A Gibberellin-Mediated DELLA-NAC Signaling Cascade ...A Gibberellin-Mediated DELLA-NAC Signaling Cascade Regulates Cellulose Synthesis in RiceOPEN Debao Huang,a,1 Shaogan Wang,a,1

A Gibberellin-Mediated DELLA-NAC Signaling CascadeRegulates Cellulose Synthesis in RiceOPEN

Debao Huang,a,1 Shaogan Wang,a,1 Baocai Zhang,a Keke Shang-Guan,a Yanyun Shi,a Dongmei Zhang,a

Xiangling Liu,a Kun Wu,b Zuopeng Xu,a Xiangdong Fu,b and Yihua Zhoua,2

a State Key Laboratory of Plant Genomics, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing100101, Chinab State Key Laboratory of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology, ChineseAcademy of Sciences, Beijing 100101, China

ORCID IDs: 0000-0002-3239-7263 (B.Z.); 0000-0003-4560-9058 (D.Z.); 0000-0003-1998-7987 (K.W.); 0000-0001-9285-7543 (X.F.);0000-0001-6644-610X (Y.Z.)

Cellulose, which can be converted into numerous industrial products, has important impacts on the global economy. It haslong been known that cellulose synthesis in plants is tightly regulated by various phytohormones. However, the underlyingmechanism of cellulose synthesis regulation remains elusive. Here, we show that in rice (Oryza sativa), gibberellin (GA) signalspromote cellulose synthesis by relieving the interaction between SLENDER RICE1 (SLR1), a DELLA repressor of GA signaling,and NACs, the top-layer transcription factors for secondary wall formation. Mutations in GA-related genes and physiologicaltreatments altered the transcription of CELLULOSE SYNTHASE genes (CESAs) and the cellulose level. Multiple experimentsdemonstrated that transcription factors NAC29/31 and MYB61 are CESA regulators in rice; NAC29/31 directly regulatesMYB61, which in turn activates CESA expression. This hierarchical regulation pathway is blocked by SLR1-NAC29/31interactions. Based on the results of anatomical analysis and GA content examination in developing rice internodes, thissignaling cascade was found to be modulated by varied endogenous GA levels and to be required for internode development.Genetic and gene expression analyses were further performed in Arabidopsis thaliana GA-related mutants. Altogether, ourfindings reveal a conserved mechanism by which GA regulates secondary wall cellulose synthesis in land plants and providea strategy for manipulating cellulose production and plant growth.

INTRODUCTION

Cellulose, comprising parallel b-1,4-glucans, is a major polymerof plant cell walls. It not only represents the most abundantsustainable resources in wide use in industrial production butalso plays essential roles in plant growth. The molecular structureof cellulose is simple, but the amount, the degree of polymeri-zation, the crystalline size, and the orientation vary across dif-ferent cell types and developmental stages. Therefore, in landplants, cellulose, synthesized by cellulose synthases (CESAs) atthe plasma membrane, has diverse physicochemical properties.CESA-related synthesis is tightly regulated (Somerville, 2006; Liet al., 2014). Multiple lines of genetic evidence have illustratedthat different sets of CESAs form complexes required for primaryand secondary wall cellulose production (Desprez et al., 2002;Doblin et al., 2002; Gardiner et al., 2003; Taylor et al., 2003). Thecorresponding genes, CESAs, are critical downstream targets incellulose regulatory networks.

Significant effort has been focused on the transcriptional reg-ulation of secondary wall biosynthesis. A detailed model consisting

of transcription factors (TFs) from the NAC and MYB families hasbeen established in the eudicot Arabidopsis thaliana (Zhong andYe, 2007; Demura and Ye, 2010; Zhao and Dixon, 2011). Cur-rently, through protein-DNA screening, an elaborate hierarchynetwork of TFs and secondary wall metabolic genes has beenrevealed in Arabidopsis, offering an opportunity for understandingthe regulatory complexity of cell wall formation under abioticstresses (Taylor-Teeples et al., 2015). However, the networks forgrasses, which possess a distinct patterning and composition ofthe secondary wall, are largely unknown (Zhong et al., 2011;Handakumbura and Hazen, 2012). The CESA genes that are re-sponsible for synthesizing cellulose, a major component of sec-ondary walls, have been proposed to be regulated by many TFs.Surprisingly few direct regulators of CESAs have been identified.Arabidopsis MYB46 and its close homolog MYB83 have beenreported to bind the promoter of secondary wall CESAs via theMYB46-responsive cis-regulatory element (Kim et al., 2012;Zhong and Ye, 2012). However, MYB46 and MYB83 may regulateoverall secondary wall synthesis, because manipulating their ex-pression alters the global cell wall composition of Arabidopsis,including the content of cellulose, hemicellulose, and lignin (Koet al., 2009; Zhong and Ye, 2012; Kim et al., 2013). The myb46myb83 double mutant showed severe abnormalities in secondarywall thickening and plant growth, suggesting that these TFs arecentral regulators of secondary wall biosynthesis in both fibersand vessels (Zhong et al., 2007a; McCarthy et al., 2009). Furtherstudies have revealed that MYB46 and MYB83 are the directtargets of several secondary wall NAC regulators, including

1 These authors contributed equally to this work.2 Address correspondence to [email protected] author responsible for distribution of materials integral to the findingspresented in this article in accordance with the policy described in theInstructions for Authors (www.plantcell.org) is: Yihua Zhou ([email protected]).OPENArticles can be viewed online without a subscription.www.plantcell.org/cgi/doi/10.1105/tpc.15.00015

The Plant Cell, Vol. 27: 1681–1696, June 2015, www.plantcell.org ã 2015 American Society of Plant Biologists. All rights reserved.

Dow

nloaded from https://academ

ic.oup.com/plcell/article/27/6/1681/6096557 by guest on 10 July 2021

http://orcid.org/0000-0002-3239-7263http://orcid.org/0000-0002-3239-7263http://orcid.org/0000-0002-3239-7263http://orcid.org/0000-0003-4560-9058http://orcid.org/0000-0003-4560-9058http://orcid.org/0000-0003-1998-7987http://orcid.org/0000-0003-1998-7987http://orcid.org/0000-0003-1998-7987http://orcid.org/0000-0001-9285-7543http://orcid.org/0000-0001-9285-7543http://orcid.org/0000-0001-9285-7543http://orcid.org/0000-0001-6644-610Xhttp://orcid.org/0000-0001-6644-610Xhttp://orcid.org/0000-0001-6644-610Xhttp://orcid.org/0000-0001-6644-610Xhttp://orcid.org/0000-0002-3239-7263http://orcid.org/0000-0003-4560-9058http://orcid.org/0000-0003-1998-7987http://orcid.org/0000-0001-9285-7543http://orcid.org/0000-0001-6644-610Xmailto:[email protected]://www.plantcell.orgmailto:[email protected]:[email protected]://www.plantcell.org/cgi/doi/10.1105/tpc.15.00015http://www.plantcell.org

SECONDARY WALL ASSOCIATED-NAC DOMAIN PROTEIN1(SND1), NAC SECONDARY WALL THICKENING PROMOTINGFACTOR1 (NST1), VASCULAR-RELATED NAC DOMAIN6 (VND6),and VND7 (Ko et al., 2007; Zhong et al., 2007a, 2008; McCarthyet al., 2009). Mounting genetic proof has characterized these NACTFs as “master switches” for secondary wall formation in Arabi-dopsis (Kubo et al., 2005; Zhong et al., 2006, 2007b; Yamaguchiet al., 2008). Although WALLS ARE THIN1 (WAT1) might bea regulator of SND1 and NST1 in fiber cells (Ranocha et al., 2010),the upstream signals for secondary wall cellulose synthesis reg-ulation remain largely unknown.

Gibberellin (GA) is an important hormone for plant growth anddevelopment throughout the whole life cycle, including seedgermination, stem elongation, floral transition, and fruit de-velopment. Genetic analyses of GA-deficient and GA-responsemutants have revealed that the central step in GA action is toturn off the repressive effects of DELLAs (Peng et al., 1997; Itohet al., 2002; Sasaki et al., 2002; Ueguchi-Tanaka et al., 2005;Harberd et al., 2009). In the presence of GA, the GA-GIBBER-ELLIN INSENSITIVE DWARF1 (GID1)-DELLA complex stim-ulates the interaction of DELLAs with an F-box protein, resultingin the degradation of DELLAs (Fu et al., 2002; Murase et al.,2008; Hirano et al., 2010; Sun, 2011) and consequently theactivation of downstream-responsive processes. DELLAs havebeen found to inhibit GA-promoted growth by interacting withkey regulatory proteins and blocking their DNA binding ortransactivation activities. The first proteins identified to interactwith DELLAs are basic/helix-loop-helix-type TFs, which arePHYTOCHROME INTERACTING FACTORS (PIFs) that functionin Arabidopsis hypocotyl elongation (de Lucas et al., 2008;Feng et al., 2008). BRASSINAZOLE RESISTANT1 (BZR1) andJASMONATE ZIM-DOMAIN are also DELLA-interacting pro-teins. These interactions integrate GA with various signals forcoordinated regulation of plant growth and defense (Hou et al.,2010; Bai et al., 2012). Although studies of recently identifiedDELLA-interacting proteins have revealed the mechanismsof several GA-mediated responses (Arnaud et al., 2010;Cheminant et al., 2011; Feurtado et al., 2011; Yu et al., 2012;Marín-de la Rosa et al., 2014), much work is required to obtaina clear understanding of the numerous roles played by GA.

Understanding the role of GA in stem height determinationsparked a great revolution in agriculture. The introduction ofthe semidwarfing genes Reduced height-1 and semi-dwarf1(sd1) into cereal crops improved crop architecture and lodgingresistance, leading to a huge increase in grain yields during the1960s (Peng et al., 1999; Hedden, 2003). Cellulose also hasmajor impacts on stem length and lodging resistance. How-ever, the genetic link between GA signaling and cellulosesynthesis remains elusive. It is unclear whether GA is an up-stream signal for cellulose synthesis and, if so, how plantstransmit GA signals downstream to the CESA genes. Here, weidentify a GA signaling cascade that modulates secondary wallcellulose synthesis in rice (Oryza sativa). SLENDER RICE1(SLR1), a key repressor of GA signaling, interacts directly withNACs, which in turn inhibits an NAC-MYB-CESA signalingcascade. These findings reveal a conserved mechanism forthe regulation of secondary wall cellulose synthesis in landplants.

RESULTS

The GA-DELLA Module Regulates Cellulose Synthesis

As the cellulose level is correlated with mechanical strength, wewere curious to learn whether the lodging-resistant sd1, a GA-deficient rice mutant harboring a mutation in GA 20-oxidase2, hasa higher amount of cellulose than the wild type. To avoid datadiscrepancies from varied developmental stages and growthhabits of plants, all of the mutants analyzed in this study share thesame genetic background with the corresponding wild-type plantsand were used for comparison during growth stages matched tothe wild type. We first compared the anatomical structure of themature internodes of sd1 and wild-type plants. Unexpectedly, theepidermal layer and sclerenchyma cell wall in sd1 were signifi-cantly thinner than in the wild type (Figure 1; Supplemental Figure1). Composition analysis revealed that sd1 had reduced cellulosecontent (Figure 2A), indicating a potential link between GA signalsand cellulose synthesis. To obtain further genetic evidence,another GA-deficient mutant (dwarf18-Akibare-waisei dwarf[d18-AD]) and two GA-response mutants (gid1 and slr1) were in-vestigated. D18 encodes GA 3-oxidase2, an enzyme for GA bio-synthesis (Sakamoto et al., 2004). Because the internodes in d18-AD are too short, we performed cellulose content and anatomicalexamination using the leaf sheaths. Anatomical analysis showedthat d18-AD had fewer sclerenchyma cells and reduced thicknessin the sclerenchyma cell walls, whereas slr1, which has a mutationin the rice DELLA gene, resulting in a GA-constitutive response(Itoh et al., 2002), had more sclerenchyma cells and increased wallthickness (Figures 1B to 1E; Supplemental Figure 1). These changesconsistently resulted in decreased cellulose content in d18-AD andan increased cellulose level in slr1 (Figure 2A). CESA4, CESA7, andCESA9 are responsible for secondary wall cellulose synthesis(Tanaka et al., 2003). We then explored the expression levels ofthree secondary wall CESAs in these mutants by quantitative RT-PCR (qRT-PCR). These CESAs were downregulated to varyingdegrees in sd1, d18-AD, and gid1 but were significantly upregu-lated in slr1 (Figure 2B). gid1 is a GA-insensitive mutant resultingfrom mutations in the gene encoding a soluble receptor for GA(Ueguchi-Tanaka et al., 2005). Most likely because the examinedgid1 is a weak allele, cellulose deficiency was not observed, al-though the wall thickness of sclerenchyma cells was slightly re-duced and CESA7 and CESA9 transcription was accordinglyrepressed (Figures 1D, 1E, 2A, and 2B).To investigate whether GA can modulate cellulose synthesis

in rice, we treated the wild-type rice plants with exogenous GAand examined CESA expression in the internodes. The expres-sion of CESAs was significantly upregulated by GA (Figure 2C).Rice PHYTOCHROME-INTERACTING FACTOR-LIKE PRO-TEIN1 (PIL1) and XYLOGLUCAN ENDOTRANSGLUCOSYLASE/HYDROLASE8 (XTH8), two genes not responsive and responsiveto GA treatment, respectively, were used as a negative and apositive control for the examination (Jan et al., 2004; Todaka et al.,2012). Our previous work showed that bc11 is a missense mutantof CESA4 and that overexpression of the mutated CESA4 in bc11increased the cellulose content (Zhang et al., 2009). The heading-stage wild-type and bc11 plants were treated with GA. Compo-sition analysis revealed that both the GA-treated wild-type and

1682 The Plant Cell

Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

the bc11 plants had increased cellulose abundance (Figure 2D),indicating that GA can indeed promote CESA expression. All ofthese data suggested that GA and the SLR1-mediated signalingpathway are required for cellulose synthesis.

SLR1 Interacts with the Top-Layer TFs for SecondaryWall Formation

We then addressed whether the role of GA in regulating cellulosesynthesis is direct or indirect. Because SLR1 is a key componentof GA signaling and mediates the downstream responses throughprotein interactions (Davière and Achard, 2013), we exploredwhether a direct interaction exists between SLR1 and the TFs thatregulate secondary wall formation. According to the hierarchicalrelationship of the Arabidopsis TFs responsible for secondary wallregulation (Zhong and Ye, 2007; Demura and Ye, 2010), we hy-pothesized that rice might have a similar transcription regulatorynetwork, in which several NACs are located in the top layer. Thecandidate NACs were screened for coexpression with the sec-ondary wall CESAs (Supplemental Table 1) and through homol-ogous search with the identified Arabidopsis NACs (SupplementalFigure 2 and Supplemental File 1). We selected two NAC candi-dates (NAC29 and NAC31) that might be involved in secondarywall formation in rice to examine the interaction with SLR1. Wefirst investigated whether SLR1 and the above two NACs are

expressed in the same cell types. Parenchyma and sclerenchymacells were harvested from young internodes via laser-capturemicrodissection and subjected to qRT-PCR analysis (Figure 3A).Although transcripts of SLR1 were detected in the parenchymacells, they were also present in the sclerenchyma cells, where thetwo NACs were preferentially expressed (Figure 3B). BecauseCESA4 is predominantly expressed in sclerenchyma cells, it wasused as a control to monitor the success of microdissection.Split-luciferase complementation assays then revealed an in-teraction between SLR1 and the two NACs in Nicotiana ben-thamiana leaves (Figures 3C and 3D). Using the transgenic riceplants that expressed green fluorescent protein (GFP)-taggedNAC29 or NAC31 (NAC29/31-GFP), we found that both NACproteins were coimmunoprecipitated with SLR1 in vivo (Figure3E). Yeast two-hybrid assays further illustrated that SLR1 inter-acts directly with both NACs via the C terminus (Figure 3F), whichcontains the GRAS domain required for repressor function (Itohet al., 2002). To narrow down the interacting domain of NACs,NAC29 and NAC31 were split into an N-terminal DNA bindingdomain (BD) and a C-terminal activation domain. Bimolecularfluorescence complementation (BiFC) analysis in N. benthamianarevealed a direct interaction between SLR1 and the BDs of thetwo NACs in the nuclei of living plant cells (Figures 3G and 3H).Therefore, GA directly modulates cellulose synthesis via SLR1-NAC interaction.

Figure 1. Rice GA-Related Mutants Exhibited Altered Secondary Walls.

(A) to (C) Comparison of the cross sections of mature sd1-8 and slr1-6 internodes and 2-month-old d18-AD leaf sheaths with the corresponding wild-type plants. Brackets indicate the width of the epidermal sclerenchyma layer. Dashed lines and arrows indicate the alterations in sclerenchyma cells(SC) and secondary wall thickness. V, vascular bundles. Bars = 50 mm.(D) Observation of the sclerenchyma cells from the internodes and leaf sheaths of the indicated GA-related mutants via scanning electron microscopy.Images are as follows: 1, cv Nanjing6 (wild type); 2, sd1-8; 3, gid1-20; 4, cv Zhonghua11 (wild type); 5, slr1-6; 6, cv Akibare (wild type); 7, d18-AD. Bars =5 mm.(E) Quantification of the thickness of sclerenchyma cell walls examined in (D). Error bars indicate SE (n = 50). *P < 0.01 determined by Student’s t test.

DELLA-NAC Modulates Cellulose Synthesis 1683

Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

Hierarchical TFs Regulate Secondary Wall CelluloseSynthesis in Rice

To determine whether the SLR1-interacting NAC29 and NAC31are in the cellulose regulation pathways, NAC29- and NAC31-overexpressing (Ox) rice plants were generated. The transgenicplants had thick internodes and significantly increased cellulosecontent (Figure 4A; Supplemental Figures 3A and 3B). The totallignin amount and xylose content were not significantly altered(Supplemental Table 2). In addition, three secondary wall CESAsin the overexpressing plants were upregulated (Figure 4B).Taken together with the transactivation activity and subcellularlocalization data (Supplemental Figures 3C to 3F), NAC29 andNAC31 are CESA regulators.

The downstream components of NAC29 and NAC31 are oftenMYBs (Zhong et al., 2011). Coexpression analysis identifiedseveral secondary wall CESA coexpressed R2R3-type MYBs(Supplemental Table 1), most of which were GA-responsive(Supplemental Figure 4A). To identify which MYBs functiondownstream of NAC29 and NAC31, the expression levels of theirrespective genes were examined in the NAC29 Ox and NAC31Ox rice plants by qRT-PCR. The upregulated levels indicated thatcertain MYBs are in the NAC29/31-mediated signaling pathways(Figure 4C). We chose MYB61 as a representative for furtherfunctional characterization. The MYB61 Ox plants had thick in-ternodes, upward curved leaves, and significantly increasedcellulose content (Figure 4D; Supplemental Figures 4B and 4C)but unchanged total lignin and xylose contents (SupplementalTable 2). Consistent with the increased cellulose content,MYB61Ox had increased secondary wall thickness (Supplemental

Figures 4D to 4F) and more secondary wall CESA transcripts andproteins (Figure 4E; Supplemental Figure 4G). Moreover, theMYB61-dominant repression plants that overexpressed a con-struct harboring an MYB61-SRDX fusion (Hiratsu et al., 2003)exhibited decreased cellulose content (Figure 4F). Thus, MYB61indeed participates in the NAC-CESA regulatory pathway.

MYB61 Directly Activated by NAC29/31 Targets GAMYBMotifs of Secondary Wall CESAs

To find molecular support for the above signaling pathway, weaddressed whether NAC29 and NAC31 regulate MYB61 tran-scription. Transactivation analysis revealed significant luciferaseactivities when the cauliflower mosaic virus (CaMV) 35S pro-moter driving either NAC29 or NAC31 was coexpressed with theMYB61 promoter driving luciferase reporter in rice protoplasts(Figure 5A), demonstrating that NAC29 and NAC31 activateMYB61 expression. Secondary wall-related NAC proteins oftenbind to the SNBE element in the targeted genes (Zhong et al.,2011; Handakumbura and Hazen, 2012). To determine whetherthe activation by NAC29 and NAC31 is direct, MYB61 promoterfragments containing two SNBE elements were examined in anelectrophoretic mobility shift assay (EMSA). SNBE2 was boundby the recombinant NAC29 and NAC31 proteins fused to glu-tathione S-transferase (GST-NAC29 and GST-NAC31), whichresulted in a mobility shift; GST alone, as a negative control, didnot cause a mobility shift (Supplemental Figures 5A and 5B). Thebinding ability to SNBE2 was gradually suppressed by the additionof increasing amounts of unlabeled probes (Figure 5B) and wascompletely abolished when SNBE2 was mutated (Supplemental

Figure 2. GA Promotes CESA Transcription and Cellulose Synthesis in Rice.

(A) Measurement of cellulose content in the GA-related mutants and corresponding wild-type plants. AIR, alcohol-insoluble residues.(B) qRT-PCR analyses indicating the altered expression levels of secondary wall CESAs in the GA-related mutants.(C) qRT-PCR analyses of secondary wall CESAs in the internodes from control plants (–GA) and wild-type plants treated for 6 h with 10 mM GA3 (+GA).PIL1 and XTH8 were used as a negative and a positive control, respectively.(D) Measurement of cellulose content of internodes from control (–GA) and 10 mM GA3-treated (+GA) wild-type and bc11 plants.Rice TP1 was used as an internal control in (B) and (C). Error bars indicate the SD of three biological repeats. *P < 0.01 determined by Student’s t test.

1684 The Plant Cell

Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

Figure 5C). Thus, NAC29 and NAC31 bind to the SNBE motif anddirectly regulate MYB61 expression.

We further investigated whether MYB61 can activate CESAtranscription. After proving that MYB61 is a functional TF,MYB61or MYB61 fused to the glucocorticoid receptor (MYB61-GR) wasplaced under the control of the CaMV 35S promoter (Figure 5C;Supplemental Figures 6A to 6C). Cotransfecting rice protoplastswith the resulting constructs and a reporter driven by individualCESA promoters resulted in significant luciferase activity(Supplemental Figure 6D). In the GR-based inducible system,luciferase activity induced by the addition of dexamethasone(DEX) was completely quenched by treatment with the protein

synthesis inhibitor cycloheximide (CHX) (Figure 5D), but thetranscription of three secondary wall CESAs was activated byDEX even in the presence of CHX (Figure 5E), demonstrating thatMYB61 directly induces the expression of secondary wallCESAs. EMSA was conducted to confirm the above result. MYB61proteins fused to maltose binding protein (MBP-MYB61) werefound to bind to the biotin-labeled CESA promoter fragments(Supplemental Figures 6E to 6G). To verify the interaction ofMYB61 with CESA promoters in vivo, we performed a chromatinimmunoprecipitation (ChIP) assay in wild-type and MYB61 Oxplants. MYB61-bound fragments enriched by immunoprecipi-tation with anti-MYB61 antibody were used for quantitative PCR

Figure 3. SLR1 Interacts with NAC29 and NAC31.

(A) Microdissection of sclerenchyma cells (SC; indicated by yellow dashed lines) and parenchyma cells (PC; indicated by the red dashed line) in thecross sections of rice internodes.(B) qRT-PCR analysis of the indicated genes in cells harvested in (A). Rice HNR and TP1 were used as internal controls for the amplification. Error barsindicate the SD of three biological repeats.(C) and (D) Split-luciferase complementation assay showing the interaction between SLR1 and Os-NAC29 or Os-NAC31 in N. benthamiana leavesagroinfiltrated with the construct combinations shown in (C). Bars = 1 cm.(E) Coimmunoprecipitation of SLR1 with NAC29 and NAC31. The proteins extracted from NAC29-GFP and NAC31-GFP transgenic plants wereimmunoprecipitated (IP) with anti-GFP antibody and blotted with anti-GFP or anti-SLR1 antibody.(F) Yeast two-hybrid assay revealing the interaction of the C terminus of SLR1 with NAC29 and NAC31. The transformed yeast cells were grown on SD-Trp/-Leu/-His/-Ade/+3AT medium.(G) and (H) BiFC analysis of the interaction between SLR1 and the BD of NAC29 and NAC31 in N. benthamiana. Infiltrations with the empty vectors wereused as negative controls (H). 49,6-Diamidino-2-phenylindole (DAPI) was used to visualize the nuclei. Merge indicates merged images of enhancedyellow fluorescent protein (YFP) and DAPI. DIC, differential interference contrast. Bars = 0.2 cm.


Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

analysis. Three CESA promoter fragments were significantly en-riched in MYB61 Ox plants (Figure 5F). Sequence searchingwithin the bound fragments identified a conserved element an-notated as GAMYB, a motif bound by GA-responsive MYBs(Woodger et al., 2003). Then, 34-bp oligonucleotides containinga wild-type motif and a single base-mutated motif (GAMYB andGAMYBm; Supplemental Figure 6E) were synthesized. TheEMSA revealed competition or abolition of binding with unlabeledGAMYB or GAMYBm (Figure 5G). Moreover, a transactivationassay performed by expressing three copies of a 20-bp wild-typecore motif of GAMYB revealed a significantly higher activity thanexpressing the mutated version in protoplasts (Figures 5H and5I). Therefore, GAMYB is an MYB61 binding element.

Taken together, the above studies have identified one regu-latory pathway, NAC29/31-MYB61-CESA, for secondary wallcellulose synthesis in rice.

The Interaction between SLR1 and NAC29/31 Blocks theNAC-MYB61-CESA Regulatory Pathway

We then addressed the effect of the SLR1-NAC29/31 interactionon the identified signaling cascade. Because NAC29 and

NAC31 interact with SLR1 via the BD, their ability to bind theMYB61 promoter may be affected. Transactivation analysisshowed that luciferase activity in the cells coexpressing a re-porter containing the MYB61 promoter driving luciferase and aneffector containing NAC29 or NAC31 was significantly re-pressed by the additional coexpression of SLR1. This repressionwas suppressed by the application of exogenous GA (Figures6A and 6B), demonstrating that SLR1 inhibits the NAC29/31-mediated regulatory pathway. Additional proof of this seques-tration role was derived from EMSAs. Affinity-purified SLR1extracted from N. benthamiana leaves agroinfiltrated with theFLAG-SLR1 construct significantly abolished the binding of riceNAC29 and NAC31 to the MYB61 promoter (Figure 6C). Theeffect of the NAC29/31-SLR1 interaction on secondary wallCESAs was further determined by examining their transcripts inrice protoplasts expressing the CaMV 35S promoter drivingNAC29 or NAC31 and SLR1. The upregulated CESA levels incells expressing NAC29 or NAC31 were attenuated by the ad-ditional coexpression of SLR1. The application of GA againactivated the transcription of CESAs (Figure 6D). Altogether, thein vitro and in vivo evidence indicate that SLR1 directly interactswith NAC29 and NAC31 to sequester these factors and inhibits

Figure 4. Identifying the TFs Involved in Secondary Wall Cellulose Synthesis in Rice.

(A), (D), and (F) Measurement of the cellulose content in NAC29 Ox and NAC31 Ox (A), MYB61 Ox (D), and MYB61-SRDX rice plants (F). AIR, alcohol-insoluble residues.(B), (C), and (E) qRT-PCR examination of the expression of three secondary wall CESAs (B) and GA-inducible MYBs (C) in NAC29 Ox and NAC31 Oxand three secondary wall CESAs in MYB61 Ox (E). TP1 was used as an internal control.Error bars indicate the SD of three biological repeats. *P < 0.01 determined by Student’s t test.

1686 The Plant Cell

Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

Figure 5. Identifying the NAC-MYB61-Secondary Wall CESA Regulatory Pathway.

(A) Luciferase activities in rice protoplasts cotransfected with the constructs shown above. The transactivation activity was monitored by assaying theluciferase activity in the rice protoplasts, with the ones transfected with an empty effector construct defined as 1.(B) EMSA showing the competing binding of the NACs to SNBE fragments with an increasing amount of unlabeled DNA probe (10-fold [+] and 50-fold[++]).(C) Diagrams of the effector and reporter constructs used in (D) and (E).(D) Luciferase activities in protoplasts cotransfected with the constructs shown in (C) in the presence or absence of 10 mM DEX and/or 2 mM CHX.(E) qRT-PCR examination of the transcripts of three secondary wall CESAs in the protoplasts expressing the effector shown in (C).TP1 was used as an internal control. The luciferase activity (D) and expression level of CESAs (E) in the protoplasts without DEX treatment were set to 1.(F) ChIP-quantitative PCR analysis showing MYB61 binding to the CESA promoter sequences in vivo compared with samples without antibodyapplication. Actin1 was used as a negative control.


Dow



their binding to MYB61, which consequently blocks CESAtranscription.

This Signaling Machinery Is Required forInternode Development

Next, the above regulatory machinery was verified in a naturalphysiological event. The rice internode contains abundant sec-ondary walls and is a typical location for GA action. Therefore,we collected the developing second internodes from wild-typeplants when they were 9 cm in length and divided the internodesinto nine sections at 1-cm intervals. Anatomical and composi-tional analyses revealed that the secondary walls were graduallydeposited in the sclerenchyma cells from the bottom up

(Supplemental Figure 7), in accordance with the feature of in-tercalary growth, a lengthwise growth due to the activity of in-tercalary meristems in monocot plants. Cell elongation almostceased after section 3 upward from the bottom (Figure 7A).Secondary wall, including cellulose and lignin, initiated formationat section 2 and rapidly accumulated above this section (Figure7B). Therefore, rice internode development includes elongation(sections 1 to 3) and secondary wall formation (sections 2 to 9). Todefine GA’s physiological roles in the two processes, the en-dogenous GA levels were examined. Because section 9 is at themature stage and the relevant genes are inactive, the examina-tions were performed in the lower eight sections. The contentprofiles of GA1, GA19, GA20, GA24, and GA53 in the developing in-ternodes were similar: the concentration was highest in section 1

Figure 5. (continued).

(G) EMSA showing the binding between MYB61 and the wild-type GAMYB or a single base-mutated GAMYB (GAMYBm).(H) Diagram of the reporter, three copies of 20-bp GAMYB (Oligo) or GAMYBm (Oligom) driving the luciferase reporter, and the effector, the CaMV 35Spromoter driving MYB61.(I) Transactivation analysis by expressing the effector and reporter shown in (H).Error bars indicate the SD of three biological repeats.

Figure 6. Effect of SLR1-NAC29/31 Interaction on the NAC-MYB61-CESA Signaling Cascade.

(A) Diagrams of the effector and reporter constructs used in (B) and (D).(B) Luciferase activities in protoplasts cotransfected with the reporter and different combinations of effectors. The transactivation activity was moni-tored by assaying the luciferase activities, with the activity in protoplasts transfected with an empty effector construct defined as 1.(C) EMSA showing that direct binding of rice NAC29 and NAC31 to MYB61 was inhibited by the addition of increasing amounts of SLR1 proteinsextracted from N. benthamiana leaves agroinfiltrated with the FLAG-SLR1 construct.(D) qRT-PCR examination of the transcripts of three secondary wall CESAs in rice protoplasts expressing the different combinations of effectors shownin (A); 10 mM GA3 was added to the reactions coexpressing SLR1.Actin1 was used as an internal control. Error bars indicate the SD of three biological repeats.

1688 The Plant Cell

Dow




and dropped to a stable level in sections 3 to 8 (Figure 7C). GA4,GA9, and GA12 were difficult to detect except in section 1, wherethe concentrations of GA4 and GA9 were 0.136 0.02 and 0.0960.02 ng/g fresh weight, respectively. The above anatomical andcompositional data on each section indicate that the relativelyhigh GA level in the lower sections is essential for cell elongationand the initiation of secondary wall synthesis; the low GA level inthe upper sections may be required to maintain cellulose syn-thesis therein at a certain level.

To verify this conclusion, we investigated the expression pat-terns of the components in the NAC29/31-MYB61-secondarywall CESA signaling pathway by qRT-PCR. The transcripts ofSLR1, NAC29, and NAC31 peaked in section 2, consistent withthe finding that this stage involves the onset of secondary wallcellulose synthesis; the peak of MYB61 and secondary wallCESAs occurred in section 3 or 4 (Figure 7D). The slightly

lagging peaks of downstream components verified the hierar-chical relationship of this signaling cascade. After being acti-vated in sections 2 and 3, this signaling pathway was graduallyrepressed through sections 3 to 8, where GA1 was maintainedat a steady low level. Therefore, it was hypothesized that thissignaling pathway is activated by a relatively high GA level inthe early stages of internode development and is graduallyrepressed by a steady low level of GA in the later developingstages. To confirm this speculation, we further compared theexpression pattern of MYB61 and CESAs in the internodes ofthe wild type and sd1, as sd1 is deficient in GA biosynthesisand has been widely used in rice breeding. The overall ex-pression patterns of MYB61 and CESAs in sd1 were lower thanin the wild type, regardless of the activation and repressionstages (Figure 7E). All of these results suggest that thissignaling transduction pathway is modulated by varied GA

Figure 7. Examination of the GA-Mediated NAC29/31-MYB61-CESA Signaling Cascade in Developing Rice Internodes.

(A) Measurement of cell length in the longitudinal direction in the developing internodes divided into nine sections, as shown in Supplemental Figure 7.Error bars indicate SD (n = 30).(B) Total yield of cellulose and lignin in each section of internodes as described above. Error bars indicate SD (n = 50).(C) Endogenous GA levels in each section of internodes as described above. F.W., fresh weight.(D) and (E) qRT-PCR analyses of the expression levels of genes in the signaling pathway in each internode section from cv Nipponbare (D) and from cvNanjing6 (wild type) and sd1 (E) plants. TP1 was used as an internal control. The expression level in section 1 was considered as 1 in (D).The numbers on the x axes indicate the internode sections from the bottom up. Error bars indicate the SD of three biological replicates.


Dow




levels, which regulate cellulose synthesis during internodedevelopment.

The Identified Signaling Machinery Is Conserved in Plants

To determine whether this signaling pathway is present in otherplant species, the role of GA in cellulose synthesis was examined inArabidopsis, a model plant with GA functional components iden-tical to the ones in rice. We found that treating Arabidopsis plantswith exogenous GA induced the expression of secondary wallCESAs and hierarchical TFs (Figure 8A; Supplemental Figure 8A).Genetic evidence was further provided by the examination of geneexpression and cellulose synthesis in GA-related Arabidopsismutants. Three secondary wall CESAs were upregulated in thedouble mutants lacking two DELLA genes, GAI and RGA (gai-t6rga-24); however, this upregulation was attenuated in a triple mu-tant (gai-t6 rga-24 ga1-3), which harbors an additional mutation inCPS (ga1-3), a gene encoding an enzyme for GA biosynthesis(Figure 8B). The three CESAs were downregulated in the DELLAgain-of-function mutant gai-1 (Figure 8B). Consistent with the geneexpression data, DELLA loss-of-function mutants exhibited in-creased secondary walls in interfascicular fibers and xylem vessels,whereas the DELLA gain-of-function mutant gai-1 and the GA-deficient mutant ga1-3 tended to exhibit compromised secondary

wall deposition in fiber and vessel cells (Figure 8C). Chemicalanalysis revealed corresponding alterations in cellulose content inthe above mutants (Figure 8D). Furthermore, split-luciferase com-plementation assays showed interaction between the ArabidopsisDELLA protein RGA and SND1, an NAC TF for secondary wallformation in N. benthamiana leaves (Figure 8E). Therefore, GAsignals and DELLA-mediated signaling are required for the regu-lation of secondary wall cellulose synthesis in Arabidopsis.Because GAMYB motifs are critical for GA signal transduction,

the promoter regions of the primary and secondary wall CESAs inArabidopsis and poplar (Populus trichocarpa) were examined. Thepresence of at least one GAMYBmotif in At-CESAs and Ptr-CESAs(Supplemental Table 3) and the action of poplar CESAs in responseto GA treatment (Supplemental Figure 8B) indicated that this sig-naling cascade for cellulose synthesis is conserved in land plants.

DISCUSSION

GA Regulates Secondary Wall Cellulose Synthesis via anSLR1-NAC Signaling Cascade

Cellulose is a basic structural polymer of plant cell walls. Itssynthesis is thought to be highly regulated by various hormones

Figure 8. GA Promotes Cellulose Synthesis in Arabidopsis.

(A) Inducing the expression of Arabidopsis secondary wall CESAs in the inflorescence stems from wild-type plants treated for 6 h with 10 mM GA3 (+GA)compared with control plants (–GA).(B) qRT-PCR analyses showing altered secondary wall CESA expression levels in the inflorescence stems from the wild type and GA-related mutants.delladm, gai-t6 rga-24; ga1-3 delladm, ga1-3 gai-t6 rga-24.(C) Observation of interfascicular fiber (If) and xylem vessel (Xv) cells from cross sections of the interfascicular stems from the wild type and GA-relatedmutants. Bars = 50 mm.(D) Measurement of cellulose content in the inflorescence stems from the wild type and GA-related mutants. AIR, alcohol-insoluble residues. *P < 0.01determined by Student’s t test.(E) Split-luciferase complementation assay showing the interaction between Arabidopsis RGA and SND1 in N. benthamiana leaves agroinfiltrated withthe constructs shown at top left. BF, bright field. Bar = 1 cm.Arabidopsis UBQ10 was used as an internal control in (A) and (B). Error bars indicate the SD of three biological repeats.

1690 The Plant Cell

Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

and environmental signals, fulfilling its roles in different cell typesand at distinct developmental stages (Xie et al., 2011; Zhonget al., 2011). Previous studies have identified a few TFs thatfunction in the transcriptional regulation of cellulose (Ko et al.,2007; Demura and Ye, 2010; Zhong and Ye, 2012; Kim et al.,2013). However, the underlying mechanism, especially a completesignal transduction pathway, has been unclear. In this study,we reveal a role for GA and the mechanism of GA-regulatedcellulose synthesis. Multiple lines of evidence from genetic,biochemical, and gene expression analyses point to a com-plete GA signaling cascade (Figure 9), in which GA signalsare transmitted through the interaction between SLR1, a keyrepressor of GA signaling (Itoh et al., 2002), and NACs, the top-layer master switches for the transcriptional regulation of sec-ondary wall formation (Zhong and Ye, 2007). In the presence ofGA, the GA-induced degradation of SLR1 releases NACs, en-abling these factors to bind to and upregulate the downstreamtarget MYB61 and consequently enhancing the transcription ofthree secondary wall CESAs (Figure 9). Critical biological prooffor this regulatory machinery came from the examination of GAcontent and the expression of key regulators in developing riceinternodes, where the behavior of this signaling pathway iscorrelated with the endogenous GA levels. Moreover, this signalingmechanism was found to be conserved in land plants, as evi-denced by the following observations: the Arabidopsis DELLAprotein RGA interacts with SND1; the Arabidopsis GA-relatedmutants exhibit alterations in CESA expression and cellulosesynthesis; and the CESAs from Arabidopsis and poplar wereGA-inducible and have at least one GAMYB element in their

promoter regions. GAMYB binding proteins are TFs belonging toa distinct subclass in the MYB superfamily, which is widelypresent in plant species, and in which many members are re-quired for modulating cell wall biosynthesis (Woodger et al.,2003; Zhong and Ye, 2007). Therefore, the GA-response path-way mediated by GAMYB binding TFs, such as MYB61, is ap-plicable to other plant species.

SLR1 May Be a Key Component That Integrates GA withOther Signals in Cellulose Synthesis Regulation

In addition to GA, other hormonal and environmental stimuli areinvolved in the modulation of cellulose synthesis. Brassinosteroidand light have been found to promote Arabidopsis hypocotylelongation and primary wall CESA gene transcription (Leivar andQuail, 2011; Xie et al., 2011). WAT1, which was characterized asa vacuolar auxin transporter, has been found to play a role insecondary wall formation in fiber cells (Ranocha et al., 2013).However, the complete signaling pathways linking CESAs andBZR1, PIFs, or WAT1 have been unclear. On the other hand, anincreasing number of studies have revealed a comprehensiveview of DELLA action by identifying the physical interactions withseveral types of TFs (Marín-de la Rosa et al., 2014), which arethought to form a central command system for integrating varioussignals (Davière and Achard, 2013). Previous work in Arabidopsishas revealed the interaction between DELLA and PIFs/BZR1(de Lucas et al., 2008; Feng et al., 2008; Bai et al., 2012). Inthis signaling cascade, SLR1 interacts with secondary wallNACs. Therefore, identification of the interdependent relationship

Figure 9. Regulatory Model for Secondary Wall Cellulose Formation in Rice.

SLR1 interacts with secondary wall NACs to repress their DNA binding activity. In the presence of GA, GA triggers the proteasomal degradation of SLR1and frees the NACs to activate the expression of downstream MYB and secondary wall CESAs, which consequently promotes cellulose synthesis. MT,microtubules; PM, plasma membrane.


Dow



between NACs and the other SLR1-interating TFs will providea better mechanistic understanding of cellulose synthesis in re-sponse to different signals.

Potential Value of This Regulatory Pathway in Crop andBiomass Production

The widely known physiological role of GA is to promote plantheight. Therefore, one common phenotype of many GA-relatedmutants is the altered stem length (Sakamoto et al., 2004). Themajor contribution of GA to the Green Revolution is based on thisrole, relying on its function in cell expansion and proliferation(Peng et al., 1997; Sasaki et al., 2002; Achard et al., 2009; Leeet al., 2012; Mimura and Itoh, 2014). The role for GA in regulatingsecondary wall cellulose synthesis, which also results from pro-moting sclerenchyma wall thickening and the proliferation ofsclerenchyma cells, extends its function directly to lodging re-sistance. Combining the results of anatomical analysis and GAcontent in the developing rice internodes, we conclude that cellelongation and secondary wall formation are modulated by variedGA levels via different SLR1-mediated signaling pathways. Ourstudy here promotes an in-depth consideration of these GreenRevolution genes, addressing how GA coordinately modulatescell elongation and secondary wall synthesis during internodedevelopment. Identifying the SLR1-mediated downstream path-ways may be a critical way to manipulate both agronomic traits.

More importantly, cellulose is the most valuable type of bio-mass. Owing to its economic importance, much attention hasbeen paid to the manipulation of cellulose properties. The acti-vation of this GA-mediated signaling cascade in rice plants pro-motes cellulose synthesis without significantly altering othermajor cell wall components. This finding was distinguished fromthe results of overexpressing the Arabidopsis CESA regulatorygenes, such as SND1, MYB46, and MYB83, in which the overallsecondary wall synthesis was altered (Zhong et al., 2006; Koet al., 2009; Zhong and Ye, 2012; Kim et al., 2013). This dis-crepancy indicates that the secondary wall regulatory network inmonocot plants may be distinct from the one in dicots in certainrespects. In this study, although total lignin content was notchanged in the transgenic plants, we cannot exclude the pos-sibility that their lignin composition was altered, as Os-NAC31and Os-MYB61 were reported to regulate Os-MYB46 and Os-CINNAMYL ALCOHOL DEHYDROGENASE2 expression (Zhonget al., 2011; Hirano et al., 2013). Moreover, cellulosic biofuel isa type of sustainable energy source that has gained popularsupport due to its potential benefit to the global environment. Inthe future, the economical use of this biofuel will depend onwhether we can manipulate cellulose biosynthesis (Li et al., 2014).This study showed that cell wall residues from TF-overexpressingplants exhibited increased cellulose content and improved sac-charification efficiency (Supplemental Figure 8D), suggesting theintriguing value of this regulatory pathway in biomass production.

METHODS

Plant Materials and Growth Conditions

The rice plants (Oryza sativa) used in this study, including the wild-typeplants, the GA-related mutants sd1-8 (E386K), gid1-20 (A110T), d18-AD,

and slr1-6 (L246P), and the relevant transgenic plants, were grown in theexperimental fields at the Institute of Genetics and Developmental Biologyin Beijing and Sanya in Hainan Province during the natural growingseasons. sd1-8 and gid1-20 are near-isogenic lines of cv Nanjing6, anindica cultivar. slr1-6 and d18-AD are of cv Zhonghua11 and cv Akibare,japonica cultivars, respectively. The Arabidopsis thaliana GA-relatedmutants, including gai-1, gai-t6 rga-24 (double mutant), and ga1-3 gai-t6rga24 (triple mutant), are of the Landsberg erecta ecotype. The typicalphenotype of GA-related mutants is altered stem length.

Phylogenetic Analysis

The secondary wall-related NAC transcription factors in rice were analyzedby BLAST searching the rice genome database (http://rice.plantbiology.msu.edu) against Arabidopsis SND1 and VND6/7. An unrooted tree of thesecondary wall-related NACs in rice and Arabidopsis was generated usingMEGA5 software (Tamura et al., 2011) with 1000 bootstrap replications.

Generation of Transgenic Rice Plants

For generation of the TF-overexpressing rice plants, the full-length cDNAsequences (CDSs) of the TFswere amplifiedbyPCRusing theprimers listedin Supplemental Table 4 and inserted into the pCAMBIA 1300 vector(Cambia) between the rice ubiquitin promoter and the nopaline synthaseterminator via BamHI and KpnI/SpeI digestion. For generation of theMYB61-SRDX plants, a sequence-confirmed MYB61 CDS was fused withthe SRDX sequence and inserted into the pCAMBIA 1300 vector (Cambia)between the rice ubiquitin promoter and the nopaline synthase terminatorvia BamHI and KpnI digestion. The resulting constructs were transfectedinto Agrobacterium tumefaciens EHA105 and introduced into the wild-typevarieties cv Nipponbare and cv Zhonghua11, respectively (Hiei et al., 1994).

Microscopy

The fresh hand-cut cross sections of second internodes from the maturerice plants of sd1-8, gid1-20, slr1-6, and their corresponding wild type, aswell as the cross sections of leaf sheaths from 2-month-old wild type andd18-AD plants, were prepared with razor blades. The fresh hand-cut crosssections of the mature inflorescence stems were also prepared from theArabidopsis GA-related mutants and the corresponding wild-type plants.The autofluorescent signals of cell walls were viewed and photographed atan excitation wavelength of 450 to 480 nmwith a fluorescence microscope(Zeiss). For the scanning microscopic analysis, the internodes and leafsheaths of GA-related mutants were cut and fixed in 4%paraformaldehyde(Sigma-Aldrich). After dehydration through a gradient of ethanol, thesamples were sprayed with gold particles and observed with a scanningelectron microscope (S-3000N; Hitachi). For the transmission electronmicroscopic analysis, the second internodes from thewild-type andMYB61Ox plants were cut and fixed in 2.5% (w/v) glutaraldehyde. The sampleswere then embedded with the Spurr Kit (Sigma-Aldrich) and sectioned withan Ultracut E ultramicrotome (Leica). The sections were stained and ob-served with a transmission electron microscope (H-7500; Hitachi) operatedat 80 kV. For laser-capture microdissection, the young internodes from cvNipponbare were cut and fixed in glacial acetic acid and ethanol (1:3). Afterthey were embedded in wax (Sigma-Aldrich), the internodeswere sectionedand applied for cell harvesting with an LMD 7000 (Leica) laser microdis-section system. The harvested samples were subjected to RNA isolationwith an RNeasy micro kit (Qiagen) and to qRT-PCR analysis.

qRT-PCR and ChIP-Quantitative PCR Assays

The second internodes or leaf sheaths were harvested from heading-stage wild-type, sd1-8, gid1-20, slr1-6, and d18-AD plants and subjectedto total RNA isolation. Four-week-old inflorescence stems of wild-type

1692 The Plant Cell

Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://rice.plantbiology.msu.eduhttp://rice.plantbiology.msu.eduhttp://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

Arabidopsis plants and the development-matched stems of GA-relatedmutants were also harvested and subjected to RNA isolation. ConcertPlant RNA Reagent (Invitrogen) and TRIzol Reagent (Invitrogen) were usedto isolate the total RNA. qRT-PCR was performed on a cycler apparatus(Bio-Rad CFX96) with the FastStart Universal SYBR Green Master (Ro-che). To examine the effect of GA on cellulose synthesis, heading-stagerice plants, 4-week-old Arabidopsis in the Landsberg erecta ecotype, and4-week-old poplar (Populus tomentosa) seedlings were treated with 10mM GA3 for 6 or 9 h. The young parts of rice internodes and Arabidopsisand poplar stems were subjected to RNA isolation and gene expressionexaminations. Primers for the expression analysis are summarized inSupplemental Table 5. For ChIP-quantitative PCR analysis, formaldehydecross-linked chromatin DNA was isolated from leaf sheaths of 2-month-old wild-type and MYB61 Ox plants. Immunoprecipitation was performedusing 5 mL of anti-MYB61 antibody with a 1:100 dilution (Immunoway).The generated ChIP DNA was used as a template for PCR amplification(Supplemental Table 5) using a cycler apparatus (Bio-Rad CFX96) with theFastStart Universal SYBR Green Master (Roche). Enrichment folds ofMYB61-bound DNA fragments were calculated by comparing the sam-ples with a sample without antibody applied. The data are presented asmeans 6 SD of three biological repeats.

Protein Interactions

The split-luciferase complementation assay was performed as described(Chen et al., 2008). The images were acquired using IndiGO software. Forcoimmunoprecipitation analysis, leaf sheaths from 2-month-old NAC29-GFP and NAC31-GFP transgenic plants were harvested and homogenizedin immunoprecipitation buffer (25 mM Tris-Cl, pH 7.5, 2 mM EDTA, 150mMNaCl, 0.1% Triton X-100, 10% glycerol, and 13 protease inhibitor). Thesamples were purified through anti-GFP immobilized protein A agarosebeads (Pierce Biotechnology) and were applied for immunoblot analysisusing anti-SLR1 and anti-GFP antibody with a 1:1000 dilution. For yeasttwo-hybrid assays, full-length SLR1 and the N-terminal and C-terminalparts of SLR1 were amplified (Supplemental Table 4) and fused with GAL4BD in the pDEST32 vector (Invitrogen). Interactions in yeast were tested onSD/-Trp/-Leu/-His/-Ade/+3AT (25 mM) medium. For the BiFC analysis, thecDNAofSLR1and theNandC termini ofNAC29 andNAC31were amplified(Supplemental Table 4) and cloned into serial pSPY vectors (Waadt et al.,2008) containing either N- or C-terminal enhanced yellow fluorescentprotein fragments via Gateway cloning technology. The resulting constructswere then introduced into Agrobacterium strain C58 and coinfiltrated intothe abaxial surface of the leaves of 4-week-old Nicotiana benthamianaplants according to Chen et al. (2008). Fluorescence was observed witha confocal laser-scanning microscope (TCS SP5; Leica).

Transactivation Assay

The CDSs of SLR1, NAC29, NAC31, andMYB61 and the promoter regions(upstream of the ATG) of MYB61 and three secondary wall CESA geneswere amplified (Supplemental Table 4) and cloned into the effector (35S-transcription factor) and reporter (promoter-luciferase) vectors (Promega)between HindIII/SacI and KpnI cleavage sites. The resulting effector andreporter constructs were cotransfected into protoplasts prepared from 2-week-old rice seedlings or 4-week-old Arabidopsis leaves (Ko et al., 2009).The Renilla luciferase gene driven by the CaMV 35S promoter was alsocotransfected to determine the transfection efficiency. Luciferase activitiesweremeasured with a dual-luciferase reporter assay system (Promega). Forthe inducible system, the CDS of MYB61 fused with the GR domain wasinserted into the p2GW7-35S-GR vector (Aoyama and Chua, 1997; Zhaoet al., 2010). The rice protoplasts transfected with the effector and reporterconstructs were treated with 10 mM DEX (Sigma-Aldrich) for 6 h, leading toMYB61-GR translocation into the nuclei. The transfected rice protoplastswere treated with 2mMCHX for 30min to inhibit new protein synthesis prior

to the addition of DEX. The transfected protoplasts were subsequentlysubjected to qRT-PCR or dual-luciferase activity analyses. To examine thetransactivation activity of GAMYB, three copies of the 20-bp GAMYB(GCGCACCAACCGCCCGTTCG) or GAMYBm (GCGCACGAACCGCCC-GTTCG) of the CESA4 promoter were synthesized and inserted into thereporter vector and transfected into Arabidopsis protoplasts. Luciferaseactivities were measured as described above. To investigate the effects ofSLR1-NAC29/31 interaction, protoplast cells coexpressing the indicatedreporters and effectors were treated with 10 mM GA3 for 6 h. Then, theprotoplast cells were subjected to qRT-PCR and luciferase activity assays.The data are presented as means 6 SD of three biological repeats.

EMSA

The amplified CDSs ofMYB61, NAC29, and NAC31 genes (SupplementalTable 4) were fused in-frame with MBP (New England Biolabs) or GST(Invitrogen) tags and transformed into Escherichia coli Rosetta (Novagen).MBP-MYB61 and GST-NAC29/31 recombinant proteins were purifiedand incubated with the biotin-11-UTP-labeled DNA fragments, includingthe CESA andMYB61 promoters and the synthesized GAMYB and SNBEoligonucleotides, for 30 min in the EMSA binding buffer (Thermo). TheDNA signals were detected by chemiluminescence (Thermo). For thecompetition assays, unlabeled oligonucleotides (10- and 50-fold of la-beled probes) were added to the EMSA reactions. To examine the effect ofSLR1 on the abilities of NAC29/31 binding to MYB61, a 35S-FLAG-SLR1construct was prepared and introduced into N. benthamiana leaves byAgrobacterium infiltration. The FLAG-SLR1 proteins were extracted fromthe transfected leaves and purified by incubation with anti-FLAG M2affinity gel (Sigma-Aldrich). Different amounts of the purified FLAG-SLR1proteins were added to EMSA reactions. The reaction mixtures wereseparated by SDS-PAGE, and the signals were analyzed as describedabove.

Cell Wall Composition Analyses

The second internodes from the mature wild-type plants, sd1-8, gid1-20,and slr1-6 mutants, and the relevant transgenic plants after seed harvestand the leaf sheaths of 2-month-old wild-type and d18-AD plants werecollected for cell wall residue preparation. The Arabidopsis cell wall residueswere prepared from the lower part of inflorescence stems of mature wild-type plants and GA-related mutants. To examine the impact of GA oncellulose synthesis, heading-stage wild-type and bc11 plants were treatedwith 10 mMGA3 until seed harvest. The second internodes from the treatedand control plants were collected to prepare cell wall residues. The cell wallsamples were treated with alcohol to obtain the insoluble residues.Monosaccharide composition was determined by gas chromatography-mass spectrometry (Agilent) as described previously (Zhang et al., 2012).For crystalline cellulose analysis, the remains after trifluoroacetic acidtreatment were hydrolyzed in Updegraff reagent. The cooled pellets werewashed and hydrolyzed with 72% sulfuric acid. The cellulose content wasquantified by the anthrone assay (Updegraff, 1969). The lignin content wasmeasured using the acetyl bromide method (Foster et al., 2010). The dataare presented as means 6 SD of three or four biological repeats.

Examination of Endogenous GAs

The 9-cm second internodes were collected from cv Nipponbare at theheading stage and divided into nine sections at 1-cm intervals. Sections 1to 8 were applied for endogenous GAmeasurement as described (Li et al.,2011). In brief, 3-g internode section tissues were frozen and ground inliquid nitrogen and extracted in the buffer containing 80% (v/v) methanolat 4°C for 12 h. The internal standards, such as [2H2]GA1 (1.00 ng/g), [

2H2]GA4 (2.00 ng/g), [

2H2]GA12 (2.00 ng/g), [2H2]GA24 (6.00 ng/g), and [

2H2]GA53(4.00 ng/g), were added to the plant samples during extraction. After


Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

a series of extraction and reaction steps, the samples were subjected tocapillary electrophoresis coupled with electrospray ionization quadrupole-time-of-flight mass spectrometry. Amounts of endogenous GAs weredetermined from established calibration curves of the peak area ratiosfor unlabeled and deuterated GAs plotted. Independent triplicate sampleswere examined in the analysis.

Cell Wall Saccharification Assay

To examine the enzyme activity of the commercial cellulase mixture(Novozymes), the filter-paper unit was used to determine the cellulase units.Fifty milligrams of filter paper was incubated with 900 mL of distilled waterfor 1 h at 100°C.After the hot-water treatment, 100mL of 0.5Mcitrate buffer,pH 4.8, was added into the tubes containing different amounts (0.01, 0.1, 1,2, and 10 mL) of cellulase mixture. The free sugars were assayed using thedinitrosalicylic acidmethod after incubation in a 50°C shaker for 1 h (Ghose,1987). The absorbance at 540 nmwas measured on a UV light-specific 96-well plate by an ELISA reader (Tecan). The filter-paper unit was calculatedaccording to the amounts of cellulase mixture used in the treatment, whichcould release 2.0 mg of glucose from 50 mg of filter paper within 60 min.According to this result, 1 to 2 mg of destarched alcohol-insoluble residuesfrom rice plants was treated with a 2-mL cellulase mixture for differentperiods and subjected to the saccharification assay as described above.The data are presented as means 6 SD of four biological repeats.

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL librariesunder the following accession numbers: Os01g66100 (SD1), Os01g08220(D18), Os05g33730 (GID1), Os03g49990 (SLR1), Os01g18240 (MYB61),Os08g02300 (NAC29), Os08g01330 (NAC31), Os01g54620 (CESA4),Os10g32980 (CESA7), Os09g25490 (CESA9), AT5G62380 (VND6),AT1G71930 (VND7), AT1G32770 (SND1), and AT2G01570 (RGA).

Supplemental Data

Supplemental Figure 1. The Phenotypes of Rice GA-Related Mu-tants.

Supplemental Figure 2. Phylogenetic Analysis of the Secondary WallNACs in Rice and Arabidopsis.

Supplemental Figure 3. NAC29 and NAC31 Are Functional TFs ThatRegulate Cellulose Synthesis.

Supplemental Figure 4. MYB61 Is Required for Cellulose Synthesis.

Supplemental Figure 5. NAC29 and NAC31 Bind to the SNBE Motif inthe MYB61 Promoter.

Supplemental Figure 6. MYB61 Regulates Secondary Wall CESATranscription.

Supplemental Figure 7. Development of Rice Internodes InvolvesSecondary Wall Accumulation.

Supplemental Figure 8. Impacts of GA on Cellulose Biosynthesis inArabidopsis and Poplar and Saccharification Assay of Cell WallResidues of Transgenic Rice Plants.

Supplemental Table 1. List of the TFs Coexpressed with SecondaryWall CESAs in Rice.

Supplemental Table 2. Composition Analysis of Sugar and LigninContent of Wall Residues of the Internodes from Wild-Type andTransgenic Rice Plants.

Supplemental Table 3. Identifying GAMYB Motif in the PromoterRegion of CESA Genes in the Rice, Arabidopsis, and Poplar Genomes.

Supplemental Table 4. The Primers Used for Generation of Trans-genic Rice Plants and Relevant Analyses in This Study.

Supplemental Table 5. qRT-PCR and ChIP-PCR Primers Used in ThisStudy.

Supplemental File 1. Alignments Used to Produce the PhylogeneticTree Shown in Supplemental Figure 2.

ACKNOWLEDGMENTS

We thank W. Yang (Institute of Genetics and Developmental Biology,Chinese Academy of Sciences) for help with laser microdissection, X.W.Deng and H. Chen (School of Life Sciences, Peking University) forproviding the BiFC vectors, J. Wei (Beijing Academy of Agriculture andForestry Sciences) for providing the poplar seedlings, and Q. Qian (ChinaNational Rice Research Institute, Chinese Academy of Agricultural Scien-ces) for providing the d18-AD mutant. This study was supported by theMinistry of Sciences and Technology of China (Grant 2012CB114501), theNational Natural Science Foundation of China (Grants 31125019 and91417303), the Ministry of Agriculture of China for Transgenic Research(Grant 2008ZX08009-003), and the State Key Laboratory of PlantGenomics.

AUTHOR CONTRIBUTIONS

Y.Z. and X.F. together designed the experiments. D.H. performed trans-activation assay, yeast two-hybrid assay, EMSA, and cell wall compositionanalysis in rice and Arabidopsis. S.W. performed qRT-PCR, split-luciferasecomplementation assay, transactivation assay, EMSA, and GA contentanalysis. B.Z. analyzed motifs in promoter regions and helped with criticaldiscussion of the work. K.S.-G. performed coimmunoprecipitation assayand gene expression analysis in Arabidopsis. Y.S. performed subcellularlocalization of TFs and BiFC assay. D.Z. performed ChIP-quantitative PCRand laser-capture microdissection assays. X.L. performed rice transfor-mation. K.W. screened and cultivated the rice and Arabidopsis GAmutants. Z.X. performed field cultivation. Y.Z. performed anatomicalanalysis and wrote the article.

Received January 6, 2015; revised April 17, 2015; accepted May 6, 2015;published May 22, 2015.

REFERENCES

Achard, P., Gusti, A., Cheminant, S., Alioua, M., Dhondt, S., Coppens,F., Beemster, G.T., and Genschik, P. (2009). Gibberellin signalingcontrols cell proliferation rate in Arabidopsis. Curr. Biol. 19: 1188–1193.

Aoyama, T., and Chua, N.H. (1997). A glucocorticoid-mediatedtranscriptional induction system in transgenic plants. Plant J. 11:605–612.

Arnaud, N., Girin, T., Sorefan, K., Fuentes, S., Wood, T.A.,Lawrenson, T., Sablowski, R., and Østergaard, L. (2010). Gibberellinscontrol fruit patterning in Arabidopsis thaliana. Genes Dev. 24:2127–2132.

Bai, M.Y., Shang, J.X., Oh, E., Fan, M., Bai, Y., Zentella, R., Sun,T.P., and Wang, Z.Y. (2012). Brassinosteroid, gibberellin and phyto-chrome impinge on a common transcription module in Arabidopsis.Nat. Cell Biol. 14: 810–817.

Cheminant, S., Wild, M., Bouvier, F., Pelletier, S., Renou, J.P.,Erhardt, M., Hayes, S., Terry, M.J., Genschik, P., and Achard, P.

1694 The Plant Cell

Dow



http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1http://www.plantcell.org/cgi/content/full/tpc.15.00015/DC1

(2011). DELLAs regulate chlorophyll and carotenoid biosynthesis toprevent photooxidative damage during seedling deetiolation inArabidopsis. Plant Cell 23: 1849–1860.

Chen, H., Zou, Y., Shang, Y., Lin, H., Wang, Y., Cai, R., Tang, X., andZhou, J.M. (2008). Firefly luciferase complementation imaging assay forprotein-protein interactions in plants. Plant Physiol. 146: 368–376.

Davière, J.M., and Achard, P. (2013). Gibberellin signaling in plants.Development 140: 1147–1151.

de Lucas, M., Davière, J.M., Rodríguez-Falcón, M., Pontin, M.,Iglesias-Pedraz, J.M., Lorrain, S., Fankhauser, C., Blázquez, M.A., Titarenko, E., and Prat, S. (2008). A molecular framework forlight and gibberellin control of cell elongation. Nature 451: 480–484.

Demura, T., and Ye, Z.-H. (2010). Regulation of plant biomass pro-duction. Curr. Opin. Plant Biol. 13: 299–304.

Desprez, T., Vernhettes, S., Fagard, M., Refrégier, G., Desnos, T.,Aletti, E., Py, N., Pelletier, S., and Höfte, H. (2002). Resistanceagainst herbicide isoxaben and cellulose deficiency caused bydistinct mutations in same cellulose synthase isoform CESA6. PlantPhysiol. 128: 482–490.

Doblin, M.S., Kurek, I., Jacob-Wilk, D., and Delmer, D.P. (2002).Cellulose biosynthesis in plants: From genes to rosettes. Plant CellPhysiol. 43: 1407–1420.

Feng, S., et al. (2008). Coordinated regulation of Arabidopsis thalianadevelopment by light and gibberellins. Nature 451: 475–479.

Feurtado, J.A., Huang, D., Wicki-Stordeur, L., Hemstock, L.E.,Potentier, M.S., Tsang, E.W., and Cutler, A.J. (2011). The Arabi-dopsis C2H2 zinc finger INDETERMINATE DOMAIN1/ENHYDROUSpromotes the transition to germination by regulating light and hor-monal signaling during seed maturation. Plant Cell 23: 1772–1794.

Foster, C.E., Martin, T.M., and Pauly, M. (2010). Comprehensivecompositional analysis of plant cell walls (lignocellulosic biomass).Part II. Carbohydrates. J. Vis. Exp. 37: e1745.

Fu, X., Richards, D.E., Ait-Ali, T., Hynes, L.W., Ougham, H., Peng,J., and Harberd, N.P. (2002). Gibberellin-mediated proteasome-dependent degradation of the barley DELLA protein SLN1 re-pressor. Plant Cell 14: 3191–3200.

Gardiner, J.C., Taylor, N.G., and Turner, S.R. (2003). Control ofcellulose synthase complex localization in developing xylem. PlantCell 15: 1740–1748.

Ghose, T.K. (1987). Measurement of cellulase activities. Pure Appl.Chem. 59: 257–268.

Handakumbura, P.P., and Hazen, S.P. (2012). Transcriptional regu-lation of grass secondary cell wall biosynthesis: Playing catch-upwith Arabidopsis thaliana. Front. Plant Sci. 3: 74.

Harberd, N.P., Belfield, E., and Yasumura, Y. (2009). The angio-sperm gibberellin-GID1-DELLA growth regulatory mechanism: Howan “inhibitor of an inhibitor” enables flexible response to fluctuatingenvironments. Plant Cell 21: 1328–1339.

Hedden, P. (2003). The genes of the Green Revolution. Trends Genet.19: 5–9.

Hiei, Y., Ohta, S., Komari, T., and Kumashiro, T. (1994). Efficient trans-formation of rice (Oryza sativa L.) mediated by Agrobacterium and se-quence analysis of the boundaries of the T-DNA. Plant J. 6: 271–282.

Hirano, K., Asano, K., Tsuji, H., Kawamura, M., Mori, H., Kitano, H.,Ueguchi-Tanaka, M., and Matsuoka, M. (2010). Characterizationof the molecular mechanism underlying gibberellin perceptioncomplex formation in rice. Plant Cell 22: 2680–2696.

Hirano, K., Kondo, M., Aya, K., Miyao, A., Sato, Y., Antonio, B.A.,Namiki, N., Nagamura, Y., and Matsuoka, M. (2013). Identificationof transcription factors involved in rice secondary cell wall forma-tion. Plant Cell Physiol. 54: 1791–1802.

Hiratsu, K., Matsui, K., Koyama, T., and Ohme-Takagi, M. (2003).Dominant repression of target genes by chimeric repressors that

include the EAR motif, a repression domain, in Arabidopsis. Plant J.34: 733–739.

Hou, X., Lee, L.Y., Xia, K., Yan, Y., and Yu, H. (2010). DELLAsmodulate jasmonate signaling via competitive binding to JAZs. Dev.Cell 19: 884–894.

Itoh, H., Ueguchi-Tanaka, M., Sato, Y., Ashikari, M., andMatsuoka, M. (2002). The gibberellin signaling pathway is regu-lated by the appearance and disappearance of SLENDER RICE1 innuclei. Plant Cell 14: 57–70.

Jan, A., Yang, G., Nakamura, H., Ichikawa, H., Kitano, H.,Matsuoka, M., Matsumoto, H., and Komatsu, S. (2004). Charac-terization of a xyloglucan endotransglucosylase gene that is up-regulated by gibberellin in rice. Plant Physiol. 136: 3670–3681.

Kim, W.C., Ko, J.H., and Han, K.H. (2012). Identification of a cis-acting regulatory motif recognized by MYB46, a master transcrip-tional regulator of secondary wall biosynthesis. Plant Mol. Biol. 78:489–501.

Kim, W.C., Ko, J.H., Kim, J.Y., Kim, J.M., Bae, H.J., and Han, K.H.(2013). MYB46 directly regulates the gene expression of secondary wall-associated cellulose synthases in Arabidopsis. Plant J. 73: 26–36.

Ko, J.H., Kim, W.C., and Han, K.H. (2009). Ectopic expression ofMYB46 identifies transcriptional regulatory genes involved in sec-ondary wall biosynthesis in Arabidopsis. Plant J. 60: 649–665.

Ko, J.H., Yang, S.H., Park, A.H., Lerouxel, O., and Han, K.H. (2007).ANAC012, a member of the plant-specific NAC transcription factorfamily, negatively regulates xylary fiber development in Arabidopsisthaliana. Plant J. 50: 1035–1048.

Kubo, M., Udagawa, M., Nishikubo, N., Horiguchi, G., Yamaguchi,M., Ito, J., Mimura, T., Fukuda, H., and Demura, T. (2005). Tran-scription switches for protoxylem and metaxylem vessel formation.Genes Dev. 19: 1855–1860.

Lee, L.Y., Hou, X., Fang, L., Fan, S., Kumar, P.P., and Yu, H. (2012).STUNTED mediates the control of cell proliferation by GA in Ara-bidopsis. Development 139: 1568–1576.

Leivar, P., and Quail, P.H. (2011). PIFs: Pivotal components ina cellular signaling hub. Trends Plant Sci. 16: 19–28.

Li, J., et al. (2011). Mutation of rice BC12/GDD1, which encodesa kinesin-like protein that binds to a GA biosynthesis gene pro-moter, leads to dwarfism with impaired cell elongation. Plant Cell23: 628–640.

Li, S., Bashlin, L., Lei, L., and Gu, Y. (2014). Cellulose synthesis andits regulation. The Arabidopsis Book 12: e0169, doi/10.1199/tab.0169.

Marín-de la Rosa, N., Sotillo, B., Miskolczi, P., Gibbs, D.J., Vicente,J., Carbonero, P., Oñate-Sánchez, L., Holdsworth, M.J., Bhalerao, R.,Alabadí, D., and Blázquez, M.A. (2014). Large-scale identification ofgibberellin-related transcription factors defines group VII ETHYLENERESPONSE FACTORS as functional DELLA partners. Plant Physiol. 166:1022–1032.

McCarthy, R.L., Zhong, R., and Ye, Z.H. (2009). MYB83 is a directtarget of SND1 and acts redundantly with MYB46 in the regulationof secondary cell wall biosynthesis in Arabidopsis. Plant CellPhysiol. 50: 1950–1964.

Mimura, M., and Itoh, J. (2014). Genetic interaction between ricePLASTOCHRON genes and the gibberellin pathway in leaf de-velopment. Rice (N. Y.) 7: 25.

Murase, K., Hirano, Y., Sun, T.P., and Hakoshima, T. (2008). Gibberellin-induced DELLA recognition by the gibberellin receptor GID1. Nature456: 459–463.

Peng, J., Carol, P., Richards, D.E., King, K.E., Cowling, R.J.,Murphy, G.P., and Harberd, N.P. (1997). The Arabidopsis GAIgene defines a signaling pathway that negatively regulates gibber-ellin responses. Genes Dev. 11: 3194–3205.


Dow



Peng, J., et al. (1999). ‘Green Revolution’ genes encode mutantgibberellin response modulators. Nature 400: 256–261.

Ranocha, P., Denancé, N., Vanholme, R., Freydier, A., Martinez, Y.,Hoffmann, L., Köhler, L., Pouzet, C., Renou, J.P., Sundberg, B.,Boerjan, W., and Goffner, D. (2010). Walls are thin 1 (WAT1), anArabidopsis homolog ofMedicago truncatula NODULIN21, is a tonoplast-localized protein required for secondary wall formation in fibers. Plant J.63: 469–483.

Ranocha, P., et al. (2013). Arabidopsis WAT1 is a vacuolar auxintransport facilitator required for auxin homoeostasis. Nat. Commun.4: 2625.

Sakamoto, T., et al. (2004). An overview of gibberellin metabolismenzyme genes and their related mutants in rice. Plant Physiol. 134:1642–1653.

Sasaki, A., Ashikari, M., Ueguchi-Tanaka, M., Itoh, H., Nishimura,A., Swapan, D., Ishiyama, K., Saito, T., Kobayashi, M., Khush,G.S., Kitano, H., and Matsuoka, M. (2002). Green revolution: a mutantgibberellin-synthesis gene in rice. Nature 416: 701–702.

Somerville, C. (2006). Cellulose synthesis in higher plants. Annu. Rev.Cell Dev. Biol. 22: 53–78.

Sun, T.P. (2011). The molecular mechanism and evolution of the GA-GID1-DELLA signaling module in plants. Curr. Biol. 21: R338–R345.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., andKumar, S. (2011). MEGA5: molecular evolutionary genetics analysisusing maximum likelihood, evolutionary distance, and maximumparsimony methods. Mol. Biol. Evol. 28: 2731–2739.

Tanaka, K., Murata, K., Yamazaki, M., Onosato, K., Miyao, A., andHirochika, H. (2003). Three distinct rice cellulose synthase catalyticsubunit genes required for cellulose synthesis in the secondary wall.Plant Physiol. 133: 73–83.

Taylor, N.G., Howells, R.M., Huttly, A.K., Vickers, K., and Turner, S.R. (2003). Interactions among three distinct CesA proteins essentialfor cellulose synthesis. Proc. Natl. Acad. Sci. USA 100: 1450–1455.

Taylor-Teeples, M., et al. (2015). An Arabidopsis gene regulatorynetwork for secondary cell wall synthesis. Nature 517: 571–575.

Todaka, D., et al. (2012). Rice phytochrome-interacting factor-likeprotein OsPIL1 functions as a key regulator of internode elongationand induces a morphological response to drought stress. Proc.Natl. Acad. Sci. USA 109: 15947–15952.

Ueguchi-Tanaka, M., Ashikari, M., Nakajima, M., Itoh, H., Katoh,E., Kobayashi, M., Chow, T.Y., Hsing, Y.I., Kitano, H., Yamaguchi, I.,and Matsuoka, M. (2005). GIBBERELLIN INSENSITIVE DWARF1encodes a soluble receptor for gibberellin. Nature 437: 693–698.

Updegraff, D.M. (1969). Semimicro determination of cellulose in bi-ological materials. Anal. Biochem. 32: 420–424.

Waadt, R., Schmidt, L.K., Lohse, M., Hashimoto, K., Bock, R., andKudla, J. (2008). Multicolor bimolecular fluorescence complemen-tation reveals simultaneous formation of alternative CBL/CIPKcomplexes in planta. Plant J. 56: 505–516.

Woodger, F.J., Millar, A., Murray, F., Jacobsen, J.V., and Gubler, F.(2003). The role of GAMYB transcription factors in GA-regulatedgene expression. J. Plant Growth Regul. 22: 176–184.

Xie, L., Yang, C., and Wang, X. (2011). Brassinosteroids can regulatecellulose biosynthesis by controlling the expression of CESA genesin Arabidopsis. J. Exp. Bot. 62: 4495–4506.

Yamaguchi, M., Kubo, M., Fukuda, H., and Demura, T. (2008).Vascular-related NAC-DOMAIN7 is involved in the differentiation ofall types of xylem vessels in Arabidopsis roots and shoots. Plant J.55: 652–664.

Yu, S., Galvão, V.C., Zhang, Y.C., Horrer, D., Zhang, T.Q., Hao, Y.H., Feng, Y.Q., Wang, S., Schmid, M., and Wang, J.W. (2012).Gibberellin regulates the Arabidopsis floral transition throughmiR156-targeted SQUAMOSA promoter binding-like transcriptionfactors. Plant Cell 24: 3320–3332.

Zhang, B., Deng, L., Qian, Q., Xiong, G., Zeng, D., Li, R., Guo, L., Li, J.,and Zhou, Y. (2009). A missense mutation in the transmembrane domainof CESA4 affects protein abundance in the plasma membrane and resultsin abnormal cell wall biosynthesis in rice. Plant Mol. Biol. 71: 509–524.

Zhang, S.J., Song, X.Q., Yu, B.S., Zhang, B.C., Sun, C.Q., Knox, J.P., and Zhou, Y.H. (2012). Identification of quantitative trait lociaffecting hemicellulose characteristics based on cell wall compo-sition in a wild and cultivated rice species. Mol. Plant 5: 162–175.

Zhao, Q., and Dixon, R.A. (2011). Transcriptional networks for lignin bio-synthesis: More complex than we thought? Trends Plant Sci. 16: 227–233.

Zhao, Q., Wang, H., Yin, Y., Xu, Y., Chen, F., and Dixon, R.A. (2010).Syringyl lignin biosynthesis is directly regulated by a secondary cellwall master switch. Proc. Natl. Acad. Sci. USA 107: 14496–14501.

Zhong, R., and Ye, Z.H. (2007). Regulation of cell wall biosynthesis.Curr. Opin. Plant Biol. 10: 564–572.

Zhong, R., and Ye, Z.H. (2012). MYB46 and MYB83 bind to the SMREsites and directly activate a suite of transcription factors and sec-ondary wall biosynthetic genes. Plant Cell Physiol. 53: 368–380.

Zhong, R., Demura, T., and Ye, Z.H. (2006). SND1, a NAC domaintranscription factor, is a key regulator of secondary wall synthesis infibers of Arabidopsis. Plant Cell 18: 3158–3170.

Zhong, R., Lee, C., McCarthy, R.L., Reeves, C.K., Jones, E.G., andYe, Z.H. (2011). Transcriptional activation of secondary wall bio-synthesis by rice and maize NAC and MYB transcription factors.Plant Cell Physiol. 52: 1856–1871.

Zhong, R., Lee, C., Zhou, J., McCarthy, R.L., and Ye, Z.H. (2008). Abattery of transcription factors involved in the regulation of secondarycell wall biosynthesis in Arabidopsis. Plant Cell 20: 2763–2782.

Zhong, R., Richardson, E.A., and Ye, Z.H. (2007a). Two NAC domaintranscription factors, SND1 and NST1, function redundantly inregulation of secondary wall synthesis in fibers of Arabidopsis.Planta 225: 1603–1611.

Zhong, R., Richardson, E.A., and Ye, Z.H. (2007b). The MYB46transcription factor is a direct target of SND1 and regulates sec-ondary wall biosynthesis in Arabidopsis. Plant Cell 19: 2776–2792.

1696 The Plant Cell

Dow



Documents

A Gibberellin-Mediated DELLA-NAC Signaling Cascade ...A Gibberellin-Mediated DELLA-NAC Signaling Cascade Regulates Cellulose Synthesis in RiceOPEN Debao Huang,a,1 Shaogan Wang,a,1