Molecular Plant Advance Access originally published online on August 24, 2009
Molecular Plant 2009 2(5):1025-1039; doi:10.1093/mp/ssp064
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The CELLULOSE-SYNTHASE LIKE C (CSLC) Family of Barley Includes Members that Are Integral Membrane Proteins Targeted to the Plasma Membrane
a Plant Cell Biology Research Centre, School of Botany, University of Melbourne Victoria 3010, Australia
b Australian Centre for Plant Functional Genomics, School of Agriculture and Wine, University of Adelaide, South Australia 5064, Australia
c Australian Centre for Plant Functional Genomics, School of Botany, University of Melbourne Victoria 3010 Australia
d Present address: Department of Biology, Institut Teknologi Bandung, Bandung, Indonesia
e Present address: Stem Cell and Cancer Institute, Jl. Jend. Ahmad Yani No.2, Pulo Mas, Jakarta 13210, Indonesia
1 To whom correspondence should be addressed. E-mail edwardjn{at}unimelb.edu.au, fax 61-3-9347-1071, tel. 61-3-8344-4871.
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
The CELLULOSE SYNTHASE-LIKE C (CSLC) family is an ancient lineage within the CELLULOSE SYNTHASE/CELLULOSE SYNTHASE-LIKE (CESA/CSL) polysaccharide synthase superfamily that is thought to have arisen before the divergence of mosses and vascular plants. As studies in the flowering plant Arabidopsis have suggested synthesis of the (1,4)-β-glucan backbone of xyloglucan (XyG), a wall polysaccharide that tethers adjacent cellulose microfibrils to each other, as a probable function for the CSLCs, CSLC function was investigated in barley (Hordeum vulgare L.), a species with low amounts of XyG in its walls. Four barley CSLC genes were identified (designated HvCSLC1–4). Phylogenetic analysis reveals three well supported clades of CSLCs in flowering plants, with barley having representatives in two of these clades. The four barley CSLCs were expressed in various tissues, with in situ PCR detecting transcripts in all cell types of the coleoptile and root, including cells with primary and secondary cell walls. Co-expression analysis showed that HvCSLC3 was coordinately expressed with putative XyG xylosyltransferase genes. Both immuno-EM and membrane fractionation showed that HvCSLC2 was located in the plasma membrane of barley suspension-cultured cells and was not in internal membranes such as endoplasmic reticulum or Golgi apparatus. Based on our current knowledge of the sub-cellular locations of polysaccharide synthesis, we conclude that the CSLC family probably contains more than one type of polysaccharide synthase.
Key Words: Cellulose synthase-like family C • plant cell wall biosynthesis • xyloglucan • cellulose • glycosyltransferase
Received for publication May 13, 2009. Accepted for publication July 13, 2009.
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Plant cell walls are highly organized composites that consist of polysaccharides and provide plants with a skeletal framework and the intercellular cohesion necessary for structural integrity (Bacic et al., 1988; Carpita and Gibeaut, 1993; Somerville, 2006). Cell walls are also highly dynamic and complex structures that give rigidity to the plant overall while providing the flexibility needed during the processes of cell expansion and differentiation. In addition, walls form a physical barrier to plant pathogens but still allow nutrients, gases, and various intercellular signals to reach the plasma membrane.
Identifying the genes and enzymes responsible for synthesizing cell wall polysaccharides is a major activity in plant research (Farrokhi et al., 2006; Lerouxel et al., 2006; Scheible and Pauly, 2004; Somerville et al., 2004). Synthesis of most β-linked cell wall polysaccharides requires the activity of a processive-type glycosyltransferase (GT) to produce the linear backbone, with current evidence suggesting that these enzymes are encoded by genes from one of two large gene families, the Cellulose Synthase/Cellulose Synthase-Like (CESA/CSL) gene family and the Glucan Synthase-Like (GSL) gene family. The CESAs code for proteins that make the (1,4)-β-glucan, cellulose, and the GSLs for proteins that make callose, a (1,3)-β-glucan (Brownfield et al., 2007; Delmer, 1999; Li et al., 2003). The CSLs are currently subdivided into nine families that are designated CSLA to CSLH and CSLJ (Fincher, 2009; Hazen et al., 2002). Consistent with the suggestion that the CSLs are processive GTs, heterologous expression studies have shown that proteins from the CSLA, CSLC, CSLF, and CSLH families are able to make the β-linked backbones for various non-cellulosic polysaccharides (heteromannans, XyGs and (1,3;1,4)-β-glucans, respectively) found in primary walls (Burton et al., 2008, 2006; Cocuron et al., 2007; Dhugga et al., 2004; Doblin et al., 2009; Liepman et al., 2007, 2005; Suzuki et al., 2006). The functions of the other CSL families are unknown, although the CSLDs are suggested to be involved in the synthesis of a non-crystalline form of cellulose (Bernal et al., 2007; Doblin et al., 2001; Manfield et al., 2004).
The proposed function of the CSLCs is synthesizing the XyG backbone (Cocuron et al., 2007). XyG is a major class of wall polysaccharide present in the primary walls of most land plants and its backbone is composed of (1,4)-β-D-glucosyl residues to which 
-D-xylosyl residues and other sugars are attached. Cocuron et al. (2007) found that long and short chains of (1,4)-β-glucan accumulated in Pichia pastoris cells that were co-expressing Arabidopsis CSLC4 (AtCSLC4) and Arabidopsis XXT1 (AtXXT1), the latter being a xylosyl transferase that adds the first side-chain xylosyl residue onto XyG (Cavalier and Keegstra, 2006; Cavalier et al., 2008; Faik et al., 2002). Although no detectable xylose was present on the β-glucan chains, this was presumably because yeast cells lack UDP-xylose, which is the AtXXT1 substrate. Since the yeast cells produced an unbranched β-glucan, it was possible that AtCSLC4 was involved in synthesizing either cellulose or the XyG backbone. However, Cocuron et al. (2007) argued that other evidence pointed most strongly to a role in XyG biosynthesis. This evidence included the highly correlated expression of AtCSLC4 and AtXXT1 and the likely targeting of stably expressed AtCSLC4 to the Golgi of BY-2 tobacco cells. The presence of AtCSLC4 in Golgi is consistent with a role in synthesizing XyG, which is believed to be synthesized in this compartment (Cosgrove, 2005). Cellulose synthesis, on the other hand, occurs at the plasma membrane (Delmer, 1999).
Although CSLC genes are found in all flowering plant genomes, XyGs are not abundant components of all flowering plant cell walls. In particular, the commelinoid monocots, of which the Poaceae or grass family is the best studied example, have a primary cell wall that is characterized by relatively low levels of XyGs (Carpita, 1996; Fry, 1989; Harris et al., 1997; Hayashi, 1989; O'Neill and York, 2003). The amount of XyG in Poaceae walls is generally lower (1–5% of wall dry weight) than that in non-commelinoid monocots, gymnosperms, and eudicots, which have cell walls containing 10–20% XyG. In commelinoid monocots, glucuronoarabinoxylans (GAXs) are proposed to be the principal polymers interlocking cellulose microfibrils (Bacic et al., 1988; Carpita and McCann, 2000; Smith and Harris, 1999).
The CSLC genes of barley (Hordeum vulgare) were studied in order to understand the function of CSLCs in the Poaceae, where the amount of XyG in walls is low. By a combination of bioinformatic searches and gene-cloning, four barley CSLCs were identified (HvCSLC1–4). As all the barley CSLC ESTs identified to date are derived from these four genes, HvCSLC1–4 represent the most actively transcribed members of this family in the barley genome. CSLCs were expressed in cells with primary and secondary walls and at various stages of the barley lifecycle. In sub-cellular fractions of barley suspension culture cell membranes, HvCSLC2 preferentially partitioned into fractions enriched in plasma membrane and was barely detectable in other membrane-containing fractions. HvCSLC2 was also detected in the plasma membrane by immuno-electron microscopy (immuno-EM) with an antibody raised against HvCSLC2. These findings are discussed in the context of the diversity that exists within the CSLC family and the possibility that it contains more than one type of polysaccharide synthase.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Cloning and Preliminary Characterization of Barley CSLC cDNAs and Genes
Barley CSLC genes were identified by iterative searching of cDNA and BAC libraries with CSLC-derived gene probes and databases with CSLC EST sequences. Library searching began with the sequence of a barley pre-anthesis spike EST (accession no. BE455720) that encodes a putative CSLC. Primers to this sequence were used to amplify a fragment from barley pre-anthesis spike cDNA, and this was used to screen a barley (cv. Schooner) cell suspension culture cDNA library. From this screen, cDNAs for two different CSLC genes (designated HvCSLC1 and HvCSLC2) were identified. Sequence analysis indicated that the HvCSLC1 cDNA was full-length (2.8 kb) but that about 1.7 kb of sequence was missing from the 5’ end of the HvCSLC2 cDNA. An EST (accession no. AV836446) that overlapped the HvCSLC2 cDNA was used to extend the sequence of this gene in the 5’ direction. The continuity of these two sequences was confirmed by amplifying an overlapping DNA fragment from suspension-cultured cell cDNA.
A fragment amplified from the HvCSLC1 cDNA was used to screen a barley (cv. Morex) BAC library. After screening over 184 000 colonies (estimated to be roughly a three-fold coverage of the barley genome), 14 positive BAC clones were identified that sequence analysis showed included the two HvCSLC genes already identified, as well as two new HvCSLC genes (designated HvCSLC3 and HvCSLC4). A 2.9-kb BAC fragment that included almost all of HvCSLC3 except for 
500 bp from the 5’ end of the open reading frame, and a 1-kb BAC fragment that contained the central portion of HvCSLC4, were sequenced. A near full-length HvCSLC4 was produced using ESTs that extended the sequence in both the 5’ and 3’ directions. Most of the intron/exon boundaries predicted by FGENESH (www.softberry.com) in the HvCSLC1 and HvCSLC4 genomic sequences were confirmed from EST data.
A schematic representation of the four polypeptides encoded by the HvCSLC genes is shown in Figure 1. The sequences for HvCSLC1–4 have been deposited in GenBank with accession numbers GQ386981 [GenBank] to GQ386984. All 50 barley CSLC ESTs in GenBank (as of April 2009) are derived from these four genes (Supplemental Table 1). HvCSLC1 is predicted to encode a 698-amino-acid polypeptide with a molecular weight of 78.2 kDa and six transmembrane domains (two at the NH2-terminus and four at the COOH-terminus). Amongst the rice and Arabidopsis CSLC proteins, HvCSLC1 is most similar to OsCSLC7 (86.2% identity, 92.1% similarity) and AtCSLC12 (66.8% identity, 80.0% similarity). The partial sequences of HvCSLC2, HvCSLC3, and HvCSLC4 are predicted to encode proteins of 535, 597, and 530 amino acids, respectively, that have the same predicted membrane topology as HvCSLC1, as deduced by comparison to their putative rice orthologs OsCSLC9, OsCSLC3, and OsCSLC1 (Figure 2). All four genes encode proteins with the D,D,D,QQHRW motif within homology (H) domains 1–3. This motif is also found in all the CSLCs from rice (Oryza sativa), poplar (Populus trichocarpa), the moss Physcomitrella patens, and in the majority of CSLCs from sorghum (Sorghum bicolor), grapevine (Vitis vinifera), and four of the five Arabidopsis CSLCs (AtCSLC6 has QQYRW instead).
|
|
Figure 2 shows a Neighbour Joining tree of the CSLC family as currently proscribed (Hazen et al., 2002; Richmond and Somerville, 2000; Roberts and Bushoven, 2007). This tree was produced from an alignment of the four barley CSLC polypeptides and full-length CSLC sequences from a number of monocots and eudicots, the moss Physcomitrella, the lycophyte Selaginella moellendorffii, and the green alga Chara globularis (Supplemental Figure 1). Trees with similar sequence groupings were produced when other phylogenetic methods were used and when only the H1-3 regions were used (data not shown). The CSLCs receive strong bootstrap support as a separate clade within the CESA/CSL superfamily and the most divergent member of this group, the CSLC from Chara, was chosen as the root of the tree shown in Figure 2. Tree topology is characterized by four well supported groups (clades I–IV), with three of these groups (clades I–III) being part of a polytomy. All the CSLCs from Physcomitrella and Selaginella are in clade III, which suggests taxonomic relationships are one source of tree structure. However, clades I and II contain monocot and eudicot CSLCs, suggesting functional specialization as another possible source of tree structure. HvCSLC1, HvCSLC2, and HvCSLC4 are in clade I and HvCSLC3 is in clade II. Clade IV contains AtCSLC6 and CSLCs from grapevine, Medicago truncatula, and poplar.
HvCSLC Transcript Levels and Correlations with Other Genes
Quantitative RT–PCR (QPCR) was used to determine the pattern and transcript level of barley CSLCs. The normalized transcript levels for each HvCSLC across a range of barley tissues and suspension-cultured cells are shown in Figure 3. In nine of the 12 tested tissues, HvCSLC2 transcript levels were either the highest or equal highest of the four genes, in agreement with the high level of representation of HvCSLC2 sequences among barley CSLC ESTs (Supplemental Table 1). In coleoptile, HvCSLC1 and HvCSLC4 transcript levels were higher than those of HvCSLC2 and, in root tip, HvCSLC1 and HvCSLC3 transcript levels were higher (Figure 3A). Root tip was the only tissue to accumulate significant levels of HvCSLC3 transcript, although low levels of this transcript were present in several other tissues.
|
The four HvCSLC genes are present on the Affymetrix 22K Barley1 GeneChip with the sequence IDs listed in Supplemental Table 2. The Barley1 microarray contains at least 21 439 genes and has been used in a number of experiments for which the datasets are publically available through both BarleyBase/PLEXdb (Shen et al., 2005; Wise et al., 2007) and ArrayExpress (Parkinson et al., 2006). In an experiment in which various vegetative and floral tissues were sampled across plant development in two barley cultivars (Druka et al., 2006), the expression pattern of the HvCSLCs in common tissues is generally consistent with those found using QPCR (Supplemental Figure 2A). For example, HvCSLC2 transcripts are present at relatively high levels in most tissues, whereas HvCSLC3 transcripts are generally low except in root tips.
Correlations were sought between the transcript profiles of individual HvCSLCs and those of other genes on the array (Figure 4). As a standard by which to assess the significance of these correlations, the lowest pairwise correlation coefficient (0.89) of the transcript profiles of the three barley primary wall CESAs (HvCESA1, HvCESA2, and HvCESA6) was used, as it is known that their expression is highly correlated (Burton et al., 2004; Supplemental Figure 2C). By this criterion, none of the HvCSLC genes showed significant co-expression with another CSLC or with another member of the CESA/CSL superfamily, with the most highly correlated pair being HvCSLC1 and HvCSLC3 (0.77). Supplemental Table 3 lists the top 20 correlations for each HvCSLC gene.
|
Part of the evidence linking the CSLCs to XyG production is the coordinate expression of AtCSLC4 and AtXXT1 (Cocuron et al., 2007). To determine whether the barley CSLCs were coordinately expressed with barley orthologs of AtXXT1, putative barley homologs were identified by iterative database searches. This yielded a total of 39 ESTs (Supplemental Table 4) that sequence alignments showed represented the partial sequences of five genes that we provisionally named HvGT1–5 (for H. vulgare glycosyl transferase). HvGT1 has the longest sequence and encodes a protein of 296 amino acids that covers the COOH-terminal half of AtXXT1 (Supplemental Figure 3). HvGT1 is also the most closely related of the five HvGTs to AtXXT1 and AtXXT2 (82.6 and 84.9% amino acid identity, respectively). All five HvGT genes are present on the Barley1 microarray and the contig IDs for these genes are listed in Supplemental Table 2. Figure 4 shows where the five HvGT genes occur in the correlation distributions of each HvCSLC gene. The transcript profiles of HvGT3 and HvCSLC3 were highly correlated (correlation coefficient = 0.99), as transcripts for both predominantly accumulate in root tips (Supplemental Figure 2A and 2B). The next highest correlation coefficient, 0.82, for the transcript profiles of HvGT1 and HvCSLC1, was below the level considered significant. The transcript profiles of 83 other genes were more highly correlated to HvCSLC1 than HvGT1.
To study the cell-type specificity of HvCSLC expression, in situ RT–PCR (Koltai and Bird, 2000) was carried out on barley coleoptiles and root tips that had been harvested 3 d after germination. As expected, 18S rRNA transcripts (positive control) were detectable in most root and coleoptile cells (Figure 5C and 5F) and no signal was detected when primers were omitted (Figure 5B and 5E). In root and coleoptile, HvCSLC1 labeling was seen in all cell types and was particularly apparent in the vascular bundles (Figure 5A and 5D), indicating that transcripts for this gene accumulate in cells with a primary wall as well as cells with a secondary wall. While less labeling was seen in cortical cells, this was probably due to their low content of cytoplasm. Similar results were obtained for the other three CSLC genes, with signal strength reflecting QPCR transcript levels (Supplemental Figure 4).
|
Transient Xyloglucan Deposition during Early Endosperm Development
Figure 3B shows the normalized expression of three HvCSLC genes (HvCSLC1, 2, and 4) during early endosperm development. HvCSLC3 transcripts were barely detectable in this tissue. HvCSLC4 transcript levels were the highest and, early in development, were well above those of any barley CSLC gene in vegetative tissues. HvCSLC4 transcript levels were relatively constant up to 5 d after pollination (DAP) and declined thereafter. HvCSLC1 and HvCSLC2 transcript levels were lower than HvCSLC4 levels at 2 DAP and were undetectable by 6 DAP.
Detecting CSLC expression in early stages of endosperm development raised questions about their proposed role in XyG synthesis, as XyG has not been detected by chemical analysis in the walls of mature barley endosperm cells (Bacic and Stone, 1981; Fincher, 1975). Immuno-EM with the LM15 monoclonal antibody (Marcus et al., 2008) was used to determine whether XyG was present in immature endosperm walls. The LM15 antibody was used because it recognizes a non-fucosylated XyG-derived oligosaccharide, whereas other antibodies, CCRC-M1 for example, bind to a fucosylated XyG epitope that is not present on barley XyG (Gibeaut et al., 2005; Puhlmann et al., 1994).
To confirm that LM15 could be used to detect XyG in barley, thin sections of 3-day-old barley coleoptiles, which are known to have XyG in their walls (Gibeaut et al., 2005), were examined. LM15 labeling was evident in the thin primary walls of cortical cells and thick secondary walls of vascular cells (Figure 6A and 6B, respectively). Labeling was abolished in coleoptile sections pre-incubated with either the non-fucosylated XyG from Nicotiana plumbaginifolia suspension-cultured cells (Figure 6C and 6D) or Tamarindus indica (tamarind) seed (data not shown; Sims and Bacic, 1995; Sims et al., 1996; York et al., 1990). Labeling was also unchanged when sections were co-incubated with two derivatives of cellulose (microcrystalline cellulose, carboxymethyl cellulose), cellohexaose (a cellodextrin), barley flour (1,3;1,4)-β-glucan, and laminarin (Supplemental Figure 5A–5E, respectively). No labeling was detected when the LM15 antibody was omitted (Supplemental Figure 5F) or when the CCRC-M1 antibody was used (data not shown).
|
At 4 DAP in barley endosperm, cellularization is taking place and the first anticlinal cell walls can be seen growing out from the central cell wall between nuclei of the multinucleate syncytium (Figure 6E; Wilson et al., 2006). Light LM15 labeling was seen in the first-formed anticlinal walls, periclinal walls, and in the wall of the central cell, as well as in the walls of the surrounding maternal tissues (Supplemental Figure 6A and 6B, respectively). By 8 DAP, the endosperm is fully cellularized (Wilson et al., 2006) and LM15 labeling was no longer evident, although a significantly reduced level persisted in the walls of the maternal cells, which served as an internal positive control for antibody labeling (Supplemental Figure 6D).
An HvCSLC2 Antibody Detects a Protein in the Plasma Membrane
To determine the sub-cellular location of the CSLCs, a rabbit antiserum was raised to a peptide covering amino acids 411–466 of HvCSLC2 (Supplemental Figure 7). This region was chosen because it lacks sequence similarity to other members of the CESA/CSL superfamily. It is, however, possible that this antiserum detects CSLCs other than HvCSLC2. In preliminary experiments, the anti-HvCSLC2 antiserum recognized a single protein band of 80 kDa in Western blots of a detergent-soluble, mixed-membrane (MM) fraction from coleoptiles, which is consistent with the expected size of a CSLC protein (data not shown). To determine which membrane type contains HvCSLC2, MM (125 000 g pellet) from barley suspension-cultured cells were fractionated by PEG/DEX two-phase partitioning (Larsson et al., 1987) into a fraction enriched in plasma membrane (PM) (the PEG phase) and a second fraction containing other membrane types and some PM (the DEX phase). The degree of membrane enrichment in each fraction (homogenate, MM, PEG, and DEX) was assessed by Western blot analysis using antisera to proteins with known sub-cellular locations and biochemical marker assays (data not shown; Dwivany, 2003).
Figure 7 shows the results of Western blots incubated with antisera to an Arabidopsis H+-ATPase (a PM marker; Chevallet et al., 1998) and two Golgi apparatus markers, pea RGP1 (Dhugga et al., 1997) and HvGlyT4 (Farrokhi, 2005). HvGlyT4 is a member of CAZy family GT47 and has highest sequence similarity to β-glucuronyltransferases (Supplemental Figure 8) that in other systems are located in the Golgi apparatus (Brown et al., 2009; Iwai et al., 2002; Wu et al., 2009). Consistent with these reports, immuno-EM with the anti-HvGlyT4 antibody detected label in the Golgi apparatus of suspension-cultured cells (Supplemental Figure 9). The anti-H+-ATPase antiserum detected a 90-kDa band in the homogenate fraction, and bands of 80 and 60 kDa in the MM, PEG, and DEX fractions. These bands correspond to sizes previously reported for H+-ATPase, with the lower MW bands presumably arising by degradation during sample processing (Chevallet et al., 1998). The H+-ATPase bands were most intense in the PEG fraction and less intense in the DEX, MM, and homogenate fractions (Figure 7B). The anti-RGP1 antiserum detected a protein of the expected size in the homogenate, MM, and DEX fractions that was much less abundant in the PEG fraction (Figure 7C). A similar pattern of labeling was also observed with the HvGlyT4 antiserum, with the 55-kDa protein larger than predicted (41 kDa), suggesting that it is post-translationally modified (Figure 7D). Collectively, these data indicated that the PEG fraction was enriched in PM proteins and largely depleted of proteins from the Golgi apparatus. Enzyme marker assays done on the same PEG and DEX fractions are consistent with these findings (data not shown; Dwivany, 2003).
|
When blots of the fractions were incubated with the anti-HvCSLC2 antiserum, the
80-kDa protein was enriched in the PEG fraction, with trace amounts present in the DEX fraction and no detectable protein in the homogenate and MM fractions (Figure 7A). Thus, the anti-HvCSLC2 antiserum detected a low-abundance integral membrane protein that preferentially partitioned into a PM-enriched fraction. To confirm this location, barley suspension-cultured cells were prepared for immuno-EM (Figure 8). The anti-HvCSLC2 antiserum gave a light level of labeling (arrowheads) that was restricted to the PM (Figure 8A, arrows). Little or no labeling was seen in intracellular organelles such as the Golgi apparatus and endoplasmic reticulum (Figure 8B).
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Here, we report the identification and characterization of four barley genes belonging to the CSLC family of polysaccharide synthases. Our data show that at least one of the barley CSLCs resembles the better characterized CESAs in being an integral membrane protein targeted to the PM. This cellular location is relevant to discussions of the proposed role of CSLCs as polysaccharide synthases and contrasts with previous work on the Arabidopsis CSLC4 that suggests some of these proteins are targeted to the Golgi apparatus and involved in XyG backbone synthesis (Cocuron et al., 2007).
The four CSLC genes described here potentially represent the full complement of functional CSLC genes in the barley genome, since all barley CSLC ESTs were derived from one or other of these four genes. However, as Arabidopsis (Richmond and Somerville, 2000), poplar (Suzuki et al., 2006), grapevine (Jaillon et al., 2007; www.genoscope.cns.fr/externe/GenomeBrowser/Vitis/), and sorghum (Paterson et al., 2009) all have five functional CSLC genes, rice has six functional CSLCs and at least four CSLC-derived pseudogenes (http://waltonlab.prl.msu.edu/research-cwb.htm), and maize has 12 CSLCs, at least one of which is a pseudogene (van Erp and Walton, 2009), it is possible that one or more barley CSLC genes remain to be discovered. For instance, there is most likely at least one more clade Ic-type sequence, as there are two rice and two Sorghum CSLC sequences within this subclade and only one barley sequence, HvCSLC2 (Figure 2). Because they are not represented among the barley ESTs, any functional CSLC genes still missing are likely to be expressed at only low levels or only by certain cell types during particular developmental stages.
Phylogenetic analysis shows the CSLC family contains four well supported but poorly resolved clusters. One of these clusters (clade III) contains all the moss and lycophyte CSLCs and the other three clusters all the flowering plant CSLCs (Figure 2). Clades I and II contain both grass and eudicot CSLCs, whereas clade IV contains only eudicot sequences. One barley CSLC is in clade II and the other three are in clade I. Within clade III, the Physcomitrella and Selaginella CSLCs are well separated, reflecting the separate evolutionary paths taken by the lycopod and moss lineages (Figure 2). Consistent with previous work, none of the Physcomitrella CSLCs is basal to any of the flowering plant clades, indicating that divergence within flowering plant CSLC lineages occurred later than the divergence of the flowering plants and moss lineages (Roberts and Bushoven, 2007). A CSLC from Chara globularis lies outside these clades and further sampling of CSLCs from other non-flowering plants and charophyte algae is required to resolve the evolutionary history of this family.
Accumulation of CSLC transcripts in barley is generally quite low and transcript abundance is usually less than 10% of the level of any CESA mRNA expressed in the same tissue (Burton et al., 2004) (compare Supplemental Figure 2A and 2C). Although three of the barley CSLC genes (HvCSLC1, HvCSLC2, and HvCSLC4) are expressed to varying degrees in most tissues examined, these genes do not appear to be coordinately regulated at the transcriptional level. This does not, however, preclude different CSLC isoforms from acting jointly in the synthesis of a particular polysaccharide. Furthermore, unlike the CESAs, the CSLCs cannot be classified by expression pattern into those that function in primary cell wall synthesis and those that function in secondary cell wall synthesis (Burton et al., 2004). Consistent with this are in situ RT–PCR results showing that HvCSLC transcripts accumulate in all cell types of the root and coleoptile.
Five barley members of the CAZy glycosyltransferase family 34 (GT34) are on the Barley1 microarray (HvGT1–5). Plant GT34s include the galactomannan -(1,6)-galactosyltransferases and the UDP-Xyl:xyloglucan -(1,6)-xylosyltransferases (Cantarel et al., 2008; www.cazy.org/). Correlation analysis revealed that the expression profiles of HvCSLC3 and HvGT3 were highly correlated (Figure 4 and Supplemental Table 3). It seems likely that HvGT3 encodes a XyG xylosyltransferase, as its partial sequence aligns most closely to those of three confirmed XyG xylosyltransferases from Arabidopsis: AtXXT1, AtXXT2, and AtXXT5 (Cavalier and Keegstra, 2006; Cavalier et al., 2008; Faik et al., 2002; Zabotina et al., 2008). As a recent study of wheat seedlings found high levels of XyG (23–39 mol%) in the cell walls of root tips (Leucci et al., 2008) and our own analysis has confirmed the presence of XyG in 3-day-old barley roots (data not shown), it seems plausible to suggest that HvCSLC3 and HvGT3 are involved in XyG biosynthesis in barley root tips. Although this conclusion needs to be confirmed experimentally, as the correlated transcript profiles of HvCSLC3 and HvGT1 are not proof that the products of these genes participate in the same pathway, it is consistent with the presence of HvCSLC3 and AtCSLC4 in the same CSLC clade (clade II), as Cocuron et al. (2007) had previously concluded that AtCSLC4 is involved in XyG backbone synthesis. Among other genes with transcript patterns highly correlated to HvCSLC3 were two XyG endotransglucosylases/hydrolases (XET/XTHs) (r = 0.98 and 0.93), which are also likely to be involved in XyG assembly or re-modeling.
Evidence for the other three barley CSLCs being involved in XyG synthesis is either lacking or equivocal. HvCSLC1, 2, and 4 are all in clade I, which has only one Arabidopsis member (AtCSLC12). The highest correlation between a clade I barley CSLC and the barley GT34s was between HvCSLC1 and HvGT1 (r = 0.82). Two other GT34s, HvGT2 and HvGT3, showed slightly lower correlations to HvCSLC1 (r = 0.71 and 0.80, respectively), and none of these correlations was above the level assigned as significant (r = 0.89). However, like HvGT3, HvGT1 and HvGT2 align better to XyG xylosyltransferases from Arabidopsis than to the galactomannan -(1,6)-galactosyltransferases, which are also in GT34 (Supplemental Figure 3). Furthermore, the HvCSLC1 transcript profile was correlated to the profile of a XET/XTH gene (r = 0.85). Together, these data suggest that HvCSLC1 may be involved in XyG backbone synthesis in tissues other than root tips, such as the coleoptile. But HvCSLC1 was most highly correlated (r = 0.90) to a gene related to SHORT HYPOCOTYL 2 (SHY2), that codes for an Arabidopsis protein that promotes cell differentiation by negatively regulating genes involved in auxin redistribution (Dello Ioio et al., 2008). Further investigation into the function of HvCSLC1 is therefore required to understand its transcriptional correlation with SHY2. The expression profiles of HvCSLC2 and HvCSLC4 were not significantly correlated to those of a HvGT or a XET/XTH (r < 0.5), suggesting that these genes have functions different from those of HvCSLC3 and possibly HvCSLC1.
An alternative approach to determine whether the barley CSLCs were involved in XyG biosynthesis was to use immuno-EM and the LM15 antibody to see whether XyG was present in the walls of barley cells that accumulated CSLC transcripts. Coleoptile was used as an example of a tissue in which both HvCSLC expression and XyG accumulation take place. To confirm that LM15 specifically detected XyG, coleoptile sections were pre-incubated with a variety of (1,4)- and (1,3)-β-glucose containing polysaccharides/oligosaccharides, including derivatives of cellulose, callose, and (1,3;1,4)-β-glucan. None of these treatments reduced LM15 labeling (Supplemental Figure 5). Both primary and secondary cell walls contained XyG and these cells also accumulated HvCSLC transcripts.
Endosperm was examined because HvCSLC1, 2, and 4 transcripts accumulated to relatively high levels during the initial stages of wall development (Figure 3B), yet XyG was not known to be present in endosperm cell walls (Bacic and Stone, 1981; Fincher, 1975). However, immuno-EM with the LM15 antibody detected light labeling in anticlinal and periclinal walls of developing endosperm cells at 4 DAP (Figure 6 and Supplemental Figure 6). These walls are known to contain callose and have been suggested to contain cellulose as well, based on labeling with gold-conjugated cellobiohydrolase II (CBH II; Wilson et al., 2006). This conclusion may need to be revised in light of these findings, as CBH II also binds to XyG. By 8 DAP, LM15 labeling was still detectable in the surrounding maternal tissues but was no longer evident in endosperm walls, implying that the XyG deposited during endosperm cellularization was rapidly turned over. XyG is thus a second example, along with callose, of a polysaccharide that is deposited and then removed from early endosperm cell walls. Concomitant with XyG disappearance was a 10-fold reduction in the levels of HvCSLC1, 2, and 4 transcripts (Figure 3B).
Although HvCSLC expression and XyG deposition in developing endosperm were correlated, experiments with a polyclonal antibody to a HvCSLC2 peptide provided evidence that the CSLCs reside in the PM and are not in the Golgi—an important observation, given previous evidence that XyG synthesis occurs in the Golgi (Becker et al., 1995; Delmer, 1999; Gibeaut and Carpita, 1994; Gordon and Maclachlan, 1989). HvCSLC2 is the most abundant CSLC transcript in barley suspension-cultured cells, although HvCSLC1 and HvCSLC4 transcripts also accumulate to varying degrees (Figure 3A). As the specificity of the HvCSLC2 antibody to other CSLC isoforms is unknown, it is possible that the band it detects contains all three isoforms found in suspension-cultured cells, with HvCSLC2 probably being the most abundant of these forms (Figure 7). Thus, one interpretation is that all CSLCs in the extract preferentially partitioned into the PM-enriched PEG fraction and that the low amount of CSLC in the Golgi-containing DEX fraction was due to the presence of some PM in this fraction. However, it is equally possible that only the most abundant CSLC isoform (presumably HvCSLC2) partitioned into the PM fraction, with the DEX fraction containing in other Golgi-targeted CSLC isoforms. According to this interpretation, CSLCs are targeted to either the PM or Golgi. Of the two interpretations, only the PM location was supported by immuno-EM with HvCSLC2 antibody (Figure 8). It therefore appears that in barley suspension-cultured cells, CSLCs are targeted to the PM, although we cannot rule out the possibility that some isoforms, present in low abundance, are targeted to the Golgi as well.
However, linkage analysis detected only trace amounts of XyG in suspension-cultured cell walls (Yulia, 2006) and LM15 labeling of the walls of these cells was very light (Supplemental Figure 10). HvCSLC2 and the other isoforms present in these cells consequently do not appear to be involved in XyG backbone synthesis and may instead play a role in synthesizing callose or cellulose, as these are the only polysaccharides known to be made at the PM. Of the two polysaccharides, it is most likely that HvCSLC2 is involved in making cellulose, as β-(1,4)-glucan synthase activity has already been shown for another member of the CSLC family (Cocuron et al., 2007), and callose, a β-(1,3)-glucan, is synthesized by proteins of the GSL gene family (Brownfield et al., 2007; Li et al., 2003).
The CSLC family therefore appears to contain proteins targeted to two distinct sub-cellular locations and participating in the synthesis of two distinct polysaccharides. Clade II proteins, such as AtCSLC4 and probably HvCSLC3, are targeted to the Golgi where they might participate in XyG backbone biosynthesis (Cocuron et al., 2007; Dunkley et al., 2006). However, clade I proteins such as HvCSLC2 are targeted to the PM, where they probably participate in cellulose biosynthesis. From a biochemical perspective, the proposed functional diversification within the CSLC family is far from implausible because XyG backbone synthases and cellulose synthases both make a β-(1,4)-glucan chain. This conclusion is also consistent with the absence of XyG in the walls of charophyte algae (Popper and Fry, 2003, 2004), suggesting that the Chara CSLC is not involved in making XyG, but rather participates in the synthesis of some other polysaccharide, possibly cellulose. However, a thorough chemical analysis of the walls of various charaophycean algae is required to confirm the absence of XyG from this taxon. An ancestral role for the CSLCs as cellulose synthases is also in keeping with this family belonging to an ancient lineage that is evolutionarily distinct from the CESA lineage to which most other CSL families belong (Nobles and Brown, 2004; Roberts and Bushoven, 2007).
The cellular locations and functions of some barley isoforms, specifically HvCSLC1 and HvCSLC4, are at this point uncertain. They are closely related to each other and phylogenetic analysis places them in a separate group within clade I (group a in Figure 2) to HvCSLC2 (group c in Figure 2). HvCSLC4 and HvCSLC1 most likely share the same function, although there is ambiguous evidence as to whether this is in cellulose or XyG synthesis. For instance, although the HvCSLC1 transcript profile shows some correlation to genes involved in XyG synthesis and deposition in the wall, the HvCSLC4 transcript profile does not and the list of genes most highly correlated to it provides few clues towards a likely function (Supplemental Table 3). However, using the experimental tools described in this paper, it should be possible to distinguish between these two functions for HvCSLC1 and HvCSLC4.
While this paper was under review, ESTs for a fifth HvCSLC gene, HvCSLC5, were identified in GenBank (accession numbers EX582174, EX582178, EX594308, EX594309, EX590401, EX590402, EX596001, EX596002). These ESTs are from a pooled tissue sample library and the HvCSLC5 contig is most similar to HvCSLC2 and thus highly likely to be the barley CSLC gene predicted to be missing from clade Ic of Figure 2. The discovery of this gene does not in any way change the conclusions drawn here.
|
METHODS |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Plant Materials, cDNA, and BAC Libraries
Grains of Hordeum vulgare L. cultivars. Schooner and Sloop were imbibed, sown into soil, and grown in the greenhouse as previously described (Burton et al., 2004). The barley cv. Schooner suspension cell culture was initiated from a seed-derived callus and has been maintained at 23°C in the dark with shaking (114 rpm) in MS8 basal nutrient salt (ICN Biomedical) medium (pH 5.8–6.0) supplemented with 3% w/v sucrose, 2 mg l–1 2,4-dichlorophenoxyacetic acid (2,4-D) and 1 mg mL–1 mixed cytokinins (333 µM each of 6-
,-dimethylallylaminopurine, 6-benzylaminopurine (BAP), and kinetin (Sigma-Aldrich). Cells were maintained by sub-culturing 50 mL of cell suspension to a flask containing 100 mL of fresh medium at weekly intervals. Preparation of a 
ZAPII cDNA library from barley suspension-cultured cells is described in Burton et al. (2004). The barley cv. Morex BAC library was obtained from the Clemson University Genomics Institute (CUGI; www.genome.clemson.edu).
Identification of HvCSLC cDNAs and Genes
The barley suspension culture cDNA library was probed with a DNA fragment amplified from a putative HvCSLC EST (accession no. BE455720). The barley BAC library was screened with PCR fragments from the 3’ ends of the HvCSLC1 and HvCSLC2 cDNAs. Hybridizations were performed as described in Burton et al. (2004).
Sequence Analysis and Bioinformatics
DNA sequencing was done at a commercial sequencing facility (Australia Genomic Research Facility, Australia) and the chromatograms analyzed using SequencherTM 3.0 (Gene Codes Corporation, Inc., Michigan, USA).
Arabidopsis, rice, and barley CSLC sequences were used in iterative searches of public databases, including the now discontinued Stanford Cell Wall website, NCBI (www.ncbi.nlm.nih.gov/), HarvEST (http://harvest.ucr.edu/), GrainGenes (http://wheat.pw.usda.gov/GG2/index.shtml), BarleyGene Index (http://compbio.dfci.harvard.edu/tgi/plant.html), and BarleyBase (www.barleybase.org) to obtain full-length cDNAs and identify other members of this gene family in barley. Sequences were assembled into contigs using either SequencerTM 3.0 (Gene Codes) or ContigExpress, a module of Vector NTI Advance 9.1.0 (Invitrogen). DNA and protein alignments were performed using CLUSTALW2 (www.ebi.ac.uk/) or CLUSTALX v 2.0.9 (Larkin et al., 2007).
CSLC sequences from other species were downloaded from TAIR (At, www.arabidopsis.org/), TIGR Gene Indices (Os, http://compbio.dfci.harvard.edu/tgi/plant.html), MIPS (Sb, http://mips.gsf.de/proj/plant/jsf/sorghum/index.jsp), JGI (Pt, www.jgi.doe.gov/poplar/; Cg, Pp, Sm, http://genome.jgi-psf.org/), TIGR Plant Transcript Assemblies (Cg, Mt, Sl, Vv, http://plantta.tigr.org/, http://genome.jgi-psf.org/), and NCBI (Zm, www.ncbi.nlm.nih.gov/).
Phylogenetic trees were constructed from sequence alignments using the distance algorithm of Paup 4.0b10 (Swofford, 2000) using the PaupUp v1.0.3.1 graphical interface (Calendini and Martin, 2005). Default distance settings were used with the Neighbour Joining clustering option. Trees were bootstrapped with 1000 replicates to assess the robustness of each node. MEGA4.0 was used to view and edit the resulting trees (Tamura et al., 2007). Sequence identities and similarities were calculated using MatGat 2.02 using default settings (Campanella et al., 2003). Transmembrane domains and protein topologies were predicted using WoLF PSORT (http://wolfpsort.org/; Horton et al., 2007).
The Affymetrix Barley1 22K GeneChip reference dataset for Experiment BB3: ‘Transcription patterns during barley development’ (Druka et al., 2006) was downloaded from PLEXdb (Wise et al., 2007; www.plexdb.org/). The downloaded text file contained the robust multi-array average (RMA) treatment means for all probe sets. The RMA treatment includes background adjustment, normalization, and log2-transformation of perfect match values calculated from triplicate hybridizations. These data were imported into Excel 2007, row-to-column transformed, and the correlation of CSLC and GT probe sets calculated using the CORREL function.
Quantitative PCR
QPCR of HvCSLC gene expression used a previously described collection of barley cDNAs (Burton et al., 2004), except for the suspension cell culture cDNA, which was prepared from the suspension culture cell line described above 1 week after subculture. QPCR amplification was performed in a total reaction volume of 20 µl using the method described by Burton et al. (2008). Relative expression levels were normalized by geometric averaging of the internal control genes calculated using geNorm (Vandesompele et al., 2002). Barley genes for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), cyclophilin, -tubulin and elongation factor 1 (EF1) were used as the internal controls. Supplemental Table 5 lists the primers used to amplify the four CSLC genes. Primers for the control genes are listed in Doblin et al. (2009).
In Situ PCR
In situ PCR was performed on coleoptile and root tissues harvested 3 d after germination using the method of Koltai and Bird (2000) with the modifications listed in Doblin et al. (2009). Synthesis of cDNA was carried out using Thermoscript RT (Invitrogen, USA) and one of the gene-specific primers listed in Supplemental Table 5. A typical PCR profile was as follows: initial denaturation period of 96°C for 2 min, 40 cycles of 94°C for 30 s, 30 s at an annealing temperature chosen based on the primer pair being used, 72°C for 2 min. The products of all primer combinations were analyzed by agarose gel electrophoresis and sequenced to ensure that they gave only the expected product (data not shown).
HvCSLC Antibody Production
The region of the HvCSLC2 cDNA coding for amino acids 411–466 was amplified by PCR and cloned into the expression vector pProEX HTa (Invitrogen). The resultant plasmid (pEVCSLC) was transformed into Escherichia coli BL21(DE3) cells. Expression was induced in a 500-ml culture by addition of isopropyl-β-D-thiogalactoside (IPTG) to 1 mM and the peptide enriched using Ni-NTA affinity chromatography (Qiagen) as described by the manufacturer. After assessment of Ni-NTA-eluate fractions by SDS–PAGE, fractions containing expressed peptide were pooled and dialyzed. Antibodies were raised by intramuscular injection of 500 µg expressed protein with Freund's complete adjuvant (Sigma) into New Zealand white rabbits (Monash University, Melbourne, Australia). Booster injections were given 21, 49, and 77 d after the initial injection with the same amount of protein but with Freund's incomplete adjuvant (Sigma) and the rabbits were exsanguinated 94 d after the initial immunization. Affinity purification was conducted using the method of Brownfield et al. (2007).
Two-Phase Partitioning
Barley suspension culture cells (20 g) were homogenized with a glass-to-glass grinder (Tenbroek, Pyrex, USA) in 80 mL homogenization buffer (50 mM potassium phosphate buffer (pH 7.5), 20 mM KCl, 0.5 M sucrose, 10 mM DTT and 400 µl plant protease inhibitor cocktail (Sigma). Disrupted cells were filtered through Miracloth (Calbiochem) and centrifuged at 10 000 g, 4°C for 10 min and the pellet discarded. The supernatant, labeled the homogenate (HM) fraction, was centrifuged at 125 000 g for 1 h at 4°C and the resultant pellet was labeled the mixed membrane (MM) fraction.
A PM-enriched fraction was prepared using the two-phase partitioning method (Larsson et al., 1987). The MM pellet was re-suspended in a solution composed of 6.5% (w/v) polyethylene glycol (PEG) 3350 (Sigma), 6.5% (w/v) dextran T500 (DEX) (Pharmacosmos A/S, Holbaek, Denmark), 3 mM KCl, 0.25 M sucrose and 5 mM potassium phosphate buffer (pH 7.5) and processed according to Natera et al. (2008). The third PEG phase was diluted two-fold with a buffer composed of 20 mM HEPES, 20 mM KCl, and 0.2 M sucrose and pelleted at 125 000 g, 4°C for 45 min. The pellet was re-suspended in the same buffer and labeled the PM fraction.
The purity of the PM fraction was assessed by Western blots with antisera to known marker proteins and by TEM (data not shown; Dwivany, 2003; Yulia, 2006).
Western Blotting
Homogeneous 12% polyacrylamide gels were prepared and Western blotting performed using standard methods (Sambrook et al., 1989) with an OSMONIC Nitropure 22 µm nitrocellulose membrane (GE Osmonics). After blocking, the membrane was incubated in the appropriate dilution of primary antibody in PBS (see below), followed by incubation in a 1:10 000 dilution of a goat anti-rabbit IgG antibody conjugated to either alkaline phosphatise (Sigma) or horseradish peroxidase (Pierce). Detection was performed using SuperSignal West Pico chemiluminescent substrate (Pierce). The protein content of all fractions was determined by the BCA protein assay (Pierce).
The plasma membrane marker Arabidopsis H+-ATPase AHA3 (P-type) antibody was generously provided by Dr Ramon Serrano (Universidad Politecnica de Valencia-CSIC, Valencia, Spain) and was used at 1:2000 dilution. The Golgi marker Pisum sativum anti-reversibly glycosylated protein 1 (RGP1, also known as UDP-arabinopyranose mutase; Konishi et al., 2007) antibody was kindly provided by Dr Kanwarpal Dhugga, Pioneer Hi-Bred International Inc., Des Moines, IA, USA, and was used at a 1:100 000 dilution. Naser Farrokhi (ACPFG, University of Adelaide, Australia) kindly provided the IgG-purified HvGlyT4 Golgi marker antibody generated towards a bacterially expressed barley GT47 family glycosyltransferase (Supplemental Figures 7 and 8) that was used at a 1:1000 dilution. The affinity-purified HvCSLC2 antibody, generated as described above, was used at a 1:1000 dilution.
Immuno-Electron Microscopy (Immuno-EM)
Coleoptile, developing grain, and suspension-cultured cells (4 d after sub-culture) of barley were fixed and processed for immuno-EM using the method described in Wilson et al. (2006). A high-pressure freezing method described in Brownfield et al. (2008) was also used to prepare the barley suspension-cultured cells. Thin sections were incubated in a 1:50 dilution of LM15 rat monoclonal antibody (PlantProbes) in PBS, pH 7.4, containing 1% w/v BSA (+/– competing polysaccharides) for 1 h at room temperature and then overnight at 4°C. Grids were washed in PBS and then incubated in a 1:20 dilution of goat anti-rat secondary antibody conjugated to 18-nm gold particles (Jackson ImmunoResearch). Sections then were washed, post-stained, and viewed by TEM as described in Burton et al. (2006). Polysaccharide/oligosaccharide solutions (1 mg mL–1) used in competitive antibody incubations were carboxymethyl cellulose, laminarin and cellohexaose (Sigma), crystalline microcellulose (Merck), barley flour (1,3;1,4)-β-D-glucan and tamarind seed XyG (Megazyme), and N. plumbaginifolia-extracted XyG (Sims and Bacic, 1995; Sims et al., 1996). For sub-cellular location studies, the affinity-purified HvCSLC2 antibody was used at a 1:10 dilution with the goat anti-rabbit secondary antibody conjugated to 18-nm gold particles (Jackson ImmunoResearch) used at a 1:20 dilution.
Accession Numbers
Sequence data from this article can be found in the EMBL/GenBank data libraries under accession numbers GQ386981
[GenBank]
to GQ386984. Accession numbers for other CSLCs are as follows: Arabidopsis thaliana (TAIR gene id. At3g28180, At4g31590, At3g07330, At2g24630, At4g07960; Richmond and Somerville, 2000), Oryza sativa (TIGR gene id. Os01g56130, Os09g25900, Os08g15420, Os05g43530, Os03g56060, Os07g03260; Hazen et al., 2002), Sorghum bicolor (SbSb02g002090.1, Sb01g006820.1, Sb09g025260.1, Sb03g035660.1, Sb07g007890.1; Paterson et al., 2009), Populus trichocarpa (Poptr1_1:763645, Poptr1_1:578365, Poptr1_1:816437, Poptr1_1:818429, Poptr1_1:830588; Suzuki et al., 2006), Vitis vinifera (CAN83466
[GenBank]
, CAN78456
[GenBank]
, CAN82135
[GenBank]
, CAN82493
[GenBank]
, AM430199
[GenBank]
), Medicago truncatula (AC171266
[GenBank]
), Solanum lycopersicum (AP009283
[GenBank]
), Tropaeolum majus (nasturtium) (Cocuron et al., 2007), Zea mays (van Erp and Walton, 2009), Physcomitrella patens (DQ898284
[GenBank]
, DQ898285
[GenBank]
, DQ898286
[GenBank]
, EF608235
[GenBank]
; Roberts and Bushoven, 2007), Chara globularis (AY995817
[GenBank]
), Selaginella moellendorffii (scaffold_92|650930, scaffold_26|411112, scaffold_57|824521, scaffold_0|6575806).
|
SUPPLEMENTARY DATA |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
Supplementary Data are available at Molecular Plant Online.
|
FUNDING |
|---|
|
TOPAbstractINTRODUCTIONRESULTSDISCUSSIONMETHODSSUPPLEMENTARY DATAFUNDING |
|---|
We gratefully acknowledge the Grains Research and Development Corporation for funding. F.M.D. was funded by a QUE Project Scholarship from the Indonesian Government and D.Y. by an Australian Development Scholarship from the Australian government.
|
Acknowledgements |
|---|
We thank Dr Andrew Harvey (ACPFG, University of Adelaide) for his assistance with designing primers for QPCR, Dr Filomena Pettolino (School of Botany, University of Melbourne) for assistance with the analysis of barley suspension-cultured cell walls, and Ms Cherie Walsh (School of Botany, University of Melbourne) for her assistance with immuno-EM. No conflict of interest was declared.
-
Bacic A, Stone BA. Chemistry and organization of aleurone cell wall components from wheat Triticum aestivum Cultivar Insignia and barley Hordeum vulgare Cultivar Clipper. Aust. J. Plant Physiol (1981) 8:475–495.
Bacic A, Harris PJ, Stone BA. Structure and function of plant cell walls. In: The Biochemistry of Plants—Priess J, ed. (1988) New York: Academic Press. 297–371.
Becker M, Vincent C, Reid JSG. Biosynthesis of (1,3) (1,4)-β-glucan and (1,3)-β-glucan in barley (Hordeum vulgare L): properties of the membrane-bound glucan synthases. Planta (1995) 195:331–338.[Web of Science][Medline]
Bernal AJ, et al. Disruption of ATCSLD5 results in reduced growth, reduced xylan and homogalacturonan synthase activity and altered xylan occurrence in Arabidopsis. Plant J (2007) 52:791–802.[CrossRef][Web of Science][Medline]
Brown DM, Zhinong Z, Stephens E, Dupree P, Turner SR. Characterization of IRX10 and IRX10-like reveals an essential role in glucuronoxylan biosynthesis in Arabidopsis. Plant J (2009) 57:732–746.[CrossRef][Web of Science][Medline]
Brownfield L, Ford K, Doblin MS, Newbigin E, Read S, Bacic A. Proteomic and biochemical evidence links the callose synthase in Nicotiana alata pollen tubes to the product of the NaGSL1 gene. Plant J (2007) 52:147–156.[CrossRef][Web of Science][Medline]
Brownfield L, Wilson S, Newbigin E, Bacic A, Read S. Molecular control of the glucan synthase-like protein NaGSL1 and callose synthesis during growth of Nicotiana alata pollen tubes. Biochemical Journal (2008) 414:43–52.[CrossRef][Web of Science][Medline]
Burton RA, et al. Cellulose synthase-like CslF genes mediate the synthesis of cell wall (1,3;1,4)-β-D-glucans. Science (2006) 311:1940–1942.
Burton RA, et al. The genetics and transcriptional profiles of the cellulose synthase-like HvCslF gene family in barley. Plant Physiol (2008) 146:1821–1833.
Burton RA, Shirley NJ, King BJ, Harvey AJ, Fincher GB. The CesA gene family of barley: quantitative analysis of transcripts reveals two groups of co-expressed genes. Plant Physiol (2004) 134:224–236.
Calendini F, Martin J-F. PaupUP v1.0.3.1: a free graphical frontend for Paup* Dos software. (2005) www.agro-montpellier.fr/sppe/Recherche/JFM/PaupUp/main.htm.
Campanella J, Bitincka L, Smalley J. MatGAT: an application that generates similarity/identity matrices using protein or DNA sequences. BMC Bioinformatics (2003) 4:29.[CrossRef][Medline]
Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Lombard V, Henrissat B. The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics. Nucl. Acids Res. (2008) gkn663.
Carpita N, Gibeaut D. Structural models of primary cell walls in flowering plants: consistency of molecular structure with the physical properties of the walls during growth. Plant J (1993) 3:1–30.[CrossRef][Web of Science][Medline]
Carpita NC. Structure and biogenesis of the cell walls of grasses. Ann. Rev. Plant Physiol. Plant Mol. Biol. (1996) 47:445–476.[CrossRef]
Carpita NC, McCann MC. The cell wall. In: Biochemistry and Molecular Biology of Plants—Buchanan BB, Gruissem W, Jones RL, eds. (2000) Rockville: American Society of Plant Physiologists. 52–108.
Cavalier DM, Keegstra K. Two xyloglucan xylosyltransferases catalyze the addition of multiple xylosyl residues to cellohexaose. J. Biol. Chem. (2006) 281:34197–34207.
Cavalier DM, et al. Disrupting two Arabidopsis thaliana xylosyltransferase genes results in plants deficient in xyloglucan, a major primary cell wall component. Plant Cell. (2008) 20:1519–1537.
Chevallet M, et al. New zwitterionic detergents improve the analysis of membrane proteins by two dimensional electrophoresis. Electrophoresis (1998) 19:1901–1909.[CrossRef][Web of Science][Medline]
Cocuron J-C, et al. A gene from the cellulose synthase-like C family encodes a β-1,4 glucan synthase. Proc. Natl Acad. Sci. U S A (2007) 104:8550–8555.
Cosgrove DJ. Growth of the plant cell wall. Nature Reviews Molecular Cell Biology (2005) 6:850–861.[CrossRef][Web of Science][Medline]
Dello Ioio R, et al. A genetic framework for the control of cell division and differentiation in the root meristem. Science (2008) 322:1380–1384.
Delmer DP. Cellulose biosynthesis: exciting times for a difficult field of study. Ann. Rev. Plant Physiol. Plant Mol. Biol. (1999) 50:245–276.[CrossRef]
Dhugga KS, et al. Guar seed β-mannan synthase is a member of the cellulose synthase super gene family. Science (2004) 303:363–366.
Dhugga KS, Tiwari SC, Ray PM. A reversibly glycosylated polypeptide (RGP1) possibly involved in plant cell wall synthesis: purification, gene cloning, and trans-Golgi localization. Proc. Natl Acad. Sci. U S A (1997) 94:7679–7684.
Doblin MS, De Melis L, Newbigin E, Bacic A, Read SM. Pollen tubes of Nicotiana alata express two genes from different β-glucan synthase families. Plant Physiol (2001) 125:2040–2052.
Doblin MS, et al. A barley cellulose synthase-like CSLH gene mediates (1,3;1,4)-β-D-glucan synthesis in transgenic Arabidopsis. Proc. Natl Acad. Sci. U S A (2009) 106:5996–6001.
Druka A, et al. An atlas of gene expression from seed to seed through barley development. Funct. Integrat. Genomics (2006) 6:202–211.[CrossRef]
Dunkley TPJ, et al. Mapping the Arabidopsis organelle proteome. Proc. Natl Acad. Sci. U S A (2006) 103:6518–6523.
Dwivany FM. Functional analysis of the barley cellulose synthase-like C (CSLC) gene family. (2003) Ph.D. thesis, School of Botany, University of Melbourne Victoria, Australia.
Faik A, Price NJ, Raikhel NV, Keegstra K. An Arabidopsis gene encoding an -xylosyltransferase involved in xyloglucan biosynthesis. Proc. Natl Acad. Sci. U S A (2002) 99:7797–7802.
Farrokhi N. Functional analysis of cell wall-related barley glycosyltransferases. Ph.D. thesis, School of Agriculture. In: Food and Wine, University of Adelaide (2005) South Australia, Australia.
Farrokhi N, et al. Plant cell wall biosynthesis: genetic, biochemical and functional genomics approaches to the identification of key genes. Plant Biotechnology J (2006) 4:145–167.[CrossRef]
Fincher GB. Morphology and chemical composition of barley endosperm cell walls. J. Inst. Brew (1975) 81:116–122.
Fincher GB. Revolutionary times in our understanding of cell wall biosynthesis and remodelling in the grasses. Plant Physiol (2009) 149:27–37.
Fry SC. The structure and functions of xyloglucan. J. Exp. Bot (1989) 40:1–11.
Gibeaut DM, Carpita NC. Improved recovery of (1
3), (14)-β-D-glucan synthase activity from Golgi apparatus of Zea mays (L.) using differential flotation centrifugation. Protoplasma (1994) 180:92–97.[CrossRef][Web of Science]
Gibeaut DM, Pauly M, Bacic A, Fincher GB. Changes in cell wall polysaccharides in developing barley (Hordeum vulgare) coleoptiles. Planta (2005) 221:729–738.[CrossRef][Web of Science][Medline]
Gordon R, Maclachlan G. Incorporation of UDP-[14C]glucose into xyloglucan by pea membranes. Plant Physiol (1989) 91:373–378.
Harris PJ, Kelderman MR, Kendon MF, McKenzie RJ. Monosaccharide compositions of unlignified cell walls of monocotyledons in relation to the occurrence of wall-bound ferulic acid. Biochemical Systematics and Ecology (1997) 25:167–179.[CrossRef][Web of Science]
Hayashi T. Xyloglucans in the primary cell wall. Ann. Rev. Plant Physiol. Plant Mol. Biol. (1989) 40:139–168.[CrossRef]
Hazen SP, Scott-Craig JS, Walton JD. Cellulose synthase-like genes of rice. Plant Physiol (2002) 128:336–340.
Horton P, et al. WoLF PSORT: protein localization predictor. Nucl. Acids Res (2007) 35:W585–W587.
Iwai H, Masaoka N, Ishii T, Satoh S. A pectin glucuronyltransferase gene is essential for intercellular attachment in the plant meristem. Proc. Natl Acad. Sci. U S A (2002) 99:16319–16324.
Jaillon O, et al. The grapevine genome sequence suggests ancestral hexaploidization in major angiosperm phyla. Nature (2007) 449:463–467.[CrossRef][Medline]
Koltai H, Bird DM. High throughput cellular localization of specific plant mRNAs by liquid-phase in situ reverse transcription-polymerase chain reaction of tissue sections. Plant Physiol (2000) 123:1203–1212.
Konishi T, et al. A plant mutase that interconverts UDP-arabinofuranose and UDP-arabinopyranose. Glycobiology (2007) 17:345–354.
Larkin MA, et al. Clustal W and Clustal X version 2.0. Bioinformatics (2007) 23:2947–2948.
Larsson C, Widell S, Kjellbom P. Preparation of high-purity plasma membranes. Methods Enzymol (1987) 148:558–568.[CrossRef][Web of Science]
Lerouxel O, Cavalier DM, Liepman AH, Keegstra K. Biosynthesis of plant cell wall polysaccharides: a complex process. Curr. Op. Plant Biol. (2006) 9:621–630.[CrossRef][Web of Science][Medline]
Leucci MR, Lenucci MS, Piro G, Dalessandro G. Water stress and cell wall polysaccharides in the apical root zone of wheat cultivars varying in drought tolerance. J. Plant Physiol (2008) 165:1168–1180.[CrossRef][Web of Science][Medline]
Li J, et al. Biochemical evidence linking a putative callose synthase gene with (13)-β-D-glucan biosynthesis in barley. Plant Mol. Biol. (2003) 53:213–225.[CrossRef][Web of Science][Medline]
Liepman AH, Nairn CJ, Willats WGT, Sorensen I, Roberts AW, Keegstra K. Functional genomic analysis supports conservation of function among cellulose synthase-like A gene family members and suggests diverse roles of mannans in plants. Plant Physiol (2007) 143:1881–1893.
Liepman AH, Wilkerson CG, Keegstra K. Expression of cellulose synthase-like (Csl) genes in insect cells reveals that CslA family members encode mannan synthases. Proc. Natl Acad. Sci. U S A (2005) 102:2221–2226.
Manfield IW, et al. Novel cell wall architecture of isoxaben-habituated Arabidopsis suspension-cultured cells: global transcript profiling and cellular analysis. Plant J (2004) 40:260–275.[CrossRef][Web of Science][Medline]
Marcus S, et al. Pectic homogalacturonan masks abundant sets of xyloglucan epitopes in plant cell walls. BMC Plant Biology (2008) 8:60.[CrossRef][Medline]
Natera SHA, Ford KL, Cassin AM, Patterson JH, Newbigin EJ, Bacic A. Analysis of the Oryza sativa plasma membrane proteome using combined protein and peptide fractionation approaches in conjunction with mass spectrometry. J Proteome Research (2008) 7:1159–1187.[CrossRef]
Nobles DR, Brown RM. The pivotal role of cyanobacteria in the evolution of cellulose synthases and cellulose synthase-like proteins. Cellulose (2004) 11:437–448.[CrossRef][Web of Science]
O'Neill MA, York WS. The composition and structure of plant primary cell walls. In: The Plant Cell Wall—Rose JKC, ed. (2003) Oxford: Blackwell Publishing Ltd. 1–54.
Parkinson H, et al. ArrayExpress: a public database of microarray experiments and gene expression profiles. Nucl Acids Res. (2006) gkl995.
Paterson AH, et al. The Sorghum bicolor genome and the diversification of grasses. Nature (2009) 457:551–556.[CrossRef][Web of Science][Medline]
Pear JR, Kawagoe Y, Schreckengost WE, Delmer DP, Stalker DM. Higher plants contain homologs of the bacterial celA genes encoding the catalytic subunit of cellulose synthase. Proc. Natl Acad. Sci. U S A (1996) 93:12637–12642.
Popper ZA, Fry SC. Primary cell wall composition of bryophytes and charophytes. Ann. Bot (2003) 91:1–12.
Popper ZA, Fry SC. Primary cell wall composition of pteridophytes and spermatophytes. New Phytol (2004) 164:165–174.[CrossRef][Web of Science]
Puhlmann J, Bucheli E, Swain MJ, Dunning N, Albersheim P, Darvill AG, Hahn MG. Generation of monoclonal antibodies against plant cell wall polysaccharides: I. Characterization of a monoclonal antibody to a yerminal -(12)-linked fucosyl-containing epitope. Plant Physiol (1994) 104:699–710.[Abstract]
Richmond TA, Somerville CR. The cellulose synthase superfamily. Plant Physiol (2000) 124:495–498.
Roberts AW, Bushoven JT. The cellulose synthase (CESA) gene superfamily of the moss Physcomitrella patens. Plant Mol. Biol. (2007) 63:207–219.[CrossRef][Web of Science][Medline]
Sambrook J, Fritsch EF, Maniatis R. Molecular cloning: a laboratory manual. (1989) (Cold Spring Harbour, New York: Cold Spring Harbour Laboratory Press).
Scheible WR, Pauly M. Glycosyltransferases and cell wall biosynthesis: novel players and insights. Curr. Op. Plant Biol. (2004) 7:285–295.
Shen L, Gong J, Caldo RA, Netteton D, Cook D, Wise RP, Dickerson JA. BarleyBase: an expression profiling database for plant genomics. Nucl Acids Res (2005) 33:D614–D618.
Sims IM, Bacic A. Extracellular polysaccharides from suspension cultures of Nicotiana plumbaginifolia. Phytochem (1995) 38:1397–1405.[CrossRef]
Sims IM, Munro SLA, Currie G, Craik D, Bacic A. Structural characterisation of xyloglucan secreted by suspension-cultured cells of Nicotiana plumbaginifolia. Carbohydrate Research (1996) 293:147–172.[CrossRef][Web of Science][Medline]
Smith BG, Harris PJ. The polysaccharide composition of Poales cell walls: Poaceae cell walls are not unique. Biochem. Systematics Ecol (1999) 27:33–53.[CrossRef]
Somerville C. Cellulose synthesis in higher plants. Ann. Rev. Cell Develop. Biol. (2006) 22:53–78.[CrossRef][Web of Science][Medline]
Somerville C, et al. Toward a systems approach to understanding plant cell walls. Science (2004) 306:2206–2211.
Suzuki S, Li L, Sun Y-H, Chiang VL. The cellulose synthase gene superfamily and biochemical functions of xylem-specific cellulose synthase-like genes in Populus trichocarpa. Plant Physiol (2006) 142:1233–1245.
Swofford DL. PAUP* Phylogenetic Analysis Using Parsimony (*and Other Methods). In: Version 4 (2000) Sunderland, Massachusetts: Sinauer Associates.
Tamura K, Dudley J, Nei M, Kumar S. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software Version 4.0. Molecular Biol. Evol. (2007) 24:1596–1599.[CrossRef]
van Erp H, Walton J. Regulation of the cellulose synthase-like gene family by light in the maize mesocotyl. Planta (2009) 229:885–897.[CrossRef][Web of Science][Medline]
Vandesompele J, et al. Accurate normalization of real-time quantitative RT–PCR data by geometric averaging of multiple internal control genes. Genome Biology (2002) 3. 0034.0031–0034.0011.
Wilson S, et al. Temporal and spatial appearance of wall polysaccharides during cellularization of barley (Hordeum vulgare) endosperm. Planta (2006) 224:655–667.[CrossRef][Web of Science][Medline]
Wise R, Caldo R, Hong L, Shen L, Cannon E, Dickerson J. BarleyBase/PLEXdb: a unified expression profiling database for plants and plant pathogens. In: Methods in Molecular Biology Plant Bioinformatics: Methods and Protocols—Edwards D, ed. (2007) Totowa, New Jersey: Humana Press. 347–363.
Wu A-M, et al. The Arabidopsis IRX10 and IRX10-LIKE glycosyltransferases are critical for glucuronoxylan biosynthesis during secondary cell wall formation. Plant J (2009) 57:718–731.[CrossRef][Web of Science][Medline]
York WS, van Halbeek H, Darvill AG, Albersheim P. Structural analysis of xyloglucan oligosaccharides by 1H-NMR spectroscopy and fast-atom-bombardment mass spectrometry. Carb. Research (1990) 200:9–31.
Yulia D. Characterisation of DCB-habituated suspension cultured cells of barley (Hordeum vulgare L. cv Schooner). (2006) Ph.D. thesis, School of Botany, University of Melbourne Victoria, Australia.
Zabotina O, Malm E, Drakakaki G, Bulone V, Raikhel N. Identification and preliminary characterization of a new chemical affecting glucosyltransferase activities involved in plant cell wall biosynthesis. Molecular Plant (2008) 1:977–989.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||