Cellulose is the most abundant biopolymer on Earth. Three cellulose synthases (CESA4, CESA7 and CESA8) are necessary for cellulose production in the secondary cell walls of Arabidopsis. Little is known about how expression of these CESA genes is regulated. We recently identified a cis-regulatory element (M46RE) that is recognized by MYB46, which is a master switch for secondary wall formation in Arabidopsis. A genome-wide survey of promoter sequences for the presence of M46REs led to the hypothesis that MYB46 may function as a direct regulator of all three secondary wall-associated cellulose synthase genes: CESA4, CESA7 and CESA8. We tested this hypothesis using several lines of experimental evidence. All three CESA genes are highly up-regulated by both constitutive and inducible over-expression of MYB46 in planta. Using a steroid receptor-based inducible activation system, we show that MYB46 directly activates transcription of the three CESA genes. We then used an electrophoretic mobility shift assay and chromatin immunoprecipitation analysis to confirm that MYB46 protein directly binds to the promoters of the three CESA genes both in vitro and in vivo. Furthermore, ectopic up-regulation of MYB46 resulted in a significant increase of crystalline cellulose content in Arabidopsis. Taken together, we have identified MYB46 as a transcription factor that directly regulates all three secondary wall-associated CESA genes. Yeast one-hybrid screening identified additional transcription factors that regulate the CESA genes. However, none of the putative regulators appears to be regulated by MYB46, suggesting the multi-faceted nature of transcriptional regulation of secondary wall cellulose biosynthesis.
Cellulose, a hydrogen-bonded β-1,4-linked glucan microfibril, is synthesized by multimeric cellulose synthase (CESA) complexes at the plasma membrane (Somerville, 2006). In plants, two distinct groups of CESAs (each consisting of at least three isoforms) are preferentially and coordinately expressed during primary and secondary cell-wall deposition (Endler and Persson, 2011). The Arabidopsis genome contains ten CESA genes (Pear et al., 1996; Richmond and Somerville, 2001). Analyses of various cellulose synthesis mutants revealed that CESA1, CESA2, CESA3, CESA5, CESA6 and CESA9 are associated with CESA complexes that are active during primary wall formation (Arioli et al., 1998; Fagard et al., 2000; Scheible et al., 2001; Desprez et al., 2002, 2007; Persson et al., 2007), while CESA4, CESA7 and CESA8 are necessary for secondary wall cellulose biosynthesis (Turner and Somerville, 1997; Taylor et al., 1999, 2000, 2003; Doblin et al., 2002; Williamson et al., 2002). Unlike the primary wall CESA complex, the three secondary wall CESA subunits appear to be equally important in the function of the complex in xylem vessels and cannot substitute for each other (Gardiner et al., 2003). Several proteins, such as KORRIGAN, COBRA and KOBITO1, are also known to negatively affect the synthesis of cellulose when mutated or mis-regulated (Endler and Persson, 2011). Nonetheless, none of the proteins has been directly associated with the CESA complex, suggesting that their effects on cellulose synthesis may be indirect. Recently, a cellulose synthase-interactive protein (CSI1) was identified as a non-CESA component of the CESA complexes (Gu et al., 2010). So far, CESAs appear to be the only group of proteins with the known ability to synthesize new cellulose molecules, which may be rate-limiting. However, little is known about how expression of secondary wall-associated CESA genes is regulated.
Formation of the secondary wall requires coordinated transcriptional activation of the genes involved in biosynthesis of secondary wall components such as cellulose, hemicellulose and lignin. Recent studies on NAC and MYB transcription factors have provided insights into the complex process of transcriptional regulation of secondary wall biosynthesis (Mitsuda et al., 2005, 2007; Ko et al., 2007, 2009; Zhong et al., 2007, 2008; Demura and Ye, 2010). Several MYB transcription factors have also been identified as important regulators of secondary wall biosynthesis in Arabidopsis. MYB46, a direct target of ANAC012/SND1, has been reported to be a master regulator of secondary wall formation in Arabidopsis thaliana (Zhong et al., 2007; Ko et al., 2009). Over-expression of MYB46 results in ectopic deposition of secondary walls in cells that are normally parenchymatous, while suppression of its function reduces secondary wall thickening (Zhong et al., 2007; Ko et al., 2009, 2012). To obtain a better understanding of MYB46-mediated transcriptional regulation, we identified a cis-acting regulatory motif (named M46RE) that is recognized by MYB46 (Kim et al., 2012). Genome-wide analysis of promoter sequences in Arabidopsis revealed that many MYB46-induced secondary wall biosynthetic genes, including all three secondary wall cellulose synthases, have one or more M46REs in their promoter region. Here, we provide experimental evidence in support of our hypothesis that MYB46 directly regulates the expression of three secondary wall cellulose synthases (CESA4, CESA7 and CESA8) in Arabidopsis plants. Furthermore, the yeast one-hybrid screening identified additional transcription factors that activate expression of the CESAs.
Expression of CESA4, CESA7 and CESA8 is up-regulated by MYB46
A genome-wide survey of promoter sequences for the MYB46-responsive cis-regulatory element M46RE led to the hypothesis that MYB46 may function as a direct regulator of all three secondary wall-associated cellulose synthase genes: CESA4 (At5g4403), CESA7 (At5g17420) and CESA8 (At4g18780) (Kim et al., 2012). This hypothesis is supported by three observations: (i) each gene has multiple M46REs in the promoter region, (ii) expression of all three genes is up-regulated by MYB46, and (iii) the genes are all co-expressed with MYB46. As a step towards verifying this hypothesis, we performed real-time PCR to examine the expression pattern of these CESA genes in transgenic Arabidopsis plants showing either constitutive or inducible over-expression of MYB46. The phenotypes of the transgenic Arabidopsis plants used in this analysis (Figure 1a) were consistent with those previously reported (Ko et al., 2009). Real-time PCR analysis showed that all three CESA genes were up-regulated more than sixfold by constitutive or inducible over-expression of MYB46 (Figure 1). The expression patterns of the CESA genes were correlated with MYB46 expression levels (Figure 1).
Transcription of CESA4, CESA7 and CESA8 is directly activated by MYB46
To investigate whether MYB46 can directly activate transcription of these three CESA genes, we used a steroid receptor-based inducible activation system. In this system, a transcription factor fused to a steroid-binding domain is sequestered in the cytoplasm by binding to a cytoplasmic complex. Upon steroid treatment, the transcription factor is released from the complex and enters the nucleus to regulate the expression of downstream target genes. Coupled with a protein synthesis inhibitor, this steroid-mediated activation system has been widely used to identify direct targets of transcription factors in plants (Sablowski and Meyerowitz, 1998; Wagner et al., 1999; Baudry et al., 2004; Zhong et al., 2008).
In this study, we fused MYB46 with the regulatory region of the glucocorticoid receptor (MYB46-GR) and constitutively expressed the MYB46-GR fusion protein under the control of the CaMV 35S promoter in Arabidopsis leaf protoplasts (Figure 2a). As a positive control, we used the promoter sequence of a known direct target of MYB46, AtC3H14 (Kim et al., 2012), to drive a GUS reporter gene. Upon dexamethasone (DEX) treatment, the MYB46-GR chimeric protein activated the AtC3H14 promoter-driven GUS reporter gene (Figure 2a, b). However, the GUS activity induced by DEX-activated MYB46-GR was completely abolished by treatment with cycloheximide (CHX), an inhibitor of protein synthesis (Figure 2b). Transcription of the positive control C3H14 was clearly induced by DEX-activated MYB46-GR even with CHX pre-treatment (Figure 2c). Likewise, DEX-activated MYB46-GR activated expression of all three secondary wall CESA genes, regardless of CHX treatment (Figure 2d). This result indicates that MYB46 directly activates transcription of all three CESA genes tested.
MYB46 binds the promoters of CESA4, CESA7 and CESA8 genes
To confirm physical interaction of MYB46 protein with the promoter regions of CESA4, CESA7 and CESA8 genes, we performed electrophoretic mobility shift assays (EMSAs) using recombinant MYB46 proteins fused to glutathione S-transferase (GST-MYB46) and a CESA promoter fragment containing a M46RE motif (Figure 3). Specific binding of MYB46 to the 32P-labeled promoter fragments ProCESA4 (−248 to −69), ProCESA7 (−662 to −486), and ProCESA8 (−525 to −358) was established using non-labeled promoter fragments (e.g. ProCESA4_wt, Figure 3a) as a competitor (Figure 3b). The binding specificity was further confirmed by using non-labeled promoter fragments with single base mutations in the M46RE (e.g. ProCESA4_m1 or m2) as a competitor. As expected, the MYB46 protein bound to the CESA promoter fragments but the GST protein alone did not (Figure 3b), demonstrating interaction of MYB46 protein with the promoters of the three CESA genes in vitro.
To further corroborate the interaction of MYB46 protein with the three CESA promoters in vivo, we performed a chromatin immunoprecipitation (ChIP) assay using transgenic Arabidopsis plants that expressed the GFP-tagged MYB46 gene under the control of the DEX-inducible promoter (Figure 4a). DEX treatment of plants expressing MYB46-GFP caused ectopic secondary wall thickening in leaf epidermal and mesophyll cells, a phenotype that is typical of ectopic MYB46 over-expression as described previously (Ko et al., 2009). This indicates that the MYB46-GFP fusion protein may be used for analysis of MYB46 binding sequences. Formaldehyde cross-linked chromatin from leaf tissues collected from 3-week-old transgenic plants with or without DEX treatment was isolated and fragmented. Chromatin fragments from plants without DEX treatment were used as a negative control. MYB46-GFP-bound DNA fragments were immunoprecipitated using GFP antibody and used as templates in quantitative real-time PCR analysis of CESA promoter sequences. All three CESA promoters were highly enriched (three to eightfold) compared to control DNA (Figure 4b). In the ChIP analysis, we used the promoter regions of C3H14 and MYB54 as positive and negative controls, respectively, as MYB54 is not directly targeted by MYB46 (Kim et al., 2012).
Together with the finding that expression of CESA4, CESA7 and CESA8 is directly activated by MYB46, these results provide both in vitro and in vivo evidence that MYB46 directly binds the promoter of all three secondary wall-associated CESA genes to activate their expression.
Increase of cellulose content by up-regulation of MYB46
As MYB46 directly regulates expression of CESA4, CESA7 and CESA8 genes, an increase in cellulose content is expected to result from MYB46-mediated up-regulation of the genes. To test this hypothesis, we measured the crystalline cellulose content of transgenic Arabidopsis plants showing either constitutive or inducible over-expression of MYB46 (Figure 5a). Compared to wild-type plants, two independent lines constitutively over-expressing MYB46 (OX8 and OX9) showed a substantial increase (approximately 30%) in crystalline cellulose content in leaf tissues of 3-week-old plants. Furthermore, 24 h induction of MYB46 resulted in an increase of up to 27% compared to non-induced plants (Figure 5a).
Crystalline cellulose accumulation in the stems of MYB46 over-expressors was visualized by immunohistological staining of cellulose using CBM3a, a carbohydrate-binding module for crystalline cellulose (Blake et al., 2006). Compared to wild-type plants, fluorescent signal driven by cellulose accumulation was more evident in the xylem and interfascicular regions of the two constitutive MYB46 over-expressors (OX8 and OX9) (Figure 5b). Furthermore, in both of the two constitutive over-expressors, fluorescent signals were detected in epidermal cells where secondary wall formation does not occur normally, and no signals were observed in wild-type plants (Figure 5b).
Taken together, these results confirm that ectopic up-regulation of MYB46 resulted in a substantial increase in cellulose content through activation of the three secondary wall CESA genes in plants.
Identification of additional transcription factors that regulate transcription of CESA4, CESA7 and CESA8
As secondary wall cellulose biosynthesis is critical in plant growth, we reasoned that there may be additional transcriptional regulators. To test this hypothesis, we performed yeast one-hybrid screening using the promoter sequences of CESA4, CESA7 and CESA8 as baits, and REGIA transcription factors (REgulatory Gene Initiative in Arabidopsis; Paz-Ares and the REGIA Consortium, 2002) that had been fused to the GAL4 activation domain (provided by Y. Kim and M.F. Thomashow, DOE-Plant Research Laboratory, Michigan State University, East Lansing, MI) as prey. Positive candidates were isolated under high-stringency conditions (SD medium lacking His, Ura and Trp and containing 40 mm 3-aminotriazole), and tested for β-galactosidase expression. The screening identified a total of 13 transcription factors (Table 1). MYB46 is not on the list because the REGIA transcription factor library does not include MYB46. All of the candidates on the list were identified by using the CESA4 promoter as bait, except ZAT7 (At3g46090), which bound to the CESA7 promoter (Table 1).
Table 1. Putative regulators of secondary wall CESA genes identified by yeast one-hybrid screening
Promoter regions used in this analysis were −668 to −1 bp (CESA4), −1000 to −1 bp (CESA7), and −1000 to −1 bp (CESA8) upstream of ATG.
To confirm physical interaction of the candidate transcription factors with the promoter of CESA4, we performed EMSAs using three randomly selected transcription factors from Table 1: MYB112 (At1g48000), WRKY11 (At4g31550) and ERF6 (At4g17490). Each of the three recombinant proteins fused to GST specifically bound the −666 to −294 bp region of the CESA4 promoter (Figure 6a). Furthermore, a transcriptional activation assay showed that all of the transcription factors activated transcription of CESA4 (Figure 6b, c). Interestingly, MYB112 activated all three CESA genes, even though it does not bind to CESA7 and CESA8 promoters, suggesting that this activation may be indirect.
Plasma membrane-bound CESA complexes are known to be the sites of cellulose biosynthesis (Delmer, 1999; Somerville, 2006; Taylor, 2008; Endler and Persson, 2011). Although non-CESA proteins may be directly associated with the complexes (Gu et al., 2010), CESA proteins appear to be the most essential component for cellulose synthesis. In the secondary walls of Arabidopsis plants, three CESA proteins (CESA4, CESA7 and CESA8) are involved in cellulose biosynthesis. Theire genes are specifically expressed in secondary wall-forming tissues (Taylor et al., 1999, 2000, 2003; Doblin et al., 2002; Williamson et al., 2002; Ko et al., 2004), suggesting that they may be under strict transcriptional control. However, little is known about the regulation of transcription of these CESA genes.
The MYB46 transcription factor and its orthologs have been shown to be a master switch that activates the entire secondary wall biosynthetic program (e.g. cellulose, hemicellulose and lignin biosynthesis) in Arabidopsis, poplar, rice and maize (Ko et al., 2009; McCarthy et al., 2010; Zhong et al., 2011). Therefore, it is reasonable to consider that transcription factors downstream of MYB46 may function as direct regulators of the biosynthetic pathways of cellulose, lignin and xylan. For instance, MYB63 and MYB58, two direct targets of MYB46, have been shown to be transcriptional activators of the lignin biosynthetic pathway during secondary wall formation in Arabidopsis (Zhou et al., 2009; Zhong and Ye, 2012). However, until now, no transcription factor has been identified as a direct regulator of either the cellulose or hemicellulose biosynthetic pathway.
The MYB46 is the first transcription factor identified as a direct regulator of all three secondary wall CESA genes (CESA4, CESA7 and CESA8). Several lines of evidence support our identification of this direct regulator. First, MYB46 directly activates the transcription of the three secondary wall CESA genes (CESA4, CESA7 and CESA8). Inducible expression of MYB46 was used to show up-regulation of the CESA genes in planta (Figure 1). This up-regulation was further confirmed in a transcriptional activation assay (Figure 2). The protoplasts used in the transcriptional activation assay experiments were from mesophyll cells that do not normally form secondary walls and therefore provided a background noise-free system for MYB46-mediated activation of secondary wall CESA genes. Second, MYB46 binds to the promoters of the three secondary wall CESA genes. We used EMSA experiments to show that MYB46 actually binds to the promoters of the CESA genes. This M46RE-mediated binding was further confirmed using mutated versions of M46RE as a competitor in the EMSA experiments (Figure 3). Furthermore, ChIP experiments followed by real-time PCR were used to provide in vivo confirmation of the interaction (Figure 4). Finally, up-regulation of MYB46 resulted in a substantial increase (up to 30%) in the crystalline cellulose content in leaves of transgenic Arabidopsis plants. This increase was observed for both constitutive and inducible up-regulation of MYB46 (Figure 5).
MYB46 regulation of the transcription of the secondary wall CESA genes appears to be mediated by the M46RE cis-regulatory motif. When the core sequence of the motif was altered, the mutated motif was no longer able to compete with labeled promoter sequences for binding the MYB46 protein in EMSA experiments. The MYB83 transcription factor, a close homolog of MYB46, functions redundantly with MYB46 in regulation of secondary wall biosynthesis in Arabidopsis. Double T-DNA knockout mutations of MYB83 and MYB46 caused a lack of secondary walls in vessels, and arrested in plant growth after development of one to two pairs of small leaves, followed by wilting and subsequent death (McCarthy et al., 2009). MYB83 also recognizes and binds to the M46RE element in vitro (Figure S1), further supporting the hypothesis that M46RE may play a role in transcriptional regulation of the secondary wall CESA genes.
As secondary wall cellulose synthesis is very critical to the plant's survival, it is not likely that MYB46 is the only direct regulator for secondary cell-wall cellulose synthases. In fact, MYB46-induced expression levels of the three CESA genes are up to fivefold higher in planta than those in the protoplast-based transcriptional activation experiments (Figures 1 and 2). This finding suggests that additional regulators acting in concert with MYB46 may be involved in transcriptional regulation of secondary wall CESA genes. Indeed, yeast one-hybrid screening using the promoter sequences of CESA4, CESA7 and CESA8 as baits identified multiple transcription factors that bind to the promoter sequences (Table 1). The results of EMSA and transcriptional activation assay experiments with three candidate regulators (i.e. MYB112, WRKY11 and ERF6) confirmed that they bind to the CESA4 promoter and cause transcriptional activation of CESA4 (Figure 6). However, none of these regulators identified from the yeast one-hybrid screening appears to be involved in the MYB46-mediated regulation pathway because their expression is neither altered by MYB46 nor are they co-expressed with MYB46 (Ko et al., 2009; Kim et al., 2012). The yeast one-hybrid assay did not identify any transcription factors that bind to the CESA7 and CESA8 promoters, with the exception of ZAT7, which binds to the CESA7 promoter. As the yeast expression library used in the study covers only 976 transcription factors, approximately half of the estimated number of transcription factors in Arabidopsis, it is possible that there may be additional transcription factors that bind to the promoters of the CESA genes. Recently, Ohashi-Ito et al. (2010) performed EMSA and ChIP analysis to demonstrate binding of VND6 to the tracheary element-specific cis-element TERE (Pyo et al., 2007) in the promoter of CESA4. VND7 was also suggested to be a direct regulator of CESA4 and CESA8 (Yamaguchi et al., 2011). However, none of the CESA genes was directly induced by estradiol-activated VND7 (Zhong et al., 2010), indicating that direct regulation of CESA genes by VND7 requires further confirmation. Taken together, the presence of multiple regulators supports the notion that the transcriptional regulation of cellulose biosynthesis is multifaceted and complex.
The finding that MYB46 directly regulate all three secondary wall CESA genes through the M46RE motif will facilitate research efforts aimed at pathway-specific regulation of secondary wall biosynthesis and provide novel tools for targeted alterations of biomass in transgenic plants.
Plant materials and growth conditions
Arabidopsis thaliana ecotype Columbia (Col-0) was used in both wild-type and transgenic plant experiments. Plants were grown on soil in a growth chamber (16 h light/8 h dark) at 23°C.
RNA extraction and quantitative real-time PCR
Total RNA was extracted from liquid nitrogen-frozen samples using a plant RNeasy extraction kit (Qiagen, http://www.qiagen.com/). For quantitative real-time PCR analysis, total RNA was treated with DNase I and used for first-strand cDNA synthesis using SuperScript II reverse transcriptase (Invitrogen, www.invitrogen.com). Real-time PCR was performed using SYBR Premix Ex Taq™ (Takara, www.takara-bio.com) and an ABI Prism 7900HT sequence detection system (Applied Biosystems, http://www.appliedbiosystems.com/). The relative mRNA levels were determined by normalizing the PCR threshold cycle number of each gene to that of the ACT8 reference gene. Three biological replicates were used in the experiments.
Protein expression and purification
MYB46 was fused in-frame with GST and expressed in Escherichia coli strain Rosetta gami (Novagen, www.novagen.com). Expression of the recombinant GST-MYB46, GST-MYB112, GST-ERF6 and GST-WRKY11 proteins was induced by culturing the E. coli cells for 16 h at 16°C in LB (Luria-Bertani) medium supplemented with 0.3 mm isopropyl β-d-thiogalactopyranoside. The recombinant proteins for EMSA were purified using a MagneGST™ protein purification system (Promega, www.promega.com) according to the manufacturer's instructions.
Electrophoretic mobility shift assay (EMSA)
DNA fragments for EMSA were obtained by PCR amplification and labeled with [γ-32P]ATP using T4 polynucleotide kinase (NEB, www.neb.com). The end-labeled probes were purified using a Microspin S-200 HR column (GE Healthcare, www.gelifesciences.com). The labeled DNA fragments were incubated for 25 min with 50 ng GST-MYB46, GST-MYB112, GST-ERF6 or GST-WRKY11 in binding buffer [10 mm Tris pH 7.5, 50 mm KCl, 1 mm dithiothreitol, 2.5% glycerol, 5 mm MgCl2, 100 μg ml−1 BSA and 50 ng μl−1 poly(dI-dC)]. Polyacrylamide gel electrophoresis (5%) was used to separate recombinant protein-bound DNA fragments from the unbound ones. The gel was dried and placed in a film cassette and exposed to X-ray film (Kodak, www.kodak.com) overnight. Radioactive fragments were visualized by autoradiography.
Dexamethasone-inducible activation system for confirmation of direct targets
The full-length cDNA of MYB46 was fused to the N-terminus of the glucocorticoid receptor (GR) coding sequence and ligated between the CaMV 35S promoter and the nopaline synthase terminator in the pTrGUS vector (Ko et al., 2009). High-quality plasmids were prepared using a PureHeix™ Fast-n-Pure plasmid kit (Nanohelix, www.nanohelix.net). The MYB46–GR expression construct was introduced into Arabidopsis leaf protoplasts alone or together with the AtC3H14 promoter–GUS construct (Ko et al., 2009). The primers used for PCR amplification of the full-length MYB46, glucocorticoid receptor and AtC3H14 promoter are shown in Table S1. Preparation of Arabidopsis leaf protoplasts and transfection was performed as described previously (Sheen, 2001; Ko et al., 2009). To activate MYB46, the protoplasts were treated with 10 μm dexamethasone (DEX, Sigma, www.sigmaaldrich.com) for 5 h. The control protoplasts were mock-treated with the same concentration (0.01%) of ethanol used to dissolve DEX. To inhibit new protein synthesis, the protein synthesis inhibitor cycloheximide (2 μm) was added 30 min before addition of DEX (Zhong et al., 2008). After the treatments, the protoplasts were harvested for quantitative real-time PCR analysis and GUS activity analysis (Ko et al., 2009). The expression level of each gene in the control protoplasts without DEX treatment was set to 1, and three biological replicates were used in the experiments.
Transcriptional activation analysis
Preparation of Arabidopsis leaf protoplasts and transient transfection of reporter and effector constructs were performed as previously described (Ko et al., 2009). For the effector constructs, full-length cDNAs of the transcription factors were ligated between the CaMV 35S promoter and the nopaline synthase terminator after removing GUS from the pTrGUS vector. The reporter constructs were created by placing the CESA promoter fragments in front of the GUS reporter gene after removing the 35S promoter from the pTrGUS vector. The primers used for PCR amplification of full-length genes and promoters are listed in Table S1. Transfected protoplasts were lysed after 16 h incubation, and the soluble extract was used for the GUS analysis. In each experiment, the expression level of the GUS reporter gene in the protoplasts transfected with the reporter construct alone was used as the control. Three biological replicates were used in the experiments.
Chromatin immunoprecipitation analysis
The full-length cDNA of MYB46 was fused in-frame with GFP and ligated downstream of the GAL4 upstream activation sequence in the pTA7002 binary vector (Aoyama and Chua, 1997). The vector construct was used for Agrobacterium-mediated transformation of Arabidopsis thaliana (Col-0) plants.
The MYB46–GFP/pTA7002 transgenic plants were grown on soil for 3 weeks before DEX treatment. DEX (10 μm) was applied by spraying with 0.02% Silwet surfactant (Lehle Seeds, www.lehleseeds.com). Eight hours after DEX treatment, the above-ground portion of the plants was harvested and immediately cross-linked with 1% formaldehyde for 10 min under vacuum. The cross-linking was quenched in 0.125 m glycine for 5 min. The cross-linked samples were washed twice with deionized water and then ground into a fine powder in liquid nitrogen for extraction of chromatin. To extract chromatin, 2 g ground powder was resuspended in 30 ml of extraction buffer 1 (10 mm Tris/HCl pH 8.0, 0.4 m sucrose, 5 mm 2-mercaptoethanol, 1 mm phenylmethanesulfonyl fluoride, 1 tablet per 50 ml protease inhibitor cocktail (Roche Applied Science, www.roche-applied-science.com) and 4 μg ml−1 pepstatin A) and filtered through two layers of Miracloth (www.miracloth.com) before centrifugation at 2500 g for 20 min at 4°C. The pellet was resuspended in 1 ml of extraction buffer 2 (10 mm Tris/HCl pH 8.0, 0.25 m sucrose, 10 mm MgCl2, 1% Triton X-100, 5 mm 2-mercaptoethanol, 1 mm phenylmethanesulfonyl fluoride, 1x protease inhibitor cocktail and 4 μg ml−1 pepstatin A) and centrifuged at 14 000 g for 10 min at 4°C. The pellet was resuspended in 300 μl of extraction buffer 3 (10 mm Tris/HCl pH 8.0, 1.7 m sucrose, 0.15% Triton X-100, 2 mm MgCl2, 5 mm 2-mercaptoethanol, 1 mm phenylmethanesulfonyl fluoride, 1x protease inhibitor cocktail and 4 μg ml−1 pepstatin A), and then layered on top of a cushion of 300 μl of extraction buffer 3 and centrifuged at 14 000 g for 1 h at 4°C. The chromatin pellet was resuspended in 500 μl ice-cold nuclei lysis buffer [50 mm Tris/HCl pH 8.0, 10 mm EDTA, 1% SDS, 1 mm phenylmethanesulfonyl fluoride, 1x protease inhibitor cocktail and 4 μg ml−1 pepstatin A), and sonicated to small fragments with a mean fragment size of 600–800 bp. The sonicated chromatin was diluted tenfold in ChIP dilution buffer (16.7 mm Tris/HCl pH 8.0, 1.1% Triton X-100, 1.2 mm EDTA, 167 mm NaCl, 1 mm phenylmethanesulfonyl fluoride, 1x protease inhibitor cocktail and 4 μg ml−1 pepstatin A), and pre-cleared by incubation with Protein A agarose beads (Roche Applied Science) for 1 h at 4°C. The pre-cleared chromatin was then incubated with 2 μg GFP antibody (Abcam, www.abcam.com) overnight at 4°C. The MYB46-GFP-bound chromatin was purified by incubation with Protein A agarose beads for 1 h at 4°C. The agarose beads were washed sequentially with 1 ml each of the following wash buffers by gently rocking on a shaker for 5 min at 4°C: (i) low-salt wash buffer (20 mm Tris/HCl pH 8.0, 150 mm NaCl, 0.2% SDS, 0.5% Triton X-100 and 2 mm EDTA), (ii) high-salt wash buffer (20 mm Tris/HCl pH 8.0, 500 mm NaCl, 0.2% SDS, 0.5% Triton X-100 and 2 mm EDTA), (iii) LiCl wash buffer (10 mm Tris/HCl pH 8.0, 0.25 m LiCl, 0.5% NP-40 (Thermo Scientific, www.thermoscientific.com/pierce), 0.5% sodium deoxycholate and 1 mm EDTA), and (iv) twice with TE buffer. The purified chromatin was eluted using 500 μl of elution buffer (1% SDS and 0.1 m sodium bicarbonate) at 65°C for 15 min with gentle agitation in a gyratory shaking incubator. The eluted chromatin was incubated with 0.2 m NaCl to reverse the protein-DNA cross-linking at 65°C overnight without agitation. Chromatin DNA was further purified by incubation with proteinase K (0.2 mg ml−1) for 1 h to remove any residual proteins before quantitative PCR analysis. Chromatin samples without GFP antibody immunoprecipitation were used as the control. C3H14 and MYB54 promoters were used as positive and negative controls, respectively. Three biological replicates were used in the experiments.
Labeling for CBM3a and immunofluorescence microscopy
Arabidopsis thaliana plants of ecotype Columbia (Col-0) (wild-type) and 35S::AtMYB46 transgenics plants in the Col-0 background were grown on soil in a growth chamber (16 h light/8 h dark) at 23°C for 8 weeks. Lower parts of the stems were fixed in FAA solution (50% ethanol, 5% glacial acetic acid and 3.7% formaldehyde) for 12 h at 4°C. After fixation, the fixed stems were embedded in paraffin and sectioned into 20 μm thin sections. The stem sections were labeled with crystalline cellulose-specific carbohydrate-binding module CBM3a as described by Blake et al. (2006). In brief, the sections were incubated in PBS containing 5% w/v milk protein (MP/PBS) and 10 μg ml−1 CBM3a for 1.5 h. Samples were then washed in PBS at least three times, and incubated with a 100-fold dilution of mouse anti-His monoclonal antibody (Sigma) in MP/PBS for 1.5 h. After washing with PBS, anti-mouse antibody linked to fluorescein isothiocyanate (anti-mouse FITC; Sigma) was applied for 1.5 h as a 50-fold dilution in MP/PBS in darkness. The samples were washed with PBS, mounted in ProLong® Gold anti-fade solution (Invitrogen), and observed on an Olympus Fluoview FV1000 confocal laser scanning microscope (Olympus, www.olympus-ims.com/microscope) fitted with a 488 nm argon laser and a 505–525 nm band-pass filter.
Cell-wall crystalline cellulose compositions were determined as described previously (Ko et al., 2007). In brief, 3-week-old rosette leaves were collected from soil-grown wild-type, 35S::AtMYB46 and DEX-inducible MYB46 over-expression plants, and ground in liquid nitrogen using a mortar and pestle. The ground samples (60–70 mg) were washed using 1.5 ml of 70% ethanol and centrifuged for 10 min at 10 000 g. The pellets were washed with 1.5 ml of chloroform:methanol (1:1 v/v) and again with 500 μl acetone. The remaining pellet was considered to contain the cell walls and was dried under nitrogen gas (N2). The cell-wall materials were re-suspended in 250 μl of 2 m trifluoroacetic acid (TFA) and hydrolyzed for 90 min at 121°C. After hydrolysis, samples were centrifuged for 10 min at 10 000 g to separate a TFA-soluble fraction (non-cellulosic monosaccharides) and a TFA-insoluble fraction (cellulose). The TFA-insoluble fraction was washed with 300 μl of 2-propanol and evaporated at 40°C. The washed samples were treated with Updegraff reagent (acetic acid/nitric acid/water, 8:1:2 v/v/v) and heated in an aluminum block for 30 min at 100°C (Updegraff, 1969). Then the samples were centrifuged for 10 min at 10 000 g. The pellets were washed once with water and then three times with acetone. Air-dried pellets were Seaman-hydrolyzed using 72% sulfuric acid for 30 min at room temperature (Selvendran and O'Neill, 1987). Final samples were precipitated for 5 min at 10 000 g, and analyzed by the anthrone method.
Yeast one-hybrid screening
We used a Gateway compatible yeast one-hybrid system as described by Deplancke et al. (2004). In brief, the promoter of the CESA4, CESA7 or CESA8 gene was cloned into a yeast one-hybrid reporter destination vector (pMW#2, Invitrogen) by Gateway cloning and integrated into the genome of yeast strain YM4271. Bait strains were verified by genomic PCR using promoter-specific primers (Table S1) and subsequent sequencing of the PCR amplicons. After a self-activation test, promoter bait strains growing on SD medium lacking His and Ura and containing 3-aminotriazole at 40 mm or higher concentration were used. The promoter bait strains were then transformed with the AD-TF library (obtained from Y. Kim and M.F. Thomashow, DOE-Plant Research Laboratory, Michigan State University, East Lansing, MI) and screened on SD medium lacking His, Ura and Trp and containing 40 mm 3-aminotriazole. Positive colonies were tested for β-galactosidase expression as described previously (Deplancke et al., 2004). Yeast colony PCR was performed to identify interacting transcription factors as described previously (Walhout and Vidal, 2001).
This work was funded by the Department of Energy Great Lakes Bioenergy Research Center (Department of Energy Office of Science BER DR-FC02-07ER64494) , in part by the Ministry of Education, Science and Technology of Korea via the World Class University Project at Chonnam National University (R31-2009-000-20025-0), and in part by the Basic Science Research Program through the National Research Foundation of Korea (2012-0002648) and a grant from Kyung Hee University (KHU-20100611). The authors would like to thank Michael Thomashow and Yong Sig Kim for providing the yeast expression library of Arabidopsis transcription factors, and Daniel Keathley (Department of Horticulture, Michigan State University), Kenneth Keegstra (DOE-Great Lakes Bioenergy Research Center, Michigan State University), Chandrashekhar Joshi (Department of Biology, Michigan Technological University) and Curtis Wilkerson (Department of Plant Biology, Michigan State University) for critical reading of the manuscript.