urn. ‘ t. ,3, 1.4.... .:. 3.: ... .. LIBRARY Michigan State University This is to certify that the dissertation entitled Cell wall biosynthesis in Zea Mays presented by Henricus EG van Erp has been accepted towards fulfillment of the requirements for the Ph. D. degree in Genetics Om/cmfim MW_, U Major Professor’s Signature 11-28-2007 Date MSU is an affirmative-action, equal-opportunity employer PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 6/07 p:/CIRCIDateDue.indd-p.1 CELL WALL BIOSYNTHESIS IN ZEA MA YS By Henricus EG van Erp A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Genetics 2007 ABSTRACT CELL WALL BIOSYNTHESIS IN ZEA MAYS By Henricus EG van Erp For my thesis research I studied dark-grown maize seedlings as a model system for cell wall biosynthesis. When maize seedlings are grown in complete darkness an organ develops called the mesocotyl, which serves to position the seedling towards the light. After exposure to light, the rapid elongation of the mesocotyl is reduced and concomitant with this, certain enzymatic activities related to cell wall biosynthesis are down-regulated, including a Golgi-localized glucan synthase. I attempted to identify the glucan synthase with biochemical methods and characterized the product it synthesizes. I established that it synthesizes (1,4)-[3-glucan. Because pure (1,4)-B-glucan is not known to be synthesized in the Golgi, the glucan synthase might be involved in the synthesis of other polysaccharides, such as xyloglucan or mixed-linkage g1ucan(MLG). The glucan synthase was successfitlly solubilized with digitonin and it still synthesized (1,4)-B- glucan. However, after chromatography glucose-6-phosphate and glucose-1,6- diphosphate were synthesized instead of (1,4)-[3-g1ucan. I studied the correlation of the light-regulation of cell wall biosynthetic enzymatic activities to transcript levels of the maize Cellulose Synthase Like (CSL) gene family. The CSL gene family encodes the enzymes involved in the synthesis of the (l,4)-B-linked glycan backbones of hemicelluloses such as xyloglucan, MLG and (1,4)-B-mannan. As a first step, I annotated the maize CSL gene family. The rice and maize CSL gene families are similar, except for a few differences. The CSLC gene family is expanded from five genes in Arabidopsis and six in rice to potentially twelve in maize. Also, an atypical CSL gene was found called CSLX. The CSLX protein is closely related to the CSLG proteins fiom poplar, which suggests that the CSLX protein belongs to the CSLG subfamily. Assays for enzymatic activities related to cell wall biosynthesis such as callose synthase, xylan synthase, glucan synthase, mannan synthase, UDP-galactose incorporating activity and latent inosine diphosphatase (IDPase) were performed. Mannan synthase and UDP-galactose incorporation were strongly reduced by light- treatment. Some of the CSLA genes encode for mannan synthases. The reduction in mannan synthase activity correlated with the reduction in CSLA transcript levels after light-treatment. This suggests that mannan biosynthesis is regulated at the level of transcription. Copyright by HENRICUS E.G. VAN ERP 2007 ACKNOWLEDGEMENTS I would like to thank Jonathan Walton for accepting me in his laboratory. He is an excellent adviser who always has time for scientific discussion. I would like to thank my committee members, Ken Keegstra, Kathy Osteryoung, Andreas Weber, Jonathan Walton, Beronda Montgomery-Kaguri and Marcus Pauly for giving me guidance and support during my Ph. D. I want to thank my past and previous lab members for the good times we had and the help with my experiments, especially Peter Kuhn and Melissa Borrusch. I want to thank the past and present members of the Keegstra lab for their help. I want to thank Prof. Jurgen Abel who spent a few weeks during two summers helping me with the identification of the (1 ,4)-[3-g1ucan synthase. I want to thank our secretaries for arranging all my papers and making travel arrangements. I want to thank Jeannine Lee our genetics secretary for all her help and her nice parties. I want to thank Barb Sears, our current genetics program director, and the previous director Helmut Bertrand for leading the genetics program. I want to thank my friends for all the fim time we had and the helpful scientific discussions, especially Hiroshi Maeda. I want to thank my parents and my parents in law for their great support. I would like to thank my wife Young Nam Lee for supporting me. TABLE OF CONTENTS LIST OF TABLES - - - - -- _ IX LIST OF FIGURES - - -- - -- _- -- X CHAPTER 1: LITERATURE REVIEW - - - - ............ - ----1 Plant cell walls have diverse biological functions ..................................................... 1 Organization of the apoplast .................................................................................... 2 Different plants vary in their cell wall composition .................................................. 2 Cell wall proteins and polysaccharides ..................................................................... 3 Cellulose biosynthesis and the CESA gene family .................................................... 3 Hemicelluloses and the CSL gene family ................................................................. 5 Xyloglucan .............................................................................................................. 6 XTH proteins ........................................................................................................... 8 Xylan ....................................................................................................................... 9 Mannan ................................................................................................................. 10 Mixed-linkage glucan (MLG) ................................................................................ 11 Callose ................................................................................................................... 12 Pectin .................................................................................................................... 13 Arabinogalactan proteins (AGPs) ........................................................................... 14 Expansins .............................................................................................................. 14 Summary ............................................................................................................... 14 REFERENCES .......................................................................................................... 16 CHAPTER 2: THE LIGHT-REGULATED GOLGI-LOCALIZED GLUCAN SYNTHASE FROM MAIZE: PROPERTIES, PRODUCT ANALYSIS AND SOLUBILIZATION - - - -- - _ -26 ABSTRACT .............................................................................................................. 26 INTRODUCTION ..................................................................................................... 26 MATERIALS AND METHODS ............................................................................... 27 Growth of maize Seedlings .................................................................................... 27 Isolation of total membranes .................................................................................. 27 (1,4)-B-glucan synthase assay ................................................................................ 28 Callose synthase assay ........................................................................................... 29 Enzymatic hydrolysis of radio labeled products ...................................................... 29 Collection of hydrolysis products ........................................................................... 31 Total hydrolysis of the products ............................................................................. 31 High-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) ....................................................................................... 32 Solubilization of the (1,4)-B-g1ucan ....................................................................... 33 Selection of the optimal detergent for solubilization of the glucan synthase ........... 34 Optimized solubilization procedure for the glucan synthase ................................... 35 Hydroxyapatite chromatography ............................................................................ 35 vi Measurement of protein concentrations .................................................................. 36 RESULTS ................................................................................................................. 37 Optimization of the glucan synthase assay ............................................................. 37 Glucan synthase activity is proportional to protein concentration ........................... 38 The composition of the product is dependent on the UDP-glucose concentration ...40 A. Assays with microsomes isolated fi'om dark-grown seedlings ............................ 40 B. Assays with microsomes isolated from light-treated seedlings. .......................... 40 The product synthesized at 3.7 nM UDP-glucose is a (l,4)-B-glucan ...................... 44 At 1 mM UDP-glucose, the product synthesized is a (1,3)-B-glucan ....................... 46 The product in which [3 H] glucose is incorporated is non-crystalline (1—-)4)-B-g1ucan and is not attached to a protein. .............................................................................. 47 Glucan synthase assays in the presence of UDP-xylose .......................................... 48 The (l,4)-B-glucan synthase activity is not inhibited by cellulose synthase inhibitors .............................................................................................................................. 5 1 Solubilization of the (l,4)-B-glucan synthase ......................................................... 52 Fractionation of the solubilized (1 ,4)-B-g1ucan synthase ........................................ 56 REFERENCES .......................................................................................................... 67 CHAPTER 3: ANNOTATION OF THE ZEA MAYS CSL GENE FAMILY .......... 70 ABSTRACT .............................................................................................................. 70 INTRODUCTION ..................................................................................................... 70 MATERIALS AND METHODS ............................................................................... 71 Annotation of the maize CSL protein family .......................................................... 71 Phylogenetic analysis ............................................................................................. 72 RESULTS ................................................................................................................. 73 Phylogenetic analysis of the CSLA and CSLC protein families of different plant species ................................................................................................................... 74 Phylogenetic analysis of the CSLB, D, E, F, H and G protein families of different plant species .......................................................................................................... 81 ZmCSLX ............................................................................................................... 87 ZmCSL ESTs ......................................................................................................... 88 DISCUSSION ........................................................................................................... 92 REFERENCES .......................................................................................................... 97 CHAPTER 4: LIGHT-REGULATION OF THE MAIZE CSL GENE FAMILY AND ENZYMATIC ACTIVITIES RELATED TO CELL WALL BIOSYNTHESIS - ............ ---100 ABSTRACT ............................................................................................................ 100 INTRODUCTION ................................................................................................... 101 MATERIALS AND METHODS ............................................................................. 101 Genomic DNA isolation ...................................................................................... 101 RNA isolation ...................................................................................................... 102 Reverse transcription ........................................................................................... 103 Primer design ....................................................................................................... 103 Genomic PCR ...................................................................................................... 104 Semi-quantitative RT-PCR .................................................................................. 104 vii Agarose gel electrophoresis ................................................................................. 104 Real-time quantitative RT-PCR ........................................................................... 105 Enzyme assays ..................................................................................................... 105 Determination of the amount of MLG in maize cell walls .................................... 106 RESULTS ............................................................................................................... 107 Light-regulation of the maize CSL gene family .................................................... 107 Amplification of ZmCSL genes from genomic DNA ............................................ 107 Semi-quantitative RT-PCR .................................................................................. 108 Real-time quantitative RT-PCR ........................................................................... 111 Light-regulation of enzymatic activities related to cell wall biosynthesis. ............. 115 Analysis of the radiolabeled products ................................................................... 117 Correlation between the levels of the ZmCSLA transcripts and (1,4)—B-mannan synthase activity .................................................................................................. 119 Time-course of the light-inhibition of enzymatic activities related to cell wall biosynthesis ......................................................................................................... 1 19 Analysis of MLG in cell walls of dark-grown and light-treated seedlings ............. 122 DISCUSSION ......................................................................................................... 123 REFERENCES ........................................................................................................ 127 CHAPTER 5: FUTURE DIRECTIONS - - 129 APPENDIX 1: ANNOTATION OF THE MAIZE CSL GENE FAMILY ............... 132 APPENDIX 11: GENBAN K LOCUS NUMBERS AND JGI PROTEIN IDS ......... 149 APPENDIX III: GENE SPECIFIC PRIMERS FOR SEMI-QUANTITATIVE RT- PCR -- --------------- - 153 APPENDIX IV: GENE SPECIFIC PRIMERS FOR REAL-TIME QUANTITATIVE RT-PCR ....... - ......... - ........................ 156 viii LIST OF TABLES Table 1: Enzymes used, their catalog number, buffers used and substrates they hydrolyze .............................................................................................. 30 Table 2: Gradients used for the HPAEC-PAD analysis of different products ............... 33 Table 3: The effect of different detergents on (1,4)-B-glucan synthase activity. . . . . . . . ....53 Table 4: The number of CSL genes for each subfamily in rice and maize ................... 74 Table 5: Genomic DNA and EST sequences related to the ZmCSLX ORF in different monocot species (family Poaceae) ................................................................ 88 Table 6: PCR amplification of ZmCSL genes from genomic DNA and cDNA ............ 109 Table 7: Change in transcript levels compared to EFI a determined with quantitative RT- PCR ................................................................................................. 115 Table 8: Genomic sequences used for the annotation of the ZmCSL gene family. . . . . 132 Table 9: Detailed description of the annotation for each ZmCSL gene ........................... 135 Table 10: ESTs found for each CSL in GenBank and MaizeSeq as of November 2006.141 Table 11: GenBank locus number for Arabidopsis proteins .................................. 149 Table 12: JGI Protein IDs for the poplar CSL proteins ....................................... 150 Table 13: JGI protein IDs for Physcomitrella CSL proteins ................................. 151 Table 14: GenBank locus number for the ZmCESA proteins ............................... 152 Table 15: Gene specific primers for the ZmCSLA genes .................................... 153 Table 16: Gene specific primers for the ZmCSLC genes .................................... 154 Table 17: Gene specific primers for the ZmCSLD, E, F, H, X and ZmEFZa genes ...... 155 Table 18: Gene specific primers for the ZmCSLA, C, D, E, F and EFI (1 genes .......... 156 LIST OF FIGURES Figure 1. Effect of different MgClz concentrations on glucan synthase activity ............ 37 Figure 2. Effect of pH on glucan synthase activity ............................................. 38 Figure 3. Specific glucan synthase activity over a 5-fold range of protein concentration 39 Figure 4. Enzymatic sensitivity of the glucan synthase product made at different UDP- Glc concentrations ................................................................................. 42 Figure 5. Products synthesized with microsomes isolated from dark-grown seedlings treated with xylanase M1 and proteinase K ...................................................... 44 Figure 6. The glucan synthase products synthesized with microsomes isolated fi'om dark- grown and light-treated seedlings were treated with cellulase and the hydrolysis products were analyzed by HPAEC-PAD .................................................................. 45 Figure 7. The glucan synthase products synthesized with microsomes isolated from dark- grown and light-treated seedlings were hydrolyzed with TFA and the products were analyzed by HPAEC-PAD ......................................................................... 46 Figure 8. The glucan synthase product synthesized at 1 mM UDP-glucose was hydrolyzed with laminarinase and the products were analyzed by HPAEC-PAD ......... 47 Figure 9. Solubility of the (l,4)-B-glucan in different solvents ............................... 48 Figure 10. Glucan synthase assays in the presence of UDP-xylose .......................... 50 Figure 11. Glucan synthase assays performed in the presence of DCB and isoxaben ..... 51 Figure 12. Comparison of (1,4)-B-glucan synthase activity after solubilization with 0.5% digitonin with the activity in the microsomes ................................................... 54 Figure 13. Enzyme treatment of the products synthesized by the solubilized (l,4)-B- glucan synthase preparation ........................................................................ 55 Figure 14. Analysis of the product synthesized by the solubilized (1,4)-B-glucan synthase ............................................................................................... 56 Figure 15. Hydroxyapatite fractionation of solubilized (1,4)-B-glucan synthase ........... 57 Figure 16. The ethanol-insoluble and water-soluble product synthesized after hydroxyapatite chromatography was analyzed by HPAEC-PAD ............................ 58 Figure 17. Enzyme sensitivity of the products synthesized after hydroxyapatite chromatography. .................................................................................... 59 Figure 18. The product synthesized after hydroxyapatite chromatography was treated with cellulase and the hydrolysis products were analyzed by HPAEC-PAD ............... 60 Figure 19. The product synthesized afier hydroxyapatite chromatography was treated with alkaline phosphatase and the hydrolysis products were analyzed by HPAEC- PAD ................................................................................................... 61 Figure 20. The water soluble products synthesized afier hydroxyapatite chromatography were analyzed by I-IPAEC-PAD .................................................................. 62 Figure 21. Parsimony phylogram of Arabidopsis (At), guar (Ct), maize (Zm), moss (Pp), pine tree (Pta), poplar (Pt) and rice (Os) CSLA proteins and the CSLC proteins of Arabidopsis, maize, moss, nasturtium (Tm), poplar and rice ................................. 78 Figure 22. Parsimony phylogram of Arabidopsis (At), guar (Ct), rice (Os), maize (Zm), moss (Pp) and poplar (Pt) CSLA proteins ....................................................... 81 Figure 23. Parsimony phylogram of Arabidopsis (At), maize (Zm), moss (Pp), tobacco (Nicotiana tabacum (Nt)), poplar (Pt) and rice (Os) CSLB, D, E, F, G, H and X proteins ......................................................................................................... 84 Figure 24. Distribution of ESTs belonging to maize CSL genes in GenBank and MaizeSeq as of November 2006 .................................................................. 90 Figure 25. Ethidium-stained agarose gel showing products of the ZmCSL gene family.110 Figure 26. Real-time quantitative RT-PCR results for the maize CSLA gene family ..... 112 Figure 27. Real-time quantitative RT -PCR results for the maize CSLC, D, E and F gene family ................................................................................................ 114 Figure 28. (1,4)-B-glucan synthase, (1,4)-B-mannan synthase, UDP-Gal incorporating activity, (1,4)-B-xylan synthase, (1,3)-B-glucan synthase, and latent IDPase activities in microsomes from dark-grown and light-treated mesocotyls ................................. 116 Figure 29. The product synthesized from UDP-[14C]mannose was treated with endo- (1,4)-B-mannanase and the hydrolysis products were analyzed by HPAEC-PAD ........ 117 Figure 30. The product synthesized from UDP-[14C]xylose was treated with xylanase M6 and the hydrolysis products were analyzed by HPAEC-PAD ............................... 118 xi Figure 31. Time-course of inhibition of enzymatic activities after light-treatment for (l,4)-B-mannan synthase, the UDP-galactose incorporating activity and (l,4)-B-xy1an synthase ............................................................................................. 121 Figure 32. Amount of MLG in cell walls in dark-grown and light-treated seedlings. . ..123 xii CHAPTER 1: LITERATURE REVIEW Plant cell walls have diverse biological functions The cell wall is a defining feature of plants and has a variety of fimctions including structural support and defense against pathogens. The cell wall gives plant cells the shapes they need to perform their fimction as shown by tracheary elements or petals (Carpita and McCann, 2000). Cell walls provide the force needed to cope with the turgor pressure exerted by the vacuole. When plant cells take up water in the vacuole pressure is exerted on the cell wall and subsequent loosening of the cell wall allows plant cells to expand. Cell walls have a role in defense against pathogens. After pathogen attack papillae are formed below the site of infection in order to provide a physical barrier. They are composed of callose, proteins and other components (Aist, 1976). The cell wall is a source of signaling/regulatory molecules. Xyloglucan fragments can inhibit auxin- induced elongation grth of pea epicotyls (York et al., 1984) and homogalacturonan oligosaccharides can induce the synthesis of phytoalexins in plants (Nothnagel et al., 1983). Polysaccharides can fimction in carbon storage in seeds. Examples of storage polysaccharides are (1,4)-B-mannan, xyloglucan and galactans (Buckeridge et al., 2000). The importance of the cell wall in grth and development is shown by mutations in genes related to cell wall biosynthesis. Mutants in cellulose synthase genes involved in secondary cell wall biosynthesis show a collapsed xylem phenotype (Taylor et al., 1999, 2000, 2003). Mutations in genes involved in the synthesis of the pectic polysaccharide rhamnogalacturonan 2 (RG2), show severe growth defects and cell adhesion defects (O’Neill et al., 2001; Iwai et al., 2002). Organization of the apoplast The continuum of cell wall material and liquid between individual plant cells is called the apoplast. The cell walls of individual cells are connected with a cell wall layer composed of pectin and proteins called the middle lamella (Carpita and McCann, 2000). The next layer of cell wall material, which is deposited at the inside of the middle lamella, is called the primary cell wall. It is composed of cellulose, hemicelluloses and proteins and it provides strength but at the same time allows plant cells to expand. The secondary cell wall is deposited in certain cell types at the inside of the primary cell wall when plant cells stop elongating (Carpita and McCann, 2000). It is more rigid and generally contains more lignin. The development of xylem vessels composed of secondary cell walls was a major adaptation for land plants to assume upright growth and transport solutes over long distance (Bateman et al., 1998). Diflerent plants vary in their cell wall composition A complicating factor in the understanding of cell wall biosynthesis is that different groups of plants have different types of primary cell walls, although the main component, cellulose, is identical in all plants. Two types of primary cell walls can be distinguished: type I and type II. Dicots and group B monocots have a type I cell wall. In type I cell walls xyloglucan is the major hemicellulose cross-linking the cellulose microfibrils, and structural proteins strengthen the primary cell wall. The monocots can be distinguished fin'ther based on the presence (group A) or absence (group B) of ester- linked ferulic acids in their primary cell walls. The group A monocots have a type 11 cell wall. In the type 11 cell wall, glucuronoarabinoxylan (GAX) is the major cross linking hemicellulose, and xyloglucan and glucomannan are relatively minor component (Carpita et al., 2001, Smith and Harris, 1998). The type II cell wall of the group A monocots can be distinguished further based on the presence or absence of mixed-linkage glucan (MLG). Cell walls of the family of the Poales (grasses), which belongs to the group A monocots contains MLG (Carpita and Gibeaut, 1993, Smith and Harris, 1998). Cell wall proteins and polysaccharides Cellulose biosynthesis and the CESA gene family Cellulose is the major load-bearing polymer in the primary cell wall and is synthesized at the plasma membrane in terminal rosette complexes (Kimura et al., 1999). The rosettes consist of six globules that each synthesize six [3(1,4)-glucan chains. Together these 36 B(l,4)-glucan chains compose the cellulose microfibrils. The (1,4)-[3- glucan chains are held together by hydrogen bonds and Vanderwaals interactions (N ishiyama et al., 2002, 2003) and their length varies between the primary and secondary cell wall Generally the chains are shorter in the primary cell wall (2000-8000 glucose residues) than in the secondary cell wall (14,000-15,000 glucose residues) (Brown et al., 2004) The first plant cellulose-synthase genes (celAI and ceIA2) were identified from a cotton cDNA library based on similarity to bacterial cellulose synthases from Acetobacter xylinum and Agrobacterium tumefaciens (Pear et al., 1996). Arioli et a1. (1998) provided genetic evidence that AtCESAI is involved in cellulose biosynthesis. With immunogold labeling it was shown that the globules involved in cellulose biosynthesis consist of cellulose synthase (CESA) proteins (Kimura et al., 1999). In vitro assays for cellulose biosynthesis were established by Lai-Kee-Him et al (2002). They solubilized intact cellulose synthase complexes from Rubusfruticosus (blackberry) and synthesized cellulose in the presence of UDP-glucose. No Mg2+ was required for cellulose biosynthesis in vitro. The CESA proteins contain two conserved domains. Domain A consists of 3 widely spaced Asp (D) residues and is thought to be involved in UDP-Glc binding. Domain B consists of a conserved domain, QXXRW, hypothesized to be part of the catalytic site (Richmond et al., 2000). Domain A is conserved in both processive- and non-processive glycosyltransferases. Only processive enzymes contain domain B (Richmond et al., 2000). Different CESA proteins are involved in cellulose biosynthesis in the primary and secondary cell walls. The CESAI, 3, and 6 proteins are involved in cellulose biosynthesis in the primary cell wall (Fagard et al., 2000; Arioli et aL, 1998; Ellis et al., 2002; Burn et al., 2002). The CESA4, 7, and 8 proteins are involved in cellulose biosynthesis in secondary cell walls (Taylor et al., 1999, 2000, 2003; Zhong et al., 2003). The CESA proteins belong to the glycosyltransferase family (GT) 2 of carbohydrate active enzymes (www.cazy.org; Campell et al., 1997; Coutinho and Henrissat, 1999; Coutinho et al., 2003). For a review about cellulose biosynthesis see Somerville (2006). Other proteins associated with the cellulose synthase complex are sucrose synthase and microtubules (Salnikov et al., 2001). Microtubules can guide the cellulose synthase complexes, but movement of the CESA complexes is not random in their absence, suggesting that these complexes can also be guided by polymerization of the cellulose microfibrils (Paradez et al., 2006). Other proteins which might be involved in cellulose biosynthesis are the cellulase KORRIGAN (Nicol et al., 1998), the putative glycosylphosphatidylinositol (GPI) anchored protein COBRA (Schindelrnan et al., 2001), the kinesin—like protein FRAGILE FIBER 1 (FRAl) (Zhong et al., 2002), the novel plasma membrane protein KOBITOI (Pagant et al., 2002) and the ct-glucosidase I KNOPF (Gillrnor et al., 2002). Hemicelluloses and the CSL gene family Cross-links between the cellulose microfibrils are formed by hemicelluloses such as xyloglucan, xylan, mixed-linkage glucan, and glucomannan, and pectin. The substrates for hemicellulose biosynthesis are nucleotide sugars, which are synthesized from UDP- glucose by nucleotide sugar-converting enzymes (Seifert, 2004). They are hypothesized to be in complexes with glycosyltransferases and nucleotide-sugar transporters at the Golgi membrane (Seifert, 2004). Once the hemicelluloses are synthesized in the Golgi, they are transported to the plasma membrane in vesicles, secreted, and incorporated into the cell wall (Moore et al., 1991). Enzymes encoded by the Cellulose Synthase Like (CSL) (GT2) gene family are hypothesized to encode for the enzymes synthesizing the B(1,4)-linked hemicellulose backbones. The CSL gene family was identified based on homology to the CESA gene family and is composed of 30 members in Arabidopsis (Cutler and Somerville, 1997) and 37 members in rice (Hazen et al., 2002). In dicots such as Arabidopsis the CSL gene family consists of the CSLA, B, C, D, E and G subfamilies (Cutler and Somerville, 1997). In rice the CSL gene family consists of the CSLA, C, D, E, F and H subfamilies (Hazen et al., 2002). The CESA, CSLB, D, E, G, F and H genes most likely originated fiom a cyanobacterial ancestor. The CSLA and C genes most likely originated from a different bacterial ancestor (Nobles and Brown, 2004). For several CSL genes a fimction has been determined. AtCSLC4 is involved in the synthesis of the [3(1,4)—linked glucan backbone of xyloglucan (Cocuron et al., 2007). OsCSLF2 is involved in the synthesis of mixed-linked glucan (Burton et al., 2006) and the CSLA proteins are involved in (gluco)mannan biosynthesis (Dhugga et al., 2004; Liepman etal., 2005; Suzuki et al., 2006; Liepman et al., 2007). For several CSL genes, a biological firnction has been found using a genetics approach. AtCSLD3 is important for root hair grth in Arabidopsis (Wang et al., 2001; F avery et al., 2001) and pollen tube grth in Nicotiana alata (Doblin et al., 2001). AtCSLA 7 is involved in pollen-tube grth and embryo genesis in Arabidopsis (Goubet et al., 2003) and the AtCSLA9 mutant rat4 is resistant to transformation by Agrobacterium tumefaciens (Zhu et al., 2003). Xyloglucan Xyloglucan is present in the primary cell walls of all land plants. It is not present in charophycean green algae, indicating that xyloglucan might have been an important adaptation for plants to colonize land (Popper and Fry, 2003). The core structure of all xyloglucans is a B(1,4)-1inked glucan backbone decorated with a-(1,6)-linked xylose residues. In the dicot Arabidopsis the xylose residues are spaced very regularly. Three consecutive glucose residues are decorated with a-(1,6)-linked xylosyl residues, followed by an unsubstituted glucose. The xylo se residues in Arabidopsis can be decorated further by B-(1,2)-1inked galactose, which subsequently can be substituted with a-(1,2)-linked fucose. The xyloglucan structure in monocots is different fiom dicots. In monocots only 30-40% of the glucose residues are substituted with xylose (Kato et a1, 1982). As a result, a major difference between dicot and monocot xyloglucans is their solubility. Dicot xyloglucan is soluble in water, whereas monocot xyloglucan can only be dissolved in alkali solutions. In vitro assays for dicot xyloglucan biosynthesis were developed by several researchers (Ray et al., 1980; Hayashi and Matsuda 1981; Gordon and Maclachlan, 1989). Hayashi and Matsuda (1981) determined that UDP-glucose and UDP-xylose are both necessary in order to synthesize xyloglucan in vitro. Incorporation of UDP-xylose into xyloglucan is stimulated by Mn2+, but not by Mg2+ (Hayashi and Matsuda, 1981). It was determined that UDP-xylose stimulates the incorporation of UDP-glucose into xyloglucan and that UDP-glucose stimulates the incorporation of UDP-xylose into xyloglucan (Hayashi and Matsuda, 1981). This indicates that the synthesis of the B(1,4)- glucan backbone and the addition of xylosyl side chains need to occur simultaneously. The reason for this might be the insolubility of unsubstituted B(l,4)-glucan. A curious observation is that UDP-xylose only stimulates the incorporation of UDP-glucose into xyloglucan if the UDP-xylose concentration is lower than the UDP-glucose concentration (Hayashi and Matsuda, 1981; Gordon and Maclachlan, 1989). The reason for this is not clear. The B(1,4)—glucan backbone of xyloglucan is synthesized by members of the CSLC protein family (Cocuron et al., 2007). The xyloglucan xylosyltransferase (X1) (GT34) genes were identified by Faik et a1. (2002) and Cavalier et a1. (2006). An in vitro assay for the galactosyltransferase was developed by F aik et al., (1997). Degalactosylated xyloglucan was used as a substrate for the galactosyltransferase. A galactosyltransferase (M UR3) (GT47) was identified by Madson et a1. (2003). An in vitro assay for the fircosyltransferase was developed by Camarind and Maclachlan (1986). The fucosyltransferase (F UT) (GT37) was identified by Perrin et a1. (1999) and firrther studied by F aik et a1. (2000) and Vazin et a1. (2002). XTH proteins It is thought that cross-linking of the cellulose microfibrils by xyloglucan is important in the regulation of cell elongation in dicots (Cosgrove, 2005). The hydrolysis and the subsequent reannealing of xyloglucan allow plant cells to expand while maintaining strength. The proteins involved in this process are the xyloglucan endotransglucosylase/hydrolases (XTH) (Rose et al., 2002). In Arabidopsis there are 33 XTH genes and their expression pattern is correlated with elongation grth (Yokoyama and Nishitani, 2001). In rice there are 29 members (Yokoyama et al., 2004), which suggests that xyloglucan also has an important role in cell elongation in grasses. XTH proteins are not active on substrates other than xyloglucan (Smith and Fry, 1991; Fry et al., 1992). However, Hnnova et al (2007) showed that HvXTHS can facilitate the formation of covalent bonds between xyloglucan and cellulose or MLG, although this was a minor activity of this protein. Xylan Xylans consist of a backbone of (1,4)-B-linked xylose residues, which can be substituted with arabinose, ferulic acid, glucuronic acid or methylglucuronic acid residues. Xylan can also be highly acetylated (Teleman et al., 2000). In both dicots and monocots the xylan backbone is decorated with arabinose and glucuronic acid residues. Group A monocots have arabinose at the O-3 position, glucuronic acid at the O-2 position of the xylose residues, and feruloyl groups at O-5 position of the arabinose residues. Group B monocots and all dicots have arabinose residues at the O-2 and O-3 position and glucuronic acid at the O-2 position (Carpita and McCann, 2000). In dicots xylan is a minor component of the primary cell wall, but a major component of the secondary cell wall. In monocots xylan is a major component of both the primary and secondary cell walls (McNeil et al., 1984). In the primary cell wall of monocots it has a major role in cross-linking the cellulose microfibrils (Carpita et al., 2001). Unsubstituted xylans form a tight connection with cellulose microfibrils and highly substituted xylans form the cross links between the cellulose microfibrils (Carpita et al., 1983 and 2001). In vitro xylan synthase assays were developed by Bailey and Hassid (1966). Bioinformatics approaches have identified candidate genes for the (1,4)-B-xylan synthase. Person et al. (2005) and Brown et a1. (2005) analyzed micro-arrays for coexpression of genes with secondary cell wall CESA genes. Several of these genes showed an irregular xylem phenotype (irx) when mutated. Two examples are the irx8 (GT8) and irx9 (GT43) mutants, which have a decreased xylan and pectin content (Pena et al., 2007; Persson et al., 2007). Glucuronosyl transferase assays were developed by Waldron and Brett (1983). The IRREGULAR XYLEM 7 (IRX 7) or FRAGILE FIBER 8 (FRA8) (GT47) gene, might encode an enzyme that adds glucuronosyl residues to the xylan backbone. The irx7/fra8 mutant has no glucuronic acid side chains, a reduction in xylan and cellulose, and an increase in pectin and xyloglucan (Zhong et al., 2005). Mannan I Mannan polysaccharides are present in all plants. Algae and mosses, however, have more mannan in their cell walls than vascular plants (Popper and Fry, 2003). Mannan polysaccharides occur in several different forms, such as pure mannan, glucomannan and galactomannan. Pure mannan consists of (1,4)-B—linked mannose and I glucomannan is composed of (1,4)—B-linked glucose and mannose. These polymers can have galactose attached by an a—(1,6)-glycosidic bond and acetyl groups at the C-2 or C- 3 position of the mannose residues. Glucomannan is a minor hemicellulose in the cell wall of angiosperms, but the major hemicellulose in secondary cell walls of gymnosperms where it constitutes 16-18% of the cell wall (Maeda et al., 2000). In Arabidopsis glucomannan is present in the secondary cell walls of xylem, xylem parenchyma and interfascicular fibers (Handford et al., 2003). In maize it is a minor cross-linking hemicellulose in the primary cell wall (Carpita et al., 2001). Mannan, glucomannan and galactomannan are also present as a storage polysaccharide in seeds (Buckeridge et al., 2000). Galactomannan can be synthesized in vitro with UDP-galactose as a substrate and B(1,4)-mannan as an acceptor (Edwards et al., 1989). The first gene involved in galactomannan biosynthesis, the galactomannan galactosyltransferase (GMGT) (GT34), 10 was identified by Edwards et al. (1999). Glucomannan can be synthesized in vitro in the presence of GDP-glucose and GDP-mannose (Heller and Villemez, 1972; Liepman et al., 2005). The genes encoding for the enzymes involved in the synthesis of the mannan backbone are encoded by the CSLA gene family (Dhugga et al., 2004; Liepman et al., 2005; Suzuki et al., 2006; Liepman et al., 2007). Most likely all the members of the CSLA gene family are involved in mannan biosynthesis (Liepman et al., 2007). Recently a mannan transglycosylase activity was discovered in flowers of kiwi fruit and in tomato fruit (Schroder et al., 2004). This is an indication that mannans might have a role in cell elongation comparable to xyloglucan. Mixed-linkage glucan (MLG) MLG is specific for the cell wall of the members of the Poales family (grasses) (Smith and Harris, 1998). It is composed of (l—->3),(1—>4)-B-D-linked glucose residues followed by blocks of (l—)4)-[3-D-glucan and (l—>3)-B-D-glucan residues (Kato and Nevins, 1984). MLG is a developmentally regulated polysaccharide in certain tissues such as intemodes of deep water rice (Sauter and Kende, 1992), and the maize coleoptile (Carpita et al., 1984). Because the amount of MLG in some tissues increases during rapid cell extension, it is thought that this polysaccharide contributes to increased extensibility of the cell wall (Carpita et al., 1984; Sauter and Kende, 1992). This is supported by the fact that MLG antibodies suppress auxin-induced elongation and also inhibit the degradation of MLG (Hoson and Nevins, 1989). In vitro assays for MLG were established by Gibeaut and Carpita (1993) and Henry and Stone (1982). MLG synthase uses UDP-glucose as a substrate and requires 11 Mgz” or Mn2+ for activity (Gibeaut and Carpita, 1993). Golgi-membranes were incubated with radiolabeled UDP-glucose, the products were collected, treated with lichenase and the hydrolysis products were analyzed with HPAEC-PAD. The radiolabeled hydrolysis products eluted at the same position as the MLG standards. This analysis showed that MLG was synthesized in vitro. The CSLF proteins are involved in the synthesis of MLG (Burton et al., 2006). Arabidopsis was transformed with the rice CSLF genes. With monoclonal antibodies against MLG and MLG specific enzymes it was shown that MLG was synthesized in Arabidopsis expressing OsCSLF2. Callose Callose is composed of (l,3)—B-D-linked glucose, and it is synthesized at the plasma membrane (Turner et al., 1998). Callose is involved in many processes, such as cell plate formation (Hong et al., 2001a), blocking of sieve plates in the phloem (Furch et al., 2007), pollen tube grth (Doblin et al., 2001) and papilla formation in response to wounding and pathogen infection (Nishimura et al., 2003; Jacobs et al., 2003). Callose synthase activity was first detected in membrane preparations and requires Ca2+ (Feingold et al., 1958). Dhugga and Ray (1994) attempted to identify the callose synthase and found two proteins of 55 kDa and 70 kDa that correlated with callose synthase activity. Based on homology to the FKSI genes Hong et al. (2001a) annotated 12 putative callose synthase (CalS) genes in Arabidopsis. The FKSI genes encode for callose synthases in yeast (Douglas et al., 1994). The plant callose synthase genes were also annotated by Richmond (http://cellwall.stanford.edu/) and they named them GLUCAN SYNTHASE-LIKE (GSL) genes. GSL genes (GT48) are large and have 3 12 to 50 exons. The proteins they encode do not possess the conserved D, D, D, QXXRW domain found in the CESA protein family, which suggest a separate evolutionary origin (Hong et al., 2001a). Hong et al. (2001a) provided genetic evidence that the GSL genes encode for callose synthases. Over-expression of GSL] resulted in increased callose deposition at the cell plate. Additional genetic evidence came from the work of Nishimura et al. (2003) and Jacobs et al. (2003), who found that a mutation in GSL5 causes Arabidopsis not to deposit callose in papillae after pathogen infection. Biochemical evidence that the GSL proteins are involved in callose biosynthesis comes from the work of Li et al. (2003a). A callose synthase was purified from barley and enriched more than 60-fold. The purified protein was shown to possess callose synthase activity by performing in gel callose synthase assays. The purified protein was sequenced and the peptide fragments corresponded to the amino—acid sequence predicted for the HvGSLl protein (Li et al., 2003a). Other proteins involved in callose biosynthesis are sucrose synthase (Amor et al., 1995), UDP-glucose transferase (UGTl) and Ropl, which is a GTPase. UGTl might function to transfer UDP-glucose to the active site of callose synthase, because it does not posses a known UDP-Glc binding domain. UGTl is regulated by Ropl. Ropl only interacts with UGTl in the GTP-bound form (Hong et al., 2001b) and can therefore function as a molecular switch for callose biosynthesis. Pectin Pectin is a minor component of the cell wall of Poaceae but a major component of the cell walls of other monocots and dicots. The major pectins are homogalacturonan 13 (HG), rhamnogalacturonan I (RG-I) and rhamnogalacturonan H (RG—II) (Ridley et al., 2000). Several genes involved in pectin biosynthesis have been identified using genetic and biochemical methods (Lerouxel et al., 2006). Arabinogalactan proteins (AGPs) AGPs are a diverse class of proteoglycans which have a role in plant development including xylem formation, somatic embryogenesis and plant pathogen interactions. AGPs consist of carbohydrates which are mainly composed of galactose and arabinose, and are attached to a protein backbone. No proteins involved in the biosynthesis of the glycan chains have been identified yet, but many genes encoding for the protein backbone have been isolated (Seifert et al., 2007). Expansins Expansins are proteins involved in the process of cell expansion as shown by their influence on extension growth in cucumber hypocotyls (Li et al., 2003b). Their exact mechanism of action is not known, but they are hypothesized to break the hydrogen bonds in the cellulose/xyloglucan network, thus allowing for cell expansion to take place. Initially two expansin subfamilies were discovered, called a- and B-expansins. Recently two new families have been found, which are called y— and S-expansins (Li et al., 2003b). Summary In summary, the plant cell wall is composed of a highly complex network of polysaccharides, and proteins. There are still many unknowns, such as the regulation of 14 cell wall biosynthesis, most of the enzymes involved in pectin and AGP biosynthesis, the enzymes synthesizing the xylan backbone, and the exact mode of action of expansins. However, advanced genomics and proteomics technologies are opening the door to new discoveries in this exciting but difficult area of plant science. 15 REFERENCES Aist JR (1976) Papillae and related wound plugs of plant cells. Annu rev phytopathol 14: 145-163 Amor Y, Haigler C, Johnson S, Wainscott M, Delmer D (1995) A membrane associated form of sucrose synthase and its potential role in synthesis of cellulose and callose in plants. Proc Natl Acad Sci USA 92: 9353-9357 Arioli T, Peng L, Betzner A, Burn J, Wittke W, Camilleri C, Hofte H, Plazenski J, Birch R, Cork A, Glover J, Redmond J, Williamson R (1998) Molecular analysis of cellulose biosynthesis in Arabidopsis. Science 279: 717-720 Bailey R, Hassid W (1966) Xylan synthesis from uridine-diphosphate-D—xylose by particulate preparations from immature comcobs. Proc Natl Acad Sci USA 56: 1586-1593 Bateman RM, Crane PR, DiMichele WA, Kenrick PR, Rowe NP, Speck T, Stein WE (1998) Early evolution of land plants: phylogeny, physiology, and ecology of the primary terrestrial radiation. Annu Rev Ecol S yst 29: 263—92 Brown D, Zeef L, Ellis J, Goodacre R, Turner S (2005) Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell 17 : 2281-2295 Brown RM jr (2004) Cellulose structure and biosynthesis: What is in store for the 21St century? Journal of polymer science A 42: 487-495 Buckeridge M, Possoa dos Santos H, Tine M (2000) Mobilisation of storage cell wall polysaccharides in seeds. Plant physiology and biochemistry 38: 141-156 Burn J, Hocart C, Birch R, Cork A, Williamson R (2002) Functional analysis of the cellulose synthase genes CesAI, CesA2, and CesA3 in Arabidopsis. Plant Physiol 129: 797 -807 Burton R, Wilson S, Hrmova M, Harvey A, Shirley N, Medhurst A, Stone B, Newbigin E, Bacic A, Fincher G (2006) Cellulose synthase-like Cle genes mediate the synthesis of cell wall (1,3;1,4)-B-D-glucans. Science 311: 1940- 1942 Camirand A, Maclachlan G (1986) Biosynthesis of the fucose-containing xyloglucan nonasaccharide by pea microsomal membranes. Plant Physi0182: 379-383 16 Campbell J, Davies G, Bulone V, Henrissat B (1997) A classification of nucleotide- diphospho-sugar glycosyltransferases based on amino acid sequence similarities. Biochem J 326: 929-939 Carpita N (1983) Hemicellulosic polymers of cell walls of Zea coleoptiles. Plant Physiol 72: 515-521 Carpita N (1984) Cell wall development in maize coleoptiles. Plant Physiol 76: 205-212 Carpita N, Defemez M, Findlay K, Wells B, Shoue D, Catchpole G, Wilson R, McCann M (2001) Cell wall architecture of the elongating maize coleoptile. Plant Physiol 127: 551-565 Carpita N, Gibeaut D (1993) Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J 3: 1-30 Carpita N, McCann M (2000) The cell wall. In B Buchanan, W Gruissem, R Jones, eds, Biochemistry & Molecular Biology of Plants, pp 52-108 Cavalier D, Keegstra K (2006) Two xyloglucan xylosyltransferases catalyze the addition of multiple xylosyl residues to cellohexaose. J Biol Chem 281: 34197- 34207 Cocuron J, Lerouxel O, Drakakaki G, Alonso A, Liepman A, Keegstra K, Raikhel N, Wilkerson C (2007) A gene from the cellulose synthase-like C family encodes a B-l,4-glucan synthase. Proc Natl Acad Sci USA 104: 8550-8555 Cosgrove D (2005) Growth of the plant cell wall. Nat Rev Mol Cell Biol 6: 850-861 Coutinho P, Deleury E, Davies G, Henrissat B (2003) An evolving hierarchical family classification for glycosyltransferases. J Mol Biol 328: 307-317 Coutinho PM HB (1999) Carbohydrate-active enzymes: an integrated database approach. In Gilbert HJ, Davies G., Henrissat B, Svensson B, eds, Recent advances in carbohydrate bioengineering. The Royal Society of Chemistry, Cambridge, pp 3- 12 Cutler S, Somerville C (1997) Cloning in silico. Curr Biol 7: R108-111 Dhugga K, Barreiro R, Whitten B, Stecca K, Hazebroek J, Randhawa G, Dolan M, Kinney A, Tomes D, Nichols S, Anderson P (2004) Guar seed B-mannan synthase is a member of the cellulose synthase super gene family. Science 303: 363-366 17 Dhugga K, Ray P (1994) Purification of 1,3-B-D-glucan synthase activity from pea tissue. Two polypeptides of 55 kDa and 70 kDa copurify with enzyme activity. Eur J Biochem 220: 943-953 Doblin M, De Melis L, Newbigin E, Bacic A, Read S (2001) Pollen tubes of Nicotiana alata express two genes from different B-glucan synthase families. Plant Physiol 125: 2040-2052 Douglas C, Foor F, Marrinan J, Morin N, Nielsen J, Dahl A, Mazur P, Baginsky W, Li W, el-Sherbeini M (1994) The Saccharomyces cerevisiae F KSI (E TGI) gene encodes an integral membrane protein which is a subunit of 1,3-B-D-glucan synthase. Proc Natl Acad Sci USA 91: 12907-12911 Edwards M, Bulpin P, Deal, Reid J (1989) Biosynthesis of legume-seed galactomannans in vitro. Planta 178: 41-51 Edwards M, Dickson C, Chengappa S, Sidebottom C, Gidley M, Reid J (1999) Molecular characterisation of a membrane-bound galactosyltransferase of plant cell wall matrix polysaccharide biosynthesis. Plant J 19: 691-697 Ellis C, Karafyllidis I, Wasternack C, Turner J (2002) The Arabidopsis mutant cevl links cell wall signaling to jasmonate and ethylene responses. Plant Cell 14: 1557- 1566 Fagard M, Desnos T, Desprez T, Goubet F, Refregier G, Mouille G, McCann M, Rayon C, Vemhettes S, Hbfte 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 Faik A, Chileshe C, Sterling J, Maclachlan G (1997) Xyloglucan galactosyl- and fucosyltransferase activities from pea epicotyl microsomes. Plant Physiol 114: 245-254 Faik A, Bar-Peled M, DeRocher A, Zeng W, Perrin R, Wilkerson C, Raikhel N, Keegstra K (2000) Biochemical characterization and molecular cloning of an 0t- 1,2-fucosyltransferase that catalyzes the last step of cell wall xyloglucan biosynthesis in pea. J Biol Chem 275: 15082-15089 Faik A, Price NJ, Raikhel NV, Keegstra K (2002) An Arabidopsis gene encoding an or- xylosyltransferase involved in xyloglucan biosynthesis. Proc Natl Acad Sci USA 99: 7797-7802 Favery B, Ryan E, Foreman J, Linstead P, Boudonck K, Steer M, Shaw P, Dolan L (2001) KOJAK encodes a cellulose synthase-like protein required for root hair cell morphogenesis in Arabidopsis. Genes Dev 15: 79-89 18 Feingold D, Neufeld E, Hassid W (1958) Synthesis of a B-1,3-linked glucan by extracts of Phaseolus aureus seedlings. J Biol Chem 233: 783-788 Fry S, Smith R, Renwick K, Martin D, Hodge S, Matthews K (1992) Xyloglucan endotransglycosylase, a new wall-loosening enzyme activity from plants. Biochem J 282: 821-828 Furch A, Haflre J, Schulz A, van Bel A (2007) Ca2+-mediated remote control of reversible sieve tube occlusion in Vicia faba. J Exp Bot 58: 2827-2838 Gibeaut D, Carpita N (1993) Synthesis of (l —) 3), (1 —> 4)-B-D-glucan in the Golgi apparatus of maize coleoptiles. Proc Natl Acad Sci USA 90: 3850-3854 Gillmor C, Poindexter P, Lorieau J, Palcic M, Somerville C (2002) cL-glucosidase I is required for cellulose biosynthesis and morphogenesis in Arabidopsis. J Cell Biol 156: 1003-1013 Gordon R, Maclachlan G (1989) Incorporation of UDP-[MC] glucose into xyloglucan by Pea membranes. Plant Physiol 91: 373-378 Gouhet F, Misrahi A, Park S, Zhang Z, Twell D, Dupree P (2003) AtCSLA7, a cellulose synthase-like putative glycosyltransferase, is important for pollen tube grth and embryogenesis in Arabidopsis. Plant Physiol 131: 547-557 Handford M, Baldwin T, Gouhet F, Prime T, Miles J, Yu X, Dupree P (2003) Localisation and characterisation of cell wall mannan polysaccharides in Arabidopsis thaliana. Planta 218: 27-36 Hayashi T, Kato Y, Matsuda K (1981) Biosynthesis of xyloglucan in suspension- cultured soybean cells. An assay method for xyloglucan xylosyltransferase and attempted synthesis of xyloglucan from UDP-D-xylose. J Biochem (Tokyo) 89: 325-328 Hazen S, Scott-Craig J, Walton J (2002) Cellulose synthase-like genes of rice. Plant Physiol 128: 336-340 Heller J, Villemez C (1972) Solubilization of a mannose-polymerizing enzyme from Phaseolus aureus. Biochem J 128: 243-252 Henry R, Stone B (1982) Factors influencing B-glucan synthesis by particulate enzymes from suspension-cultured Lolium multzflorum endosperm cells. Plant Physiol 69: 632-636 Hong Z, Delauney A, Verma D (2001a) A cell plate-specific callose synthase and its interaction with phragmoplastin. Plant Cell 13: 755-768 19 Hong Z, Zhang Z, Olson J, Verma D (2001b) A novel UDP-glucose transferase is part of the callose synthase complex and interacts with phragmoplastin at the forming cell plate. Plant Cell 13: 769-779 Hoson T, Nevins D (1989) B—D-glucan antibodies inhibit auxin-induced cell elongation and changes in the cell wall of Zea coleoptile segments. Plant Physiol 90: 1353- 1358 Hrmova M, Farkas V, Lahnstein J, Fincher G (2007) A barley xyloglucan xyloglucosyl transferase covalently links xyloglucan, cellulosic substrates and (1,3;1,4)-B-D-glucans. J Biol Chem 282: 12951-12962 Iwai H, Masaoka N, Ishii T, Satoh S (2002) A pectin glucuronyltransferase gene is essential for intercellular attachment in the plant meristem. Proc Natl Acad Sci USA 99: 16319-16324 Jacobs A, Lipka V, Burton R, Panstruga R, Strizhov N, Schulze-Lefert P, Fincher G (2003) An Arabidopsis callose synthase, GSLS, Is required for wound and papillary callose formation. Plant Cell 15: 2503—2513 Kato Y, Ito S, Iki K, Matsuda K (1982) Xyloglucan and B-D-glucan in cell walls of rice seedlings. Plant and cell physiology 29: 351-364 Kato Y, Nevins D (1984) Enzymic dissociation of Zea shoot cell wall polysaccharides: II. Dissociation of (1 —9 3),(1 —> 4)-B-D-glucan by purified (1 -—) 3),(1 —> 4)-B-D- glucan 4-glucanohydrolase from Bacillus subtilis. Plant Physiol 75: 745-752 Kato Y, Nevins D (1985) A (l —> 3)-B-D-glucan isolated from Zea shoot cell wall preparations. Plant Physiol 78: 20-24 Kimura S, Laosinchai W, Itoh T, Cui X, Linder C, Brown RJ (1999) Immunogold labeling of rosette terminal cellulose-synthesizing complexes in the vascular plant Vigna angularis. Plant Cell 11: 2075-2086 Lai-Kee-Him J, Chanzy H, Miiller M, Putaux J, Imai T, Bulone V (2002) In vitro versus in vivo cellulose microfibrils from plant primary wall synthases: structural differences. J Biol Chem 277: 36931-36939 Lerouxel O, Cavalier D, Liepman A, Keegstra K (2006) Biosynthesis of plant cell wall polysaccharides - a complex process. Curr Opin Plant Biol 9: 621-630 Liepman A, Naim C, Willats W, Sarensen I, Roberts A, Keegstra K (2007) Functional genomic analysis supports conservation of function among CslA gene family members and suggests diverse roles of mannans in plants. Plant Physiol 143: 1881-1893 20 Liepman A, Wilkerson C, Keegstra K (2005) Expression of cellulo se synthase-like (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proc Natl Acad Sci USA 102: 2221-2226 Li J, Burton R, Harvey A, Hrmova M, Wardak A, Stone B, Fincher G (2003a) Biochemical evidence linking a putative callose synthase gene with (1 -) 3)-B-D- glucan biosynthesis in barley. Plant Mol Biol 53: 213-225 Li Y, Jones L, McQueen-Mason S (2003b) Expansins and cell growth. Curr Opin Plant Biol 6: 603-610 Madson M, Dunand C, Li X, Verma R, Vanzin G, Caplan J, Shoue D, Carpita N, Reiter W (2003) The M UR3 gene of Arabidopsis encodes a xyloglucan galactosyltransferase that is evolutionarily related to animal exostosins. Plant Cell 15: 1662-1670 Maeda Y, Awano T, Takabe K, Fujita M (2000) Immunolocalization of glucomannans in the cell wall of differentiating tracheids in Chamaecyparis obtusa. Protoplasma 213: 148-156 McNeil M, Darvill A, Fry S, Albersheim P (1984) Structure and fimction of the primary cell walls of plants. Annu Rev Biochem 53: 625-663 Mitchell R, Dupree P, Shewry P (2007) A novel bioinformatics approach identifies candidate genes for the synthesis and feruloylation of arabinoxylan. Plant Physiol 144: 43-53 Moore P, Swords K, Lynch M, Staehelin L (1991) Spatial organization of the assembly pathways of glycoproteins and complex polysaccharides in the Golgi apparatus of plants. J Cell Biol 112: 589-602 Nicol F, His 1, Jauneau A, Vemhettes S, Canut H, Hofte H (1998) A plasma membrane-bound putative endo-1,4-B-D-glucanase is required for normal wall assembly and cell elongation in Arabidopsis. EMBO J 17 : 5563-5576 Nishimura M, Stein M, Hou B, Vogel J, Edwards H, Somerville S (2003) Loss of a callose synthase results in salicylic acid-dependent disease resistance. Science 301: 969-972 Nishiyama Y, Langan P, Chanzy H (2002) Crystal structure and hydro gen-bonding system in cellulose 13 from synchrotron X-ray and neutron fiber diffraction. J Am Chem Soc 124: 9074-9082 Nishiyama Y, Sugiyama J, Chanzy H, Langan P (2003) Crystal structure and hydrogen bonding system in cellulose Ia fi'om synchrotron X-ray and neutron fiber diffraction. J Am Chem Soc 125: 14300-14306 21 Nobles DR, Brown RM (2004) The pivotal role of cyanobacteria in the evolution of cellulose synthases and cellulose synthase-like proteins. Cellulose 11: 437-448 Nobles D, Romanovicz D, Brown RJ (2001) Cellulose in cyanobacteria. Origin of vascular plant cellulose synthase? Plant Physiol 127: 529-542 Nothnagel E, McNeil M, Albersheim P, Dell A (1983) Host-patho gen interactions: XXII. A galacturonic acid oligosaccharide from plant cell walls elicits phytoalexins. Plant Physiol 71: 916-926 O'Neill M, Eberhard S, Albersheim P, Darvill A (2001) Requirement of borate cross- linking of cell wall rhamnogalacturonan II for Arabidopsis growth. Science 294: 846-849 Pagant S, Bichet A, Sugimoto K, Lerouxel O, Desprez T, McCann M, Lerouge P, Vemhettes S, Hiifte H (2002) KOBITOl encodes a novel plasma membrane protein necessary for normal synthesis of cellulose during cell expansion in Arabidopsis. Plant Cell 14: 2001-2013 Paradez A, Wright A, Ehrhardt DW (2006) Microtubule cortical array organization and plant cell morphogenesis. Curr Opin Plant Biol 9: 571-578 Pear J, Kawagoe Y, Schreckengost W, Delmer D, Stalker D (1996) Higher plants contain homologs of the bacterial celA genes encoding the catalytic subunit of cellulose synthase. Proc Natl Acad Sci USA 93: 12637-12642 Per'ia M, Zhong R, Zhou G, Richardson E, O'neill M, Darvill A, York W, Ye Z (2007) Arabidopsis irregular xylem8 and irregular xylem9: implications for the complexity of glucuronoxylan biosynthesis. Plant Cell 19: 549-563 Perrin R, DeRocher A, Bar-Peled M, Zeng W, Norambuena L, Orellana A, Raikhel N, Keegstra K (1999) Xyloglucan fucosyltransferase, an enzyme involved in plant cell wall biosynthesis. Science 284: 1976-1979 Persson S, Caffall K, Freshour G, Hilley M, Bauer S, Poindexter P, Hahn M, Mohnen D, Somerville C (2007) 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, Wei H, Milne J, Page G, Somerville C (2005) Identification of genes required for cellulose synthesis by regression analysis of public microarray data sets. Proc Natl Acad Sci USA 102: 8633-8638 Popper Z, Fry S (2003) Primary cell wall composition of bryophytes and charophytes. Ann Bot (Lond) 91: 1-12 22 Ray P (1980) Cooperative action of B-glucan synthetase and UDP-xylose xylosyl transferase of Golgi membranes in the synthesis of xyloglucan-like polysaccharide. Biochim Biophys Acta 629: 431-444 Richmond T (2000) Higher plant cellulose synthases. Genome Biol 1: R3001 . 1—3001 .6. Ridley B, O'Neill M, Mohnen D (2001) Pectins: structure, biosynthesis, and oligogalacturonide-related signaling. Phytochemistry 57: 929-967 Rose J, Braam J, Fry S, Nishitani K (2002) The XTH family of enzymes involved in xyloglucan endotransglucosylation and endohydrolysis: current perspectives and a new unifying nomenclature. Plant Cell Physiol 43: 1421-1435 Salnikov V, Grimson M, Delmer D, Haigler C (2001) Sucrose synthase localizes to cellulose synthesis sites in tracheary elements. Phytochemistry 57: 823-833 Sauter M, Kende H (1992) Levels of B-glucan and lignin in elongating intemodes of deepwater Rice. Plant cell physiology 22: 1089-1097 Schindelman G, Morikami A, Jung J, Baskin T, Carpita N, Derbyshire P, McCann M, Benfey P (2001) COBRA encodes a putative GPI-anchored protein, which is polarly localized and necessary for oriented cell expansion in Arabidopsis. Genes Dev 15:1115-1127 Schrfider R, Wegrzyn T, Bolitho K, Redgwell R (2004) Mannan transglycosylase: a novel enzyme activity in cell walls of higher plants. Planta 219: 590-600 Seifert G (2004) Nucleotide sugar interconversions and cell wall biosynthesis: how to bring the inside to the outside. Curr Opin Plant Biol 7: 277-284 Seifert G, Roberts K (2007) The biology of arabinogalactan proteins. Annu Rev Plant Biol 58: 137-61 Smith BG, Harris PJ (1998) The polysaccharide composition of Poales cell walls: Poaceae cell walls are not unique. Biochemical systematics and ecology 27: 33-53 Smith R, Fry S (1991) Endotransglycosylation of xyloglucans in plant cell suspension cultures. Biochem J 279 (Pt 2): 529-535 Somerville C (2006) Cellulose synthesis in higher plants. Annu Rev Cell Dev Biol 22: 53-78 Suzuki S, Li L, Sun Y, Chiang V (2006) The cellulose synthase gene superfanrily and biochemical functions of xylem-specific cellulose synthase-like genes in Populus trichocarpa. Plant Physiol 142: 1233-1245 23 Taylor N, Howells R, Huttly A, Vickers K, Turner S (2003) Interactions among three distinct CesA proteins essential for cellulose synthesis. Proc Natl Acad Sci USA 100: 1450-1455 Taylor N, Laurie S, Turner S (2000) Multiple cellulose synthase catalytic subunits are required for cellulose synthesis in Arabidopsis. Plant Cell 12: 2529-2540 Taylor N, Scheible W, Cutler S, Somerville C, Turner S (1999) The irregular xylem3 locus of Arabidopsis encodes a cellulose synthase required for secondary cell wall synthesis. Plant Cell 11: 769-780 Teleman A, Lundqvist J, Tjemeld F, Stalbrand H, Dahhnan O (2000) Characterization of acetylated 4-O-methylglucuronoxylan isolated from aspen employing 1H and 13C NMR spectroscopy. Carbohydr Res 329: 807-815 Turner A, Bacic A, Harris P, Read S (1998) Membrane fractionation and enrichment of callose synthase from pollen tubes of Nicotiana alata Link et Otto. Planta 205: 380-388 Vanzin G, Madson M, Carpita N, Raikhel N, Keegstra K, Reiter W (2002) The mur2 mutant of Arabidopsis thaliana lacks fucosylated xyloglucan because of a lesion in fucosyltransferase AIF UT 1 . Proc Natl Acad Sci USA 99: 3340-3345 Waldron K, Brett C (1983) A glucuronyltransferase involved in glucuronoxylan synthesis in pea (Pisum sativum) epicotyls. Biochem J 213: 115-122 Wang X, Cnops G, Vanderhaeghen R, De Block S, Van Montagu M, Van Lijsebettens M (2001) AtCSLD3, a cellulose synthase-like gene important for root hair grth in Arabidopsis. Plant Physiol 126: 575-586 Yokoyama R, Nishitani K (2001) A comprehensive expression analysis of all members of a gene family encoding cell-wall enzymes allowed us to predict cis-regulatory regions involved in cell-wall construction in specific organs of Arabidopsis. Plant Cell Physiol 42: 1025-1033 Yokoyama R, Rose J, Nishitani K (2004) A surprising diversity and abundance of xyloglucan endotransglucosylase/hydrolases in rice. Classification and expression analysis. Plant Physiol 134: 1088-1099 York W, Darvill A, Albersheim P (1984) Inhibition of 2,4-dichlorophenoxyacetic acid- stimulated elongation of Pea stem segments by a xyloglucan oligosaccharide. Plant Physiol 75: 295-297 Zhong R, Burk D, Morrison Wr, Ye Z (2002) A kinesin-like protein is essential for oriented deposition of cellulose microfibrils and cell wall strength. Plant Cell 14: 3101-3 1 17 24 Zhong R, Morrison Wr, Freshour G, Hahn M, Ye Z (2003) Expression of a mutant form of cellulose synthase AtCesA7 causes dominant negative effect on cellulose biosynthesis. Plant Physiol 132: 786-795 Zhong R, Pefia M, Zhou G, Naim C, Wood-Jones A, Richardson E, Morrison Wr, Darvill A, York W, Ye Z (2005) Arabidopsis fragile fiber8, which encodes a putative glucuronyltransferase, is essential for normal secondary wall synthesis. Plant Cell 17: 3390-3408 Zhu Y, Nam J, Carpita N, Matthysse A, Gelvin S (2003) Agrobacterium-mediated root transformation is inhibited by mutation of an Arabidopsis cellulose synthase-like gene. Plant Physiol 133: 1000-1010 25 CHAPTER 2: THE LIGHT-REGULATED GOLGI-LOCALIZED GLUCAN SYNTHASE FROM MAIZE: PROPERTIES, PRODUCT ANALYSIS AND SOLUBILIZATION ABSTRACT In maize, a Golgi-localized enzymatic activity uses UDP-glucose as a substrate and incorporates it into ethanol-insoluble products. This enzymatic activity is light and auxin regulated (Walton and Ray 1982a, 1982b). We hypothesized that this enzymatic activity might be involved in the biosynthesis of a hemicellulosic glucan. We determined that this UDP-glucose incorporating enzymatic activity is a (1,4)-B-glucan synthase and attempted to identify it using biochemical methods. We solubilized the (l,4)-B-glucan synthase activity using digitonin. After subsequent hydroxyapatite chromatography glucose was no longer incorporated into (l,4)-B-glucan but in glucose-6-phosphate (G6P) and glucose-1,6-diphosphate (G1,6dP) instead. INTRODUCTION Glucan synthases synthesize starch from ADP-glucose or (1,4)-B-glucans and (1,3)-B-g1ucans from UDP-glucose or GDP-glucose (Barber et al, 1964; Villemez et al., 1967; Ray et al., 1969; Tsai and Hassid, 1971; Ball et al., 2003). Biochemical identification of enzymes involved in cell wall biosynthesis is difficult due to their hydrophobic nature, instability and low levels (Meikle et al., 1991; Dhugga et al., 1994). However, this approach has been successfully used to identify galactomannan galactosyltransferase (Edwards et al., 1999), xyloglucan fucosyltransferase (Perrin et al., 26 1999), callose synthase (Li et al., 2003) and galacturonan synthase involved in pectin biosynthesis (Sterling et al., 2006). In this part of my research I attempted to identify a Golgi-localized glucan synthase using biochemical methods. As a first step I optimized the enzyme assay. As a second step, I determined the product it synthesizes. As a third step, I solubilized and partially purified the glucan synthase. MATERIALS AND METHODS Growth of maize Seedlings Maize seeds (variety DK355, Monsanto, or variety FR1061 X FR9661, Midwest Seed Genetics, St Carroll, IA) were imbibed at room temperature for 24 h in an Erlenmeyer flask shaking at 150 rpm under white light. The seeds were spread on top of trays of fine vermiculite soaked in water, covered with a thin layer of vermiculite and covered with plastic. The trays were kept in a completely dark room at 24°C and after two days the plastic cover was removed. On the evening of day three, one tray of seedlings was given a 15 min white light exposure (97 uMol s'l m’z), after which it was placed back in the dark room. Sixteen hours later the second cm of the mesocotyl below the first node was harvested under white light. Isolation of total membranes An equal weight of mesocotyl segments isolated from dark-grown and light- treated maize seedlings were ground in buffer (50 mM Tris buffer pH 8, 1 mM EDTA, 0.01 mM MgSO4 and 250 mM sucrose). All steps of this procedure were performed at 27 4°C. The amount of buffer added was 3 ml/g of fresh weight. The mesocotyl segments were ground in a mortar and filtered through two layers of cheesecloth. The flow-through was centrifuged for 5 min at 10,000 rpm in a Sorval SS34 rotor. The supernatant was centrifuged for 20 min at 146,000x g in a Ti-50 rotor (Beckman Coulter). The pellet was resuspended by homogenization in grinding buffer at a final concentration of l ml/g of original fresh weight. (1,4)-B-glucan synthase assay One hundred ul of microsomal suspension (100-200 ug of protein) was added to a 15 ml Pyrex glass tube, MgSO4 was added to a concentration of 20 mM and UDP- [3H]glucose (34 Ci/mmol, Sigma) to 3.7-49 nM (below 49 nM the product synthesized was similar). The reaction mixture was adjusted with grinding buffer to a volume of 110- 200 pl, vortexed, and incubated for 1 h at room temperature (21°C). Five ml of 70% ethanol at 4°C was added to the tube and vortexed. This mixture was filtered through a glass fiber filter (GF/A, 2.5 cm diameter, Cat. No. 1820025, Whatman) with a vacuum manifold (Millipore) to collect the radiolabeled ethanol-insoluble products. The tube was rinsed with 5 ml of ethanol and this was also filtered. Finally it was washed with 5 ml of ethanol and dried. The radioactivity on the dried filter was measured using a scintillation counter (1.85000 TDC, Beckman Coulter). 28 Callose synthase assay Callose synthase assays were performed in the same manner as described for the (1,4)-B- glucan synthase except that the UDP-glucose concentration was 1 mM and no Mg2+ and Ca2+ were added. Enzymatic hydrolysis of radiolabeled products One ml of buffer was added to the dried filters containing the radiolabeled products. The specificity of the enzymes and buffers used are described in Table 1. Five ul of the respective enzyme solution was added to the filters soaked in buffer, after which they were incubated overnight at room temperature. Nine ml of 100% ethanol at 4°C was added, and this solution was filtered through GF/A filters. The filter remaining in the vial in which the enzyme treatment was performed was collected on the new filter. The vial in which the enzyme treatment was performed was rinsed once with 90% ethanol at 4°C. Finally the two filters were washed once with ethanol, dried and radioactivity was measured. The radioactivity remaining on the filter was compared to a control that was not treated with hydrolytic enzymes. 29 TABLE 1: Enzymes used, their catalog number, buffers used and substrates they hydrolyze *obtained from Megazyme, "Roche, ***Sigma Enzyme name Cat. No Buffer used Substrates hydrolyzed alkaline P7640 50 mM tris-glycine buffer phosphate attached to many phosphatase*** pH 8.8 types of molecules a-amylase (10 A6380 50 mM MOPS pH 7 starch mg/ml)*** cellulase* Lot 30201 50 mM acetate buffer pH 5 cellulose xyloglucan MLG ((1—)3,1—>4)—B-D-glucan) glucomannan xylan Driselase from D8037 50 mM acetate buffer pH 5 callose Basidiomycetes xylan sp.*** cellulose endo- l ,4-B-D- Lot 00901 50 mM acetate buffer pH 4 galactan galactanase* polygalacturonic acid arabinan birch-wood xylan citrus pectin cellulose B—l-(3,4,6)- G1288 50 mM acetate buffer pH 5 releases B-l—>3, (34—>4, and B- galactosidase*** l—+6-linked galactose from the non-reducing end of complex oligosaccharides 30 TABLE 1 Continued larninarinase* Lot 90601 50 mM acetate buffer pH (l—>3)-B-D—glucan 4.5 lichenase* Lot 30501 50 mM MES, pH 6 MLG B-mannanase* Lot 21101 50 mM tris-glycine buffer Carob galactomannan pH 8.8 pectinase from P2736 50 mM acetate buffer pH 4 pectin Aspergillus hemicelluloses niger*** cellulose proteinase K** 12222300 50 mM Tris pH 8 proteins (20mg/ml) xylanase M1* Lot 70502 50 mM acetate buffer pH 5 xylan CM-cellulose 4M CM-cellulose 6M xylanase M6* Lot 51206 50 mM acetate buffer pH 5 xylan Collection of hydrolysis products The same procedure was followed as described for the enzymatic digestion of the radiolabeled products, except that after the addition of ethanol to a final concentration of 90%, the solution was collected in eppendorf tubes. The buffer/ethanol solution was dried under vacuum and the radiolabeled hydrolysis products were redissolved in 100 pl of water. Total hydrolysis of the products Two ml of 2 M trifluoroacetic acid (TFA) was added to a screw cab tube containing a filter with ethanol-insoluble product. The tube was heated for 2 h at 121°C 31 and subsequently cooled to room temperature. The TFA was evaporated and the filter was rinsed twice with 90% ethanol. The ethanol rinses were collected in an eppendorf tube and dried. High-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) The HPAEC-PAD analysis (Dionex) was equipped with a 25 ul injection loop and an ED50 electrochemical pulsed amperometric detector (PAD). The radiolabeled digestion products were mixed with oligosaccharide standards (1-5 pg each) and analyzed using a CarboPac PA-l column (4 x 250 mm) run at a flow rate of 1 ml a min. One or two min fractions were collected and neutralized with HCl before scintillation counting. The identity of the radiolabeled compounds was determined by co-elution with the standards. Table 2 describes the solutions and gradients used for each type of analysis. 32 TABLE 2: Gradients used for the HPAEC-PAD analysis of different products Type of products analyzed (program name) Gradient Monosaccharides Isocratic at 18 mM NaOH for 20 min Oligosaccharides (cellobiose, cellotriose, Isocratic at 90 mM NaOH for 30 min laminaribiose, xylobiose, xylotriose) Cellooligosaccharides 1 (glucose, cellobiose, Isocratic at 100 mM NaOH for 30 min, cellotriose, cellotetraose, ce110pentaose, gradient of 100-500 mM NaOH in 10 min, cellohexaose) isocratic at 500 mM NaOH + 500 mM NaOAc Cellooligosaccharides 2 (glucose, cellobiose, A gradient from 0-500 mM N aOAc in 30 min and cellotriose, cellotetraose, cellopentaose, then constant 500 mM NaOH cellohexaose) Glucose-phosphates Isocratic at 39% buffer B for 10 min, gradient from 39%-100% buffer B in 30 min. Buffer A: 59 mM NaOH, 196 mM NaOAc Buffer B: 50 mM NaOH, 75 mM NaOAc Manno/xylooligosaccharides Isocratic at 20 mM NaOH for 20 min, gradient from 20-90 mM NaOH in 20 min, isocratic at 500 mM NaOH Mixed linkage glucan (MLG) 15 min gradient from 0.5 M NaOH to 0.5 NaOH and 0.25 M NaOAc Solubilization of the (l,4)-B-glucan Reactions were performed and the filters with the products were dried. Solubility of the (1,4)-B-glucan was tested by adding different solvents to filters containing ethanol- insoluble reaction products and incubating them overnight. Next day the solvents were removed and the radioactivity remaining on the filter was determined. Solutions used 33 were 0.2 N, 1 N, 2 N and 4 N NaOH, 0.5% ammonium oxalate, 15.8 M acetic / 1.5 M nitric acid, water, dimethyl sulfoxide (DMSO), DMSO plus 4.2 M l-methylimidazole (MI) (Cat. No.: 336092, Sigma) and DMSO plus 0.5 M tetrabutylammonium fluoride (TBAF) (Cat. No.: 86843, Sigma). Ammonium oxalate (0.5% w/v) dissolves non- crystalline cellulose attached to a protein by heating cell wall material in this solution for 1 hour at 100°C (Peng et al., 2001). Acetic/nitric acid dissolves non-cellulosic (1,4)-B- glucan (Updegraff, 1969). MI and TBAF will dissolve crystalline cellulose (Heinze et al., 2001; Lu et al., 2003). DMSO plus TBAF dissolves cellulose up to a degree of polymerization of 650 within 15 minutes without any pretreatment at room temperature (Heinze et al., 2001). The acetic/nitric acid solution was prepared by mixing 150 ml 80% acetic acid and 15 ml concentrated nitric acid. The DMSO plus TBAF solution was prepared by mixing 6.6 g of TBAF and 33 ml of DMSO. The DMSO plus MI solution was prepared by adding 5 ml of MI to 10 ml of DMSO. Selection of the optimal detergent for solubilization of the glucan synthase (1,4)-B-glucan synthase activity was determined in the presence of different detergents. One hundred 111 of microsomes were incubated with 0.01%, 0.1% or 1% detergent for 30 min on ice and assayed. Detergents used were Brij35 (Cat. No. 430A9-6, Sigma), digitonin (Cat. No. D141, Sigma), Triton X100 (Cat. No. 111037, Research Products International), Tween 20 (Cat. No. P7949, Sigma), CHAPS (Cat. No. C5070, Sigma), Zwittergent (Cat. No. 693030, Calbiochem) and n-octoglucoside (Cat. No. 494460, Calbiochem). Detergents that inhibited the glucan synthase activity the least were tested for their ability to solubilize this enzymatic activity. Microsomes were 34 incubated with different detergent concentrations and for different time periods. The detergent / microsome mixture was centrifuged at 146,000 rpm in a Ti-50 rotor for 20 min, and the supernatant and pellet were separated and both were assayed. Solubilization efficiency was determined by comparing the amount of ethanol-insoluble product synthesized by the supernatant to the ethanol-insoluble product synthesized by the pellet. Optimized solubilization procedure for the glucan synthase Microsomal membranes were resuspended in solubilization buffer (10 mM Tris pH 8, 1 mM MgSO4, 4 mM DTT, 20% glycerol) by homogenization at a concentration of 1 ml/g of fresh weight. Digitonin was added to a final concentration of 0.5%, mixed with the microsomal solution and kept on ice for 20 min. This mixture was centrifuged for 20 min at 146,000 g and the supernatant was collected. The resulting pellet was resuspended in half the fresh weight volume of solubilization buffer and homogenized. Digitonin was added to a concentration of 0.5%; the mixture was incubated on ice for 20 min and centrifuged for 20 min at 146,000 g. The supernatant was collected, combined with the supernatant of the first solubilization and stored at -80°C for later use. Hydroxyapatite chromatography Hydroxyapatite chromatography (5 ml Econo-Pac CHT-II Cartridge, Cat. No. 732-0081, Biorad) was performed under the following conditions: a 30 min gradient was run from 1-100 mM potassium phosphate buffer, pH 8.0, at a flow rate of 0.5 ml/min, followed by a wash with 400 mM phosphate buffer pH 8.0. Two min (1 ml) fractions were collected and 100 pl was used for glucan synthase assays. 35 Measurement of protein concentrations Protein concentrations were measured with the Bradford assay (Bradford, 1976). Five ml of Bradford solution was added to 10-100 ul of sample, vortexed and incubated at room temperature for 10 min. The Bradford stock solution contained 300 mg Coomassie Brilliant Blue G-250 (Cat. No. 27815, Fluka) dissolved in 150 ml 95% ethanol and 300 ml 85% phosphoric acid. The Bradford working solution was 15 ml of stock solution diluted to 100 ml with water. The OD595 was measured and the protein concentration was determined with bovine immunoglobulin as standard. 36 RESULTS Optimization of the glucan synthase assay The optimal MgClz concentration for glucan synthase activity is 15 mM (Figure 1). Therefore all the glucan synthase assays were performed in the presence of 20 mM Mgz". 0.8 0.6 ‘ 0.4 ‘ 0.2 1 Enzyme activity (fmol/ug) 0.0 I l l I I 0 5 10 15 20 25 30 Mgcr2 (mM) Figure 1. Effect of different MgClz concentrations on glucan synthase activity (measured as fmol [3H] glucose incorporated into ethanol-insoluble products per ug of protein) using microsomes isolated from dark-grown seedlings. This experiment was repeated several times and a representative result is shown in this figure. The pH of the microsomal solution was adjusted by adding either KH2P04 [pH 6.5, 7 and 7.5] or Tris [pH 7.5, 8 and 8.5]. The glucan synthase has a broad pH optimum from 7.580 (Figure 2). Therefore all the glucan synthase assays were performed at pH 8. 37 p—s N 1 Enzyme actrvrty (fmol/ug) P . 00 p A 1' : I I I I I I I u I I l I l l ‘ .' : : I I I l .9 O .4 Figure 2. Effect of pH on glucan synthase activity; microsomes isolated from dark-grown (solid line), and light-treated (dashed line) seedlings. This experiment was repeated several times and a representative result is shown in this figure. Glucan synthase activity is proportional to protein concentration Glucan synthase assays were performed by adding increasing amounts of microsomes to the assay. Figure 3 shows that the specific activity is constant over a 5- fold range of protein concentration for microsomes isolated from both the dark-grown and light-treated seedlings. 38 A 1.0 on _ 3. g 0.8 - i’ N “E _ 5,, 0.6 7 3.5 ‘ ------- i ---------- {~- 8 0 4 ‘ """ i --------- f E _ g. 0.2 r m _ 0.0 l I T r r 20 40 60 80 100 Microsomes added (ul) Figure 3. Specific glucan synthase activity over a 5-fold range of protein concentration. Data represent the mean (n = 3) t SE for microsomes isolated from dark-grown (solid line) and (n = 2) i SE for microsomes isolated from light-treated (dashed line) seedlings. Researchers in the past have shown that glucose from UDP-glucose can be incorporated into several different products which can contain B(1,4), B(l,3) and both B(1,4) and B(1,3) linked glucose (Tsai et al., 1971; Gibeaut et al., 1993; Kudlicka et al., 1997). In order to determine in which products glucose is incorporated in the case of the glucan synthase, we performed assays at different UDP-glucose concentrations and treated the products with cellulase, laminarinase, or lichenase. 39 The composition of the product is dependent on the UDP-glucose concentration A. Assays with microsomes isolated from dark-grown seedlings. When assays were performed with 20 nM UDP-[3H] glucose, the product was mainly solubilized by cellulase (Figure 4). Xylanase, proteinase K, or (rt-amylase did not solubilize the radiolabeled product by more than 25% (Figure 5 and data not shown). When the UDP-glucose concentration was 10 11M the products were partially solubilized by cellulase and partially by laminarinase. At 1 mM UDP-glucose the products were mainly solubilized by laminarinase (Figure 4). B. Assays with microsomes isolated from light-treated seedlings. After light-treatment there was a ~70% reduction in activity at 20 nM UDP-Glc, but the product was still mainly cellulase-sensitive (Figure 4). After light-treatment there was a ~50% reduction in activity when the UDP-glucose concentration was 10 11M and the products were mainly solubilized by laminarinase. After light-treatment there was a ~15% increase in activity when the UDP-glucose concentration was 1 mM and the product was also mainly solubilized by laminarinase. In summary, these results show that at low UDP-glucose concentration the product is mainly solubilized by cellulase and light has a large effect on activity. At high UDP-glucose concentration the product is mainly solubilized by laminarinase and light has no effect. These results can be explained in the following way. At low UDP-glucose concentrations a (1,4)-B-glucan synthase is active which has a low Km and a low Vmax. This enzyme is down-regulated by light. At high UDP-glucose concentrations a different 40 enzyme, callose synthase is active, which has a high Km and a high Vmax. Callose synthase is not affected by light. 41 Figure 4. Enzymatic sensitivity of the glucan synthase product made at different UDP- Glc concentrations. Solid bars: assays with microsomes from dark-grown seedlings; white bars: assays with microsomes from light-treated seedlings. Data represent the mean (n = 3) i SE for assays performed at 20 nM or 10 uM UDP-glucose and (n = 2) i SE for assays performed at 1000 uM UDP-glucose. 42 Figure 4 20 nM UDP-glucose 3 2 J 0 0 0 22:8 mEEaEE 8.68m 0. O 10 M UDP-glucose d u q u - a 0 0 6 2 1 1 £083 wEEaEB 8.695 m w 0 1000 [AM UDP-glucose iWi-q-q in. 06 21 29:3 wEEmEE 8.695 2. 8 4 1 Wm. 8.2.23: 03:55:8— 822—8 35:8 Enzyme treatment 43 JL §..§.§ § Product remaining (fmol) o l control xylanase proteinase K Figure 5. Products synthesized with microsomes isolated from dark-grown seedlings treated with xylanase M1 and proteinase K. Data represent the mean (n = 4) i SE for the control and (n =3) i SE for xylanase M1 and proteinase K. The cellulase that was used for solubilizing the products was also active on other substrates (see Table 1). Therefore, to determine the chemical nature of the glucan synthesized by the light-regulated glucan synthase, further characterization of the product was necessary. The product synthesized at 3.7 nM UDP-glucose is a (l,4)-B-glucan The product solubilized by cellulase was analyzed by HPAEC-PAD using the oligosaccharide program (see Table 2). The major hydrolysis products have the same elution time as cellobiose and glucose (Figure 6), which indicates that the product in which [3H] glucose is incorporated is a (l,4)-B-g1ucan both for microsomes isolated from dark-grown and light-treated seedlings. The monosaccharide composition of the products was also analyzed. The products were hydrolyzed with TFA and analyzed by HPAEC-PAD using the monosaccharide program (see Table 2). This showed that the radiolabel was in glucose for assays performed with microsomes isolated both from dark-grown and light-treated seedlings (Figure 7). 3 A G1 Dark C X2 C2 E 3‘3 23 3000 ~ B .l g, 2000 - ..l 1000 - 0 3 C GI Light I: o G. if) C2 Q X2 X3 35 l i L2 can 3000 - D E2000 - 1000 - O - A A4 A 10 20 Time (min) Figure 6. The glucan synthase products synthesized with microsomes isolated from dark- grown (B) and light-treated (D) seedlings were treated with cellulase and the hydrolysis products were analyzed by HPAEC-PAD. A and C, Oligosaccharide standards: G1 (glucose), X2 (xylobiose), C2 (cellobiose), X3 (xylotriose), L2 (laminaribiose). B and D, Radioactivity. 45 A Dark 8 A1 = o s- ?! Gall a l Glx1 a. 2000- B E "l Q. Q J 1000- l \. l O 8 Light I: o 8‘ 2 Gall a G1 X1 m A 20004 D g- - 1000- A .4 N 0 v T 10 20 Time (min) Figure 7. The glucan synthase products synthesized with microsomes isolated from dark- grown (B) and light-treated (D) seedlings were hydrolyzed with TFA and the products were analyzed by HPAEC-PAD. A and C, Monosaccharide standards: A1 (arabinose), Gall (galactose), Gl (glucose), X1 (xylose). B and D, Radioactivity. At 1 mM UDP-glucose, the product synthesized is a (1,3)-B-glucan The product synthesized at 1 mM UDP-glucose was analyzed in a similar way as described for the product synthesized at 3.7 nM UDP-[3 H] glucose. The hydrolysis products elute with larninaribiose and glucose (Figure 8), which indicates that the product in which UDP-[3H] glucose is incorporated is a (1,3)-B—glucan. A Q a G1 § 2 x2 3 L A C2 x3 L2 .1 r _/\/\_/\- 1000-B g .. a 500- A /\ 0 . . - .. 10 20 30 Time (min) Figure 8. The glucan synthase product synthesized at 1 mM UDP-glucose was hydrolyzed with laminarinase and the products were analyzed by HPAEC-PAD. A, Oligosaccharide standards. B, Radioactivity. The product in which [3H]glucose is incorporated is non-crystalline (1—94)-B-glucan and is not attached to a protein. Important properties of the (1,4)—B—glucan are its size and solubility. The (1,4)-l3- glucan synthesized at low UDP-Glc concentration was solubilized by 4 N NaOH, but not by 0.2 N or 2 N NaOH. It became insoluble again after dilution of 4 N NaOH to 1 N (data not shown). This solubilization experiment indicates that the product has the properties of a hemicellulose. The product was not soluble in 0.5% ammonium oxalate (data not 47 shown) (Peng et al., 2001). The product was completely dissolved by Updegraff’ s reagent. The product was solubilized partially by DMSO, and DMSO plus MI (Figure 9). It was almost completely solubilized by DMSO plus TBAF (Figure 9). The product was reprecipitated again when the DMSO/'1‘ BAF solution was diluted tenfold with water (data not shown). These results are consistent with the product being non-crystalline (1,4)-B- glucan and not attached to a protein. A 100 '8 - .3 80 ~ 2" 60 " "-1 -l = .a . E 40- 8 20 _ ‘3 _ a: 0 - i control water DMSO DMSO + DMSO + MI TBAF Figure 9. Solubility of the (1,4)-B-glucan in different solvents. Filters containing ethanol- insoluble products were incubated with the above mentioned solutions, filtered and the remaining radioactivity was determined. Glucan synthase assays in the presence of UDP-xylose (1,4)-B-glucan is found in xyloglucan and in cellulose (Haysashi, 1989). To test whether the glucan synthase is involved in xyloglucan biosynthesis assays were performed in the presence of both UDP-glucose and UDP-xylose. It has been shown that plant membranes can synthesize xyloglucan when UDP-glucose and UDP-xylose are 48 both added to the enzyme assay (Hayashi and Matsuda, 1981). If the maize (1,4)-B- glucan synthase is involved in xyloglucan biosynthesis, there is a possibility that addition of UDP-glucose and UDP-xylose to the enzyme assay together might result in xyloglucan biosynthesis in vitro. However, in vitro xyloglucan assays have been developed only for dicots and not for monocots. Glucan synthase assays were performed in the presence of constant concentration of UDP-[3H] glucose (49 nM) and varying concentrations of UDP-xylose. UDP-xylose inhibited incorporation of UDP-[3H] glucose (half maximum inhibition at ~50 nM). The effect was similar for assays performed with microsomes isolated from dark-grown and light-treated seedlings (Figure 10). The inhibition of UDP-glucose incorporation by UDP-xylose has also been observed for xyloglucan biosynthesis in dicots. However, there is a stimulation of xyloglucan biosynthesis when the UDP-xylose concentration is lower than the UDP-glucose concentrations in the enzyme assays (Gordon and Maclachlan, 1989; Hayashi and Matsuda, 1981). I did not observe any stimulation in my enzyme assays by UDP-xylose, even at a concentration as low as 25 nM. 49 1.2 1.0 - 0.8 ‘ 0.6 ‘ 0.4 s 0.2 t L .— ll 53 530 control 5.3 Enzyme activity (fmol/ug) 11M UDP—xylose Figure 10. Glucan synthase assays in the presence of UDP-xylose. Assays were performed at constant UDP-Glc concentrations (49 nM) with microsomes isolated from dark-grown (gray bars), and light-treated (white bars) seedlings. No UDP-xylose was added to the control. To test further the hypothesis that the maize (1,4)-fi-glucan synthase is involved in xyloglucan biosynthesis, the products synthesized in the presence of UDP- [3H] glucose and UDP-xylose (50 nM) were treated with Driselase. Driselase is known to release isoprimeverose from xyloglucan (Popper and Fry, 2003). The solubilized products were collected, mixed with oligosaccharide and isoprimeverose standards and analyzed by HPAEC-PAD with the oligosaccharide program (see Table 2). The radiolabeled hydrolysis products coeluted with cellobiose and glucose, but not with isoprimeverose (data not shown). No in vitro xyloglucan biosynthesis was observed. 50 The (l,4)-B-glucan synthase activity is not inhibited by cellulose synthase inhibitors In order to determine if perhaps the (l,4)-B-glucan synthase is a cellulose synthase in transit to the plasma membrane, (l,4)-B-glucan synthase assays were performed in the presence of 2,6-dichlorobenzonitrile (DCB) and isoxaben. DCB and isoxaben are known to inhibit cellulose biosynthesis although it is not known if this inhibition is a direct or an indirect effect (Peng et al., 2001; Desprez et al., 2002, Scheible et al., 2001). DCB inhibits cellulose biosynthesis, but causes no accumulation of non- crystalline cellulose (Peng et al., 2001). DCB inhibits cellulose biosynthesis in barley cell cultures, but not hemicellulose biosynthesis (Sheletzky et al., 1992). Figure 11 shows that the (l,4)-B-glucan synthase activity is not inhibited by DCB or isoxaben. Enzyme activity (fmol/ug) N l control DMSO DCB ISO Figure 11. Glucan synthase assays performed in the presence of DCB and isoxaben (ISO). DCB and isoxaben were dissolved in DMSO and used at a concentration of 8 11M and 8 nM, respectively. The DMSO concentration was 0.8%. 51 From these results no definitive conclusion can be made if the (l,4)-B-glucan synthase is a cellulose synthase or not, because the mode of action of these herbicides is not known. If they inhibit cellulose biosynthesis directly then the (l,4)-B-glucan synthase is most likely not a cellulose synthase. Solubilization of the (1,4)-B-glucan synthase In order to fractionate proteins using chromatographic methods they have to be in solution. The (1,4)-B-glucan synthase is a membrane protein and therefore insoluble. In order to solubilize it I tested several different detergents for their effect on (1,4)-B-g1ucan synthase activity. The detergents which had the least inhibitory effect were tested for their ability to solubilize the (l,4)-B-glucan synthase. Most detergents when used at 0.01% concentration did not inhibit the activity. At a concentration of 1.0% most of the detergents inhibited the enzyme strongly, except Brij35, digitonin and Tween 20 (Table 3). 52 Table 3. The effect of different detergents on (1,4)-B-glucan synthase activity. The activity remaining after detergent treatment is given as % compared to the control (n = nonionic, z = zwitterionic; 100% activity = 10,000 dpm). Detergent Activity compared to control (%) Detergent concentration 1% 0.1% 0.01% Brij35 (n) 100 100 95 Digitonin (n) 63 44 143 Triton X100 (11) 2 8 28 Tween 20 (n) 35 50 91 CHAPS (z) 3 65 94 Zwittergent (z) 2 3 22 n-octoglucoside (n) 2 37 71 Brij35, digitonin, and Tween 20 were tested for their ability to solubilize the (1,4)-B-glucan synthase. Digitonin solubilized it most efficiently. Several different digitonin concentrations were tested and it was concluded that digitonin solubilized the activity most efficiently at a concentration of 0.25%-0.5% (0.0025-0.005 pg digitonin/ pg protein). At this concentration 19% of (1,4)-B-glucan synthase activity which was present in the total microsomes was recovered (Figure 12). 53 120 100‘ 80‘ 60' activity (%) 40‘ 20 — total solubilized pellet after microsomes solubilization Figure 12. Comparison of (l,4)—B-glucan synthase activity after solubilization with 0.5% digitonin with the activity in the microsomes. Data represent the mean (11 = 3) i SE. Enzyme assays were performed with the solubilized (1,4)-B-glucan synthase preparation and the products were treated with different hydrolytic enzymes. The products were solubilized more than 80% by cellulase. Laminarinase, xylanase, and lichenase appeared to solubilize the product but to a much lesser degree (Figure 13). 54 853 L l Product remaining (frnol) 9.5.8 \ as?“ . 4r (9&0 \V§ . .efi’y Ag ~96? 6° ‘53 V . 't-v q, \ {‘6' Figure 13. Enzyme treatment of the products synthesized by the solubilized (1,4)-B- glucan synthase preparation. This graph shows the remaining product after enzyme treatments. The solubilized products were mixed with oligosaccharide standards and analyzed by HPAEC-PAD. Only cellulase gave any detectable product, which co-eluted with cellobiose and glucose (Figure 14). The other enzymes gave no detectable products (data not shown). Therefore, after solubilization the glucan synthase still synthesizes a (1,4)-B— glucan. 55 8 g Gl a. i3 3 X2 C2 L2 a. l A X3 1000 r B Time (min) Figure 14. Analysis of the product synthesized by the solubilized (1,4)-B-glucan synthase. The product was treated with cellulase and the hydrolysis products were analyzed by HPAEC-PAD. A, Oligosaccharide standards. B, Radioactivity. Fractionation of the solubilized (l,4)-B-glucan synthase We attempted to fractionate the solubilized (1,4)-B-glucan synthase using anion exchange, hydrophobic interaction, size exclusion and hydroxyapatite chromatography. Apparent activity was recovered after hydroxyapatite chromatography (Figure 15). The total amount of ethanol-insoluble products synthesized after hydroxyapatite chromatography was comparable to the total amount of ethanol-insoluble products synthesized in the solubilized glucan synthase preparation before chromatography. 56 I 14000- — =PO4 concentration 400mM ' - , -0.6 8.10000~ :‘i - a ". . g _ I‘ .0.4 I .2 : ‘. a :2 6000- : ‘-. ‘ g 1?: l l - -0.2 8 : E. .' § time (min) Figure 15. Hydroxyapatite fractionation of solubilized (l,4)-B-glucan synthase. A 30- minute gradient was run from 1 to 100 mM potassium phosphate, pH 8, followed by 400 mM potassium phosphate, pH 8 (solid line). The Ongo trace (dashed line) shows that the majority of protein is separated from the fractions containing the activity (diamonds). The products synthesized after hydroxyapatite chromatography were precipitated with 70% ethanol, collected by centrifugation, and dried. Water was added to redissolve the products, cellooligosaccharide standards were added, and the products were analyzed by HPAEC-PAD using the cellooligosaccharide 1 program (see Table 2). Although the product before hydroxyapatite chromatography was not water-soluble (Figure 9), after hydroxyapatite chromatography the product was completely water-soluble. The radiolabeled product eluted near the cellopentaose and cellohexaose standards (Figure 16). 57 A Glc C6 C3 C4C5 C2 § PAD response me 500‘ O l l . I 0 3O 6O Time (min) Figure 16. The ethanol-insoluble and water-soluble product synthesized after hydroxyapatite chromatography was analyzed by HPAEC-PAD. A, Oligosaccharide standards: Glc (glucose), C2 (cellobiose), C3 (cellotetriose), C4 (cellotetraose), C5 (cellopentaose), C6 (cellohexaose). B, Radioactivity. To further characterize the product synthesized by the hydroxyapatite fractions, it was treated with different hydrolytic enzymes. The product was solubilized almost completely by xylanase M1. Cellulase, laminarinase, lichenase and proteinase K appeared to solubilize the product to some degree (Figure 17). This indicates that the product after hydroxyapatite chromatography is different from the product made before hydroxyapatite chromatography (Figs. 4 and 5). 58 1 200 800 ~ 400- Product remaining (fmol) Figure 17. Enzyme sensitivity of the products synthesized after hydroxyapatite chromatography. The solubilized products obtained after cellulase, laminarinase, or lichenase treatments were analyzed by HPAEC-PAD with the cellooligosaccharide 2 program. The products eluted mainly near the cellopentaose and cellohexaose standards. A minor change in elution time of the radiolabeled products was observed after cellulase treatment of the products (Figure 18). This change was not specific for cellulase, however, because the same observation was made after treating the product with endo-(1,3)-B-glucanase or lichenase (data not shown). From these data it was concluded that none of these enzymes actually hydrolyzed the products, and therefore after hydroxyapatite chromatography the product synthesized is not a (1,4)-B-glucan. 59 A “.3 Glc : f- “.3 Q < C6 o. . _l 1000 4 B E .. 8‘ 500 - AA 0 . . f‘. r“ Tr. 0 15 30 Time (min) Figure 18. The product synthesized after hydroxyapatite chromatography was treated with cellulase and the hydrolysis products were analyzed by HPAEC-PAD. A, Oligosaccharide standards. B, Radioactivity. After xylanase M1 treatment, glucose was the only product observed (data not shown). Because xylanase M6 did not hydrolyze the product, its not xylan. Alkaline phosphatase also hydrolyzed the product and glucose was released (Figure 19). These data suggested that after hydroxyapatite chromatography, the product might be a glucose- phosphate. Its hydrolysis by xylanase M1 could be due to contamination of this enzyme preparation by phosphatase. 6O o A a Glc E C2 C6 Q ,3 C3 C4C5 a" .l n. 1000 " B E. - Q 500 ‘ 0 I I ' . I 10 20 30 Time (min) Figure 19. The product synthesized after hydroxyapatite chromatography was treated with alkaline phosphatase and the hydrolysis products were analyzed by HPAEC-PAD. A, Oligosaccharide standards. B, Radioactivity. The product synthesized after hydroxyapatite chromatography was further characterized. Figure 20 shows that the products synthesized after hydroxyapatite chromatography eluted with glucose-6-phosphate (G6P) and glucose-1,6-diphosphate (G1,6dP). Therefore we conclude that after hydroxyapatite chromatography UDP-glucose is converted to G6P and G1,6dP. G6P and G1,6dP might form a complex with Mg2+ in the glucan synthase assay, which would make them insoluble in ethanol. Support for this comes from the fact that EDT A at a concentration greater than the Mg2+ concentration solubilizes the ethanol-insoluble products synthesized after hydroxyapatite chromatography (data not shown). 61 a A G6P G 0 g e E Gl,6dP On 4000 a B a .. O. Q 2000' 0 W I I ' l 10 20 ' 30 ° 40 Time (min) Figure 20. The water soluble products synthesized after hydroxyapatite chromatography were analyzed by HPAEC-PAD. A, glucose-6-phosphate (G6P) and glucose-1,6- disphosphate (G1,6dP) standards. B, Radioactivity. DISCUSSION In this study the biochemical properties of a Golgi-localized (1,4)-B-glucan synthase in the maize mesocotyl were studied, the product it synthesizes was characterized, and an attempt was made to purify the responsible protein. It was determined that the pH optimum of the (1,4)-B-glucan synthase is 7.5-8 and that it requires a high (>15 mM) Mg2+ concentration for maximal activity. Glucan synthases have been studied in the past and there are several possible products in which glucose could be incorporated, including cellulose (Kudlicka et al., 1997), mixed-linkage-glucan (Gibeaut et al., 1993), xyloglucan (Bauer et al., 1973) or callose (Kudlicka et al., 1997). 62 I determined that microsomes isolated from both dark-grown and light-treated seedlings synthesize a (l,4)-B-glucan at a low UDP-Glc concentration and a (1,3)-B-glucan at a UDP-Glc concentration of 1 mM. The (1,4)-B-glucan synthase activity is affected by exposure of the seedlings to light 16 hours before extraction, in contrast to the (1,3)-B- glucan synthase activity, which is only affected somewhat by light. Further analysis of the B(1,4)-glucan showed that the product is non-crystalline cellulose because the product was soluble in 4 N NaOH. Most likely the [3(1,4)—glucan synthesized in our reactions is not attached to a protein because it could not be solubilized in 0.5% ammonium-oxalate. However, from these experiments we can not definitively conclude that the B(1,4)-glucan synthesized in our reactions is not attached to a protein. The reason being that the method we used for dissolving B(1,4)-glucan in ammonium-oxalate was different than the method Peng et al., (2001) used. Peng et al., (2001) boiled cell wall material for 1 hour in 0.5% ammonium oxalate while in our case we overnight incubated the B(1,4)-glucan in 0.5% ammonium oxalate at room temperature. Its possible that in our case the B(1,4)-glucan is attached to a protein, but that our method did not dissolve the B(1,4)-glucan because of the lower temperature at which we performed the incubation. Another factor which could influence the results we obtained in comparison to the results other researchers obtained is that in our case the B(1,4)-glucan was synthesized in vitro and was subsequently ethanol precipitated on a glass fiber filter. Other researchers used cell wall material synthesized in vivo for their experiments. Attempts were made to determine if the (l,4)-B-glucan synthase is involved in xyloglucan or cellulose biosynthesis, but the results were inconclusive. A possible 63 explanation for these inconclusive results is that the assay conditions used were optimized for in vitro xyloglucan biosynthesis in dicots, because no assay for in vitro xyloglucan biosynthesis in monocots has been established. In order to determine if the (l,4)-B-glucan synthase is a cellulose synthase on its way to the plasma-membrane, enzyme assays were performed in the presence of the cellulose synthase inhibitors DCB and isoxaben. Neither of these compounds inhibited (l,4)-B-glucan synthase activity in vitro, which might indicate that this enzyme is not a cellulose synthase. It is not known, however, what the exact mechanism of actions of these herbicides is, so no definitive conclusions can be made. Most likely these herbicides inhibit cellulose biosynthesis indirectly. This is supported by the observation that resistance to isoxaben is mediated by mutations which are not in the active site of the CESA3 and CESA6 proteins (Scheible et al., 2001; Desperez et al., 2002) and by the observation that loss of function mutations in genes related to cellulose biosynthesis such CESA2, COBRA, KORRIGAN, and CESA6 resulted in increased sensitivity to DCB and isoxaben (Somerville, 2006). The (1,4)-B-glucan synthase was successfully solubilized with digitonin and the product synthesized was still (l,4)-B-glucan. The (l,4)-B-g1ucan synthase activity recovered after solubilization by digitonin was low, however. A possible explanation is that digitonin might inhibit (1,4)-B-glucan synthase activity by changing the conformation of the enzyme or by removing other factors which might stimulate activity. After hydroxyapatite chromatography G6P and G1,6dP were synthesized, but no (l,4)-B-glucan. GlP can be synthesized from UDP-glucose by UDP-glucose pyrophosphorylase (UGPase). G6P can be synthesized from GlP by phosphoglucomutase (Sowokinos et al., 1993). UGPase is a PPi and Mg2+-dependent enzyme activity which catalyzes the following reversible reaction: UDP-glucose + PPi c) glucose-l-P + UTP (Sowokinos et al., 1993). One plausible explanation is that UDP-glucose is first converted into GlP by UGPase. GlP is subsequently converted into G6P by phosphoglucomutase. The PPi could have leaked from the hydroxyapatite column, which consists of calcium phosphate. In vitro G1,6dP is an activator of phosphoglucomutase, but G1,6dP is not found in plants, however (Galloway et al., 1985). In animals and bacteria G1,6dP can be synthesized by phosphoglucomutase and is an intermediate in the conversion of GlP into G6P (Naught et al., 2005). In plants G1,6dP might also be synthesized by phosphoglucomutase and serve as an intermediate in the conversion of GlP into G6P. The UDP-glucose pyrophosphorylase and phosphoglucomutase proteins were present in the hydroxyapatite fractions with activity. This supports the hypothesis that UDP-glucose is converted in GlP and subsequently in G6P by these enzymes. Under standard glucan synthase assay conditions no UDP-glucose pyrophosphorylase and phosphoglucomutase activity was detected in the solubilized (1,4)-[3-glucan synthase preparation. However, after hydroxyapatite chromatography this is the only activity present under standard glucan synthase assay conditions. One possibility is that there is no PPi present in the solubilized glucan synthase fraction, which is necessary for UDP- glucose pyrophosphorylase activity. After chromatography UDP-glucose pyrophosphorylase might be activated by leaking of PPi from the hydroxyapatite column. Under standard glucan synthase assay conditions the pmol of [3H]-glucose precipitated by ethanol after hydroxyapatite chromatography is similar to the pmol [3H]-glucose precipitated by ethanol in the solubilized (l,4)-B-glucan synthase preparation. In case 65 G6P and G1,6dP are synthesized by a different enzyme than the (1,4)-B-glucan synthase I would expect the pmol of [3H]-glucose precipitated to be different. An alternative possibility is that the glucan synthase becomes damaged or is separated from an acceptor or necessary auxiliary protein following chromatography and that it converts UDP- glucose into GIP instead of synthesizing (l,4)-B-glucan. 66 REFERENCES Ball S, Morel] M (2003) From bacterial glycogen to starch: understanding the biogenesis of the plant starch granule. Ann Rev of Plant Biol 54: 207-233 Barber G, Elbein A, Hassid W (1964) The synthesis of cellulose by enzyme systems from higher plants. J of Biol Chem 239: 4056-4061 Bauer W, Talmadge K, Keegstra K, Albersheim P (1973) The structure of plant cell walls. Plant Physiol 51: 171-187 Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254 Desprez T, Vemhettes S, Fagard M, Refregier G, Desnos T, Estelle A, Py N, Pelletier S, Hofte H (2002) Resistance against herbicide isoxaben and cellulose deficiency caused by distinct mutations in same cellulose synthase isoform CESA6. Plant Physiol 128: 482-490 Dhugga K, Ray P (1994) Purification of 1,3-B-D-glucan synthase activity from pea tissue. Two polypeptides of 55 kDa and 70 kDa copurify with enzyme activity. Eur J Biochem 220: 943-953 Edwards M, Dickson C, Chengappa S, Sidebottom C, Gidley M, Reid J (1999) Molecular characterisation of a membrane-bound galactosyltransferase of plant cell wall matrix polysaccharide biosynthesis. Plant J 19: 691-697 Galloway C, Dugger W, Black C (1985) In vitro activation of phosphoglucomutase by fructose 2,6-bisphosphate. Plant Physiol 79: 920-922 Gibeaut D, Carpita N (1993) Synthesis of (1 —> 3), (1 —> 4)-B-D-glucan in the Golgi apparatus of maize coleoptiles. Proc Natl Acad Sci USA 90: 3850-3854 Gordon R, Maclachlan G (1989) Incorporation of UDP-[ l"’C] glucose into xyloglucan by pea membranes. Plant Physiol 91: 373-378 Hayashi T, Matsuda K (1981) Biosynthesis of xyloglucan in suspension-cultured soybean cells. Occurrence and some properties of xyloglucan 4-B-D- glucosyltransferase and 6-cL-D-xylosyltransferase. J Biol Chem 256: 11117-11122 Kudlicka K, Brown Jr R (1997) Cellulose and callose biosynthesis in higher plants. (I. Solubilization and separation of (1 —-> 3)- and (1 —> 4)-B-glucan synthase activities from Mung bean). Plant Physiol 115: 643-656 67 Liebert T, Heinze T (2001) Exploitation of reactivity and selectivity in cellulose functionalization using unconventional media for the design of products showing new superstructures. Biomacromolecules 2: 1124-1132 Li J, Burton R, Harvey A, Hrmova M, Wardak A, Stone B, Fincher G (2003) Biochemical evidence linking a putative callose synthase gene with (1 ——> 3)-B-D- glucan biosynthesis in barley. Plant Mol Biol 53: 213-225 Lu F, Ralph J (2003) Non-degradative dissolution and acetylation of ball-milled plant cell walls: high-resolution solution-state NMR. Plant J 35: 535-544 Meikle P, Ng K, Johnson E, Hoogenraad N, Stone B (1991) The B—glucan synthase from Lolium mulriflorum. Detergent solubilization, purification using monoclonal antibodies, and photoaffinity labeling with a novel photoreactive pyrimidine analogue of uridine 5'-diphosphoglucose. J Biol Chem 266: 22569-22581 Naught L, Tipton P (2005) Formation and reorientation of glucose 1,6-bisphosphate in the PMM/PGM reaction: transient-state kinetic studies. Biochemistry 44: 6831- 6836 Peng L, Xiang F, Roberts E, Kawagoe Y, Greve L, Kreuz K, Delmer D (2001) The experimental herbicide CGA 325'615 inhibits synthesis of crystalline cellulose and causes accumulation of non-crystalline B-1,4-glucan associated with CesA protein. Plant Physiol 126: 981-992 Perrin R, DeRocher A, Bar-Peled M, Zeng W, Norambuena L, Orellana A, Raikhel N, Keegstra K (1999) Xyloglucan fucosyltransferase, an enzyme involved in plant cell wall biosynthesis. Science 284: 1976-1979 Popper Z, Fry S (2003) Primary cell wall composition of bryophytes and charophytes. Ann Bot (Lond) 91: 1-12 Ray P, Shininger T, Ray M (1969) Isolation of B-glucan synthetase particles from plant cells and identification with Golgi membranes. Proc Natl Acad Sci USA 64: 605- 612 Scheible W, Eshed R, Richmond T, Dehner D, Somerville C (2001) Modifications of cellulose synthase confer resistance to isoxaben and thiazolidinone herbicides in Arabidopsis Ixrl mutants. Proc Natl Acad Sci U S A 98: 10079-10084 Shedletzky E, Shmuel M, Trainin T, Kalman S, Delmer D (1992) Cell wall structure in cells adapted to grth on the cellulose-synthesis inhibitor 2,6- dichlorobenzonitrile. A comparison between two dicotyledonous plants and a Grarninaceous monocot. Plant Physiol 100: 120-130 68 Somerville C (2006) Cellulose synthesis in higher plants. Annu Rev Cell Dev Biol 22: 53-78 Sowokinos J, Spychalla J, Desborough S (1993) Pyrophosphorylases in Solanum tuberosum. (IV. Purification, tissue localization, and physicochemical properties of UDP-Glucose pyrophosphorylase). Plant Physiol 101: 1073-1080 Sterling J, Atmodjo M, Inwood S, Kolli KV, Quigley H, Hahn M, Mohnen D (2006) Functional identification of an Arabidopsis pectin biosynthetic homogalacturonan galacturonosyltransferase. Proc Natl Acad Sci USA 103: 5236-5241 Tsai C, Hassid W (1971) Solubilization and separation of uridine diphospho-D-glucose: B-l -—> 4) glucan and uridine diphospho-D-glucose:B-(l —> 3) glucan glucosyltransferases from coleoptiles of Avena sativa. Plant Physiol 47: 740-744 Updegraff D (1969) Sernirnicro determination of cellulose in biological materials. Anal Biochem 32: 420-424 Villemez C, Franz G, Hassid W (1967) Biosynthesis of alkali insoluble polysaccharide from UDP-D-glucose with particulate enzyme preparations from Phaseolus aureus. Plant Physiol 42: 1219-1223 Walton J, Ray P (1982a) Inhibition by light of grth and Golgi-localized glucan synthase in the maize mesocotyl. Planta 156: 302-308 Walton J, Ray P (1982b) Auxin controls Golgi-localized glucan synthase activity in the maize mesocotyl. Planta 156: 309-313 69 CHAPTER 3: ANNOTATION OF THE ZEA MAYS CSL GENE FAMILY ABSTRACT The maize (Zm) CSL gene family was annotated based on orthology to the rice (Os) CSL gene family using NCBI (public) and MaizeSeq (proprietary) EST collections and available genome sequence (public). Maize has all of the CSL subfamilies which are present in rice. There are, however, several differences between the two. The CSLC subfamily, for instance is expanded from six members in rice and five in Arabidopsis to potentially twelve members in maize. The CSLH subfamily is reduced from three members in rice to one in maize. Maize has one atypical CSL, called ZmCSLX, which is related to Populus m'chocarpa (poplar) CSLGl and G3. Several ESTs related to the ZmCSLX ORF were found in other monocot species such as Sorghum bicolor (sorghum), Saccharum sp (sugar cane) and Hordeum vulgare subsp vulgare (barley). INTRODUCTION The first cellulose synthase genes were discovered in Acetobacter xylinum (Wong et al., 1990). Based on sequence similarity to these bacteria] cellulose genes the plant cellulose synthase (CESA) genes were discovered (Pear et al., 1996; Arioli et al., 1998). Based on sequence similarity to the CESA family, a related gene family called the CSL family was described in Arabidopsis (Richmond and Somerville, 2000). Arabidopsis has 10 CESA genes and 29 CSL genes, which together compose the CESA superfamily. The Arabidopsis (At) CSL gene family is divided into 6 subfamilies called CSLA, B, C, D, E and G. The rice CSL gene family was annotated by Hazen et al (2002) and is composed 70 of the CSLA, C, D, E, F and H subfamilies and it has 34 members. The rice CSL gene family is missing the CSLB and G subfamilies but has two cereal specific subfamilies called CSLF and H. All the members of the CESA superfamily have two conserved domains, several widely spaced aspartate residues (D) and the QXXRW motif. These two domains are hypothesized to be the sugar—nucleotide binding site and the active site, respectively (Richmond and Somerville, 2000). The CSL proteins are involved in the synthesis of the sugar backbones of at least some of the hemicelluloses. The CSLA proteins, for instance, are (l,4)-,6-(gluco)mannan synthases (Dhugga et al., 2004; Liepman et al., 2005; Suzuki et al., 2006; Liepman et al., 2007), AtCSLC4 and TmCSLC synthesize the (1,4)-,6-1inked glucose backbone of xyloglucan (Cocuron et al., 2007), the CSLD proteins might be involved in cellulose biosynthesis in tip growing cells (Doblin et al., 2001; Wang et al., 2001; Roberts and Bushoven, 2007) and OsCSLF2 is involved in mixed-linkage-glucan (MLG) biosynthesis (Burton et al., 2006). It is not known however, if all the members of a CSL subfamily catalyze the synthesis of the same polymer. The function of the other CSL proteins is not known. MATERIALS AND METHODS Annotation of the maize CSL protein family The ZmCSL gene family was annotated based on homology to the OsCSL protein family (Hazen et al., 2002; http://waltonlab.prl.msu.edu//CSL_updates.html). The Iowa State Maize Assembled Genome Island 4 (MAGI4) database (http://magi.plantgenomics.iastate.edu/; Fu et al., 2005) was searched (tblastn) using OsCSLAl, C1, D1, E1, F1 and H1 proteins as queries. Proteins encoded by the MAGIs 71 were determined using the FGENESH monocot ab initio gene prediction program (Yao et al., 2005 ; www.softberry.com). The predicted proteins were aligned with the OsCSL proteins using ClustalW, MegAlign (DNASTAR). MAGIs belonging to the same CSL subfamily were aligned (criteria: match size 12, mismatch percentage 80, minimum sequence length 100) using SeqMan (DNASTAR). Sequences which aligned were assembled into contigs. These genomic contigs were used to search for additional genomic sequence in the MAGI4 database, at the public database at TIGR (http://www.tigr.org/), and at the Plant Genome Database (http://www.plantgdb.org/). The genomic contigs were also used to search the NCBI EST database (http://www.ncbi.nlm.nih.gov/. blastn, database est_others, Zea mays) or the ESTs in the MaizeSeq database (http://www.maizeseqorgL blastn). The ESTs were assembled in contigs and aligned with the genomic contigs using Spidey (http://www.ncbi.nlm.nih.gov/spidev0 in order to determine the intron-exon structure of the genes. The annotation was refined by comparing the ZmCSL proteins with their closest rice orthologs using ClustalW. Images in this dissertation are presented in color. Phylogenetic analysis Proteins were aligned with the ClustalX using default settings (Multiple Alignment Mode) (Thompson et al., 1997). Phylograms were constructed with the heuristic search method in PAUP and a phylogram corresponding to 1000 bootstrap replicates was constructed (Version 4.0b10, Sinauer Associates, Sunderland, MA). The phylogenetic trees were modified with the Treeview program (Page, 1996). 72 RESULTS The genomic, cDNA and amino acid sequences for each ZmCSL gene can be found at http://waltonlab.prl.msu.edu//. The list of genomic sequences used for the annotation can be found in Appendix I, Table 8. A detailed description of the annotation of each CSL gene can be found in Appendix I, Table 9. The number of genes in the maize and rice CSL gene families is comparable for the CSLA, D, E, and F subfamilies. The ZmCSLC gene family is expanded from six in rice to twelve in maize including one gene, ZmCSLC9, which is most likely a pseudogene. ZmCSIAII, E2 and E4 are probably also pseudogenes, because they have stop codons in their coding regions and there are no ESTs in the NCBI and MaizeSeq EST databases. Rice has three CSLH genes, maize only one. One atypical CSL, ZmCSLX, was found which does not have a homolog in rice (Table 4). 73 Table 4: The number of CSL genes for each subfamily in rice and maize Rice Maize (Full length: full length genomic DNA and/or cDNA is available for this gene. Partial: partial genomic DNA and/or cDNA). CSL subfamily Full length Partial A 9 6 5 (ZmCSL/11, 3, 6, 7, 8, 9, 10) (ZmCSL/12, 4, 5, II (pseudogene)) C 6 7 5 (ZmCSLC 1, 3, 4, 5, 6, 7, 11) (ZmCSLCZ, 8, 9 (pseudogene), 10, 12) D 5 4 l (ZnCSLDI, 2, 4, 5) (ZJnCSLD3) E 3 2 2 (ZmCSLE 1, 3) (ZmCSLEZ (pseudogene), E4 (pseudogene)) F 8 5 2 (ZmCSLFI, 2, 4, 6, 7) (ZmCSLF3, 5) H 3 l 0 (ZmCSLHI) X 0 1 0 (ZmCSlX) Phylogenetic analysis of the CSLA and CSLC protein families of different plant species Figure 21 shows a phylogram constructed with the full length CSLA proteins of maize, rice (www.pr1.m_su.edu/walton/CSL updateshtm), poplar (Populus trichocarpa) 74 (Suzuki et al., 2006), moss (P. patens) (Roberts and Bushoven, 2007), Arabidopsis, guar (Cyamopsis tetragonoloba) (Dhugga et al., 2004) and loblolly pine (Pinus taeda) (Liepman et al., 2007) and the full length CSLC proteins of Arabidopsis, nastursium (Tropaeolum majus, Tm), maize, moss, rice and poplar. The ZmCESAl, 2, 3 and 4 proteins were used as an outgroup. In Appendix II, Table 11 the Genbank locus numbers of the Arabidopsis CSL proteins are mentioned, in Table 12 the JGI protein IDs of the poplar CSL proteins, in Table 13 the JGI protein IDs for the Physcomitrella CSL proteins and in Table 14 the GenBank locus number for the ZmCESA proteins. For almost every rice CSLA or C protein, there are one or two corresponding maize orthologs. Conversely, there are one or two rice orthologs for each maize protein. Therefore it can be concluded that there is a strong conservation between these two protein families in rice and maize. Nine CSLA and CSLC genes in maize do not have full length genomic or cDNA sequence available (Table 4). Only after these genes are completely annotated can a definitive conclusion about the similarity between these two gene families be made. The same can be observed for the Arabidopsis and poplar CSLA and CSLC protein families. For almost every Arabidopsis CSLA or C protein, there are one or two corresponding poplar orthologs. Conversely, there are one or two poplar orthologs for each Arabidopsis protein. Therefore it can be concluded that there is a strong conservation between these two protein families in Arabidopsis and poplar. The strong conservation between maize and rice on the one hand and the Arabidopsis and poplar CSLA and CSLC protein families on the other hand reflects the evolutionary divergence between cereals and dicots. This evolutionary divergence of the CSLA and 75 CSLC protein families might reflect the structural differences between cereal and dicot cell walls. Some of the CSLA proteins from rice and maize (OsCSLA2, 3, 4, 5, 7, 11 and ZmCSLAl, 6, 7, 10), on the one hand, and Arabidopsis (AtCSLAl, 3, 7, 10, 11, 14, 15), on the other, form distinct clades. These cereal and Arabidopsis specific CSLA clades are weakly supported, however (bootstrap value of <50). AtCSLA2; OsCSLAl, 9; PtCSLA4, 5 and ZmCSLA3, 9 cluster in a mixed clade, which is weakly supported (bootstrap value of <50). The remainder of the CSLA proteins from rice, maize, poplar, pine tree, and Arabidopsis (AtCSLA9; OsCSLA6; PpCSLAl, 2; PtCSLAl, 2, 3; PtaCSLAl, 2; CtMan and ZmCSLA8) do not cluster in any specific clade. In conclusion, dicots and cereals might have some CSLA proteins in common but might also have unique CSLA proteins. This might reflect biochemical specialization due to the different structures of their cell walls. The AtCSLC4 and TmCSLC proteins are involved in synthesizing the B(1,4)- glucan backbone of xyloglucan (Cocuron et al., 2007). Figure 21 shows that some of the CSLC proteins from rice and maize (OsCSLCl, 7; ZmCSLC3, 4, 6) on the one hand, and moss (PpCSLCl, 2 3) on the other, form distinct clades. The rice and maize-specific clade and the moss-specific clades are weakly supported (bootstrap value of <50). The clustering in these distinct clades might reflect biochemical specialization due to the different structures of moss, dicot and cereal cell walls. Xyloglucan in cereals is less xylosylated than xyloglucan in dicots (Kato et al., 1980b, 1981b, 1982; Zablackis et al., 1995; Gibeaut et al., 2005). The cereal specific group of CSLC proteins might be involved in the synthesis of the less xylosylated type of xyloglucan found in cereals. Two 76 other major clades consists of both dicot and cereal CSLC proteins. Clade 1 consists of the AtCSLCS, 8; OsCSLC2, 3; PtCSLCl, 2 and ZmCSLC], 11 proteins. Clade 2 consists of the AtCSLC12; OsCSLC9, 10; PtCSLC3, 4; ZmCSLCS, 7 proteins. These clades are weakly supported, however (bootstrap value <50). The rice and maize CSLC proteins that belong to these mixed clades might be involved in the synthesis of the B(1,4)-glucan backbone of galactose-containing xyloglucan such as found in rice endosperm and barley cell walls (Shibuya and Misaki, 1978; Kato et al., 1981). In maize the presence of galactosylated xyloglucan has not been shown, however. There is one minor clade consisting of AtCSLC6 and PtCSLCS, which is weakly supported (bootstrap value 56). 77 Figure 21. Parsimony phylogram of Arabidopsis (At), guar (Ct), maize (Zm), moss (Pp), pine tree (Pta), poplar (Pt) and rice (Os) CSLA proteins and the CSLC proteins of Arabidopsis, maize, moss, nasturtium (Tm), poplar and rice. The colors in this figure correspond to different species. The ZmCESAl, 2, 3 and 4 proteins were used as an outgroup. Four ZmCSLA proteins and five ZmCSLC proteins were excluded because full length sequences are not available. 78 Figure 21 .-\t(‘Sl A] 5 CSLA 100 anCS LAB OsCSLA6 AtCSLClZ ZmCSLC3 ZmCSLC-4 CSLC OsCSLCl Cereals ZmCSlL‘6 OsCSL£7 .XICSIL‘G PtCSLCS 100 PPCSIL‘Z 97 ppcsug moss CSLC PpCSLCl AtCSLCS rmcsuc .csrc AtCSLC4 Dicot/med ZmCSLCl OsCSLC2 ZmCSLCll Oscsrcs ZmC ESA4 79 Figure 22 shows which CSLA proteins encode for mannan synthases. None of the CSLA proteins belonging to the cereal-specific clade of CSLA proteins has been shown to encode mannan synthase, while many members of the other clades of CSLA proteins are proven mannan synthases (Dhugga et al., 2004; Liepman et al., 2005 and 2007; Suzuki et al., 2007). None of the CSLA proteins in the cereal-specific clade have been tested for mannan synthase activity. This raises the question if the CSLA proteins cereal- specific clade are mannan synthases or if they have other biochemical activities. 80 100 PtCSLA-l 100 PICSLAS AtCSLAZ 100 anCSLA3 OsCSLAl 54 anCSLA9 OsCSLA9 AtCSLAlO AtCSLAlS 99 AtCSLAll AtCSLA] AtCSLA7 65 AtCSLA3 AtCSLAl4 lOO PtCSLAl PtCSLAZ 8' AtCSLA9 PtCSLA3 - PtaCSLAl PtaCSLAZ 100 PpCSLAl PpCSLA2 CtMan 95 ZmCSLAl 100 OsCSLA7 73 OsCSLAS 70 OsCSLA4 75 100 OsCSLAZ ZmCSLA7 51 OsCSLA3 76 ZmCSLA6 100 ZmCSLAlO OsCSLAll 100 ZmCSLAS OsCSLAfi plant species Figure 22. Parsimony phylogram of Arabidopsis (At), guar (Ct), rice (Os), maize (Zm), moss (Pp) and poplar (Pt) CSLA proteins. The tree is identical to figure 21, but colored by function rather than taxonomy. Those that have been determined to encode for (1 ,4)- B-mannan synthases are shown in red and blue means that they have not been tested. Phylogenetic analysis of the CSLB, D, E, F, H and G protein families of different Figure 23 shows a parsimony phylogram of the Arabidopsis, maize, moss, 81 Nicotiana tabacum (Nt), poplar and rice CSLB, D, E, F, G, H and X proteins. For almost every rice CSLD, E, F and H protein there are one or two maize orthologs. Conversely, there are one or two rice orthologs for each maize gene. Therefore it can be concluded that there is a strong conservation between these four protein families in rice and maize. For almost every Arabidopsis CSLB, D, E and G protein, there are one or two corresponding poplar orthologs. Conversely, there are one or two poplar orthologs for each Arabidopsis protein. Therefore it can be concluded that there is a strong conservation between these two protein families in Arabidopsis and poplar. The CSLD protein family clusters in four major clades. Three of these clades consist of cereal and dicot CSLD proteins and one clade is specific for moss. Clade 1 consists of the AtCSLD2, 3; OsCSLDl, 2; ZmCSLD2, 5 and PtCSLDS, 6 proteins and is weakly supported (bootstrap value <50). Clade 2 consists of the AtCSLDl, 4; OsCSLD3, 5; ZmCSLD4 and PtCSLD7, 8, 9, 10 proteins and is weakly supported (bootstrap value 56). Clade 3 consists of the AtCSLDS; OsCSLD4; ZmCSLDl and PtCSLDl, 2 proteins and is strongly supported (bootstrap value 89). The moss specific clade consists of PpCSLDl, 2, 3, 4, 5, 6, 7 and 8 and is strongly supported (bootstrap value of 89). The biological significance of these clades is not known, however. Besides these four major clades PtCSLD3 and 4 cluster together in a specific clade which is weakly supported (bootstrap value of 60) and AtCSLD6 clusters by itself (bootstrap value of 100). The analysis of the maize CSL protein family supports the conclusion that the CSLF proteins are found only in cereals (Fig. 23). The maize and rice CSLF proteins cluster in a separate clade which is weakly supported (bootstrap value of 53). The CSLB family is not present in rice or maize, but only in dicots and is strongly supported (bootstrap value of 98). The CSLB family is divided into two different clades. 82 One consists of Arabidopsis CSLB proteins and the other consists of poplar CSLB proteins. These clades are weakly supported (bootstrap value of 58). Rice and maize have a CSLB-like family, which was named CSLH by Hazen et al. (2002) and is strongly supported (bootstrap value of 98). The CSLE protein family consists of a cereal specific clade (OsCSLEl, 2, 6; ZmCSLEl, 3), a poplar specific (PtCSLE2, 3) clade and a mixed Arabidopsis and poplar clade (AtCSLEl, PtCSLEl). The cereal specific clade and the poplar specific clade are weakly supported (bootstrap value of 58). The Arabidopsis and poplar specific clade is strongly supported (bootstrap value of 100). The CSLG family is also not present in rice (Hazen et al., 2002). Maize however, has a protein, which we call ZmCSLX, which clusters with all of the other members of the CSLG family. This clade is strongly supported (bootstrap value of 86). 83 Figure 23. Parsimony phylogram of the Arabidopsis (At), maize (Zm), moss (Pp), tobacco (Nicotiana tabacum (Nt)), poplar (Pt) and rice (Os) CSLB, D, E, F, G, H and X proteins. The ZmCSLAl, 3 and ZmCSLC], 3 proteins were used as an outgroup. One ZmCSLD, two ZmCSLE and'two CSLF proteins were excluded because full length sequences are not available. The colors in this figure indicate the different species. 84 Figure 23 99 P CSLDt 100 p p LD4 100 in“: P LDS 100 L08 CSLD 10° P L03 Pjicsr. 7 ““058 10° cgcsmz 89 p80 L06 99‘ "as“ 1°° AtCSLD5 CSLD 10° ZmCSLDr Cereal/dicot 56 OsCSL 100100 'CSLD7 PfCSLDB m are“ 57 OgggLosHCSLDQ CSLD. 99 99 FIGS 10 Cereal/dicot ArcsLLor L02 7 OsCSLD2 6° 00 ZmCSL05 951" AtCSLODECSI-m CSLD 36 AtCSLDS CGl‘eal/drcot 10° "= PtCSLDS PtCSLDG 1" PtCS'LD3 AtCSLD6 100 OsCSLF1 5’ -- 100 ‘°° " wears. 52 mo ozflcgiéi" CSLF 1w99 szSLF1 cereal ZmCSL OsCSLFG 99 ZmCSLF7 OsCSLF7 85 Figure 23 continued CSLG Cereal/dicot C SLE Cereal 10° ZmCS LE 1 sCSLEZ ~ CSLE poplar CSLB poplar and Arabidopsis CSLB Arabidopsis AtCSLBS PtCSLB1 PtCSLBZ} CSLB p091ar OSCS LH1 0‘03”“ CSLH cereal 86 ZmCSLX The properties of the ZmCSLX protein suggest that it is a member of the CSL family. It has the D, D, D and the QXXRW motifs present in all of the members of the CESA superfamily. It is predicted to have four transmembrane domains by the TMHMM Server v. 2.0 (magi/WWW.cbs.dtuLk/servic4es/TMI-IIVIM-2.Ol). When a blast search (GenBank, tblastn, nr) was performed using CSLX as a query, a CSLG protein from Nicotiana tabacum (tobacco) had the highest score (261”) (as of November, 2006). Figure 23 shows that ZmCSLX clusters with the PtCSLGl and 3 proteins. The reason why the highest blast scores were not for poplar CSLGl or G3, but for a tobacco CSLG, is that the poplar proteins are not available in Genbank but only at the Joint Genome Institute (JGI) website. To determine if other cereals have orthologs of ZmCSLX the SAMI (Sorghum Assembled genoMic Islands) database was searched (http://magi.plantgenomics.iastate.edu/, tblastn) with the ZmCSLX ORF as query (Table 5). Several related genomic sequences were found. Two ESTs related to this gene were found in the Sorghum EST database (GenBank, est_others, Sorghum bicolor, tblastn). For rice and wheat, no ESTs were found. Sugarcane has one EST and barley three (Table 5). The high percentage of identity (Table 5) suggests that these are true orthologs. The proteins encoded by these ESTs cluster most closely with ZmCSLX when aligned with the maize CSLA, C, D, E, F and H proteins, and with the Arabidopsis, poplar and tobacco CSLG proteins (data not shown). It cannot yet be concluded, however, that these monocots have CSLX proteins, because no full length sequences are available yet. In conclusion, the CSLG protein family, which was previously thought to be specific for dicots, might also be present in some monocots. 87 Table 5: Genomic DNA and EST sequences related to the ZmCSLX ORF in different monocot species (family Poaceae) *% maximal identity at nucleotide level **% coverage at nucleotide level Species Genomic DNA ESTs Sorghum CW233534, CW233535, CF430961 (*92%, **19%) CW457534, CW457535, CF431079 (*97%, ** 13%) CW494817, CW494818, fsbeOlf286e18.R, SAMIv2_25399 Barley CA000498 (*85%, **9%) CA002678 (*97%, **5%) CA011599 (*95%, **13%) Sugar cane CAO96252 (*95%, **16%) ZmCSL ESTs The ESTs for each ZmCSL are described in Table 6. The number of ESTs in GenBank (http://www.ncbi.nlm.nih.gov/. blastn, est_others, Zea mays) and MaizeSeq (http://www.maizeseqorgl. blastn) databases was determined by searching with ZmCSL cDNAs as queries. An interesting observation is that the numbers of ESTs for each CSL gene varies significantly for each CSL subfamily and within each subfamily (Figure 24). A total of 828 ESTs belonging to maize CSL genes were found in GenBank and MaizeSeq as of November, 2006. Six hundred and six of these ESTs (66%) belong to five 88 members of the ZmCSL gene family (ZmCSLA3, C5, C6, F1 and F2). ZmCSLFI and F2 account for 36% of all the maize ESTs in GenBank and MaizeSeq. The accession numbers of the ESTs for each ZmCSL are in Appendix I, Table 10. 89 Figure 24. Distribution of ESTs belonging to maize CSL genes in GenBank (blue bar) and MaizeSeq (purple bar) as of November 2006. F1/F2 indicates that these ESTs aligned with identical regions in the CSLF] and F2 genes, which makes it impossible to determine from which gene they originate. 90 Figure 24 A11 A10 A8 A7 teeter. ZmCSL 50 100 Number of ESTs 91 150 200 DISCUSSION The annotation of the ZmCSL gene family showed that it is very similar to the OsCSL gene family. For almost every maize CSL protein there are one or two rice homologs, and there are one or two maize homologs for every rice CSL protein. When the rice and maize CSL proteins were aligned with CSL proteins from other plant species, clustering in monocot specific clades was observed for the CSLA, C and E proteins. The CSLD proteins did not show this type of clustering, instead the monocot and dicot CSLD proteins clustered in several mixed clades. My phylogenetic analysis confirms that CSLB proteins are only present in dicots and that the CSLH and F proteins are only present in monocots (Hazen et al., 2002). Many of the CSLA proteins are mannan synthases (Dhugga et al., 2004; Liepman et al., 2005, 2007; Suzuki et al., 2006) but none of the CSLA proteins in the cereal- specific clade have been shown to encode for mannan synthases. This raises the question if the CSLA proteins in this cereal-specific clade are mannan synthases. Cereal and dicot cell walls have a different structure. Could it be that the cereal specific CSLA proteins are involved in synthesis of a cereal-specific polymer? This could be tested by expressing them and determining their biochemical activity. Handford et al. (2003) showed that mannan epitopes are abundantly present in Arabidopsis. In maize mannan is only a minor component of the cell wall however (Carpita et al., 2001). Investigation of mutants indicates that mannan has a role in embryo development in Arabidopsis (Goubet et al., 2003) and is important for Agrobacterium infection of Arabidopsis (Zhu et al., 2003). Several members of the CSLC protein family (AtCSLC4 and TmCSLC) are involved in the synthesis of the B(1,4)-linked glucose backbone of xyloglucan (Cocuron 92 et al., 2007). The ZmCSLC protein family has twelve members in maize in comparison to six in rice and five in Arabidopsis and poplar. Xyloglucan is a minor hemicellulose in maize and other cereals (Carpita et al., 2001; Gibeaut et al., 2005) in comparison to dicots (Zablackis et al., 1995), so it is surprising that maize has so many CSLC proteins. The CSLC proteins cluster in several different clades. Five of the rice and maize CSLC proteins cluster in a cereal-specific clade. The members of this clade might be involved in the synthesis of the B(1,4)-glucan backbone of the cereal type of xyloglucan. The majority of xyloglucan in cereals is less xylosylated than dicot xyloglucan and contains no galactose and fucose residues on the xylosyl side chains of the B(1,4)-glucan backbone (Bauer et al., 1973; Kato et al., 1980b, 1981b, 1982). The remainder of the maize and rice CSLC proteins clusters with dicot CSLC proteins. These maize and rice CSLC proteins might be involved in the synthesis of the galactosylated xyloglucan found in, for instance, rice endosperm and barley cell walls (Shibuya and Misaki, 1978; Kato et al., 1981). Galactosylated xyloglucan has not yet been reported in maize, however. The three moss CSLC proteins cluster in a separate clade. The clustering of these moss CSLC proteins in a separate clade might reflect the evolutionary distance between mosses and other land plants. Moss cell walls also contain xyloglucan (Popper and Fry, 2002), so these CSLC proteins might be involved in the synthesis of the B(1,4)-linked glucose backbone of xyloglucan in mosses. The CSLD proteins cluster in four major clades. Three of these clades consist of dicot and monocot CSLD proteins. One clade consists only of moss CSLD proteins. It is not known what the biological significance of the separate clades with dicot and monocot CSLD proteins is. It could be that the CSLD proteins belonging to the three monocot and 93 dicot specific clades are expressed in specific tissues or that they have a different biochemical function. A knock-out mutant in Kojak/AtCSLD3 showed reduced elongation of root hairs and the root hair tips leaked cytoplasm. This indicates tensile strength of the root hair cell walls was reduced (Favery et al., 2001; Wang et al., 2001). Based on these observations the authors suggested that the KOJAK/AtCSLD3 protein might be involved in the synthesis of B(1,4)-glucan in tip-growing cells such as roothairs (Favery et al., 2001; Wang et al., 2001). Pollen tubes in Nicotiana alata consist of callose and cellulose. Two of the major glycosyltransferases expressed in these pollen tubes were a callose synthase and NaCSLDl. This suggests that NaCSLDl is involved in cellulose biosynthesis in pollen tubes in N. alata (Doblin et al., 2001). CSLD genes are also highly expressed in auxin-treated cultures of P. patents (Roberts et al., 2007). Tip grth is important in P. patents, so this supports the hypothesis that the CSLD proteins are involved in cell wall biosynthesis in tip growing cells. The biochemical functions of the CSLE proteins are not known. The CSLE protein family consists of a cereal specific clade, a poplar specific clade and a mixed Arabidopsis and poplar clade. The biological significance of these different clades is not known. The CSLF proteins, which are specific for cereals (Hazen et al., 2002) are involved in mixed-linkage glucan biosynthesis (Burton et al., 2006). Arabidopsis was transformed with the OsCSLF2 gene and MLG was detected in the cell walls with monoclonal antibodies and enzymatic analysis. One atypical ZmCSL was found which was provisionally called ZmCSLX. Rice apparently does not have a CSLX, but sorghum, sugarcane and barley do. It cannot yet be 94 concluded, however, that ZmCSLX genes are present or absent in other monocot species besides maize because the complete coding region is currently available only for ZmCSLX. The ZmCSLX protein clustered with the CSLG proteins of Arabidopsis, poplar and tobacco, so ZmCSLX probably belongs to the CSLG protein family. Previously it was thought that the CSLG family was specific for dicots because no CSLG ESTs and genomic sequences were found for rice or maize (Richmond and Somerville, 2001; Hazen et al., 2002). My results indicate that CSLG proteins might also be present in monocots. The biochemical function of the CSLG family is not known. Expression analysis with GUS fusions of AtCSLGl and G2 showed their expression is consistent with a role in xylem cell wall formation (Richmond and Somerville, 2001). The ZmCSLH protein family has one member in maize in comparison to three in rice. We do not know the function of the CSLH proteins. The CSLH subfamily is not found in dicots (Hazen et al., 2002), which is confirmed by my phylogenetic analysis. CSLH proteins might be involved in the synthesis of a monocot-specific polysaccharide such as MLG. The number of ESTs in GenBank and MaizeSeq are unequally distributed among the ZmCSL genes of every CSL subfamily. ZmCSLA3, C5, C6, F1 and F2, for instance, are represented by a large number of ESTs, especially in GenBank. One possible reason for this is that the ESTs originate from libraries that are constructed from pooled multiple tissues, and it is not known what percentage of the RNA came from what tissue. ESTs from CSL genes that are highly expressed in certain tissues might dominate the libraries. The abundance of the ESTs seems to be more equal in MaizeSeq than in GenBank. One 95 reason for this could be that the ESTs in the MaizeSeq database mainly consist of full length cDN As, which might be assembled from individual ESTs. For predicting the structure of the genes encoded by the genomic sequences, the FGENESH gene prediction program trained for monocots was used (Yao et al., 2005). FGENESH predicted the gene structure of 17 (41%) of the 41 CSL genes correctly. Yao et al. (2005) found that FGENESH predicted 50% of the gene structures of maize genes correctly. Several reasons why the gene predictions by FGENESH are not always correct are that it is not suitable for the prediction of genes with very small exons (<100 bp), large introns, or non-canonical splice sites. The major source of problems in the FGENESH prediction for the ZmCSL gene family was related to incomplete genomic sequence, large intron size, small exon size, and incorrect prediction of the start or end of an exon. Alternative splice sites were not a major cause of the incorrect predictions of FGENESH in the case of the ZmCSL gene family. There is only one intron in the 41 genes of the ZmCSL family which has a non-canonical splice site. The second intron of ZmCSLF4 has a non-canonical splice site, GC/AG instead of GT/AG, which was predicted incorrectly by FGENESH. One rice gene, 0sCSLA4, also has a non-canonical splice site. 96 REFERENCES Arioli T, Peng L, Betzner A, Burn J, Wittke W, Canrilleri C, Hofte H, Plazenski J, Birch R, Cork A, Glover J, Redmond J, Williamson R (1998) Molecular analysis of cellulose biosynthesis in Arabidopsis. Science 279: 717-720 Bauer W, Talmadge K, Keegstra K, Albersheim P (197 3) The structure of plant cell walls. Plant physiology 51: 171-187 Burton R, Wilson S, Hrmova M, Harvey A, Shirley N, Medhurst A, Stone B, Newbigin E, Bacic A, Fincher G (2006) Cellulose synthase-like Cle genes mediate the synthesis of cell wall (1,3;1,4)-B—D-glucans. Science 311: 1940- 1942 Carpita N, Defemez M, Findlay K, Wells B, Shoue D, Catchpole G, Wilson R, McCann M (2001) Cell wall architecture of the elongating maize coleoptile. Plant Physiol 127: 551-565 Dhugga K, Barreiro R, Whitten B, Stecca K, Hazebroek J, Randhawa G, Dolan M, Kinney A, Tomes D, Nichols S, Anderson P (2004) Guar seed B-mannan synthase is a member of the cellulose synthase super gene family. Science 303: 363-366 Doblin M, De Melis L, Newbigin E, Bacic A, Read S (2001) Pollen tubes of Nicotiana alata express two genes from different B-glucan synthase families. Plant Physiol 125: 2040-2052 Favery B, Ryan E, Foreman J, Linstead P, Boudonck K, Steer M, Shaw P, Dolan L (2001) KOJ AK encodes a cellulose synthase-like protein required for root hair cell morphogenesis in Arabidopsis. Genes Dev 15: 79-89 Fu Y, Emrich S, Guo L, Wen T, Ashlock D, Aluru S, Schnable P (2005) Quality assessment of maize assembled genomic islands (MAGIs) and large-scale experimental verification of predicted genes. Proc Natl Acad Sci USA 102: 12282-12287 Gibeaut D, Pauly M, Bacic A, Fincher G (2005) Changes in cell wall polysaccharides in developing barley (Hordeum vulgare) coleoptiles. Planta 221: 729-738 Gouhet F, Misrahi A, Park S, Zhang Z, Twell D, Dupree P (2003) AtCSLA7, a cellulose synthase-like putative glycosyltransferase, is important for pollen tube growth and embryogenesis in Arabidopsis. Plant Physiol 131: 547-557 97 Hazen S, Scott-Craig J, Walton J (2002) Cellulose synthase-like genes of rice. Plant Physiol 128: 336-340 Holland N, Holland D, Helentjaris T, Dhugga K, XoconostIe-Cazares B, Delmer D (2000) A comparative analysis of the plant cellulose synthase (CesA) gene family. Plant Physiol 123: 1313-1324 Kato Y, Matsuda K (1980b) Structure of oligosaccharides obtained by hydrolysis of mung bean xyloglucan with Trichoderma viride cellulase. Agric Biol Chem 44: 1759-1766 Kato Y, Iki K, Matsuda K (1981b) Cell-wall polysaccharides of immature barley plants 11. Characterization of a xyloglucan. Agric Biol Chem 45: 2745-2753 Kato Y, Ito S, Iki K, Matsuda K (1982) Xyloglucan and B-D-glucan in cell walls of rice seedlings. Plant cell phys 29: 351-364 Liepman A, Nairn C, Willats W, Sarensen I, Roberts A, Keegstra K (2007) Functional genomic analysis supports conservation of function among CslA gene family members and suggests diverse roles of mannans in plants. Plant Physiol 143: 1881-1893 Liepman A, Wilkerson C, Keegstra K (2005) Expression of cellulose synthase-like (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proc Natl Acad Sci USA 102: 2221-2226 Page RDM (1996) TREEVIEW: An application to display phylogenetic trees on personal computers. Comput Appl Biosci 12: 357-358 Pear J, Kawagoe Y, Schreckengost W, Delmer D, Stalker D (1996) Higher plants contain homologs of the bacterial celA genes encoding the catalytic subunit of cellulose synthase. Proc Natl Acad Sci USA 93: 12637-12642 Popper Z, Fry S (2003) Primary cell wall composition of bryophytes and charophytes Ann Bot (Lond) 91: 1-12 Richmond T, Somerville C (2000) The cellulose synthase superfamily. Plant Physiol 124: 495-498 Roberts AW, Bushoven JT (2007) The cellulose synthase (CESA) gene superfamily of the moss Physcomitrella patens. Plant Mol Biol 63: 207—219 Richmond T, Somerville C (2001) Integrative approaches to determining Csl function. Plant Mol Biol 47: 131-143 98 Shibuya N, Misaki A (1978) Structure of hemicellulose isolated from rice endosperm cell wall: Mode of linkages and sequences in xyloglucan, glucan and arabinoxylan. Agric. Biol. Chem. 42: 2267-2274 Suzuki S, Li L, Sun Y, Chiang V (2006) The cellulose synthase gene superfamily and biochemical functions of xylem-specific cellulose synthase-like genes in Populus trichocarpa. Plant Physiol 142: 1233-1245 Thompson J, Gibson T, Plewniak F, Jeanmougin F, Higgins D (1997) The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25: 4876-4882 Wang X, Cnops G, Vanderhaeghen R, De Block S, Van Montagu M, Van Lijsebettens M (2001) AtCSLD3, a cellulose synthase-like gene important for root hair grth in Arabidopsis. Plant Physiol 126: 575-5 86 Wong H, Fear A, Calhoon R, Eichinger G, Mayer R, Amikam D, Benziman M, Gelfand D, Meade J, Emerick A (1990) Genetic organization of the cellulose synthase operon in Acetobacter xylinum. Proc Natl Acad Sci USA 87: 8130- 8 134 Yao H, Guo L, Fu Y, Borsuk L, Wen T, Skibbe D, Cui X, Scheffler B, Cao J, Emrich S, Ashlock D, Schnable P (2005) Evaluation of five ab initio gene prediction programs for the discovery of maize genes. Plant Mol Biol 57: 445-460 Zhu Y, Nam J, Carpita N, Matthysse A, Gelvin S (2003) Agrobacterium-mediated root transformation is inhibited by mutation of an Arabidopsis cellulose synthase-like gene. Plant Physiol 133: 1000-1010 99 CHAPTER 4: LIGHT-REGULATION OF THE MAIZE CSL GENE FAMILY AND ENZYMATIC ACTIVITIES RELATED TO CELL WALL BIOSYNTHESIS ABSTRACT To gain a better understanding of the developmental regulation of cell wall biosynthesis, light-regulation of cell-wall-related enzymatic activities were analyzed in the maize mesocotyl. (1,4)-B-glucan synthase activity was reduced by ~50%, (1,4)-B— mannan synthase by ~85%, and galactose incorporating activity by ~80% by exposure of mesocotyls to a 15 min light pulse (97 uEinstein s'1 m'z) 16 hour before harvest. The decline of the (1,4)-B-mannan synthase and the galactose incorporating activities was gradual (half life of ~4 hours), suggesting that these enzyme activities are regulated by transcription of the corresponding genes. An attempt was made to correlate these activities with the light-regulation of the mRN A levels of the ZmCSL genes in the mesocotyl. Many of the ZmCSLA and C genes, but only a few of the ZmCSLD, E and F genes are expressed in the mesocotyl. Many but not all of them are light-regulated. The transcript levels of only two of the nine CSZA genes expressed in the mesocotyl are reduced by light-treatment, whereas the mannan synthase activity is reduced by >85% by light-treatment. Quantitative RT-PCR showed that the major CSLA expressed in the maize mesocotyl is CSLA3. The transcript levels of CSLA3 are reduced by seven-fold after light-treatment, which correlates with the strong reduction in mannan synthase activity after light-treatment. 100 INTRODUCTION An excellent system for studying the relation of plant growth by an environmental factor such as light is the maize seedling. When seedlings are grown in complete darkness, 3 structure called the mesocotyl develops, which serves to position the meristem at the soil surface. Once the dark-grown seedlings receive a short light- treatment, the rapid elongation of the mesocotyl is reduced. One enzymatic activity in the mesocotyl, which is light- and auxin-regulated, is the Golgi-localized glucan synthase of maize (Walton and Ray 1982a and b). In Chapter 2 it was shown that this enzyme synthesizes (1,4)-B-glucan. In this chapter, the regulation of glycan synthases related to cell wall biosynthesis, including the Golgi-localized (1,4)-B-glucan synthase, (l,4)-B—mannan synthase, a galactose-incorporating activity, (1,4)-B-xylan synthase, and callose synthase were studied. An attempt was made to correlate this with the light-regulation of transcription of the ZmCSL gene family. MATERIALS AND METHODS Genomic DNA isolation One gram of tissue that had been frozen in liquid nitrogen and stored at -80°C was ground, resuspended in 10 ml hexadecyltrimethylammonium bromide (CT AB) buffer and incubated at 65°C for 30 min. CT AB buffer (2%) was prepared by dissolving 2 g CT AB, 8.2 g NaCl, and 1 ml B-mercaptoethanol in 100 ml 100 mM Tris, pH 8.0. The CT AB buffer was heated to 65°C in order to dissolve it after the addition of NaCl. The genomic 101 DNA was extracted twice with an equal volume of chloroform: isoamyl alcohol (24 : 1) for 5 min, followed by centrifugation at 3000 g for 10 min. The aqueous phase was collected, isopropanol was added to 0.8 volumes, and the genomic DNA was precipitated by centrifugation at 3000 g for 10 min. The pellets were washed with 70% ethanol, and the genomic DNA was air dried and resuspended in 10 ml TE buffer (10 mM Tris, pH 7.5, 1 mM EDTA). RNA isolation Tissue frozen in liquid nitrogen and stored at -80°C was ground, transferred to eppendorf tubes (100 mg per tube) and 1 ml of TRIzole reagent (Cat. No 15596, Invitrogen) was added. All following incubation steps were performed at room temperature and the centrifugations at 4°C, unless otherwise mentioned. The mixture was incubated for 30 min, centrifuged for 10 min at 12,000 g and 0.2 ml of chloroform was added. The tubes were shaken for 15 s, incubated for 2 min, and centrifuged at 12,000 g for 15 min. The aqueous phase was transferred to a new tube and the RNA was precipitated with isopropyl alcohol by centrifugation at 12,000 g for 10 min. The RNA was washed with 75% ethanol, precipitated by centrifugation at 7,500 g for 5 min, air dried, and redissolved in 243 pl diethylpyrocarbonate (DEPC)-treated water. The RNA was DNAse treated by adding 27 1110f 10X DNAse buffer (100 mM Tris pH 7.5, 25 mM MgC12,5 mM CaClz) and 1 ul of RNase free DNAse (Cat. No. 85897570-36, Roche), incubated for 30 min, and heated for 10 min at 85°C in order to inactivate the DNAse. The RNA concentration was calculated by measuring OD260. 102 Reverse transcription RNA (l-3.4 ug) isolated from the second cm of the mesocotyl below the first node of dark-grown and light-treated seedlings was reverse-transcribed into cDNA with the Superscript III kit (Cat. No. 18080-051, Invitrogen). The volume of the RNA solution was adjusted to 8 111 using DEPC-treated water. One ul of 50 11M oligodT primer and 1 ul of 10 mM dNT P mixture were added followed by incubation at 65°C for 5 min. Two u] of 10X reverse transcriptase buffer, 4 til 25 mM MgC12, 2 1110.1 mM DTT, 1 pl RNAse OUT and 1 ul of Superscript III reverse transcriptase were added followed by incubation at 50°C for 50 min. The reaction was terminated by heating at 85°C for 5 min. Primer design Primers for ZmCSL genes were designed based on the open reading frames using the PrimerSelect program (DNAstar, Madison, WI) (parameters for primer design were the following: primer length of 17-24 hp; product length of 400-600 bp; melting temperature of 39.1-70.2 °C). The primers were tested on other ZmCSL genes using the PrimerSelect program in order to determine their specificity. Only primers that did not give a predicted PCR product with the other ZmCSL genes were synthesized (Integrated DNA Technologies, Coralville, IA). The sizes of the amplified DNA fragments were determined by agarose gel electrophoresis. Primers amplifying the correct size fragments from genomic DNA were selected for PCR on cDNA. 103 Genomic PCR The PCR reaction mixture was prepared by mixing 10 ul 10X PCR buffer minus MgClz, 2 111 10 mM dNT P mixture (Cat. No. 12048400, Roche), 3 til 50 mM MgC12, 2.5 ul of the forward and reverse primer (50 nM), 9 ul of genomic DNA solution, 0.33 111 Taq polymerase (5 U / ul) (Cat. No. 18038-018, Invitrogen) and 71 ul ddH20. PCR was performed under the following conditions on a Gradient 40 Robocycler (Stratagene): 3 min at 94°C (1 cycle), 1 min 94°C, 1 min 55°C, 1 min 72°C (33 cycles) and 10 min at 72°C (1 cycle). Semi-quantitative RT-PCR The PCR reaction mixture was prepared by mixing 10 ul 10X PCR buffer minus MgClz, 2 pl 10 mM dNT P mixture, 3 pl 50 mM MgC12, 2.5 it] of forward and reverse primer (50 11M), 1 ul of cDN A solution (50 ng / til RNA equivalent), 0.33 ul Taq polymerase (5 U / ul) and 79 ul ddeO. PCR was performed under the following conditions on a Gradient 40 Robocycler: 3 min at 94°C (1 cycle), 1 min 94°C, 1 min 55°C, 1 min 72°C (30 cycles) and 10 min 72°C (1 cycle). Ten ul samples were taken after 22, 26, and 30 cycles and analyzed by agarose gel electrophoresis in order to determine if the PCR reaction was in the linear range. Agarose gel electrophoresis Agarose gel electrophoresis was performed with 150 ml 1.3% Tris-Acetate-EDTA (TAE) agarose gels stained with ethidium bromide (15 u] of 5 mg/ml stock solution). A stock of 10X TAE buffer, pH 8.2, was prepared by mixing 48.4 g Tris, 10.9 g glacial 104 acetic acid, 2.9 g EDTA, and deionized water to a final volume of 1 l. The agarose gels were run in TAE buffer for 30 min at 94 V and imaged with a Gel Doc EQ System (BioRad). Real-time quantitative RT-PCR Primers were designed as described under primer design in this materials and methods section. Sequences of the primers used for quantitative PCR can be found in supplementary Table 18. Primers with an efficiency between 90-110% were selected and the primer concentration in the PCR reactions was 300 nM. An equivalent of 5 ng of reverse transcribed RNA and 7.5 ul of 2X SYBR® Green I dye (Cat. No. 4309155, Applied Biosystems) was added to each reaction. RNA was isolated from the second centimeter of the maize mesocotyl of dark and light-treated maize seedlings and reverse transcription into cDNA was performed as described under “Reverse transcription” in this Materials and Methods section. The final reaction volume was 15 pl. PCR reactions were performed on the ABI PRISM® 7700 Sequence Detection System (PE Biosystems) with the following cycling parameters: 2 min at 50 °C, 1 cycle, 10 min 95 °C, 1 cycle and 15 sec 95 °C, 1 min 60 °C for 40 cycles. The relative expression compared to EFI awas determined with the 2'“. method (Livak and Schrnittgen, 2001). Enzyme assays Enzyme assays with 100 pl of microsomal solution (prepared as described in the materials and methods of chapter 1) were performed in 15 ml Corex conical glass tubes. Glucan synthase assays were performed with 20 mM Mg“ and 49 nM UDP-[3H] glucose 105 (34 Ci I mmol, Sigma). Callose synthase assays were performed with 49 nM UDP- [3H] glucose and 1 mM UDP-glucose. Xylan synthase assays were performed with 2.0 mM MgSO4, 4.1 mM MnClz and 1.2 uM UDP-[14C]xylose (238.4 mCi / mmol, Perkin- Elmer). Mannan synthase assays were performed with 2.0 mM MgSOa, 4.1 mM MnClz and 1.4 nM GDP-[14C]mannose (260 mCi / mmol, Amersham). Assays with 1.37 nM UDP-[14C]galactose (367 mCi / rnmol, Perkin-Elmer) were performed with 1.2 mM MgSO4. Inosine diphosphatase (IDPase) assays were performed by mixing 0.4 ml assay mixture and 50 ul microsomal solution. The assay mixture was prepared by mixing 13.2 ml stock solution and 1.65 ml 3% digitonin. The stock solution was prepared by mixing 30 ml 80 mM Tris, pH 7.5, 200 ill 1 M MgClz, 66 mg IDP and 6.2 ml water. This mixture was incubated for 45 min; the reactions were stopped by adding 1 ml 14% trichloroacetic acid (TCA) followed by centrifugation for 20 min at 1000 g. One ml of supernatant was removed and 1 ml of freshly prepared Taussky-Shorr reagent was added. Taussky-Shorr reagent was prepared by mixing 5% (w/v) FeSO4, 1% (w/v) NH4MO4 and l N HZSO4 (T aussky and Shorr, 1953). This mixture was vortexed and OD710 was measured after 10 min. Determination of the amount of MLG in maize cell walls Ten second centimeter mesocotyl segments of dark-grown and light-treated maize seedlings were collected in eppendorf tubes and frozen in liquid nitrogen. The samples were boiled for 20 minutes in 1 ml of ddH20, ground and centrifuged for 6 min at 14,000 rpm at 4°C. The supernatant was removed and the pellet was washed three times with water, once with ethanol and once with acetone. The pellet was dried and weighed. One 106 ml of 25 uM Bis-Tris (pH 6.5), 4 pl of 2% sodium azide and 10 ul of lichenase were added to the pellet and this mixture was incubated overnight at 30°C. Ethanol was added to a final concentration of 70%, centrifuged for 10 min at 14,000 g and the supernatant was collected. The pellet was washed twice with 70% ethanol and the supematants were combined. The supernatant was dried under vacuum and the pellet was dissolved in one ml of water. Twenty ul of sample was analyzed by HPAEC-PAD with the MLG program (see Table 2, Chapter 2). When MLG is hydrolyzed by a MLG-specific enzyme such as lichenase characteristic hydrolysis products including G4G3G, G4G4G3G and G4G4G4G3G are released. The PAD response peak of each of these oligosaccharides was integrated and taken as a measure for the amount of MLG present in the maize mesocotyl. RESULTS Light-regulation of the maize CSL gene family Amplification of ZmCSL genes from genomic DNA Most (83%) of the ZmCSL genes could be amplified from genomic DNA using gene-specific primers (Table 6; Appendix III, Tables 15, 16 and 17). The following ZmCSL genes could not be amplified from genomic DNA although for each of these genes multiple primer pairs were tested: C2, C10, C12, D4, F1 and F7. In the case of CSLC2, C10 and C12 this could be due to the incomplete annotation of these genes. The CSLD4, F1 and F7 genes were completely annotated however, so the reason why these genes could not be amplified from genomic DNA is not clear. 107 Semi-quantitative RT-PCR Nine ZmCSLA genes, nine ZmCSLC genes, four ZnCSLD genes, two ZmCSLE genes, two ZmCSLF genes, and none of the ZmCSLH and ZmCSLX genes could be amplified from cDNA prepared from RNA from dark-grown mesocotyls (Table 6). 108 Table 6: PCR amplification of ZmCSL genes from genomic DNA and cDNA. n.d. means not determined. ZmCSL Amplified from genomic DNA Amplified from cDNA Al Yes Yes A2 Yes No A3 Yes Yes A4 Yes Yes A5 Yes Yes A6 Yes Yes A7 Yes Yes A8 Yes Yes A9 Yes Yes A10 Yes Yes C1 Yes Yes C2 No No C3 Yes Yes C4 Yes Yes C5 Yes No C6 Yes Yes C7 Yes Yes C8 Yes No C9 No No C 10 Yes No Cl 1 Yes No C12 No No D1 Yes No D2 Yes Yes D3 Yes No D4 N o No D5 Yes No E1 Yes Yes E2 n.d. n.d. E3 Yes No E4 n.d. n.d. F1 N o No F2 Yes No F3 Yes No F4 Yes No F5 Yes Yes F6 Yes Yes F7 No No Hi Yes No X1 Yes No The transcript levels of ZmCSLA], 4, 5, 6, 7, 8 and 10 were not or weakly affected by light, and the transcript levels of ZmCSLA3 and 9 were strongly increased by light (Figure 25). The transcript levels of ZmCSLC] were not affected by light, ZmCSLC3 and 6 weakly, and ZmCSLC4 and 7 strongly. The transcript levels of ZmCSLE] and 109 ZmCSLF 5 were not affected by light whereas the transcript levels of Zm CSLD2 and ZmCSLF 6 were only weakly affected by light. The transcript levels of Elongation Factor 2a (EFZa), a control gene, were not affected by light. I meant -~:2mcst.ca _ ZmCSLC4 ZmCSLC” -'32mCSI.D2 ZmCSLEl 21116811", E1“ ZmCSLAlO Ed?” 9:?“I ZmCSLPG .- .i W“ Figure 25. Ethidium-stained agarose gel showing RT-PCR products of the ZmCSL gene family. PCR was performed for 30 cycles as described in materials and methods. This number of cycles was not saturating for any of the genes. The source of mRN A was the second centimeter of the mesocotyl of dark-grown (D) or light-treated (L) seedlings. 110 Real-time quantitative RT-PCR Quantitative RT—PCR was performed with cDNA reverse transcribed from RNA isolated from dark and light-treated maize seedlings using gene specific primers (See Appendix IV Table 18). The relative expression compared to EFI awas determined with the Z‘Mct method (Livak and Schmittgen, 2001). Figure 26 shows that the major CSLA expressed in the maize mesocotyl is CSLA3 (57% of total CSLA transcript levels) and its transcript levels were reduced approximately seven-fold by light (Table 7). The transcript levels of CSLA7 were reduced approximately 13-fold by light. The total reduction in CSLA transcript levels after light-treatment was approximately 60%. The transcript levels of CSLA], A6 and A9 were not affected by light (Fig. 26). The transcript levels of CSLA4, 5, 8 and 10 were weakly affected by light (Fig. 26). The transcript levels of CSLA4 were increased approximately 2-3 fold and the transcript levels of CSLA8 and A10 were reduced by approximately 2-3 fold by light. 111 0.25 0.2 ' 0.15 - 0.1 0.05 Normalization to EF] 0' 0.1 0.08 0.06 0.04 0.02 Normalization to EFI a CSL genes Figure 26. Real-time quantitative RT-PCR results for the maize CSLA gene family. The relative expression level was normalized to EFI 0. Grey bar: dark-treated maize seedlings. White bar: light-treated seedlings. A: CSLA3 included; B: the same figure but excluding CSLA3 (expanded y-scale). Data represent the mean (n = 2-3) i SE. 112 Figure 27 shows the quantitative RT-PCR results for the CSLC, D, E and F genes. Table 7 shows the fold change in gene expression compared to EFI a The transcript levels of CSLC], E1 and F5 were increased by approximately 4, 6 and 9-fold respectively. The transcript levels of CSLC3, C7 and F6 were reduced by 2-3, 2 and 2-3-fold respectively. The transcript levels of CSLC4 were strongly reduced after light-treatment. No conclusion could be made about the light-dependence of CSLC6 transcript levels because the efficiency of the primer pair used was not in the acceptable range. 113 0.14 0.12 J 0.1 “ 0.08 4 0.06 0.04 - 0.02 ‘ lfl fl nf lfl O ‘ 1 fl _,_i=_ C1 C3 C4 C6 C7 D2 E1 F5 F6 Normalization to BF] 0' 0.035 0.03 ' 0.025 a 0.02 ' 0.015 ‘ 0.01 ' °°°°ii l J . L C1 C3 C4 C6 C7 D2 E1 F5 F6 Normalization to EFI a CSL genes Figure 27. Real-time quantitative RT-PCR results for the maize CSLC, D, E and F gene family. The relative expression was compared to EFI a Grey bar: dark-treated maize seedlings. White bar: light-treated seedlings. A: CSLF5 included; B: the same figure but excluding CSLF5 (expanded y-scale). Data represent the mean (it = 2-3) i SE. 114 Table 7: Change in transcript levels compared to EFI adetermined with real-time quantitative RT-PCR ZmCSL Fold change A1 -l.1 A3 -7.0 A4 +2.5 A6 +1.2 A7 -13.3 A8 -2.5 A9 +1.1 A10 -2.4 Cl +3.7 C3 -2.5 C4 —l9.7 C6 n.d. C7 -1.7 D2 -1.l E1 +6.1 F5 +9.4 F6 -2.4 Light-regulation of enzymatic activities related to cell wall biosynthesis. Time course experiments were performed to determine if there was a correlation between the light-regulation of ZmCSL transcripts and enzymatic activities related to cell wall biosynthesis. Enzymes assayed were (l,4)-B-glucan synthase, (1,4)-B-mannan synthase, UDP-galactose incorporating activity, (1,4)-B-xylan synthase and (l,3)-B- glucan synthase. As a control, latent IDPase, a Golgi-localized enzymatic activity which is not affected by light, was used (Ray et al., 1969; Walton and Ray 1982a; Mitsui et al., 1994). The enzyme assays were stopped after 0, 15, 30, 45 and 60 min and radioactive incorporation into ethanol-insoluble products was determined. Walton and Ray (1982a) found that the IDPase activity was reduced by around 15% after light-treatment and callose synthase by 25%. In the experiments described here an increase in callose synthase activity of 15-55% was observed. Mannan synthase, xylan synthase, glucan 115 synthase and the galactose-incorporating activity were reduced by 85-90%, 15%-25%, 48%-53% and 75%-83% respectively (Figure 28). ’cTo ”£3 :1. :3. = 8‘ A: UDP-Glc (49 nM) ‘—‘ E * E V 64 V 3‘ b '5' ‘ '5 § 4; ,,,,, z 32‘ a a - 2 m 0‘ I I I r m 0 15 30 45 60 Time(min) 30160 22 = ‘ —- E 120. E 2:: E‘: a ‘ 29 IE 80. E Q . a g 401 g m 0. Lu 0 - . . . . O 15 30 45 60 Time(min) E 120 ’30 > 3 3100- :a 12‘ F: IDPase u— 5 J E; 80‘ b 35 60- LE. 8‘ E40” “5’ 4- ''''' :3 204 5‘ - m m 0 o I I f I I 0 15 30 45 60 Time(min) Figure 28. (1,4)—B-glucan synthase (A), (1,4)-B-mannan synthase (B), UDP-Gal incorporating activity (C), (l,4)—B-xylan synthase (D), (l,3)—B-glucan synthase (E), and latent IDPase (F) activities in microsomes from dark-grown (solid line) or light-treated mesocotyls (dashed line). Reactions were stopped after 0, 15, 30, 45 or 60 min. Data represent the mean (n = 2) i SE. 116 Analysis of the radiolabeled products The products in which glucose was incorporated at 4 to 49 nM was a (1,4)-B- glucan (Figure 6 and 7, Chapter 2) and at 1 mM a(1,3)-B-glucan (Figure 8, Chapter 2). The products synthesized in the presence of GDP-[14C]mannose were fully solubilized by commercial mannanase (see Table 1, Chapter 2). The hydrolysis products were analyzed by HPAEC-PAD using the manno/xylooligosaccharide program (see Table 2, Chapter 2). Figure 29 shows that the radiolabeled hydrolysis products coeluted with mannose, mannobiose and mannotriose, which indicates that the product is (1,4)-B-mannan. M2 M3 § PAD response DPM u. - gm O is ' ' 50 time (min) Figure 29. The product synthesized from UDP-[I4C1mannose was treated with endo- (l,4)-[3-mannanase and the hydrolysis products were analyzed by HPAEC-PAD. A, Oligosaccharide standards: M1 (mannose), M2 (mannobiose), M3 (mannotriose). B, Radioactivity. 117 The products synthesized in the presence of UDP-[14C]xylose were treated with endo-(1,4)-B-xylanase and the hydrolysis products were analyzed by HPAEC-PAD with the manno/xylooligosaccharide program (see Table 2, Chapter 2) (Figure 30). Figure 30 shows that the radiolabeled hydrolysis products coeluted with xylose and xylobiose, which indicates that UDP-[14C]xylose is incorporated into a (l,4)-B-xylan. d) g A g X2 0 x1 E 52 X3 L2 IOOO‘B E . E‘ 500 0 . - .. 25 50 Time(min) Figure 30. The product synthesized from UDP-[14C]xylose was treated with xylanase M6 and the hydrolysis products were analyzed by HPAEC-PAD. A, Oligosaccharide standards: X1 (xylose), X2 (xylobiose), G2 (cellobiose), X3 (xylotriose), L2 (laminaribiose). B, Radioactivity. The product in which [14C]galactose was incorporated was treated with several different hydrolytic enzymes such as endo-l,4-B—D-galactanase and B-l-(3,4,6)- galactosidase (see Chapter 2, Table 1), none of which hydrolyzed the product. 118 The only enzyme mixture that hydrolyzed this product was pectinase from Aspergillus niger (See Table 1, Chapter 2) (data not shown). Monosaccharide analysis of the hydrolyzed product showed that the radiolabel was still in galactose (data not shown). Attempts to further characterize the galactose-containing product were unsuccessful. Correlation between the levels of the Zm CSLA transcripts and (1,4)-B-mannan synthase activity Whereas (1,4)—B-mannan synthase activity was reduced by 85-90% by light, the transcript levels of only two of nine ZmCSLA genes were strongly reduced by light. This apparent discrepancy can be explained by the difference in expression levels of the different CSLA proteins. CSLA3 accounts for 57% of the CSLA transcript levels and its transcript levels are reduced seven-fold after light-treatment. This reduction in CSLA3 transcript levels could result in a comparable reduction in the amount of CSLA3 mRN A translated into CSLA3 protein. This could explain the strong reduction in mannan synthase activity after light-treatment (Fig. 28). Time-course of the light-inhibition of enzymatic activities related to cell wall biosynthesis In order to get a better insight into the mechanism of regulation of the glycan synthase activities by light, time-course experiments for inhibition of enzyme activity after light-treatment were performed for the (1,4)-B-mannan synthase, (l,4)-B-xylan synthase and the UDP-galactose incorporating activities (Figure 31). A similar time- course for the (1,4)-B-g1ucan synthase was performed by Walton and Ray, (1982a). Five 119 trays of dark-grown maize seedlings were 1i ght-treated for 15 minutes and second centimeter mesocotyl segments were collected immediately from one tray of seedlings. The other four trays of light-treated seedlings were placed back in complete darkness and the mesocotyl segments were collected after 1, 3, 6 and 12 hours. Figure 31 shows that the (l,4)-B-mannan synthase and the UDP-galactose incorporating activities, gradually declined after light-treatment. The (l,4)-B—xylan synthase activity was not affected by light-treatment. These data suggest that the (l,4)-B-mannan synthase and the UDP- galactose incorporating activity are probably regulated at the level of transcription rather than by a rapid post-translational process such as phosphorylation. This conclusion is consistent with the quantitative RT-PCR results showing light-induced reduction in CSL mRNA levels (Figures 26 and 27). 120 Enzyme activity (fmoVug) Enzyme activity (fmol/ttg) Enyme activity (fmol/ug) 120 « , . 1 30+ “““ f _ ...... ‘ ..... 4o . ------------------ "I (l,4)-fl-mannan synthase 0 I 1 I l I I I Y I I I I I 0 l 3 6 12 Time after light-treatment (hours) 0.16 ‘ 0.12 - I " IA - f T: """ 1"“. 0.08 q “““““ I ....... 0.04 ‘ i l UDP-Gal incorporating activity 0 I I I I r I I I I I I I I 0 l 3 6 12 Time after light-treatment (hours) ‘ L 400 pm} ........... i 300 . i ---------- I """" 200 ‘ 100 ‘ ~ (l,4)-flxylan synthase 0 7 If I l T T l I I I I l I O l 3 6 12 Time after 1i ght-treatment (hours) Figure 31. Time-course of inhibition of enzymatic activities after light-treatment for (1,4)-B-mannan synthase, the UDP-galactose incorporating activity and (1,4)-B-xylan synthase. Solid line: enzyme assays performed with microsomes isolated from dark- grown seedlings. Dashed line: enzyme assays performed with microsomes isolated from light-treated seedlings. Data represent the mean (n = 2) :t SE. 121 Analysis of MLG in cell walls of dark-grown and light-treated seedlings The CSLF gene family is involved in MLG synthesis (Burton et al., 2006). In order to determine if there is a correlation between the CSLF transcript levels and the amount of MLG, MLG levels were determined. Figure 32 shows the result of the MLG analysis of the cell walls of second centimeter mesocotyl segments of dark-grown and light-treated maize seedlings. There is no difference in MLG levels between the dark- grown and light-treated maize seedlings. The transcript levels of CSLF6 correlate with the MLG levels (Fig. 27). The transcript levels of CSLF5, however, do not correlate with the MLG levels. The transcript levels of CSLF5 are strongly increased after light- treatment in contrast to the MLG levels. This discrepancy can be explained in several ways. It could be the CSLF5 is not involved in MLG biosynthesis or that MLG biosynthesis might not be regulated at the level of transcription. Another possibility is that the turnover of MLG might be slow, so it is hard to see newly synthesized MLG against the background of MLG which is synthesized previously. 122 80 l l l 20‘ Peak areal mg of tissue 3 0 .. L G4G3G G4G4G3G G4G4G4G3G MLG oligosaccharides Figure 32. Amount of MLG in cell walls in dark-grown and light-treated seedlings. The integrated peak area of the PAD response per mg of dried cell wall material for each of three different oligosaccharides (G4G3G, G4G4G3G, G4G4G4G3G) released after lichenase treatment is shown. Data represent the mean (n = 6) i SE. DISCUSSION In this part of my research I investigated the light-regulation of expression of the ZmCSL genes and enzymatic activities related to cell wall biosynthesis. Expression of 18 of the 34 ZmCSL genes was detected in the maize mesocotyl. Nine of them are ZmCSLA, 5 are CSLC, 1 is CSLD, 1 is CSLE and 2 are CSLF genes. It is curious that maize has 11 ZmCSLA genes, although (1,4)-B-mannan is a minor polysaccharide in maize (Carpita et al., 2001). After light-treatment of the dark- grown seedlings there was a more than 85-90% reduction in (1,4)-B-mannan synthase activity. This is the first time to my knowledge that mannan synthase activity has been shown to be regulated by light. In contrast, transcript levels of most of the CSLA genes were not affected by light-treatment. There are several possible explanations for the 123 observed discrepancy between CSLA transcript levels and mannan synthase activity. The most likely explanation is that CSLA3 is expressed more highly than the other CSLA genes and is strongly light-regulated. In this case it is assumed that there is a strong correlation between CSLA3 transcript levels, CSLA3 protein levels and CSLA3 mannan synthase activity. It has not yet been shown that CSLA3 encodes a mannan synthase. However, CSLA3 is most likely involved in mannan biosynthesis because it clusters with CSLA proteins from other plant species that have been demonstrated to be mannan synthases. The closest ortholog of ZmCSLA3 is OsCSLAl, which is a mnnan synthase (Liepman et al., 2007). The gradual decrease of the (1,4)-B-mannan synthase activity after light- treatment of dark-grown seedlings supports the hypothesis the mannan biosynthesis is regulated at the level of transcription. In case mannan biosynthesis would be regulated at the post-transcriptional level, for example by reversible phosphorylation, I would not expect this gradual decrease in mannan synthase activity after light-treatment, but a more abrupt change in activity. Possible candidates genes for encoding the light-regulated (1,4)-B-glucan synthase are the members the CSLC gene family, because AtCSLC4 and TmCSLC are involved in the synthesis of the (l,4)-B—glucan backbone of xyloglucan (Cocuron et al., 2007). ZmCSLC] transcript levels are not affected by light, ZmCSLC3 and 6 transcript levels are weakly and ZmCSLC4 transcript levels are strongly reduced after light- treatment. The reduction in transcript levels of the CSLC genes correlates with the reduction in (1,4)-B-glucan synthase activity after light-treatment. 124 The CSLD proteins might be involved in (l,4)-B-glucan biosynthesis in tip- growing cells (Doblin et al., 2001; Favery et al., 2001; Wang et al., 2001; Roberts et al., 2007). Doan et al. (2001) showed that the main glycosyltransferases expressed in tobacco pollen tubes were NtCSLDI and NtGSLI . Pollen tubes cell walls mainly consist of cellulose and callose. From these data they concluded that CSLD proteins might be involved in B(1,4)—glucan biosynthesis in tip-growing cells. Support for this hypothesis comes from the work of Favery et al. (2001) and Wang et al. (2001). They showed that a mutation in AtCSLD3 caused the root hair tips to leak cytoplasm and to rupture, which indicates the tensile strength is changed. Root hairs elongate by tip growth. One ZmCSLD, ZmCSLD2, was expressed in the second centimeter of the maize mesocotyl. It is possible that ZmCSLD2 encodes for the light-regulated Golgi-localized (1,4)-B-glucan synthase. However, the transcript levels of ZmCSLDZ were not affected by light treatment, which does not correlate with the down-regulation of the (1,4)-B-glucan synthase activity after light-treatment. Of the ZmCSLF genes only F5 and F6 are expressed in the maize mesocotyl. The CSLF proteins are involved in MLG biosynthesis (Burton et al., 2006). We analyzed the MLG content of the second centimeter of the maize mesocotyl before and after light- treatment of dark-grown maize seedlings. No difference in MLG content was found (Figure 34). The transcript levels of ZmCSLF5 were strongly increased after light- treatment. This is not consistent with the observation that MLG levels do not change after light-treatment. If ZmCSLF5 encodes for a MLG synthase 1 would not expect that the transcript levels of this gene would be strongly increased after light-treatment. One possibility is that CSLF5 is involved in the synthesis of a different polysaccharide. 125 Another possibility is that changes in MLG levels caused by the increase in CSLF5 transcript and protein levels are not visible against the background of already synthesized MLG. Of the CSLE genes, only CSLE] is expressed in the mesocotyl, and its transcript levels are strongly increased after light-treatment. The biochemical function of the CSLE proteins is not known. ZmCSLX and H are not expressed in the maize mesocotyl. The CSLH subfamily is specific for cereals (Hazen et al., 2000) and is hypothesized to be involved in the synthesis of a cereal-specific polysaccharide such as MLG. Because ZmCSLH is not expressed in the maize mesocotyl it cannot be involved in the synthesis of MLG in the mesocotyl. Whether ZmCSLH is involved in the synthesis of MLG in other tissues in maize remains to be determined. The biochemical function of ZmCSLX is also not known. The semi-quantitative PCR results do not correspond very well to the results obtained with quantitative RT-PCR. A major discrepancy for instance was CSLA9. The semi-quantitative PCR results showed that the transcript levels of CSLA9 are strongly light-regulated. However, the quantitative RT-PCR experiments showed no difference in transcript levels. The reason for this discrepancy is not known. Semi-quantitative PCR also does not give accurate information about the real transcript levels. A good example of this is CSLF5. From the semi-quantitative PCR data it is hard to conclude that there is an increase in transcript levels after light-treatment in contrast to quantitative RT-PCR, which showed a nine fold increase. 126 REFERENCES Burton R, Wilson S, Hrmova M, Harvey A, Shirley N, Medhurst A, Stone B, Newbigin E, Bacic A, Fincher G (2006) Cellulose synthase-like Cle genes mediate the synthesis of cell wall (1,3;1,4)—B-D-glucans. Science 311: 1940- 1942 Carpita N, Defemez M, Findlay K, Wells B, Shoue D, Catchpole G, Wilson R, McCann M (2001) Cell wall architecture of the elongating maize coleoptile. Plant Physiol 127: 551-565 Cocuron J, Lerouxel O, Drakakaki G, Alonso A, Liepman A, Keegstra K, Raikhel N, Wilkerson C (2007) A gene from the cellulose synthase-like C family encodes a B-l,4—glucan synthase. Proc Natl Acad Sci USA 104: 8550-8555 Dhugga K, Barreiro R, Whitten B, Stecca K, Hazebroek J, Randhawa G, Dolan M, Kinney A, Tomes D, Nichols S, Anderson P (2004) Guar seed B-mannan synthase is a member of the cellulose synthase super gene family. Science 303: 363-366 Doblin M, De Melis L, Newbigin E, Bacic A, Read S (2001) Pollen tubes of Nicotiana alata express two genes from different B-glucan synthase families. Plant Physiol 125: 2040-2052 Favery B, Ryan E, Foreman J, Linstead P, Boudonck K, Steer M, Shaw P, Dolan L (2001) KOJAK encodes a cellulose synthase-like protein required for root hair cell morphogenesis in Arabidopsis. Genes Dev 15: 79-89 Hazen S, Scott-Craig J, Walton J (2002) Cellulose synthase-like genes of rice. Plant Physiol 128: 336-340 Liepman A, Nairn C, Willats W, Serensen I, Roberts A, Keegstra K (2007) Functional genomic analysis supports conservation of function among CslA gene family members and suggests diverse roles of mannans in plants. Plant Physiol 143: 188 1-93 Liepman A, Wilkerson C, Keegstra K (2005) Expression of cellulose synthase-like (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proc Natl Acad Sci USA 102: 2221—2226 Livak K, Schmittgen T (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25: 402-408 127 Mitsui T, Honma M, Kondo T, Hashimoto N, Kimura S, Igaue I (1994) Structure and function of the Golgi complex in rice cells (II. Purification and characterization of Golgi membrane-bound nucleoside diphosphatase). Plant Physiol 106: 119-125 Ray P, Shininger T, Ray M (1969) Isolation of B-glucan synthetase particles from plant cells and identification with Golgi membranes. Proc Natl Acad Sci USA 64: 605- 612 Roberts AW, Bushoven JT (2007) The cellulose synthase (CESA) gene superfamily of the moss Physcomitrella patens. Plant Mol Biol 63: 207—219 Suzuki S, Li L, Sun Y, Chiang V (2006) The cellulose synthase gene superfamily and biochemical functions of xylem-specific cellulose synthase-like genes in Populus trichocarpa. Plant Physiol 142: 1233-1245 Taussky H, Shorr E (1953) A microcolorimetric method for the determination of inorganic phosphorus. J Biol Chem 202: 675-685 Walton J, Ray P (1982a) Inhibition by light of growth and Golgi-localized glucan synthase in the maize mesocotyl. Planta 156: 302-308 Walton J, Ray P (1982b) Auxin controls Golgi-localized glucan synthase activity in the maize mesocotyl. Planta 156: 309-313 Wang X, Cnops G, Vanderhaeghen R, De Block S, Van Montagu M, Van Lijsebettens M (2001) AtCSLD3, a cellulose synthase-like gene important for root hair growth in Arabidopsis. Plant Physiol 126: 575-586 128 CHAPTER 5: FUTURE DIRECTIONS The filture for research in cell wall biosynthesis is bright. Biofuels research is attracting a large amount of money for cell wall research. Progress is also being made in the sequencing of the genomes of grasses. Rice is sequenced and other grasses in the process of being sequenced are maize and Brachypodium distachyon (Brachypodium). This will create good conditions for studying cell wall biosynthesis in grasses. An important question to be answered in the field of cell wall biosynthesis is which genes encode for the enzymes synthesizing the cell wall polysaccharides and how their synthesis is regulated. Sequencing of the maize genome will allow us to finish the annotation of the maize CSL gene family and determine the number of CSL genes. Pseudo genes can be determined by investigating if there are stop codons in their open reading frames. Alternative splice variants can be found by comparing EST sequences with the genomic DNA sequences. We might be able to establish the biochemical functions of some of the maize CSL genes by expressing them in heterologous expression systems. It would be interesting to determine the biochemical functions of the CSLA proteins of maize by heterologous expression, especially the members of the cereal specific clade of CSLA proteins. None of the CSLA proteins in this cereal specific clade has been tested for mannan synthase activity yet and all the CSL proteins which are mannan synthases cluster separately fi'om these cereal CSLA proteins. Knock-outs in the CSLA3 and A 7 genes could be investigated The transcript levels of CSLA3 and 7 are strongly reduced after light-treatment, which correlates with the reduction in mannan synthase activity after light-treatment. There might be a reduction in mannan synthase 129 activity in a CSIA3 knock-out and in case mannan is important for elongation growth, there might be a growth related phenotype. The levels of mannan in the maize mesocotyl are low however, so any phenotype related to elongation grth might be subtle. An interesting question is why there are so many CSLC genes in maize in comparison to Arabidopsis and rice. AtCSLC4 and TmCSLC are involved the synthesis of the glucan backbone of xyloglucan. Some of the CSLC genes could be pseudogenes or not all of them encode for xyloglucan glucan synthases. The transcript levels of CSLCI were increased after light-treatment and the transcript levels of CSLC4 were strongly reduced after light-treatment. Knock—outs in these genes could be investigated for phenotypes related to growth. The CSLC genes could be heterologously expressed in order to determine if they encode for xyloglucan glucan synthases. Other candidates for investigating knockouts are the CSLD2 and E1 genes. Only one CSLD and E gene are expressed in the maize mesocotyl and the transcript levels of CSLE] are strongly upregulated after light-treatment. OsCSLF2 has been shown to be involved in MLG biosynthesis. CSLF5 and F6 might be involved in MLG biosynthesis in the maize mesocotyl, but the strong increase in transcript levels of ZmCSLF 6 after light-treatment contradicts this hypothesis. The MLG levels are not changed after light-treatment of the maize mesocotyl, so the question is why the transcript levels of CSLF5 would be increased if this gene encodes for a MLG synthase? It would be interesting to investigate knockouts in CSLF5 and F6. Knockouts of these genes could give interesting phenotypes related to grth or mesocotyl elongation. By heterologously expressing these proteins we might be able to determine their function. 130 Once the maize genome is sequenced, promoters of the CSL genes can be analyzed for the presence of transcription factor binding sites which could be involved in auxin- or light-regulation. Some of the promoter elements which are expected to be found are auxin response elements, and promoter elements involved in light-regulation such as GATA boxes, GT1 and G-boxes. Especially investigation of the promoters of the CSL genes whose transcript levels are strongly affected by light could be interesting. 131 APPENDIX I: ANNOTATION OF THE MAIZE CSL GENE FAMILY Table 8: Genomic sequences used for the annotation of the ZmCSLA subfamily Maize gene Genomic sequences MAGI_... : genomic sequence comes from MAGI3 database MAGI4_... : genomic sequence comes from MAGI4 database ZmCSLA] MAGI4_62945, MAGI4_1 19251, MAGI4_1 19252, MAGI4_1 19253 ZmCSLAZ MAGI4_126476, MAGI4_138561, ZmGSStucl 1-12-04. 14465.2 ZmCSL/l3 MAGI4_122017 ZmCSL/l4 MAGI4_90697. MAGI4_90698, MAGI4_90700 ZmCSLAS MAGI4_96931, MAGI4_96932. AZM5_99705 ZmCSLA6 MAGI4_143329. MAGI4_143330 ZmCSLA 7 MAGI_21246, MAGI_79371, ZmGSStucl l- l2-04.l l6. 1 ZnCSLA8 MAGI4_125022 ZmCSL/l9 MAGI4_106644 ZmCSLA 10 MAGI4_146469, MAGI4_157680, MAGI4_157681, AZM5_15751, AZM5_18549 ZmCSLA II MAGI4_103458 132 Table 8 Continued ZmCSLC] MAGI4_123263, MAGI4_123264 ZnCSLCZ MAGI4_70285, ZMMBBC0414A15. ZmCSLC3 MAGI_106507, MAGI_106508, MAGI_106509, MAGI4_72417, MAGI4_7241 8. MAGI4_72419, AZM5_86073, BH227815, CW003073, CW007807, ZmGSStucl 1-12- 04.7424. 1 ZmCSLC4 MAGI4_121535 ZJnCSLC5 MAGI4_77380, MAGI4_77381, MAGI4_77382, MAGI4_103479, CZ3 86891 ZmCSLC6 MAGI4_7 8784, MAGI4_163203 ZmCSLC 7 MAGI4_7 2790, MAGI4_72791, MAGI4_72792, ZmGSStucl 1-12-04.6288.2 ZmCSLC8 MAGI4_1 19258 ZmCSLC9 MAGI4_1 18095 ZJnCSLCIO MAGI4_19169, APZT30938.R, ZMMBLb0010H15.R ZmCSLC] 1 MAGI4_34342, MAGI4_94745, MAGI4_1 13710 , APT A6 13 86R, CZ330204. CZ400294, ZmGSStucl 1-12-04.21281.1 ZmCSLCIZ MAGI4_91218 ZmCSLDI MAGI4_83198, MAGI4_83199, ZmGSStucl l-12-04.6504.1 ZmCSLDZ MAGI4_67442, MAGI4_1 10509, MAGI4_123679, MAGI4_151659, PUHOBO7.R, ZmGSStucl l- 12-04.5 1425. l, ZmGSStucl 1-12-04.91725 ZmCSLD3 MAGI4_965 19, MAGI4_965 20 ZnCSLD4 MAGI_100286, MAGI4_408 90, MAGI4_40891, MAGI4_40892, AZM5_30134, ZmGSStucl 1- 1 2-04. 10994.1 ZmCSLD5 MAGI4_1 54948, MAGI4_154949 133 Table 8 Continued ZmCSLE] MAGI_62646, MAGI_1449OS, MAGI4_126139, MAGI4_144906, AZM_5329, AZM5_95772, CC784651 ZmCSLEZ MAGI4_14602, AZM4_23029, AZM_5330, CC428870, PUHUD52.F ZmCSLE3 MAGI4_116066, MAGI4_153810, CZ323910 ZmCSLE4 MAGI4_48170, ZmGSStucl 1-12-04.3014l.1, CC984189, CG338953, CZ359872 ZnCSLFI MAGI_7632, MAGI_7634, MAGI_7635, MAGI4_93926, MAGI4_93927, MAGI4_93920, AZM5_9281 ZmCSLF 2 MAGI4_93921, MAGI4_93922, MAGI4_93923, MAGI4_93924 ZmCSLE? MAGI4_10379 ZmCSLF4 MAGI4_105268 ZmCSLF5 MAGI4_105266, MAGI4_105270 ZmCSLF6 MAGI4_159174 ZmCSLF7 MAGI_38566, MAGI4_91023, MAGI4_91024, CC386831 ZmCSLH MAGI4_89273 ZmCSLX MAGI_82244, MAGI4_4386, MAGI4_4387, MAGI4_4388, ZmGSStucl 1-12- 04.91351 134 Table 9: Detailed description of the annotation for each ZmCSL gene. Maize gene Description of annotation In some cases the genomic DNA or cDNA for a CSL was divided into several pieces. In these cases the genomic or cDNA was numbered according to the orientation in regards to the 5’ end of the gene with number 1 the part of the genomic DNA or cDNA closest to the 5’ end of the gene. The following fragments of genomic or cDNA are labeled 2 or 3. Proteins encoded by this genomic DNA were predicted using the FGENESH gene monocot gene prediction program unless otherwise mentioned. ZmCSLA] FGENESH does not predict exon 5 which is from bp 3922 to 4032 in the genomic DNA. The genomic DNA of exon 8, intron 8, and exon 9 is missing. The cDNA sequence is full length. ZmCSL/12 The 3’ part of genomic DNA and cDNA is missing. The DNA for exon 1, intron 1 and part of exon 2 is complete. The FGENESH prediction is correct. ZmCSLA3 FGENESH incorrectly predicts 3 genes, (bp 2578-3587, 4420—5003 and 7335-8897) in the genomic DNA. These 3 genes are in fact 1 gene. The start of exon 3 is at bp 4417 instead of bp 4420. FGENESH incorrectly predicts an exon from bp 4996-5003 in the genomic DNA. The FGENESH prediction for the start site of exon 4 is incorrect. The start site is at bp 7325 in genomic DNA instead of bp 7335. The genomic and cDNA sequences are full length. ZmCSLA4 Part of the genomic DNA of intron 3 is missing. The FGENESH prediction for the start site of exon 4 start is incorrect. The start site is not at bp 391 in genomic DNA ZmCSLA4.2, but at bp 429. Exon 6 starts at bp 922 in ZmCSLA4.2. Part of the genomic DNA and cDNA of exon 6 and intron 6 are missing. The FGENESH prediction for the start site of exon 7 is incorrect. The start site is not at bp 258 in genomic DNA ZmCSlA4.3, but at bp 183. 135 Table 9 Continued ZmCSLA5 The FGENESH prediction for the start site of exon 1 is incorrect. The correct start site is at bp 1559 in genomic DNA ZmCSlASJ, not at bp 1933. FGENESH incorrectly predicts 3 extra exons (bp 2575-2661, 2991-3146, 3226-3354) in genomic DNA ZmCSLA5. 1. The genomic DNA for intron 2 and intron 3 is missing and the genomic DNA and cDNA for exon 3 is missing. The FGENESH prediction for the 3’ end of exon 8 is incorrect. The 3’ end of exon 8 is at bp 1922 in genomic DNA ZnCSLA5.2 instead of bp 1910. ZmCSLA6 FGENESH dicot incorrectly predicts the start site of exon 1 at bp. 224. The correct start site is at bp 344 in the genomic DNA. FGENESH dicot predicts 5 extra exons (bp 1337-1768, 1918-2103, 2321-2568, 2765-3080, 3157-3235 in the genomic DNA) of intron 2. The FGENESH dicot prediction for the start site of exon 3 is incorrect. The start site is at bp 3445 instead of bp 3372 in the genomic DNA. FGENESH dicot does not predict exon 4, which is from bp 3793-3913 in the genomic DNA. The FGENESH dicot prediction for the end of exon 8 is at bp 548 instead of bp 5051 in the genomic DNA. FGENESH dicot does not predict exon 9, which is from bp 5410 to 5572 in the genomic DNA. The genomic and cDNA sequences are full length. ZmCSL/l7 FGENESH predicts an extra exon (bp 1809 tol895 in the genomic DNA) in intron 1. FGENESH does not predict exon 6, which is from bp 3642-3744 in the genomic DNA. The genomic and cDNA sequences are full length. ZmCSLA8 FGENESH incorrectly predicts an exon from bp 11843-11849. The start site of exon 1 is unclear. The genomic and cDNA sequences are not full length. ZmCSL/l9 FGENESH incorrectly predicts an exon from bp 2574-2596 in the genomic DNA. It does not predict exons 5 (bp 2972-3109) and exon 6 (bp 3230-3345). The genomic and cDNA sequences are full length. 136 Table 9 Continued ZmCSLAII FGENESH does not predict the exon which is from bp 1088-1197 in the genomic DNA. The exon which is from bp 1542-1656 in the genomic DNA has a stopcodon at bp 1585. The genomic and cDNA sequences are not full length. This might be a pseudogene. ZmCSLC] FGENESH is correct. The genomic DNA and cDNA sequences are full length. ZmCSLCZ FGENESH is correct. Part of the 5’ end of the genomic DNA and cDNA is missing. The genomic and cDNA sequences are not full length. ZmCSLC3 FGENESH is correct. The genomic and cDNA sequences are full length. ZmCSLC4 FGENESH combines 2 genes into 1 gene. The correct stopcodon is at bp 6951 in the genomic DNA. The genomic and cDNA sequences are full length. ZmCSLC5 FGENESH misses exon 3 (bp 3644-3935 in genomic DNA). The genomic DNA for exon 5 is missing, but the cDNA sequence is full length. ZmCSLC6 The FGENESH prediction for exon 3 is incorrect. The start site of this exon is at bp 3566 in the genomic DNA instead of bp 3671. The genomic and cDNA sequences are full length. ZmCSLC7 The FGENESH prediction for exon 3 is incorrect. The start site of this exon is at bp 2880 in the genomic DNA instead of bp 2958. The genomic and cDNA sequences are full length. ZmCSLC8 The FGENESH prediction is incorrect for the exon which is from bp 1632 to 1940 in the genomic DNA. The end of this exon is at hp 1940 instead of bp 2030. The genomic and cDNA sequences are not full length. 137 Table 9 Continued ZmCSLC9 The FGENESH prediction for exon 2 is incorrect. The end of this exon is at bp 965 in the genomic DNA instead of bp 937. This exon contains a stopcodon in exon 2 at bp 940 in the genomic DNA. FGENESH makes two exons out of exon 3 which is from bp 1056 to 1362 in the genomic DNA. This exon contains two stopcodons. One at bp 1180 in the genomic DNA and one at bp 1233 in the genomic DNA. FGENESH incorrectly predicts an exon from bp 1482-1739 in the genomic DNA. The genomic and cDNA sequences are not full length. The 3’ end of this gene is missing. This gene might be a pseudogene because of the stopcodons in the genomic DNA. Another possibility is that the errors in the genomic DNA are due to low coverage (IX) of this region in the maize genome. ZmCSLCIO The FGENESH prediction for the exon which is from bp 283 to 645 in the genomic DNA is incorrect. The end of this exon is at bp 645 instead of bp 666. The genomic and cDNA sequences are not full length. ZnCSLCI I The FGENESH prediction is incorrect for exon 1 and 2. The start of exon 1 is at bp 3209 in the genomic DNA instead of hp 3275. The end of exon 2 is at bp 4475 in the genomic DNA instead of hp 4600. A part of the genomic DNA of intron 2 is missing. The predicted start site of exon 3 is incorrect. The start site is at bp 543 instead of bp 495. FGENESH predicts two extra exons at the 3’ end of the gene. This gene ends at bp 1857 in the genomic DNA. The genomic DNA is not full length, but the cDNA sequence is. ZmCSLCIZ FGENESH incorrectly predicts the end of the exon encoded by this genomic DNA at bp 890 instead of bp 896. The genomic and cDNA sequences are not full length. ZmCSLDI The FGENESH prediction for the start site of exon 2 is incorrect. The start site of exon 2 is at bp 2004 in instead of hp 1980 in the genomic DNA. The genomic DNA and cDNA sequences are full length. 138 Table 9 Continued ZmCSLDZ The FGENESH prediction is incorrect for the end of exon 1. The end of exon 1 is at bp 1520 in the genomic DNA instead of hp 1523. Part of the genomic DNA of exon 2 and intron 1 is missing. The FGENESH prediction is incorrect for the end of exon 2. This exon ends at bp 460 instead of bp 232 in genomic DNA ZmCSLDZ.2. Part of the genomic DNA of exon 3 and intron 3 is missing. The cDNA for this gene is full length. ZJnCSLD3 The FGENESH prediction is correct. Part of the 5’ end of the genomic and cDNA sequences are missing. ZmCSLD4 The FGENESH prediction is correct. The genomic and cDNA sequences are full length. ZmCSLD5 The FGENESH prediction is correct. The genomic and cDNA sequences are full length. ZmCSLE] The FGENESH prediction is correct. The cDN A sequence is full length, but part of genomic DNA of the middle part of the gene is missing. Genomic DNA ZmCSLE 1. I has a deletion of a thymidine base at bp 2748 which causes a frameshift in exon 3. This deletion is not present in the ESTs. Genomic DNA ZnCSLEIJ contains a stopcodon at bp 300 due to a basepair change from A to G. This stopcodon is not present in the ESTs. ZmCSLEZ FGENESH incorrectly predicts the end of the exon which is from bp 647 to 685 in the genomic DNA at bp 647 instead of 749. The genomic and cDNA sequences are not full length. There is a stopcodon in the genomic DNA at bp 724. This might be a pseudogene. ZmCSLE3 FGENESH is correct. The genomic and cDNA sequences are full length. 139 Table 9 Continued ZmCSLE4 The FGENESH prediction is incorrect for the exon which is from hp 3325 to 3464 in the genomic DNA. The start site of this exon is at bp 3325 instead of bp 3327. At bp 3303 there is a stopcodon in the genomic DNA. This might be a pseudogene. The genomic and cDNA sequences are not full length. ZmCSLF] FGENESH is correct. The genomic and cDNA sequences are full length. ZJnCSLF2 FGENESH is correct. The genomic and cDNA sequences are full length. ZmCSLF] These ESTs align with both CSLF 1 and CSLF2. and ZmCSLFZ ZmCSLF3 The predicted start site of the exon which is from bp 834-1502 in the genomic DNA is incorrect. The correct start site it at bp 759. The end of the exon which is from bp 1580-2317 in the genomic DNA is incorrect. The genomic and cDNA sequences for the end of this exon are missing. The genomic and cDNA sequences are not full length. ZnCSLF 4 The FGENESH prediction is incorrect for exon 2 which is predicted to end at hp 1882, instead the correct end is at bp 1956 in the genomic DNA. FGENESH incorrectly predicts an exon from bp 2819-2877. The genomic and cDNA sequences are full length. ZmCSLF5 FGENESH is correct. Part of the 5‘ genomic and cDNA sequence is missing. ZmCSLF 6 FGENESH is correct. The genomic and cDNA sequences are full length. ZmCSLF7 FGENESH is correct. The genomic and cDN A sequences are full length. ZmCSLH FGENESH is correct. The genomic DNA is of a piece of exon 9 is missing, the cDNA sequence is full length. ZmCSLX FGENESH is correct. The genomic and cDNA sequences are full length. 140 Table 10: ESTs found for each CSL in GenBank and MaizeSeq as of November 2006 Maize gene ESTs ZmCSLA] AYl 11707, BE509763, BE639162, CB179595, CF637647, CX054069, DN219895, 13 ESTs EB819007EC877128, EC901950 , MRT4577_113780, 61012.1, 112651.] ZmCSLAZ AI712277, BG267241, CO459150, C0529865, EB822275, MRT4577_160794, 11 ESTs MRT4577_178541, 40199, 92880, 395289 ZmCSLA3 BE345572, 31543108, 81542892, B1502667, BM337360, CB885661, CB605095, 85 ESTs CD986535, CD986776, CF623308, CF647714, CF646849, CK827009, C0534845. C0521720, DN203737, DN210538, DN220659, DR785175, DR793529, DR796360, DR802009, DR810267, DR819129, DR824882, DR827942, DR830824, DR959515, DR960996, DR962628, DR972154, DT645077, DT652063, DT944501, DT947934, DVO29937, DV032430, DV508950, DV521309, DV522106, DV528686, DV530970. DV533083, DY232094, DY236276, DY618629, DY686234, DY689456, DY689950. EB403285, EB403767, EB404299, EB405782, EB676269, EB701629, EB705890, EB816589, EC875141, EC882568, EC885731, EC886357, EC897088, EE017356, EE036839, EE040917, EE041738, EE164658, EE16S758, EE167563, EE185452, EE187985, EE190120, EE287464, EE291877, MRT4577_29888C.1, 46987.2, 12762.2, 3447.3, 220844, 306862, 318020, 44615.1, 46986.1, 73611.4, 129440.] ZmCSLA4 CF650426, MRT4577_170140C.1, MRT4577_122156C.1, 112873.1 4 ESTs ZmCSLAS DR808454, EE156170, MRT4577_95044C.1 3 ESTs ZmCSLA6 AI966916, AI629471, AWl46710, BM331755, BM337273, BM339419, CO439336, 17 ESTs DN206568, DR804900, DR959523, DR960322, DV028649, DV491468, DV512882. MRT4577_44468C. 1, 6799.1, 148484. 1 141 Table 10 Continued ZnCSLA7 BM080245, CF040341, CO448087, CO451898, DN225331, DR970075, DV534833. 15 ESTs EC885334, EC885995, EC891529, EE160654, MRT4577_40487C.1. MRT4577_40488C.1, 939.1, 1369031 ZmCSL/l8 DR972654, DT645477, DT646829, DY537262, DY623006, EB642842, EE019504, ll ESTs EE188078, EE287562, MRT4577_105165C.1, 69248.1 ZmCSLA9 CO441048, CO454463, MRT4577_167180C.1, 55086.1, 55087.2, 125237.l 6 ESTs ZmCSLA 10 CK368643, CK827890, C0526201, MRT4577_145084C.1, MRT4577_152087C.1, 7 ESTs 324543, 35933.1 ZmCSLA] I 0 ESTs ZmCSLC] AI491631, AI782918, AI967143, AI999860, BM349218, CB351253, CF004102, 23 ESTs DR806804, DV020708, DV520752, DV523387, DV942945, DW961997, EE177147. EE287163, EE289125, MRT4577_280C.1, MRT4577_74050C.1, 15495.1, 503149. 51996.2, 51997.1,116485.1 ZmCSLC2 CF637490, CK371550, EEO42682 3 ESTs ZmCSLC3 CA828196, BQ279719, DN204842, EE025542, MRT4577_77404C.1 6 ESTs ZnCSLC4 BU092694, BU092832, BQ487075, CA831369, EC892727, EE020946, EE175501. 11 ESTs DY402635, MRT4577_177513C.1, 36147.1, 1167631 142 Table 10 Continued ZmCSLC5 90ESTs AW134434, BF727705, BM498825, BQ578082, BQ048622, BT018527, CD440675, CD443934, CFO44437, CF650782, CK144562, CO441868, CO457702. C0523572, COS27804, C0528077, COS31797, DR791573, DR791826. DR798460, DR805276, DR822059, DR825906, DR961560, DR970669. DT649058, DT941621, DVO23358, DV024073, DV031784, DV165615, DV168144, DV170219, DV170453, DV170486, DV171001, DV508366. DV511482, DV529562, DV530255, DV536340, DV549730, DY535126, DY538898, DY539559, DY690607, EB163654, EB400980, EB406179, EB674364. EB676456, EB702758, EC874460, EC881320, EC881465, EC883184, EC893104. EC899426, EC901681, EC903967, EC904132, EE013636, EE018349, EE019352. EE020315, EE021227, EE042191, EE044776, EE046495, EE153400, EE156704. EE156726, EE159137, EE164327, EE172128, EE174453, EE176046, EE285929. EE291585, EE294171,MRT4577__41982C.1, MRT4577_53439C. l, MRT4577_53441C.1, MRT4577_99427C.1, 31567.2, 45744.1, 64532.2, 76758.3. 145181.],152721.1 ZmCSLC6 67 ESTs CD986188, CF650598, CF919899, CF649783, C0526243, C0532188, CO461994. DR787544, DR797174, DR791495, DR802201, DR809519, DR812129. DR813649, DR829637, DR960832, DT942330, DT653933, DT654270, DV170447, DV503853, DV506512, DV514162, DV514533, DV514967, DV534763, EB158905, EBl64184, EB402572, EB673906, EB813228, EB821678. EC874262, EC874577, EC875492, EC876358, EC877605, EC878679, EC889570. EC889068, EC894672, EE010659, EE012971, EE018864, EE024111, EE024645, EE026179, EE033570, EE033766, EE033832, EE033894, EE036521, EE040400. EE043759, EE156566, EE162034, EE165420, EE168545, EE170986, EE171280, EE178640, EE179176, EE288312, EE291735, MRT4577_112624, 61530.]. 139048.] 143 Table 10 Continued ZmCSLC 7 DV033286 1 EST ZmCSLC8 0 ESTs ZmCSLC9 0 ESTs ZnCSLCIO DR969222 ZmCSLC] l CF635266, EE022110, EE186486, EE186309, MRT4577_144971C.1 5 ESTs ZJnCSLCIZ 0 ESTs ZJnCSLDI C0533241, DR960265, DR828710, DR789710, DR786107, DT649020 , DT945255, DV164427, DV165563, DV025799, DV527723, DW467865. EB70160, 133818933, EC899381, EE024752, EE038233, EC887069. MRT4577_16017C.1, MRT4577_40058C.1, MRT4577_168633C.1, 22837.1. 31393.1, 39225.1,11716l.l ZmCSLDZ AI657474, CA404833, CD439402, CD445140, CD940648, CD957767, CD997962. 27 ESTs CF024310, CK348127, DN225988, DN 229925, DV495290, DV520872, DY622790, EB158461, EB167441, EE021092, EE030414, MRT4577_37569C. 1, MRT4577_62600C.1, MRT4577_83752C.1, 207 88. 1, 28076.1, 35884.1, 49466.1. 139902.], 148928.] ZmCSLD3 MRT4577_23057C.1, 38389.1 2 ESTs ZmCSLD4 DR970587, DV534771, DY621894, EB637344, EE179918, MRT4577_24124C.1. 9 ESTs MRT4577_78194C.1, 38192.1, 118619.] 144 Table 10 Continued ZmCSLD5 AI857200, CK828171, DV491157, EE679307, 87531.1, 92873.1 6 ESTs ZmCSLE] BE761748, BM073901, DN209030, CO441675, CF051386, CD964407, 11 ESTs CD964287, DY539742, EC881460, MRT4577_63661C.1, 38622.1 ZmCSLEZ 0 ESTs ZmCSLE3 CF627793, DR819529, DR964066, EC893493, EE011424, EE016689, EE017015, l4 ESTs EE163100, EE170793, EE286791, MRT4577_144142C.1, 38833.1, 66548.2, 153926.] ZmCSLE4 0 ESTs 145 Table 10 Continued ZmCSLF] 153 ESTs AI783230, AI795546, AI973329, AI999933, AW017656, AWO42388, AW146811. BG265599, BM268865, BM333387, BM335019, BM335870, BM338286. BM347364, BM347399, BM347728, BM348968, BM349583, CD439674. CF647371, CO447585, C0520666, C0520904, COS29267, DN215438. DR821260, DR822182, DR826199, DR791112, DR791522, DR797652, DR796453, DR799904, DR799649, DR805715, DR814323, DR816712, DR817841, DR818247, DR820048, DR821261, DR822183, DR826200. DR954675, DR954860, DR956075, DR957779, DR969573, DT644307. DT649889, DT650208, DT652176, DT653968, DT938200, DT939993, DT941682. DT941786, DT944692, DV028925, DV029265, DV032475, DV167670. DV174682, DV504205, DV504499, DV504537, DV506225, DV506644, DV511477, DV512176, DV513987, DV519602, DV526195, DV527458. DV529942, DY530223, DV532277, DV534084, DV539057, DY235230. DY237207, DY237682, DY532292, DY53621 1, DY538305, DY619806, DY622103, DY624109, DY685856, DY690078, EBl61045, EB163591, EB164004, 133399654, EB406715, EB637507, EB640049, EB641560, EB675223. EB702598, EB816752, EB820888, EB822621, EC874350, EC877358, EC885025, EC888233, EC892638, EC893592, EC893771, EC898151, EC899578, EC899898. EE021776, EE022726, EE023779, EE023856, EE024627, EE026187, EE027359. EE038746, EE040298, EE041443, EE043103, EE044201, EE045406, EE047322. EE153371, EE153947, EEl69422, EE169470, EE173970, EE174283, EE174942. EE179257, EE185535, EElS8221, EE189572, EE18991 1, EE287777, EE287836. EE287876, EE681777, 1522.2, 1751.6, 209037, 48889.1, 63576.3, 63577.4. 67452.5, 69791.1, 76585.8, 125340.] 146 Table 10 Continued ZnCSLFZ 135 ESTs AI673968, AW065348, B1135345, BM173697, BM417109, BM417198. BU049237, BU049339, CF058879, CF245147, CF244964, CD439166, CD976130, CF245147, CF623166, CF638045, CF648617, CK368125, CN071362, C0522999. C0533529, C0530432, C0519007, DR789101, DR792651, DR796147. DR805904, DR810237, DR821815, DR822597, DR829412, DR958628, DR969118, DT651821, DT652126, DT939385, DT947848, DV020239. DV020568, DV025107, DV027989, DV029421, DV033990, DV163120. DV164891, DV168557, DV172188, DV502129, DV518408, DV523120. DV528761, DV530271, DY235522, DY238535, DY530979, DY532999, DY533783, DY538486, DY539155, DY540904, DY542917, DY621079, DY685480, EB166098, EB167134, EBl67397, EB401995, EB408246, EB637641, EB639474, EB701422, EB702456, EB705868, EB708252, EB819329, EC874832. EC877563, EC879225, EC879243, EC879391, EC879735, EC882382, EC889264, EC890410, EC894591, EC900348, EC900569, EC902928, EC903762, EE019913. EE020366, EEO21716, EEO27053, EEO27323, EE032940, EE033079, EE033159. EE033838, EE033977, EE040866, EE041414, EE041512, EE042092, EE043783, EE046599, EE153305, EE153672, EE153921, EE156356, EE158385, EE160408. EE162564, EE164474, EE165945, EE167050, EE168402, EE170861, EE176042, EE179789, EE183456, EE183688, EE185850, EE186177, EE186178, EE188636. EE189897, EE190081, EE285694, EE290853, EE293194, EE295301, 67453.6. MRT4577_132210C.1 ESTs which are similar between ZmCSLF] and ZmCSLFZ CF058879, EC898208, EE012676, EE017551, EE018725, EEO21716, EE038868, EE047285, EE163229, EE189814, EE285619, EE286893, EE289400, EE293869, ZmCSLF3 0 ESTs 147 Table 10 Continued ZmCSLF4 EE172734, EE031146, MRT4577_91428C.1 3 EST 5 ZmCSLF5 EE292309, EC885590, EC886228, MRT4577_153693C. 1, MRT4577_160801C.1 5 EST 5 ZmCSLF6 EC893050, EE042523, MRT4577_108931C.1. 5 ESTs MRT4577_1855]2C.1, 24052.1 ZmCSLF7 EE014434, MRT4577_11227C.] 2 ESTs ZmCSLH] DV518628, EC891386, EE012543, EE018922, EE186369, MRT4577_12320C.1 8 ESTs MRT4577_16629C.1, MRT4577_119932C.1 ZmCSLX BM382035, BT017533, CA401527, CD443089, CD953565, CF648387. 20 EST s DT644979, DV519225, EB701720 , EB676365 , EC901471 , EC878414, EC877425, EE014664, EE048062, EE171665, EE289753, MRT4577_56702C. 1, 92886.1, 92889.1 148 APPENDIX 11: GENBANK LOCUS NUMBERS AND J GI PROTEIN IDS Table 11: GenBank locus number for Arabidopsis CSL proteins. Arabidopsis protein Locus AtCSLA2 ATSG22740 AtCSLA3 ATlG23480 AtCSLA7 AT2G35650 AtCSLA9 AT5G03760 AtCSLAIO ATlGZ407O AtCSLA] 1 ATSGl6190 AtCSLAl4 AT3GS6000 AtCSLA15 AT4G13410 AtCSLC4 AT3G28180 AtCSLCS AT4G3 1590 AtCSLC6 AT3G07330 AtCSLC8 AT2024630 AtCSLC12 AT4G07960 AtCSLBl AT2632610 AtCSLB2 AT2G32620 AtCSLB3 AT2G32530 AtCSLB4 AT2G32540 AtCSLBS AT4G15290 AtCSLB6 AT4G15320 AtCSLC4 AT3628180 AtCSLCS AT4GB 1590 AtCSLC6 AT3GO7330 AtCSIJC8 AT2G24630 AtCSLC12 AT4GO796O 149 Table 11 continued AtCSLDl ATZG33100 AtCSLD2 AT5G16910 AtCSLD3 AT3G03050 AtCSLD4 AT4G38190 AtCSLD5 AT1002730 AtCSLD6 AT1G32180 AtCSLEl AT1655850 AtCSLGl AT4GZ4010 AtCSLG2 AT4G24000 AtCSLG3 AT4G23990 Table 12: JGI protein IDs for the poplar CSL proteins Poplar protein Protein ID PtCSLAl 686549 PtCSLA2 687416 PtCSLA3 589559 PtCSLA4 594843 PtCSLA5 556940 PtCSLB] 572982 PtCSLBZ 684214 PtCSLCl 353078 PtCSLC2 578365 PtCSLC3 692569 PtCSLC4 694461 PtCSLCS 692052 PtCSLD] 552489 150 Fa... Table 12 continued PtCSLD2 700418 PtCSLD3 554065 PtCSLD4 590064 PtCSLDS 573858 PtCSLD6 703843 PtCSLD7 595034 PtCSLD8 78520 PtCSLD9 350683 PtCSLDlO 48556 PtCSLE] 550222 PtCSLE2 343986 PtCSLE3 560094 PtCSLGl 698018 PtCSLG2 554513 PtCSLG3 80778 PtCSLG4 350373 PtCSLGS 350372 Table 13: JGI protein IDs for the Physcomitrella CSL proteins Physcomr’trella protein Protein 1]) PpCSLAl DQ417756 PpCSLA2 DQ417757 PpCSLCl DQ898147 PpCSLC2 DQ898148 PpCSLC3 DQ898149 PpCSLC4 DQ898150 151 Table 13 continued PpCSLCS DQ898151 PpCSLC6 DQ898152 PpCSLDl DQ898147 PpCSLD2 DQ898148 PpCSLD3 DQ898149 PpCSLD4 DQ898150 PpCSLDS DQ898151 PpCSLD6 DQ898152 PpCSLD7 DQ898153 PpCSLD8 DQ898154 Table 14: GenBank locus number for the ZmCESA proteins ZmCESA Locus ZmCESA] AAF89961 ZmCESA2 AAF 89962 ZmCESA3 AAF 89963 ZmCESA4 AAF 89964 152 ff. APPENDIX III: GENE SPECIFIC PRIMERS FOR SEMI-QUANTITATIVE RT- PCR Table 15: Gene specific primers for the ZmCSLA genes Gene name Forward primer (5’ —-) 3’) Reverse primer (5’ -—> 3’) ZnCSIA 1 ATCCCGGAGCTTT ACCT ACCAGT AAGAACAATTGCCAGTGAAGT ZmCSlAZ CGTATATCATCGTAGAAGAATCCA CGCAGCCAGAGCCCCGTGACC ZmCSLA3 CCGTCTGCGCCTTCITTGGATTC GGTGATGGCCGAGGGGATGTAGAC ZmCSLA4 TGAACCAGTGAAGCCAACAGAATG AACCGCCCCCTACCCACAC ZnCSlAS TCCCAAAGAATTCCTGATGACAA ATGGGCGACAACCTTCCTAATG CATATCGCCGCCAACAACATC TCTACCAAGCTTCTCCGTGACAAC GAGGTGTCGCTGTGGAGGAA AGAAGAGGAAGATGGCAACT AAA A ZmCSLA6 GAGAGTGTCGCTGTGGAGTAAAAT ACCCAAAACCAACAACAAGGAACG ZmCSLA 7 ATTGGGGCGACTGTATGGAAGAA AGCGAAGGCCTGGAGGAAGATGTA GGCCAGCAAGAAAATCAACATAAA TGCCCGCACAGCCAAGTC TCGGTGAAGCAGGAGGATG ATTGGCACCAGATGGATAGACC ZnCSLA8 TTGCCCGGAGAATCGTAGG TGGGGAAGAACAAAGAGGTAAAC A GACCTGCCCI'CCTGTTCAAGA ATAAGTAGTCAAAGCACGCAGAGG GTGAATGCCAACGACTGCT AAGGTGCCTACGATTCT CC AGCTGGCTCTTTCTTATGC GCGGTACTGCTAGGACTGGT CTGCTGGAGTATGGAGAACG GCGGTACTGCTAGGACTGGT ZnCSLA9 CGCCGTGACGTTCGTGTTTTACF G ATAGCCAATGCCGACGATGAAGAA ZnCSLAI 0 GGAAGGTCGCTGCCCACACG GCCGACATACCCAAAGCCAACAAC GCTGCAGGGCTGGAAGTTTGTTT ACTCGTCCACCAGCATCCAGAAGA CTAGCACGTTCAAGGCATAC ACGAGGGGTTCTGTTAGC 153 Table 16: Gene specific primers for the ZmCSLC genes. Gene name Forward primer Reverse primer ZnCSLCI GACCT'I'I‘TCCGGCTGTGCT CATGCGAGGAATCAACI‘TATCTGT ZmCSLCZ No primers ZmCSLC3 TGCTTTGTGGATATTATCAAATCGA CCCCGATGAAACI ATATAACCCTT C A ZmCSLC4 GCTTIGTGGATATTATCAAGTCGAA CGTGTI'I‘GCGTTGAATCATAT G ZmCSLC5 AGGCATTGGAGGACTCAGGTGGAT TACGGGATGATGAAGGGGAACGAC ZmCSLC6 AA'I'I‘GT'ITAGGCTCI‘GCTITGTGG AACCCATTATI‘CAACTGCI‘ATCAAT CC ZmCSLC7 GTGCCT'I‘GGCFGCTTCTACATCC CCGGGTCAGCAGGTI‘CTCATC ZmCSLC8 GGCCATCCAGAAGCTGTCCAG CTTGAGGTTGCCGGCC'ITGTA ZmCSLC9 CAACCTTGACTGCCCGAAATC GAATCATGGCAGAGCAACAAAC'IT ATCCTATGCCTT GGCT ACT'I‘CI‘ AC ATGGGGTGTI‘ATCI'I‘GGCTTCC CGCCTACG'I'I‘GTGCTCT'I‘CCI‘ ATGGGGTGTTATCI‘I‘GGCTI‘CC ZmCSLC] 0 No primers ZmCSLCI 1 GCGGGCGGCTTCTGGGGTGTC ACGCGGGGATACGAGGCTTGATGC CGACAACGGTGCAGGAGAAC CCGCAGCAGGAAGAAGAGCAT CGACAACGGTGCAGGAGAAC ATGAGCACGGGGATGTAGCA CGACAACGGTGCAGGAGAAC GCAGAAGAGCGTGAAGGAGTAGA ZmCSLC] 2 No primers 154 Table 17: Gene specific primers for the ZmCSLD, E, F, H, X and ZmEFZagenes. n.d. means not determined Gene name Forward primer Reverse primer ZmCSLDI GGCTGCCGATGCI‘GGTGTA TTGTTCICCCTGTCCGTCTTCI‘TT ZmCSLD2 CGATACGCCGACGCT CT GA TGAAGGGGCCATITGACAT GGAGGAGGGTCAAGAGGGAGTATG GCGGAAAGCTTGCGAGTTGTAG ZmCSLD3 GCACGGCGCCCATCAACCTCA CGTACACGGCGCGGGACACA ZmCSLD4 No primers ZmCSLD5 GCGCCTGTCGCTGGTCA ATCCGACTI‘GCCI‘GTTGGATTGT ZmCSLE] GGATGGATGGGGTGGAATGTGTTA . CGGGCTGTACI‘TTGAGAGGGAGAT GGTGGAGCGAGAATGCAAGT’I'I'I‘A TTCCTCGCCACCCI’AGTCCAATCT ZJnCSLEZ No primers ZmCSLE3 TCACGGGGCTGGCGATACACI‘GC TCTGCCCGTTCCACCACCCT CI’ CA ZmCSLE4 n.d. ZmCSLF] No primers ZmCSLF? CGTTCACGGCGATCTTCCTCAT CCGCCGGCCACCITTAGC GGCTCGTGAACCCCGTCCCGTAAT GGAGAGGCCGTCGTCGCI‘GAGGTC CGCGCATCAACGGGCTGGAGAA CGCCTGCGAGTTGTTGATGTAGTG ZmCSLF 3 GGCACGGGCAGCATGAGA CAGCCCAGCAGCAGCACC ZmCSLF 4 GCGGCGCCGACGACGAGAG GCCGGGCAGCTTGTTGTTGGACI‘ CGCATGCGCAGGGAATACGAAGAG GCGGAAAGCCTGCGAGTI’GTI‘GAC TCCGGCGGCGAGTCCAACAA CCACGTCGCCTI’CATCACCATACC ZnCSLF 6 GCTGCGGGCCI’ CTGCI CT CCI‘ TTGGTGGCGGTGTGATTGTCCI‘ GT TGTGCGCIGGTCTGGTGGGTCTI'I’ GTCA'ITGTCGTCCGCGGCTGTTTG TGGCCCGGCACATGGATTGAT AGCTT‘CGGCCGTGATGTT‘ITCTGC ZmCSLF7 No primers ZmCSLF 8 GGAACATGGTTTGACCCI‘ GCT G CAAACTTGCTGGGGTTGTCC 155 TABLE 17 CONTINUED ZmCSLH CGAGTCGGCGAGGAGCATCATCA ACAGCGCCAGCGGGACGACGAACC ZmCSLX CGGCGACGGCGTTGAGGAGA'I'I‘ GCCGGGGCTGCGTAAGGATGC CGCGCCCTACGTGCTGGTCCTC GCGGTGCCGGCCI'CGTATGC ZmEFZa TGGCCAGACCCGTGAGCAT ACGGACAGCAAATCGACCAAGAG APPENDIX IV: GENE SPECIFIC PRIMERS FOR REAL-TIME QUANTITATIVE RT-PCR Table 18: Gene specific primers for the ZmCSLA, C, D, E, F and EFla genes Gene name Forward primer (5’ —> 3’) Reverse primer (5’ —) 3’) ZmCSLA I ATCCCGGAGCT'ITACCIACCAGT AAGAACAATI’GCCAGTGAAGT ZmCSLA3 CTCGGCCATCACCCTCCICAACI‘ CCCAAGACACAATGCGAAAACCAG C ZmCSLA4 TGAACCAGTGAAGCCAACAGAAT AACCGCCCCCTACCCACAC G ZmCSLAS TCCCAAAGAATTCCTGATGACAA ATGGGCGACAACC'ITCCTAATG ZmCSLA6 GAGAGTGTCGCTGTGGAGTAAAA ACCCAAAACCAACAACAAGGAACG T ZmCSL/17 TCGGTGAAGCAGGAGGATG ATTGGCACCAGATGGATAGACC ZmCSL/18 TTGCCCGGAGAATCGTAGG TGGGGAAGAACAAAGAGGTAAACA ZmCSL/19 CGCCGTGACGTTCGTGTTTTACIG ATAGCCAATGCCGACGATGAAGAA ZmCSLA 10 GCTGCAGGGCTGGAAGT'I‘TGTTT ACTCGTCCACCAGCATCCAGAAGA ZmCSLC] TATCGGCATCGGGTGTTGAGGAC AGAATGCCACCGGTGTTGCIGTTT ZmCSLC3 CCCCGATGAAACI’ATATAACCCI‘ CCCCGATGAAACTATATAACCCI‘T C TC ZmCSLC4 CGTGT'ITGCGTTGAATCATAT CGTGTTTGCGTTGAATCATAT 156 Table 18 continued ZmC SLC 6 AACCCATTA'ITCAACTGCTATCA AACCCATTAT'I‘CAACTGCI‘ATCAATC ATCC C ZmCSLC 7 GTGCCTTGGCTGCTTCTACATCC CCGGGTCAGCAGGTTCTCATC ZmCSLDZ CGATACGCCGACGCT CT GA TGAAGGGGCCATTTGACAT ZmCSLE] GGATGGATGGGGTGGAATGTGTT CGGGCTGTACI'TTGAGAGGGAGAT A ZmCSLF 6 GCTGCGGGCCT CT GCT CI‘ CCT TT‘GGTGGCGGTGTGATTGTCCT GT ZnCSLF 8 CAAACTTGCTGGGGTTGTCC CAAACTTGCI‘GGGGTI‘GTCC 211115 F 1 a TGCGGAGCTCATTACCAAGATT GCTCACCAGATGT'I‘CGGATAAGTC 157 AAAAAAAAAAAAAAAAA 1111111111111111111111111111111? 3 1293 02956 3313 ll