S.-Y., and Himmel, M. E. (2006). The Maize Primary Cell Wall Microfibril: A New Model Derived from Direct Visualization. Journal of Agricultural and Food Chemistry 54, 597-606. Djerbi, S., Aspeborg, H., Nilsson, P., Sundberg, B., Mellerowicz, E., Blomqvist, K., and Teeri, T. (2004). Identification and expression analysis of genes encoding putative celluose synthases (CesA) in the hybrid aspen, Populus tremula (L.) x P. tremuloides (Michx.). Cellulose 11, 301-312. Djerbi, S., and Mats Lindskog, L. A., Fredrik Sterky and Tuula T. Teeri (2005). The genome sequence of black cottonwood (Populus trichocarpa) reveals 18 conserved cellulose synthase (CesA) genes. Planta 221, 739 - 746. Doblin, M. S., Kurek, I., Jacob-Wilk, D., and Delmer, D. P. (2002). Cellulose Biosynthesis in Plants: from Genes to Rosettes. Plant Cell Physiol 43, 1407-1420. Donaldson, L. (2007). Cellulose microfibril aggregates and their size variation with cell wall type. Wood Science and Technology 41, 443-460. Donaldson, L. (2008). Microfibril angle: measurement, variation and relationsips \u00E2\u0080\u0093 a review. IAWA Journal 29, 345-386. Fagard, M., Desnos, T., Desprez, T., Goubet, F., Refregier, G., Mouille, G., McCann, M., Rayon, C., Vernhettes, S., and Hofte, H. (2000). PROCUSTE1 Encodes a Cellulose Synthase Required for Normal Cell Elongation Specifically in Roots and Dark-Grown Hypocotyls of Arabidopsis. Plant Cell 12, 2409-2424. Fang, S. Z., Yang, W.Z., and Fu, X.X. (2004). Variation of microfibril angle and its correlation to wood properties in poplars. Journal of Forestry Research 15, 261-267. 171 Fisher, D. D. and Cyr, R. J. (1998). Extending the Microtubule/Microfibril Paradigm . Cellulose Synthesis Is Required for Normal Cortical Microtubule Alignment in Elongating Cells. Plant Physiol 116, 1043-1051. Franke, R., Hemm, M. R., Denault, J. W., Ruegger, M. O., Humphreys, J. M., and Chapple, C. (2002). Changes in secondary metabolism and deposition of an unusual lignin in the ref8 mutant of Arabidopsis. The Plant Journal 30, 47-59. Frey-Wyssling, A. (1968). The ultrastructure of wood. In Wood Science and Technology (Springer Berlin / Heidelberg), pp. 73-83. Fujii, S., Hayashi, T., and Mizuno, K. (2010). Sucrose Synthase is an Integral Component of the Cellulose Synthesis Machinery. Plant Cell Physiol 51, 294-301. Fujita, M. (2008) Cortical microtubules and physical properties of cellulose microfibrils during primary cell wall formation in Arabidopsis thaliana, University of British Columbia, Vancouver. Gardiner, J. C., Taylor, N.G., and Turner, S.R. (2003). Control of cellulose synthase complex localization in developing xylem. The Plant Cell 15, 1740-1748. Giddings, T. H. J., Brower, D.L., and Staehelin, A.L. (1980). Visualization of particle complexes in the plasma membrane of Micrasterias denticulata associated with the formation of cellulose fibrils in primary and secondary cell walls. Journal of Cell Biology 84, 327-339. Gu, Y., Kaplinsky, N., Bringmann, M., Cobb, A., Carroll, A., Sampathkumar, A., Baskin, T. I., Persson, S., and Somerville, C. R. (2010). Identification of a cellulose synthase- associated protein required for cellulose biosynthesis. PNAS 107, 12866-12871. Guerriero, G., Fugelstad, J., and Bulone, V. (2010). What Do We Really Know about Cellulose Biosynthesis in Higher Plants? Journal of Integrative Plant Biology 52, 161- 175. Gutierrez, R., Lindeboom, J. J., Paredez, A. R., Emons, A. M. C., and Ehrhardt, D. W. (2009). Arabidopsis cortical microtubules position cellulose synthase delivery to the plasma membrane and interact with cellulose synthase trafficking compartments. 11, 797-806. Ha, M.-A., MacKinnon, I. M., Sturcova, A., Apperley, D. C., McCann, M. C., Turner, S. R., and Jarvis, M. C. (2002). Structure of cellulose-deficient secondary cell walls from the irx3 mutant of Arabidopsis thaliana. Phytochemistry 61, 7-14. Haigler, C., and Brown, R. J. (1986). Transport of rosettes from the golgi apparatus to the plasma membrane in isoalted mesophyll cells of Zinnia elegans during differentiation to tracheary elements in suspension culture. Protoplasma 134, 111-120. 172 Hamann, T., Osborne, E, Youngs, H.L., Mission, J., Nussaume, L., and Somerville, C. (2004). Global expression analysis of CESA and CSL genes in Arabidopsis. Cellulose 11, 279-286. Handford, M., Baldwin, T., Goubet, F., Prime, T., Miles, J., Yu, X., and Dupree, P. (2003). Localisation and characterisation of cell wall mannan polysaccharides in Arabidopsis thaliana. Planta 218, 27-36. Harmer, S. L., Hogenesch, J. B., Straume, M., Chang, H.-S., Han, B., Zhu, T., Wang, X., Kreps, J. A., and Kay, S. A. (2000). Orchestrated Transcription of Key Pathways in Arabidopsis by the Circadian Clock. Science 290, 2110-2113. Hellens, R., Mullineaux, P., and Klee, H. (2000). Technical Focus:A guide to Agrobacterium binary Ti vectors. Trends in Plant Science 5, 446-451. Henriksson, M., and Berglund, L.A. (2007). Microfibrillated Cellulose Films. In: Navi, P., and Guidoum, A. (Ed.), Proceedings of the Third International Symposium on Wood Machining. Presses Polytechniques et Universitaires Romandes, Lausanne, Switzerland, pp. 17-20. Herth, W. (1983). Arrays of plasma-membrane rosettes involved in cellulose microfibril formation of Spirogyra. In Planta (Springer Berlin / Heidelberg), pp. 347-356. Himmelspach, R., Williamson,Richard E.,Wasteneys,Geoffrey O. (2003). Cellulose microfibril alignment recovers from DCB-induced disruption despite microtubule disorganization. The Plant Journal 36, 565-575. His, I., Driouich, A., Nicol, F. Jauneau, A., and Hofte, H. (2001). Altered pectin composition in primary cell walls of korrigan, a dwarf mutant of Arabidopsis deficient in a membrane-bound endo-1,4-\u00EF\u0081\u00A2-glucanase. Planta 212, 348-358. Holland, N., Holland, D., Helentjaris, T., Dhugga, K. S., Xoconostle-Cazares, B., and Delmer, D. P. (2000). A Comparative Analysis of the Plant Cellulose Synthase (CesA) Gene Family. Plant Physiol 123, 1313-1324. 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. Molecular Cell 9, 789-798. Hu, W.-J., Harding, S. A., Lung, J., Popko, J. L., Ralph, J., Stokke, D. D., Tsai, C.-J., and Chiang, V. L. (1999). Repression of lignin biosynthesis promotes cellulose accumulation and growth in transgenic trees. Nature Biotechnology 17, 808-812. Jeffries, T. W. (1990). Biodegradation of lignin-carbohydrate complexes. In Biodegradation (Springer Netherlands), pp. 163-176. Jones, L., Ennos, A. R., and Turner, S. R. (2001). Cloning and characterization of irx4: a severely lignin-deficient mutant of Arabidopsis. The Plant Journal 26, 205-216. 173 Joshi, C. (2003). Xylem-specific and tension stress-responsive expression of cellulose synthase genes from aspen trees. Applied Biochemistry and Biotechnology 105-108, 17-25. Joshi, C. P., Bhandari, S., Ranjan, P., Kalluri, U.C., Liang, X., Fujino, T., and Samuga, A. (2004). Genomics of cellulose biosynthesis in poplars. New Phytologist 164, 53-61. Joshi, C. P., and Mansfield, S. D. (2007). The cellulose paradox -- simple molecule, complex biosynthesis. Current Opinion in Plant Biology Physiology and Metabolism - Edited by Clint Chapple and Malcolm M Campbell 10, 220-226. Kerppola, T. K. (2008). Bimolecular Fluorescence Complementation (BiFC) Analysis as a Probe of Protein Interactions in Living Cells. Annual Review of Biophysics 37, 465- 487. Kim, H. J., and Triplett, B. A. (2001). Cotton Fiber Growth in Planta and in Vitro. Models for Plant Cell Elongation and Cell Wall Biogenesis. Plant Physiol 127, 1361-1366. Kimura, S., Laosinchai, W., Itoh, T., Cui, X., Linder, C. R., and Brown, Jr. R. M. (1999a). Immunogold Labeling of Rosette Terminal Cellulose-Synthesizing Complexes in the Vascular Plant Vigna angularis. Plant Cell 11, 2075-2086. Kimura, S., Sakurai, N., and Itoh, T. (1999b). Different Distribution of Cellulose Synthesizing Complexes in Brittle and Non-Brittle Strains of Barley. Plant Cell Physiol 40, 335-338. Kolosova, N., Miller, B., Ralph, S., Ellis, B.E., Douglas, C., Ritland, K., and Bohlmann, J. (2004). Isolation of high-quality RNA from gymnosperm and angiosperm trees. BioTechniques 36, 821-824. Krauskopf, E., Harris, P.J., and Putterill, J. (2005). The cellulose synthase gene PrCESA10 is involved in cellulose biosynthesis in developing tracheids of the gymnosperm Pinus radiata. Gene 350, 107-116. Kumar, M., Thammannagowda, S., Bulone, V., Chiang, V., Han, K.-H., Joshi, C. P., Mansfield, S. D., Mellerowicz, E., Sundberg, B., Teeri, T., and Ellis, B. E. (2009). An update on the nomenclature for the cellulose synthase genes in Populus. Trends in Plant Science 14, 248-254. Kurek, I., Kawagoe, Y., Jacob-Wilk, D., Doblin, M., and Delmer, D. (2002). Dimerization of cotton fiber cellulose synthase catalytic subunits occurs via oxidation of the zinc- binding domains. PNAS 99, 11109-11114. Lane, D. R., Wiedemeier, A., Peng, L., Hofte, H., Vernhettes, S., Desprez, T., Hocart, C. H., Birch, R. J., Baskin, T. I., Burn, J. E., Arioli, T., Betzner, A. S. and Williamson, R. E., (2001). Temperature-Sensitive Alleles of RSW2 Link the KORRIGAN Endo-1,4-{beta}- 174 Glucanase to Cellulose Synthesis and Cytokinesis in Arabidopsis. Plant Physiol 126, 278-288. Ledbetter, M.C. and Porter, K.R. (1963). A \"MICROTUBULE\" in plant cell fine structure. The Journal of Cell Biology 19:239-250. Liang, X., and Joshi, C.P. (2004). Molecular cloning of ten distinct hypervariable regions from cellulose synthase gene superfamily in Aspen trees. Tree Physiology 24, 543. Lichtenegger, H., Reiterer, A., Stanzl-Tschegg, S. E. and Fratzl, P. (1999). Variation of Cellulose Microfibril Angles in Softwoods and Hardwoods--A Possible Strategy of Mechanical Optimization. Journal of Structural Biology 128, 257-269. Liepman, A. H., Wightman, R., Geshi, N., Turner, S. R., and Scheller, H. V. (2010). Arabidopsis: a powerful model system for plant cell wall research. The Plant Journal 61, 1107-1121. Liepman, A. H., Wilkerson, C. G., and Keegstra, K. (2005 ). Expression of cellulose synthase-like (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. PNAS 102, 2221-2226. Lu, S., Li, L., Yi, X., Joshi, C. P., and Chiang, V. L. (2008). Differential expression of three eucalyptus secondary cell wall-related cellulose synthase genes in response to tension stress. J Exp Bot 59, 681-695. Lukowitz, W., Nickle, T. C., Meinke, D. W., Last, R. L., Conklin, P. L., and Somerville, C. R. (2001). Arabidopsis cyt1 mutants are deficient in a mannose-1-phosphate guanylyltransferase and point to a requirement of N-linked glycosylation for cellulose biosynthesis. PNAS 98, 2262-2267. MacKinnon, I., \u00C5\u00A0turcov\u00C3\u00A1, A., Sugimoto-Shirasu, K., His, I., McCann, C., and Jarvis, M. (2006). Cell-wall structure and anisotropy in procuste, a cellulose synthase mutant of Arabidopsis thaliana. Planta 224, 438-448. Maloney, V. J. (2010). Function, functional conservation and interactions of the membrane-bound endo-1,4-beta-glucnases orthologous to korrigan (Vancouver, The University of British Columbia). Maloney, V. J., and Mansfield, S. D. (2010). Characterization and varied expression of a membrane-bound endo-\u00EF\u0081\u00A2-1,4-glucanase in hybrid poplar. Plant Biotechnology Journal 8, 294-307. Matzke, A. J. M., and Matzke, M. A. (1998). Position effects and epigenetic silencing of plant transgenes. Current Opinion in Plant Biology 1, 142-148. Megraw, R., Leaf, G., Bremer, D. (1998). Longitudinal shrinkage and microfibril angle in loblolly pine. In Microfibril angle in wood, B. G. Butterfield, ed. (Christchurch, New Zealand, University of Canterbury Press), pp. 27-61. 175 Mellerowicz, E. J., Baucher, M., Sundberg, B. r., and Boerjan, W. (2001). Unravelling cell wall formation in the woody dicot stem. Plant Molecular Biology 47, 239-274. Mellerowicz, E. J., and Sundberg, B. (2008). Wood cell walls: biosynthesis, developmental dynamics and their implications for wood properties. Current Opinion in Plant Biology 11, 293-300. Mirashige, T., and Skoog, F. (1962). A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15, 473-497. Nairn, C. J., and Haselkorn, T. (2005). Three loblolly pine CesA genes expressed in developing xylem are orthologous to secondary cell wall CesA genes of angiosperms. New Phytologist 166, 907-915. Nairn, C. J., Lennon, D. M., Wood-Jones, A., Nairn, A. V., and Dean, J. F. D. (2008). Carbohydrate-related genes and cell wall biosynthesis in vascular tissues of loblolly pine (Pinus taeda). Tree Physiol 28, 1099-1110. Nicol, F., His, I., Jauneau, A., Vernhettes, S., Canut, H. and H\u00C3\u00B6fte, H. (1998). A plasma membrane-bound putative endo-1,4--D-glucanase is required for normal wall assembly and cell elongation in Arabidopsis. EMBO 17, 5563\u00E2\u0080\u00935576. Oomen, R. J. F. J., Tzitzikas, E. N., Bakx, E. J., Straatman-Engelen, I., Bush, M. S., McCann, M. C., Schols, H. A., Visser, R. G. F., and Vincken, J.-P. (2004). Modulation of the cellulose content of tuber cell walls by antisense expression of different potato (Solanum tuberosum L.) CesA clones. Phytochemistry 65, 535-546. Paredez, A. R., Persson, S., Ehrhardt, D. W., and Somerville, C. R. (2008). Genetic Evidence That Cellulose Synthase Activity Influences Microtubule Cortical Array Organization. Plant Physiol 147, 1723-1734. Paredez, A. R., Somerville, C. R., and Ehrhardt, D. W. (2006). Visualization of Cellulose Synthase Demonstrates Functional Association with Microtubules. Science 312, 1491- 1495. Paux, E., Carocha, V., Marques, C., Mendes de Sousa, A., Borralho, N., Sivadon, P., and Grima-Pettenati, J. (2005). Transcript profiling of Eucalyptus xylem genes during tension wood formation. New Phytologist 167, 89-100. Pear, J., Kawagoe, Y, Schreckengost, WE, Delmer, DP, Stalker, DM (1996). Higher Plants Contain Homologs of the Bacterial celA Genes Encoding the Catalytic Subunit of Cellulose Synthase. PNAS 93, 12637-12642. Peng, L., Hocart, C. H., Redmond, J. W. and Williamson, R. E. (2000). Fractionation of carbohydrates in Arabidopsis root cell walls shows that three radial swelling loci are specifically involved in cellulose production. Planta 211, 406-414. 176 Perrin, R. M. (2001). Cellulose: How many cellulose synthases to make a plant? Current Biology 11, R213-R216. Persson, S., Caffall, K. H., Freshour, G., Hilley, M. T., Bauer, S., Poindexter, P., Hahn, M. G., Mohnen, D., and Somerville, C. (2007a). The Arabidopsis irregular xylem8 Mutant Is Deficient in Glucuronoxylan and Homogalacturonan, Which Are Essential for Secondary Cell Wall Integrity. Plant Cell 19, 237-255. Persson, S., Paredez, A., Carroll, A., Palsdottir, H., Doblin, M., Poindexter, P., Khitrov, N., Auer, M., and Somerville, C. R. (2007b). Genetic evidence for three unique components in primary cell-wall cellulose synthase complexes in Arabidopsis. PNAS 104, 15566-15571 Persson, S., Wei, H., Milne, J., Page, G. P., and Somerville, C. R. (2005). Identification of genes required for cellulose synthesis by regression analysis of public microarray data sets. PNAS 102, 8633-8638. 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. 20, 744-749. Pfaffl, M. W. (2001). A new mathematical model for relative quantification in real-time RT-PCR. Nucl Acids Res 29, 45. Plomion, C., Leprovost, G., and Stokes, A. (2001). Wood Formation in Trees. Plant Physiology 127, 1513-1523. Rajangam, A. S., Kumar, M., Aspeborg, H., Guerriero, G., Arvestad, L., Pansri, P., Brown, C. J.-L., Hober, S., Blomqvist, K., Divne, C., et al. (2008). MAP20, a Microtubule-Associated Protein in the Secondary Cell Walls of Hybrid Aspen, Is a Target of the Cellulose Synthesis Inhibitor 2,6-Dichlorobenzonitrile. Plant Physiol 148, 1283-1294. Ranik, M., and Myburg, A. (2006). Six new cellulose synthase genes from Eucalyptus are associate with primary adn secondary cell wall biosynthesis. Tree Physiology 26, 545-556. Reiterer, A., Lichtenegger, H., Tschegg, S., and Fratzl, P. (1999). Experimental evidence for a mechanical function of the cellulose microfibril angle in wood cell walls. Philosophical Magazine 79, 2173-2184. Richmond, T. (2000). Higher plant cellulose synthases. Genome Biology 1, 3001.3001- 3001.3006. Richmond, T., and Somerville, C. (2000). The Cellulose Synthase Superfamily. Plant Physiology 124, 495-498. 177 Robert, S., Mouille, G., and Hofte, H. (2004). The mechanism and regulation of cellulose synthesis in primary walls: lessons from cellulose-deficient Arabidopsis mutants. Cellulose 11, 351-364. Roberts, A., and Bushoven, J. (2007). The cellulose synthase (CESA) gene superfamily of the moss Physcomitrella patens. Plant Molecular Biology 63, 207-219. Roberts, A. and Roberts, E., 2007. Evolution of the Cellulose Synthase (CesA) Gene Family: Insights from Green Algae and Seedless Plants. In: Cellulose: Molecular and Structural Biology), pp. 17-34. Roudier, F., Fernandez, A. G., Fujita, M., Himmelspach, R., Borner, G. H. H., Schindelman, G., Song, S., Baskin, T. I., Dupree, P., Wasteneys, G. O., and Benfey, P. N. (2005). COBRA, an Arabidopsis Extracellular Glycosyl-Phosphatidyl Inositol- Anchored Protein, Specifically Controls Highly Anisotropic Expansion through Its Involvement in Cellulose Microfibril Orientation. Plant Cell 17, 1749-1763. Samuga, A., and Joshi, C. P. (2002). A new cellulose synthase gene (PtrCesA2) from aspen xylem is orthologous to Arabidopsis AtCesA7 (irx3) gene associated with secondary cell wall synthesis. Gene 296, 37-44. Sato, S., Kato, T., Kakegawa, K., Ishii, T., Liu, Y.-G., Awano, T., Takabe, K., Nishiyama, Y., Kuga, S., Sato, S., Nakamura, Y., Tabata, S. and Shibata, D. (2001). Role of the Putative Membrane-Bound Endo-1,4-\u00C3\u008E\u00C2\u00B2-Glucanase KORRIGAN in Cell Elongation and Cellulose Synthesis in Arabidopsis thaliana. Plant and Cell Physiology 42, 251-263. Saxena, I. M., and Brown, Malcolm Jr. (2005). Cellulose Biosynthesis: Current Views and Evolving Concepts. Annals of Botany 96, 9-21. Saxena, I. M., Kudlicka, K., Okuda, K., and Brown, Jr., R.M. (1994). Characterization of genes in the cellulose-synthesizing operon (acs operon) of Acetobacter xylinum: implications for cellulose crystallization. Journal of Bacteriology 176, 5735-5752. Scheible, W.-R., Eshed, R., Richmond, T., Delmer, D., and Somerville, C. (2001). Modifications of cellulose synthase confer resistance to isoxaben and thiazolidinone herbicides in Arabidopsis Ixr1 mutants. PNAS 98, 10079-10084. Scheller, H. V., and Ulvskov, P. (2010). Hemicelluloses. Annual Review of Plant Biology 61, 263-289. Seagull, R. W. (1992). A Quantitative Electron Microscopic Study of Changes in Microtubule Arrays and Wall Microfibril Orientation During in vitro Cotton Fiber Development. J Cell Sci. 101, 561-577. Solomon, O. L., Berger, D. K., and Myburg, A. A. (2010). Diurnal and circadian patterns of gene expression in the developing xylem of Eucalyptus trees. South African Journal of Botany 76, 425-439. 178 Song, D., Shen, J., and Li, L. (2010). Characterization of cellulose synthase complexes in Populus xylem differentiation. New Phytologist 187, 777-790. Stoppin-Mellet, V., Gaillard, J., and Vantard, M. (2006). Katanin's severing activity favors bundling of cortical microtubules in plants. The Plant Journal 46, 1009-1017. Stork, J., Harris, D., Griffiths, J., Williams, B., Beisson, F., Li-Beisson, Y., Mendu, V., Haughn, G., and DeBolt, S. (2010). CELLULOSE SYNTHASE9 Serves a Nonredundant Role in Secondary Cell Wall Synthesis in Arabidopsis Epidermal Testa Cells. Plant Physiol 153, 580-589. Subramanian, C., Woo, J., Cai, X., Xu, X., Servick, S., Johnson, C. H., Nebenf\u00C3\u00BChr, A., and Arnim, A. G. v. (2006). A suite of tools and application notes for in vivo protein interaction assays using bioluminescence resonance energy transfer (BRET). The Plant Journal 48, 138-152. Subramanian C, X. Y., Johnson CH, von Arnim AG. (2004). In vivo detection of protein- protein interaction in plant cells using BRET. Methods in Moelcular Biology 284, 271- 286. Suzuki, S., Li, L., Sun, Y.-H., and Chiang, V. L. (2006). The Cellulose Synthase Gene Superfamily and Biochemical Functions of Xylem-Specific Cellulose Synthase-Like Genes in Populus trichocarpa. Plant Physiol 142, 1233-1245. Szyjanowicz, P. M., McKinnon, I., Taylor, N.G., Gardiner, J., Jarvis, M.C., and Turner, S.R. (2004). The irregular xylem 2 mutant is an allele of korrigan that affects the secondary cell wall of Arabidopsis thaliana. Plant Journal 37, 730-740. Tanaka, K., Murata, K., Yamazaki, M., Onosato, K., Miyao, A., and Hirochika, H. (2003). Three Distinct Rice Cellulose Synthase Catalytic Subunit Genes Required for Cellulose Synthesis in the Secondary Wall. Plant Physiol 133, 73-83. Taylor, J. G., Owen, T. P., Koonce, L. T., and Haigler, C. H. (1992). Dispersed lignin in tracheary elements treated with cellulose synthesis inhibitors provides evidence that molecules of the secondary cell wall mediate wall patterning. The Plant Journal 2, 959- 970. Taylor, N. (2007). Identification of cellulose synthase AtCesA7 (IRX3) in vivo phosphorylation sites\u00E2\u0080\u0094a potential role in regulating protein degradation. Plant Molecular Biology 64, 161-171. Taylor, N. G. (2008). Cellulose biosynthesis and deposition in higher plants. New Phytologist 178, 239-252. Taylor, N. G., Gardiner, J.C., Whiteman, R., and Turner, S.R. (2004). Cellulose synthesis in the Arabidopsis secondary cell wall. Cellulose 11, 329-338. 179 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. PNAS 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, 2529-2540. 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. Timmers, J., Vernhettes, S., Desprez, T., Vincken, J.-P., Visser, R. G. F., and Trindade, L. M. (2009). Interactions between membrane-bound cellulose synthases involved in the synthesis of the secondary cell wall. FEBS letters 583, 978-982. Tsekos, I. (1999). The sites of cellulose synthesis in algae: diversity and evolution of cellulose-synthesizing enzyme complexes. Journal of Phycology 35, 635-655. Turner, S. R., and Somerville, C. R. (1997). Collapsed Xylem Phenotype of Arabidopsis Identifies Mutants Deficient in Cellulose Deposition in the Secondary Cell Wall. Plant Cell 9, 689-701. Vergara, C., and Carpita, N. (2001). \u00C3\u009F-D-Glycan synthases and the CesA gene family: lessons to be learned from the mixed-linkage (1-3),(1-4)\u00C3\u009F-D-glucan synthase. Plant Molecular Biology 47, 145-160. Vonk, C. (1973). Computerization of Ruland's X-ray method for determination of the crystallinity in polymers. Journal of Applied Crystallography 6, 148-152. Wang, J., Howles, P. A., Cork, A. H., Birch, R. J., and Williamson, R. E. (2006). Chimeric Proteins Suggest That the Catalytic and/or C-Terminal Domains Give CesA1 and CesA3 Access to Their Specific Sites in the Cellulose Synthase of Primary Walls. Plant Physiol 142, 685-695. Wasteneys, G. O. (2004). Progress in understanding the role of microtubules in plant cells. Current Opinion in Plant Biology 7, 651-660. Wasteneys, G. O., and Fujita, M. (2006). Establishing and maintaining axial growth: wall mechanical properties and the cytoskeleton. Journal of Plant Research 199, 5-10. Whittington, A.T., O. Vugrek, K.J. Wei, N.G. Hasenbein, K. Sugimoto, M.C. Rashbrooke and G.O. Wasteneys. (2001). MOR1 is essential for organizing cortical microtubules in plants. Nature 411, 610-613. Wightman, R., and Turner, S.R. (2008). The roles of the cytoskeleton during cellulose deposition at the secondary cell wall. The Plant Journal 54, 794-805. 180 Wightman, R., Marshall, R., and Turner, S. R. (2009). A Cellulose Synthase-Containing Compartment Moves Rapidly Beneath Sites of Secondary Wall Synthesis. Plant Cell Physiol 50, 584-594. Wightman, R., and Turner, S. (2010). Trafficking of the Plant Cellulose Synthase Complex. Plant Physiol 153, 427-432. Wong, H., Fear, A., Calhoon, R., Eichinger, G., Mayer, R., Amikam, D., Benziman, M., Gelfand, D., Measde, J., Emerick, A., et al. (1990). Genetic Organization of the Cellulose Synthase Operon in Acetobacter xylinum. PNAS 87, 8130-8134. 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. Wu, L., Joshi, Chandrashekhar P., and Chiang, Vincent L. (2000). A xylem-specific cellulose synthase gene from aspen (Populus tremuloides) is responsive to mechanical stress. The Plant Journal 22, 495-502. Xu, Y., Piston, D.W., and Johnson, C.H. (1999). A bioluminescence resonance enrergy transfer (BRET) system: Application to interacting circadian clock proteins. PNAS 96, 151-156. Yamaguchi, M., Goue, N., Igarashi, H., Ohtani, M., Nakano, Y., Mortimer, J. C., Nishikubo, N., Kubo, M., Katayama, Y., Kakegawa, K., et al. (2010). VASCULAR- RELATED NAC-DOMAIN6 (VND6) and VND7 Effectively Induce Transdifferentiation into Xylem Vessel Elements under Control of an Induction System. Plant Physiol 153, 906-914. Zhang, B., Deng, L., Qian, Q., Xiong, G., Zeng, D., Li, R., Guo, L., Li, J., and Zhou, Y. (2009). A missense mutation in the transmembrane domain of CESA4 affects protein abundance in the plasma membrane and results in abnormal cell wall biosynthesis in rice. In Plant Molecular Biology (Springer Netherlands), pp. 509-524. Zhong, R., Burk, D. H., Morrison, W. H., III, and Ye, Z.-H. (2002). A Kinesin-Like Protein Is Essential for Oriented Deposition of Cellulose Microfibrils and Cell Wall Strength. Plant Cell 14, 3101-3117. Zhong, R., Morrison, W. H., III, Freshour, G. D., Hahn, M. G., and Ye, Z.-H. (2003). Expression of a Mutant Form of Cellulose Synthase AtCesA7 Causes Dominant Negative Effect on Cellulose Biosynthesis. Plant Physiol 132, 786-795. Zhong, R., Pena, M. J., Zhou, G.-K., Nairn, C. J., Wood-Jones, A., Richardson, E. A., Morrison, W. H., III, Darvill, A. G., York, W. S., and Ye, Z.-H. (2005). Arabidopsis Fragile Fiber8, Which Encodes a Putative Glucuronyltransferase, Is Essential for Normal Secondary Wall Synthesis. Plant Cell 17, 3390-3408. 181 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 19, 2776-2792. 182 Appendix A: AtCesA gene expression in mature stem tissue of Arabidopsis thaliana plants. Transcript abundance of AtCesA4, AtCesA7, and A8CesA8 in wild-type plants (Landsberg ecotype) was measured using real-time PCR analysis. RNA was harvested from stem tissue of 21-day-old plants that had been grown under long-day lighting conditions (16 hours light). Expression levels are relative to the control gene, AtUBQ5. Error bars represent standard deviation, n represents 3 biological replicates. 0 20 40 60 80 100 120 140 160 180 200 R e la ti v e e x p re s s io n Gene AtCesA4 AtCesA7 AtCesA8"@en . "Thesis/Dissertation"@en . "2011-05"@en . "10.14288/1.0071497"@en . "eng"@en . "Forestry"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "Attribution-NonCommercial-NoDerivatives 4.0 International"@en . "http://creativecommons.org/licenses/by-nc-nd/4.0/"@en . "Graduate"@en . "Investigating the role of cellulose synthases in the biosynthesis and properties of cellulose in secondary cell walls"@en . "Text"@en . "http://hdl.handle.net/2429/30534"@en .