Zhiying Zho,Sh Tng,Wei Li,Xiorui Yng,Ruiju Wng,Xinmin Dio,,*,Wenqing Tng,*

a Ministry of Education Key Laboratory of Molecular and Cellular Biology,Hebei Collaboration Innovation Center for Cell Signaling,Hebei Key Laboratory of Molecular and Cellular Biology,College of Life Sciences,Hebei Normal University,Shijiazhuang 050024,Hebei,China

b Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,China

c Foxtail Millet Improvement Center of China,Institute of Millet Crops,Hebei Academy of Agricultural and Forestry Science,Shijiazhuang 050031,Hebei,China

d College of Bioscience &Bioengineering,Hebei University of Science and Technology,Shijiazhuang 050018,Hebei,China

Keywords:SiBZR1 SiPLT-L1 Brassinosteroids Drought tolerance Foxtail millet

ABSTRACT Two potential BRASSINAZOLE RESISTANT 1 (BZR1) homologs were downregulated by brassinosteroids(BRs) in Setaria italica roots.Functional analysis showed that BR regulates the dephosphorylation and nuclear localization of SiBZR1 and that SiBZR1 binds conserved BZR1-recognizing cis elements.In comparison with the wild type, SiBZR1-overexpressing S.italica seedlings were more sensitive to BR-inhibited primary root growth and drought stress,indicating that SiBZR1 is a positive regulator of BR signaling and a negative regulator of drought tolerance in S.italica. PLETHORA-LIKE 1 (SiPLT-L1) was found to be a direct target gene of SiBZR1 in S.italica roots.The expression of SiPLT-L1 was downregulated by SiBZR1.SiPLT-L1-overexpressing S.italica was less sensitive to BR-inhibited root growth and more tolerant to drought stress,possibly owing to the upregulation of drought-inducible Dehydrin-family genes.

1.Introduction

Brassinosteroids (BRs) play essential roles in regulating diverse plant growth and developmental processes [1].BRs are perceived by the membrane-localized receptor BRASSINOSTEROID INSENSITIVE 1,which initiates a signal transduction pathway to activate downstream transcription factors known as BRASSINAZOLE RESISTANT 1 (BZR1)-family proteins (BZRs) [1,2].When the BR level is low,BZRs are phosphorylated by BRASSINOSTEROID INSENSITIVE 2 (BIN2)-family protein kinases [3,4].BIN2 phosphorylation prevents BZRs from binding with DNA and promotes the translocation of BZRs from the nucleus to the cytoplasm[5,6].Increasing the BR level inactivates BIN2 [7,8].BZRs are dephosphorylated by protein phosphatase 2A (PP2A) and accumulated in the nucleus,where they bind to the promoters of BR target genes and regulate their expression to modulate plant growth and development [9,10].

Foxtail millet(Setaria italica)is a model C4panicoid crop known for its drought tolerance.It has been proposed [11] that the drought tolerance ofS.italicais associated with the expression of genes involved in BR biosynthesis and signaling.BR was also found[12]to regulate the upright leaf architecture ofS.italica,and foliar application of BR increased the plant height,leaf size,and fresh weight ofS.italicaseedlings[13].BR also regulates spikelet meristem determination during inflorescence development inS.viridis,the wild ancestor ofS.italica[14].These few studies suggest that BRs function in regulating the growth and development of foxtail millet.However,our understanding of the BR signaling mechanism and of the role of BRs in regulating the growth and development as well as adaptation to abiotic and biotic stresses in foxtail millet is still limited.

To understand the mechanism by which BRs regulate root growth and drought tolerance in foxtail millet,RNA-Seq analysis was performed to identify the BR regulated genes in foxtail millet root.From these genes,we cloned aBZR1homolog (SiBZR1) and aPLETHORA-LIKE 1(SiPLT-L1),and generated overexpression transgenic foxtail millet plants to analyze the function of these genes.

2.Materials and methods

2.1.Growth conditions and drought treatment

TheS.italicacultivars Yugu-1 and Ci846 were used.Seeds were germinated in darkness at 28°C for 2–3 days,after which seedlings with similar root and coleoptile lengths were selected and transferred to new growth medium or to soil and grown at 28°C under long-day (16 h light/8 h dark) conditions.For root growth assays,seeds were chlorine vapor-sterilized for 6–8 h,germinated on 1/2 MS agar medium,transferred to a 6 cm (diameter) × 20 cm(height) glass cylinder containing 1/2 MS agar medium supplied with 24-epi-brassinolide (eBL) or the BR biosynthesis inhibitor propiconazole (PCZ),and allowed to grow for 6 days before being imaged,removed for root length and lateral root number measurements,or cleared for microscopic observation.Averaged measurements from at least three biological replicates (13–30 plants per experiment) are presented.

To clear the roots for microscopic observation,～0.5 cm lengths of primaryS.italicaroot tips were removed and soaked in a clearing solution (Visikol for Plant Biology;Visikol Inc.,Hampton,NJ,USA) for 12 h and then observed under a Zeiss Axio Imager M2 microscope.

For drought treatment,S.italicaseedlings were grown in individual pots containing soil and vermiculite (1:1,v/v) for 10 days.Seedlings showing uniform growth were either watered as usual or not watered for 14 days before being photographed for phenotypic comparison or harvested for quantitative real-time PCR(qRT-PCR).

2.2.RNA-Seq analysis

Roots of 8-day-old light-grown seedlings showing similar root and coleoptile lengths were immersed in 1 μmol L-1eBL or a mock solution for 2 h before being harvested and ground to a fine powder in liquid nitrogen.For total RNA extraction,0.1 g of root tissue powder was mixed with 1.2 mL of extraction buffer(100 mmol L-1Tris-HCl,pH 9.5,150 mmol L-1NaCl,1.0% sarkosyl,0.5% βmercaptoethanol),vortexed for 5 min,and centrifuged at 11,000×gfor 5 min.The supernatant was mixed with 0.5 vol of chloroform and vortexed for 2 min.After addition of a 0.5 vol of water-saturated phenol and vortexing for 2 min,the mixture was centrifuged at 11,000×gfor 15 min.The upper aqueous phase was transferred to a new RNase-free Eppendorf tube and total RNA was precipitated by addition of 90 μL of 3 mol L-1sodium acetate (pH 5.2) and 600 μL of isopropanol at room temperature for 10 min.Total RNA was pelleted,rinsed twice with 75%ethanol,briefly air-dried,and resuspended in diethyl pyrocarbonatetreated water.The RNA was further purified using Trizol reagent(Invitrogen,Carlsbad,CA,USA) so that the quality would meet the requirement for RNA-Seq.

RNA-Seq was performed with the Illumina HiSeq X-ten platform.Paired-end reads were aligned to theS.italicareference genome(JGIv2.0.34)using TopHat(version 2.0.12)[15],and gene read counts were obtained using htseq-count from the Python package HTSeq (version 0.61) [16].The RNA-Seq reads were submitted to NCBI (http://www.ncbi.nlm.nih.gov/sra/) under accession number PRJNA587050.Differential expressed genes (DEGs) are defined usingP<0.05 and a fold-change >1.5 or <0.67 as cutoffs.

2.3.qRT-PCR analysis

First-strand cDNA was synthesized using M-MLV Reverse Transcriptase (Takara Bio Inc.,Otsu,Japan).qRT-PCR was performed according to a standard protocol using the SYBR Premix Ex Taq system (Takara Bio Inc.) withActin(SiActin) as an internal reference.The primers used are listed in Table S3.Means of at least three biological replicates are presented.

2.4.Electrophoretic mobility shift assays (EMSAs)

EMSAs were performed using recombinant MBP-SiBZR1 and MBP-AtBZR1 purified fromEscherichia coliand biotin-labeled probes.Probe binding was performed using a LightShift EMSA Optimization and Control Kit (ThermoFisher Scientific Inc.,Waltham,MA,USA).

2.5.Cloning and generation of transgenic S.italica

The coding sequences ofSiBZR1andSiPLT-L1were cloned intopENTR/SD/D-TOPOvector(Invitrogen)and subcloned into the destination binary vectorpCAMBIA1305orpGWB5by LR cloning (Invitrogen).The constructs were introduced intoAgrobacterium tumefaciensstrainGV3101orEHA105and transformed intoArabidopsisby the floral-dipping method or intoS.italica(cultivar Ci846) by callus-based gene transformation.

2.6.Chromatin immunoprecipitation (ChiP)

ChIP-qPCR was performed according to the procedures described previously [17],using GFP-Trap agarose (ChromoTek,Planegg-Martinsried,Germany) and primers listed in Table S3.

3.Results

3.1.Identification of BR-regulated root genes by RNA-Seq

WhenS.italicaseedlings (cultivar Yugu-1) were grown in the presence of eBL,the primary root growth and lateral root formation of the seedlings were stimulated by low concentrations but inhibited by higher concentrations of eBL (Fig.S1).By RNA-Seq,115 BR-regulated genes (57 upregulated and 58 downregulated)were identified fromS.italicaroots (Fig.S2A;Table S1).Among these genes,homologs of several BR biosynthesis genes:SiD2,SiCPD,SiDWARF,andSiDWF4,were downregulated by BR treatment,suggesting that the eBL treatment and RNA-Seq experiments were successful.qRT-PCR confirmed the differential expression of 15 selected DEGs using newly isolated total RNA (Fig.S2B).

Using the Phytozome (http://phytozome.jgi.doe.gov) and Setaria italica Functional Genomics Database (http://structuralbiology.cau.edu.cn/SIFGD/) as references,combined with sequence similarity BLAST at the NCBI and The Arabidopsis Information Resource sites,functional annotations were retrieved for 87 of the 115 BR-regulated root genes (Fig.S2C;Table S1).

3.2.Biochemical and molecular characterization of SiBZR1

TwoBZR1homologs were downregulated in BR-treated Yugu-1 roots(Fig.S2B).Searching theS.italicagenome revealed two additional BZR1 homologs (Figs.S3 and S4).Of these homologs,the SiBZR1 protein sequence was 86% identical to OsBZR1 and 54%identical to AtBZR1.Motifs corresponding to DNA binding,14-3-3 binding,and PP2A binding were all conserved in SiBZR1 (Fig.S5).In agreement with the presence of these conserved protein motifs in SiBZR1,SiBZR1 interacted directly withArabidopsisBIN2 and PP2A and bound strongly to an AtBZR1 recognizing aciselement from theAtDWF4promoter (Figs.S6 and S7A).These results suggested that SiBZR1 possessed biochemical and molecular properties similar to those of AtBZR1.

To investigate the biological function of SiBZR1in vivo,pUbi:SiBZR1-eGFPtransgenicS.italica(SiBZR1-OX) was generated using cultivar Ci846,because of its high gene transformation efficiency(Fig.S7B).WhenSiBZR1-OXseedlings were grown in medium containing 50 μmol L-1PCZ,SiBZR1-eGFP signals in root cells were observed in both the cytosol and the nucleus,and a great amount of SiBZR1-eGFP was dephosphorylated (Fig.S7C,D).BR treatment promoted both dephosphorylation and nuclear localization of SiBZR1-eGFP(Fig.S7C,D).Similar results were observed inSiBZR1-overexpressingArabidopsisplants (Fig.S8).

3.3.Overexpression of SiBZR1 activated BR signaling in S.italica

At the two-leaf stage,SiBZR1-OX S.italicaleaves were wider than wild-type leaves (Fig.1A,B).When plants were grown to the four-leaf stage,theSiBZR1-OXleaves were twisted along the lengthwise direction(Fig.1C).When the plants were fully mature,the twisted-leaf phenotype was no longer evident and the height of the fully grownSiBZR1-OXplants was similar to that of control plants (Fig.S9).SiBZR1-OXpanicles were longer (Fig.1D),and theSiBZR1-OXkernels were on average 59.5% ± 8.6% larger and 47.3% ± 5.3% heavier than the wild-type control (Fig.1E–G).However,possibly owing to the reduced number of primary branches per panicle (Fig.1H),overall grain weight produced per panicle in theSiBZR1-OXplants was not increased (Fig.1I).

The primary roots ofSiBZR1-OXseedlings were slightly shorter than those of the wild-type control (Fig.S10),and were hypersensitive to BR with respect to primary-root growth and lateral-root initiation(Fig.S11A–C).qRT-PCR analysis showed that the expression levels ofSiDWARFandSiDWF4were reduced inSiBZR1-OXplants regardless of the presence of BRs (Fig.S11D).SiBZR1-OXplants were less sensitive to PCZ-inhibited primary root growth(Fig.S11E,F).

3.4.SiBZR1-OX plants are less tolerant to drought stress

WhenSiBZR1-OXand wild-type plants were subjected to drought treatment (Fig.2A),12 days after the cessation of watering,all leaves on the wild-type plants were still green and healthy,but the first leaf of theSiBZR1-OXplants had become yellow(Fig.2B).At 15 days after the cessation of watering,mostSiBZR1-OXleaves were wilted and/or dried,whereas a great number of leaves on the wild-type plants were still upright and green(Fig.2C,D).These results suggested that SiBZR1 negatively regulates drought tolerance inS.italica.

3.5.SiPLT-L1 is a direct target of SiBZR1

The RNA-Seq results showed that the expression of twoPLETHORA(PLT)-LIKEgenes(SiPLT-L1andSiPLT-L4)was downregulated by BR (Figs.S2B and S12).SiPLT-L1is expressed predominantly in roots (Fig.S13A),and its expression was significantly decreased as soon as 30 min after BR treatment (Fig.S13B),and inSiBZR1-OXroots (Fig.S13C).TheSiPLT-L1promoter and 5′-untranslated region (UTR) contains multiple BZR1-recognizing BRRE (CGTGT/CG) and E-box (CANNTG)cis-elements (Fig.S13D),EMSA showed that SiBZR1 bound specifically to aSiPLT-L1promoter fragment containing a BRRE and a nearby E-box(Fig.S13E).ChIP-qPCR assay revealed significant enrichment of the BRRE-containing fragments (–782 to–689 or–782 to–673)from theSiPLT-L1promoter usingpUbi:SiBZR1-eGFProots(Fig.S13F).These findings suggested thatSiPLT-L1is a direct target of SiBZR1.

Fig.1.Characterization of SiBZR1-eGFP-overexpressing S.italica. pUbi:SiBZR1-eGFP plants at the two-leaf stage (A),four-leaf stage (C),and fully mature stage (D–E).Leaf (A and C),panicle (D),and seed (E) phenotypes are shown.(B) Quantitation of leaf width for the seedlings shown in A.(F–I) Quantitation of 1000-kernel weight,kernel area,primary branch number per panicle,and mass of seeds produced per panicle in SiBZR1 overexpressing S.italica. OX1, OX2, OX3, OX5,and OX8 are different transgenic lines.Ci846 is the wild-type control.The arrows in(C)indicate twisted young leaves.The bars in(D)and(E)represent 1.5 cm and 1 cm,respectively.Error bars indicate mean±SD.Statistically significant differences are indicated by different lowercase letters (P <0.05,one-way ANOVA with Tukey’s significant difference test).

Fig.2. SiBZR1 overexpression reduces the drought tolerance of S.italica.(A) Relative soil water content changes during 12 days of drought treatment.(B,C) Phenotypes of wild-type (Ci846) and SiBZR1-OX plants subjected to drought or well-watered (control) conditions for 12 days (B) or 15 days (C) are shown.Arrows in (B) indicate yellow leaves of SiBZR1-OX plants.(D) Survival rates of the drought-treated plants shown in (C) after 8 days of re-watering.Error bars indicate mean ± SD.Statistically significant differences are indicated by different lowercase letters (P <0.05,one-way ANOVA with Tukey’s significant difference test).

3.6.BR-inhibited root growth and drought tolerance is mediated by SiPLT-L1

Using length/width ratio ≤1 as a criterion to define root cortical meristem cells revealed that BR inhibited cortical meristem development inS.italicaroots (Fig.S14A,B),and this phenomenon was more pronounced inSiBZR1-OXplants (Fig.S14C,D) and less pronounced inSiPLT-L1-OXplants (Fig.S14E–G),suggesting that down-regulated expression ofSiPLT-L1by SiBZR1 contributed to BR-inhibited primary root growth inS.italica.

SiPLT-L1overexpression did not change the kernel size ofS.italica(Fig.S15).When subjected to drought stress,time needed for the first and second leaves of theSiPLT-L1-OXplants to turn yellow is increased (Fig.S16).Drought-associated leaf wilting and drying were more pronounced in the wild type plants (Fig.3A),and theSiPLT-L1-OXplants showed better recovery after re-watering(Fig.3B),indicating thatSiPLT-L1overexpression increased drought tolerance inS.italica.

Comparison of AtPLT1/2-targeted genes with BR-regulatedArabidopsisroot genes [18,19] revealed 1511 BR and AtPLT1/2 coregulated genes (Table S2).Among these were severalDehydrinandAquaporingenes,which are known to function in regulation of plant drought response [20].A sequence search against theS.italicagenome using the protein sequences of theseArabidopsisdehydrins and aquaporins yielded threeDehydrinhomologs(Si5G286000,Si8G115200,andSi1G267200) and twoAquaporinhomologs (Si2G123000andSi1G241900) for further analysis.qRTPCR analysis showed that eBL significantly inhibited the expression of all exceptSi1G241900(Fig.3C).Drought stress induced the expression ofDehydrinsbut inhibited the expression ofAquaporins(Fig.3D–H),and drought-induced expression ofDehydrinswas more pronounced inSiPLT-L1-OXplants than in the wild-type control (Fig.3D–F).These results suggested that SiPLT-L1-mediated expression ofDehydrinsmight contribute to BR-regulated drought responses inS.italica.

4.Discussion

SiBZR1,aS.italicaBZR1 homolog,positively regulated BR signaling inS.italica.However,unlike AtBZR1,a great amount of SiBZR1 was maintained in a dephosphorylated form and localized to the nucleus even in the absence of BR,suggesting that the mechanism by which BR controls the activation of SiBZR1 inS.italicais not identical to that of AtBZR1 inArabidopsis.Identifying the unique mechanism that regulates the activity of SiBZR1 inS.italicawill expand our knowledge of the BR signaling mechanism and may shed light on the evolution of BR signaling in plants.

SiBZR1-OX S.italicakernels were larger and heavier than those of wild-type plants.However,the mean grain weight perSiBZR1-OXplant was not significantly increased,possibly owing to the limited space between spikelets or insufficient nutrient supply during grain filling because of the larger kernels produced.These results suggest identifying the BR signaling mechanism inS.italicaand using this knowledge to alter BR signaling in a tissue-specific manner is a valuable approach for increasing the productivity ofS.italica.

SiBZR1-OX S.italicaplants were hypersensitive to drought stress.This result is in agreement with our previous study [11],which showed that the BR concentration and drought-induced expression ofSiBZR2were increased in the drought-sensitiveS.italicacultivar An04.In general,on drought stress exposure,plants develop a larger root system to access available water.Recently[21],it was reported that root growth inS.viridisshifts from crown root-derived to primary root-derived branches under drought stress conditions.Compared with the wild-type control,SiBZR1-OXplants had shorter primary roots and slightly decreased lateral root numbers and were hypersensitive to BR-inhibited primary root growth and lateral root initiation.It is possible that BR signaling-inhibited primary root growth contributed to the drought-hypersensitive phenotype ofSiBZR1-OXplants.In contrast toSiBZR1-OXplants,SiPLT-L1-OX S.italicahad a larger root meristematic zone,longer primary roots,and reduced sensitivity to BR-inhibited primary root growth.More interestingly,SiPLT-L1-OXplants were more tolerant to drought stress.Downstream of SiPLTL1,the expression of severalDehydrinscould be downregulated by BR and upregulated by drought orSiPLT-L1overexpression.Thus,our study provides the first evidence that PLT family transcription factors not only are key regulators of root development but also regulate plant drought resistance (Fig.S17).Determining whether SiPLT-L1 increases the drought tolerance ofS.italicaby promoting primary root growth,increasing the expression of droughtregulated genes,or both,will shed new light on the mechanisms by which BR regulate plant drought responses inS.italica.

Fig.3.Overexpression of SiPLT-L1 increased drought tolerance in S.italica.(A,B) Phenotypes and survival rates of SiPLT-L1-OX plants after 8 days of re-watering.(C–H)Relative expression levels of Dehydrin and Aquaporin genes in S.italica roots treated with 1 μmol L-1 eBL for 2 h(C),or in SiPLT1-L1-OX plants treated with or without drought stress (D–H). OX39 and OX40 are two independent transgenic lines.Error bars indicate mean ± SD.Statistically significant differences are indicated by asterisks;*P <0.05,**P <0.01 (Student’s t-test).

CRediT authorship contribution statement

Xianmin Diao and Wenqiang Tang:conceived and designed the project.Zhiying Zhao:performed most of the experiments and analyzed the result with the help fromWei Li,Xiaorui Yang and Ruiju Wang:.Sha Tang:generated the transgenic S.italica plants.Xianmin Diao and Wenqiang Tang:wrote the manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by the National Key Research and Development Program of China (2018YFD1000706,2018YFD1000700),the National Natural Science Foundation of China (91417313,31970313),and the Department of Education of Hebei Province (SJ2016012).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.02.006.