Junying Zho, Lei Zheng, Jitong Wei, Yixun Wng, Jun Chen, Yongin Zhou, Ming Chen,Fengzhi Wng, Youzhi M, Zho-Shi Xu,

a Institute of Crop Sciences, Chinese Academy of Agricultural Sciences/National Key Facility for Crop Gene Resources and Genetic Improvement, Key Laboratory of Biology and Genetic Improvement of Triticeae Crops, Ministry of Agriculture, Beijing 100081, China

b Hebei Key Laboratory of Crop Salt-alkali Stress Tolerance Evaluation and Genetic Improvement/Cangzhou Academy of Agriculture and Forestry Sciences, Cangzhou 061000,Hebei, China

Keywords:Soybean PLATZ transcription factor Protein interaction Regulation mechanism Drought response

A B S T R A C T Plant AT-rich sequence and zinc binding (PLATZ) transcription factors are a class of plant specific zincdependent DNA-binding proteins that function in abiotic stress response and plant development. In this study, 31 GmPLATZ genes were identified in soybean. GmPLATZ17 was down-regulated by drought and exogenous abscisic acid. Transgenic Arabidopsis and soybean hairy roots overexpressing GmPLATZ17 showed drought sensitivity and inhibition of stress-associated gene transcription.In contrast,suppressed expression of GmPLATZ17 led to increased drought tolerance in transgenic soybean hairy roots. The GmPLATZ17 protein was verified to interact physically with the GmDREB5 transcription factor,and overexpression of GmDREB5 increased drought tolerance in soybean hairy roots. Interaction of GmPLATZ17 with GmDREB5 was shown to interfere with the DRE-binding activity of GmDREB5, suppressing downstream stress-associated gene expression.These results show that GmPLATZ17 inhibits drought tolerance by interacting with GmDREB5. This study sheds light on PLATZ transcription factors and the function of GmPLATZ17 in regulating drought sensitivity.

1. Introduction

Drought stress limits the growth and development of plants,reducing crop yields [1]. Transcription factors (TFs) function in stress signal transduction by affecting the expression of downstream target genes in plants [2-6]. Zinc finger proteins, one of the largest TF families, are key regulatory components in environmental stress responses, especially drought stress [7-9]. The rice zinc finger protein geneOsMSR15has transcriptional activation ability and has been shown [10] to increase drought tolerance in transgenicArabidopsisby increasing proline content and reducing electrolyte leakage.AnArabidopsis ZAT18knockout mutant showed reduced drought tolerance, whereasZAT18-overexpressing plants showed increased drought tolerance with less leaf water loss and higher antioxidant enzyme activity when compared with the wild-type(WT). [11]. The rice zinc finger transcriptional repressorOsDRZ1increased drought tolerance by inhibiting the promoter activity ofOsGLP1, an inhibitor of stress response [12,13]. Other zinc finger proteins, including TaZFP1B [14], GsZFP1 [15], OsTZF5[16],MtPHD6[17],and IbZFP1[18],are positive regulators of plant drought tolerance. However, some zinc finger proteins have the opposite effect: the chrysanthemum zinc finger transcriptional repressor CmBBX19 interacts with ABF3 to reduce drought tolerance via an abscisic acid (ABA)-dependent pathway [19].

Based on their conserved cysteine(C)and histidine(H)residues,zinc finger proteins are classified into four categories: C6, C2C2,C2H2,and C3HC4[20-22].Plant AT-rich sequence and zinc binding(PLATZ)proteins are a novel class of zinc finger proteins[23].They contain two highly conserved zinc finger motifs (C-x2-H-x11-C-x2-C-x(4-5)-C-x2-C-x(3-7)-H-x2-H and C-x2-C-x(10-11)-C-x3-C) that are necessary for DNA binding. PLATZ TFs were first isolated in pea,and PsPLATZ1 binds nonspecifically to A/T-rich sequences and suppresses the expression of target genes, acting as a transcriptional repressor [23].Arabidopsis AtPLATZ1andAtPLATZ5have been shown to be involved in regulating stress tolerance [25,26]. Overexpression ofAtPLATZ1partially rescued the desiccationintolerance phenotype of theabi3-5mutant, an ABA-insensitiveArabidopsismutant [24,25]. In contrast, overexpressingAtPLATZ5increased sensitivity to salt stress inArabidopsis[26].TheArabidopsisPLATZ geneORESARA15(ORE15) mediates leaf growth and senescence [27]. MaizeFl3(ZmPLATZ12) regulates endosperm development and storage-substance synthesis by participating in the transcriptional regulation of tRNA and 5S rRNA [28]. The rice PLATZ protein GL6/SG6 positively controls cell proliferation,increases cell number in spikelet shells, and affects grain length and yield [29,30]. CottonGhPLATZ1reduces the osmotic and saltstress sensitivity of transgenicArabidopsisduring germination and seedling establishment [31]. Thus, PLATZ genes function in abiotic stress response and plant development.

Soybean(Glycine max),a worldwide source of seed protein and oil, fixes atmospheric nitrogen via symbioses with soilborne microorganisms [32]. PLATZs as novel plant-specific transcription factors, its function has not yet fully understood, specifically in soybean. Overexpression of soybeanGmPLATZ1delayed germination and cotyledon development in transgenicArabidopsisunder ABA and osmotic stresses [33].This is the only report on the function of PLATZ genes in soybean. In this study, the PLATZ family genes were identified in the soybean genome and the mechanism of GmPLATZ17 was proposed in regulating drought tolerance. It will provide valuable information for a more comprehensive understanding of PLATZ family responses to abiotic stress.

2. Materials and methods

2.1. Plant materials and growth conditions

The soybean cultivar Zhonghuang 39 was used. Seedlings were grown in a greenhouse at 25 °C with a photoperiod of 14 h light/10 h dark and 60% relative humidity. For drought treatment,two-week-old soybean seedlings were transferred to filter paper to induce drought. For salt and ABA treatments, seedlings were watered with 200 mmol L-1NaCl or 100 μmol L-1ABA solutions.After stress treatments, seedlings were harvested at 0, 1, 2, 4, 8,12, and 24 h, immediately frozen in liquid nitrogen, and stored at-80 °C prior to RNA isolation.

Transgenic and WTArabidopsisseeds were germinated and grown on 1/2 Murashige and Skoog (MS) medium. At the twoleaf stage, seedlings were transferred to 1/2 MS medium with 9%or 12% polyethylene glycol (PEG) and placed in an incubator with a 16 h light/8 h dark cycle and cultured vertically. To evaluate drought tolerance in soil, seedlings at the four-leaf stage were transferred to mixed soil(humus:vermiculite=1:1)for further cultivation. Water was withheld starting two weeks after transfer to soil.

2.2. Gene isolation and sequence analysis

Soybean PLATZ genes were identified by searching Phytozome(https://phytozome-next.jgi.doe.gov/) (Glycine max Wm82.a2.v1)usingArabidopsisPLATZ protein sequences. Amino acid multiple sequence alignments were constructed with ClustalW. A phylogenetic tree of PLATZ proteins was constructed using the maximumlikelihood method in MEGA7 [34].

2.3. Subcellular localization

The full-length coding sequence (CDS) ofGmPLATZ17was amplified from ‘Zhonghuang 39’ and fused to the N-terminus of green fluorescent protein (GFP) under the control of the CaMV 35S promoter. The recombinant plasmidGmPLATZ17-GFP and nuclear markerEHD4-mCherry [35] were co-transformed intoArabidopsisprotoplasts and GFP fluorescence was observed with a Zeiss LSM700 microscope (Carl Zeiss AG, Jena, Germany) after 18 h [36].

2.4. Generation of transgenic plants

To generate transgenicArabidopsisplants, the open reading frames (ORF) ofGmPLATZ17was cloned into the pCAMBIA1302 vector driven by the CaMV 35S promoter.Genetic transformations were performed using anAgrobacterium-mediated transformation system [37]. Transgenic T3seeds were used for phenotypic analyses.

Transgenic soybean plants (of the cultivar ‘Williams 82’) were generated usingAgrobacterium rhizogenes. The full-lengthGmPLATZ17CDS was fused to the plant transformation vector pCAMBIA3301 under the control of the CaMV 35S promoter to generate the overexpression recombinant plasmid pCAMBIA3301-GmPLATZ17. For the RNAi suppression vector pCAMBIA3301-GmPLATZ17-RNAi, a 604-bp sequence was synthesized(AuGCT,Beijing,China)and inserted into pCAMBIA3301.This sequence contained a 229-bp fragment ofGmPLATZ17(from positions 70 to 298 bp), its reverse complement sequence, and a 146-bp intron fragment of maize alcohol dehydrogenase between the twoGmPLATZ17fragments.The recombinant vector and empty vector (pCAMBIA3301) were introduced intoA.rhizogenesstrain K599, which was then used to infect the cotyledon nodes of soybean as previously described[38].All experiments were performed in biological triplicate.

2.5. Yeast two-hybrid system (Y2H)

The CDS of theGmPLATZ17gene was cloned into the C-terminal of the Gal4 DNA-binding domain (Gal4-BD) in the pGBKT7 vector.The recombinant plasmidGmPLATZ17-BD was used as the bait to identify interacting proteins from the soybean cDNA library. The ORF of the interaction candidate geneGmDREB5was cloned into the pGADT7 vector.GmPLATZ17-BD andGmDREB5-AD plasmids were co-transformed into yeast cells using the lithium acetate method [39]. The transformation efficiency was measured on synthetic-defined (SD) medium lacking tryptophan and leucine(SD/-Trp/-Leu) and the interaction relationship was verified on the synthetic complete medium SD/-Leu/-Trp/-His/-Ade. Empty pGBKT7 or pGADT7 vector was transformed withGmDREB5-AD orGmPLATZ17-BD plasmids as negative controls.Yeast transformation was performed using the Yeastmaker Yeast Transformation System 2 (TaKaRa, Shiga, Japan) according to the manufacturer’s instructions.

2.6. Luciferase complementation assay (LCA)

For assessing transcriptional activity, transient LCA was performed in tobacco as previously described [40]. Genes encoding TFs were cloned into pCAMBIA1305-GFP and target gene promoters were cloned into pGreenII-0800. Recombinant plasmids were constructed and transformed intoAgrobacteriumstrain GV3101.Bacteria containing plasmids were infiltrated into tobacco leaves[3].After infiltration,plants were grown at 22°C with a 16 h light/8 h dark photoperiod for two days.D-luciferin(LUC)(Biovision,Milpitas, CA, USA) was then applied to the backs of tobacco leaves before imaging. LUC activity was measured using anin vivoplant imaging system (NightSHADE LB 985, Berthold Technologies, Bad Wildbad, Germany).

2.7. Bimolecular fluorescence complementation (BiFC)

The ORFs ofGmPLATZ17andGmDREB5were cloned into the Nterminus and C-terminus, respectively, of the yellow fluorescent protein (YFP) coding region to construct the nYFP-GmPLATZ17and cYFP-GmDREB5plasmids. Each recombinant vector (5 μg)was mixed and co-transformed intoArabidopsisprotoplasts as described previously [3].After incubation in the dark for 18 h,fluorescence was observed with a Zeiss LSM700 microscope.

2.8. Pull-down assay

The GST-GmDREB5vector was constructed and fusion-protein purified using Glutathione-Sepharose 4B according to the manufacturer’s instructions (GE Healthcare, Uppsala, Sweden). The maltose-binding protein (MBP)-GmPLATZ17recombinant protein was purified on amylose resin (NEB, Ipswich, MA, USA). Equal volumes of GST or GST-GmDREB5protein were incubated with MBP-GmPLATZ17protein beads in pull-down buffer (50 mmol L-1Tris-HCl at pH 7.5,100 mmol L-1NaCl,0.2%glycerol,and 0.6%Triton X-100)at 4 °C for 6 h.Bound proteins were eluted at 95 °C for 10 min with loading buffer and then separated on 10% SDSpolyacrylamide gel electrophoresis(PAGE)gels.Anti-GST antibody was used as the primary antibody and the chemiluminescent signal was detected using the EasySee Western Blot Kit (TransGen,Beijing, China).

2.9. Electrophoretic mobility shift assay (EMSA)

For the DNA binding activity assay, EMSA was conducted using the Light Shift Chemiluminescent EMSA Kit (Thermo Fisher Scientific, Waltham, MA, USA) and biotin-labeled probes as previously described [35]. The CDSs ofGmPLATZ17andGmDREB5were inserted into the pGEX-4T-1 vector. The glutathione S-transferase(GST)-fused proteins were expressed inEscherichia coli(BL21)and purified using Glutathione Sepharose 4B (GE Healthcare)according to the manufacturer’s instructions. The sequence of the probe for the GmDHN promoter (5′-GTTTATACCGACATGTGGTTTA TACCGACATGTGGTTTATACCGACATGTG-3′) was synthesized with biotin labeling by AuGCT; the same unlabeled fragment was used as a competitor. Purified GST-fused proteins were incubated at 25 °C for 30 min with 2 μL of biotin-labeled probes. The reaction mixture was separated by 6% PAGE and the signal was visualized on nylon membranes (Millipore, Burlington, MA, USA) according to the manufacturer’s protocol.

2.10. Measurements of proline, MDA, SOD and O2- contents

Soybean andArabidopsisleaves that had been drought-stressed for 7 days were frozen in liquid nitrogen.Proline content was measured in the leaves using the proline assay kit (Solarbio, Beijing,China) following the manufacturer’s protocol. Similarly, malondialdehyde (MDA), superoxide dismutase (SOD), and superoxide anion(O2-)contents were measured with MDA,SOD,and O2-assay kits (Solarbio) respectively. All measurements were repeated in triplicate and statistical analysis was performed using analysis of variance (ANOVA).

2.11. DAB and NBT staining

Detached leaves of transgenic soybean with hairy roots under drought treatment for 5 days were stained. The leaves were immersed into 3,3-diaminobenzidine (DAB) or nitroblue tetrazolium (NBT) solution (Solarbio) in darkness for 14 h, then transferred to 75% ethanol for decolorization until the samples turned white [41]. Images were acquired with a Canon 700D camera(Canon, Japan).

2.12. RNA extraction and quantitative real-time PCR (qRT-PCR)

Total RNA was extracted with a KKFast Plant RNApure Kit(ZOMANBIO, Beijing, China) from soybean hairy roots. RNase-free DNase I (TaKaRa) was used to remove genomic DNA contamination. Quality and purity were determined from the OD260/OD280ratio (1.8-2.0) and integrity was verified by 1.2% agarose gel electrophoresis. First-strand cDNA was synthesized using Fastking gDNA Dispelling RT SuperMix (Tiangen, Beijing, China). qRT-PCR was performed using Talent qPCR PreMix (Tiangen) in the ABI 7500 Real-Time PCR system (Thermo Fisher Scientific) following the manufacturer’s protocol. The primers are shown in Table S1.

CYP2(GmCYP2) (Glyma.12G024700) (Glycinemax Wm82.a2.v1)was used as an endogenous control. qRT-PCR data were analyzed using the 2-ΔΔCT method [42]. There were three technical replicates for each sample.

2.13. Statistical analysis

All experiments were repeated at least three times independently. Data were analyzed using Microsoft Excel 2016 (Microsoft Corporation, Albuquerque, NM, USA) and GraphPad Prism 5(GraphPad Software, San Diego, CA, USA). ANOVA was used to determine significant differences.

3. Results

3.1. Characterization of PLATZ transcript factors in soybean

Thirty-one PLATZs were identified in the soybean genome and were namedGmPLATZ1-GmPLATZ31based on their chromosomal locations (Table S2). Sequence alignment (Fig. S1) showed that all of the GmPLATZ proteins contained the conserved zinc finger motifs C-x2-H-x11-C-x2-C-x(4-5)-C-x2-C-x(3-7)-H-x2-H and C-x2-Cx(10-11)-C-x3-C.

31 soybean GmPLATZs and 12ArabidopsisAtPLATZs were categorized into four discrete groups (I to IV) using the maximumlikelihood method (Fig. 1). In the zinc finger domain C-x2-H-x11-C-x2-C-x(4-5)-C-x2-C-x(3-7)-H-x2-H, the vicinity of the zinc-binding site showed clear differences between the four groups, suggesting that there might be functional differences amongGmPLATZgenes(Fig. S2).

3.2. GmPLATZ17 encoding a nuclear localized protein was downregulated by multiple abiotic stresses

It has been previously reported [25,26] that PLATZ group III genes affect stress tolerance inArabidopsis. To identify stressrelated PLATZ genes in soybean,we accordingly selected the group III genes for further investigation. RNA-seq data (Fragments per kilobase million, FPKM) were retrieved from SoyBase (https://www.soybase.org/) (Table S3) and visualized as a heat map.GmPLATZs in group III showed varying expression in soybean leaves, flowers, pods, seeds, roots, and nodules (Fig. 2A).GmPLATZ17was more highly expressed than other genes in several tissues, especially roots and nodules. qRT-PCR confirmed thatGmPLATZ17showed the highest transcript levels in roots(Fig. 2B), suggesting thatGmPLATZ17might function in root stress responses. Accordingly,GmPLATZ17was selected for further verification.

Fig. 1. Phylogenetic tree of PLATZ transcription factors in soybean and Arabidopsis.

Fig. 2. Tissue-specific expression of GmPLATZs in group III. (A) Heat map of gene expression (in FPKM) of six GmPLATZs in leaves, flowers, pods, seeds, roots, and nodules.(B)Quantification of gene expression of six GmPLATZs in roots by qRT-PCR.

As GmPLATZ17 contains a putative nuclear localization signal(Fig. S3), we investigated its subcellular localization. A GmPLATZ17-GFP recombinant plasmid driven by the CaMV 35S promoter was constructed and co-transformed intoArabidopsisprotoplasts with the nuclear marker geneEHD4,and control plants were transformed with 35S::GFP vector.Fluorescence signals were observed using confocal microscopy.GmPLATZ17-GFP fluorescence was detected in the nucleus and cytoplasm, whereas the control vector was uniformly distributed throughout the protoplast(Fig. 3A).

qRT-PCR was performed to measureGmPLATZ17expression under abiotic stresses.GmPLATZ17was suppressed under drought stress (Fig. 3B) and was also suppressed under salt and ABA treatments(Fig.3C,D).These results suggest thatGmPLATZ17may play a negative regulatory role in abiotic stress responses.

3.3. GmPLATZ17 showed transcriptional inhibitory activity

The transcription activation or inhibition activity of GmPLATZ17 were evaluated. Like the empty BD vector, yeast strains harboring theGmPLATZ17-BD fusion construct could not grow on SD/-Trp/-His/-Ade selective medium, suggesting that GmPLATZ17 has no transcriptional activation ability (Fig. 4A). A dual luciferase reporter system was used for transient expression assays inArabidopsisprotoplasts to further determine whether GmPLATZ17 was a transcriptional repressor [12,43]. The ORF ofGmPLATZ17was fused to Gal4-BD as an effector plasmid(Fig.4B);the reporter plasmid contained firefly LUC,and the internal control plasmid contained renilla LUC. The effector and two reporter plasmids were co-transformed intoArabidopsisprotoplasts and LUC activity was measured.ERF11(AT1G28370),which has been shown[44]to be a transcriptional repressor,was used as a negative control. As shown in Fig. 4C, relative LUC activity was lower in GmPLATZ17 and ERF11 than in the empty vector control.These results indicated that GmPLATZ17 has transcriptional repressor activity and suppressed transcription of the reporter genes.

3.4. Overexpression of GmPLATZ17 increased drought sensitivity in transgenic Arabidopsis

Two independent transgenicArabidopsislines were chosen to further investigate the function ofGmPLATZ17at high expression levels(Fig.S4).Two-leaf seedlings were transferred to 1/2 MS medium with or without PEG and cultured vertically.There was no visible difference in the growth of transgenic lines compared to the WT under normal conditions (Fig. 5A). Plant growth was suppressed by PEG, an effect that was more severe inGmPLATZ17transgenic seedlings than in the WT after 7 days. In the PEG medium, the primary roots ofGmPLATZ17-overexpression lines were shorter than those of the WT (Fig. 5B). The fresh weight (FW) of aboveground tissue also showed differences between transgenic and WT seedlings.The mean FWs of the two transgenic lines were only 8.7 and 10.3 mg respectively,whereas that of the WT was over 12.3 mg in the 12% PEG treatment (Fig. 5C).

We tested the tolerance of transgenicArabidopsisgrowing in soil under drought stress (Fig. 5D). Before drought treatment, all plants were grown uniformly. After nine days of dehydration, theGmPLATZ17-overexpressing seedlings began to wilt and turn chlorotic, whereas the growth of WT seedlings was minimally affected.After 13 days of dehydration,most of the transgenic seedlings had died,but WT plants grew well and their survival rate was higher (Fig. 5E).

Proline prevents cell water loss by reducing cellular osmotic potential and restoring intracellular concentrations of solute.MDA content is greatly increased when the enzyme and membrane systems of plant tissues are destroyed [45,46]. Proline and MDA contents are accordingly frequently used as indicators of stress resistance in plants. In the present study, the proline content was significantly lower inGmPLATZ17-overexpressing plants (69.44-7 3.54 μg g-1FW) than in the WT (93.72 μg g-1FW) under drought treatment (Fig. 5F), whereas the MDA content was significantly higher in transgenic plants than in WT seedlings(Fig.5G).We conclude thatGmPLATZ17affected drought tolerance in transgenicArabidopsis.

3.5. Overexpression of GmPLATZ17 increased drought sensitivity in soybean hairy roots

To investigate the function ofGmPLATZ17in soybean,GmPLATZ17-overexpressing (GmPLATZ17-OE) andGmPLATZ17-RNAi transgenic hairy root plants were generated byA.rhizogenesmediated transformation. This method allows rapid study of gene function in soybean but does not produce stably inheritable transgenic traits [38].GmPLATZ17-OE hairy roots accumulated higher levels ofGmPLATZ17transcripts compared to the empty vector control (EV-control), whereas the transcriptional level ofGmPLATZ17in RNAi soybean hairy roots was significantly inhibited(Fig. S5). Water was withheld starting three weeks after infection of hairy roots.Morphologically,all plants grew well and there were no marked differences before drought treatment, but there were visible differences after seven days. The leaves ofGmPLATZ17-OE plants became chlorotic first and were the most sensitive to drought stress; the EV control was less sensitive, and theGmPLATZ17-RNAi hairy roots were least susceptible(Fig.6A).After six days of rehydration, theGmPLATZ17-OE plants died and their survival rate was significantly lower than that of the empty vector control, and the survival rate of RNAi plants was highest (Fig. 6B).Under normal growth conditions, DAB and NBT leaf staining showed no differences. Under drought treatment, leaves of theGmPLATZ17-OE plants were stained more deeply, butGmPLATZ17-RNAi leaves were lighter than those of the EV control (Fig. 6C, D).The content ofwas significantly higher inGmPLATZ17-OE hairy roots than inGmPLATZ17-RNAi plants under drought treatment(Fig. 6E), indicating that theGmPLATZ17-OE plants accumulated more. SOD accumulation was lower in leaves ofGmPLATZ17-OE plants than in those ofGmPLATZ17-RNAi plants (Fig. 6F). These results showed thatGmPLATZ17could reduce the drought tolerance of transgenic soybean hairy roots.

Fig. 4. Transcriptional activity analysis of GmPLATZ17. (A) Transcriptional activity analysis of GmPLATZ17 in yeast cells. (B) Schematic representation of the recombinant plasmid. (C)Relative LUC activity assay of reporter genes. A transcriptional repressor, ERF11,was used as a negative control. Values are means ±standard deviation (n=5).

Fig.3. Localization and expression of GmPLATZ17.(A)GmPLATZ17 was localized to the nucleus and cytoplasm.Scale bars,10 μm.(B-E)Expression profile of GmPLATZ17 in 2-week-old soybean seedlings under drought, salt, and ABA treatments. GmCYP2 was used as the endogenous control.

To elucidate the drought-response mechanism ofGmPLATZ17,the expression levels of several stress-responsive genes were measured in soybean hairy root plants under drought and normal conditions.After drought stress,the stress-responsive genesGmCOR47,GmDHN,GmLEA3, andGmMYB84were significantly downregulated inGmPLATZ17-OE hairy root plants but up-regulated in RNAi plants compared with the EV control (Fig. 7). Based on DAB and NBT staining and oncontent(Fig.6B,C),we speculated thatGmPLATZ17-OE plants reduce drought tolerance by accumulating more reactive oxygen species(ROS)than EV-control.To verify this hypothesis, we measured the expression of genes involved in ROS scavenging. qRT-PCR showed that the catalase and peroxidase genesGmCAT1andGmPODwere significantly down-regulated inGmPLATZ17-OE plants and up-regulated inGmPLATZ17-RNAi plants(Fig. 7). It appears thatGmPLATZ17may reduce soybean drought tolerance by altering the expression of stress genes involved in stress response.

Fig. 5. Response of transgenic GmPLATZ17 Arabidopsis plants to drought treatment. (A) Phenotypes of WT and transgenic seedlings on medium with or without PEG. (B, C)Primary root length(B)and fresh weight(C)of transgenic lines and WT plants.(D)Phenotypes of WT and transgenic plants under drought treatment in soil.(E)Survival rate of WT and transgenic plants under drought stress.(F,G)Proline(F)and MDA(G)contents in WT and transgenic lines.Values are mean±standard deviation(n=3).*,P ＜0.05;**, P ＜0.01 (ANOVA).

3.6. GmDREB5, the interaction candidate of GmPLATZ17

We next investigated how GmPLATZ17 affects drought tolerance in transgenic plants. This was accomplished by screening the soybean cDNA library using GmPLATZ17 as the bait in a yeast two-hybrid system. One candidate interaction gene, a DREB transcription factorGmDREB5(Glyma.13G298600), was identified. All of the yeast strains grew normally on SD/-Trp/-Leu medium,whereas only the strain with AD-GmDREB5 and BD-GmPLATZ17 co-expressed grew well on selective medium (SD/-Trp/-Leu/-His/-Ade) (Fig. 8A).

A pull-down assay was conductedin vitrowith the purified recombinants MBP-GmPLATZ17 and GST-GmDREB5. MBPGmPLATZ17, but not MBP,bound the GST-GmDREB5 protein, confirming a direct interaction between GmPLATZ17 and GmDREB5(Fig. 8B). To verify the GmPLATZ17 and GmDREB5 interactionin vivo, we performed a BiFC assay. YFP fluorescence was detected in the nucleus when nYFP-GmPLATZ17 and cYFP-GmDREB5 were co-expressed (Fig. 8C). In contrast, no fluorescence was observed when nYFP-GmPLATZ17 and cYFP or when nYFP and cYFPGmDREB5 were co-expressed. These results suggest the presence of physical interactions between GmPLATZ17 and GmDREB5.

Fig. 6. GmPLATZ17 function under drought stress in soybean. (A) Phenotypes were evaluated in transgenic soybean plants before and after drought treatment. (B) Survival rate of soybean hairy roots grown with or without drought treatment. (C,D)DAB (C)and NBT(D) staining of GmPLATZ17-RNAi,EV control, and GmPLATZ17-OE plant leaves under drought treatment. (E, F) Contents of O2- (E) and SOD (F) in GmPLATZ17-RNAi, EV control, and GmPLATZ17-OE plants under drought or normal conditions. Values are means ± standard deviation (n = 3). *, P ＜0.05; **, P ＜0.01 (ANOVA).

Fig. 7. Expression levels of stress-responding genes in transgenic soybean hairy root plants. GmCYP2 was used as internal control. Values are means of three technical replicates. *, P ＜0.05; **, P ＜0.01 for transgenic lines compared to the control.

3.7. Overexpression of GmDREB5 increased drought tolerance in soybean hairy roots

Many DREB transcription factors have been shown to be involved in drought tolerance [47-49]. In agreement with these prior observations,GmDREB5was induced by drought (Fig. S6).To investigate the function ofGmDREB5, we generatedGmDREB5overexpression (GmDREB5-OE) and RNAi (GmDREB5-RNAi) transgenic soybean hairy roots. Before treatment, there was no significant difference in growth or physiology betweenGmDREB5-RNAi,GmDREB5-OE,and EV-control hairy root plants (Fig.9A). However,under drought treatment,GmDREB5-OE hairy root soybean showed delayed leaf shriveling and a higher survival rate compared with the EV control andGmDREB5-RNAi plants (Fig. 9A, B). Thus,GmDREB5appeared to increase drought tolerance of transgenic soybean hairy roots. We measured the expression of stressresponse genes inGmDREB5-RNAi, EV-control, andGmDREB5-OE soybeans. qRT-PCR results showed thatGmCOR47,GmDHN,GmLEA3, andGmMYB84were up-regulated inGmDREB5-OE soybean plants and down-regulated inGmDREB5-RNAi plants compared to the control (Fig. 9C, F).

Fig. 8. Interaction between GmPLATZ17 and GmDREB5. (A) Interaction analysis of GmPLATZ17 and GmDREB5 using a yeast two-hybrid system. (B) Pull-down assay of GmPLATZ17 and GmDREB5. (C) Interaction of GmPLATZ17 and GmDREB5 as determined with a BiFC assay. Scale bars, 20 μm.

3.8. GmPLATZ17 interferes with DRE-binding affinity of GmDREB5

Previous study[50]revealed that DREB family transcription factors bind to DREcis-elements (CCGAC) to regulate the expression of target genes. Promoter sequences analysis showed that theGmDHNandGmLEA3promoters contained the DREcis-element.In the promoter region ofGmDHN, the DRE element was located 337 bp upstream of the start codon (Fig. 10A). We used a 51-bp fragment containing three repeats of the DRE element of theGmDHNpromoter (position -332 to -349) as a probe in an EMSA.The result showed that the DRE element in theGmDHNpromoter was directly bound by the GST-GmDREB5 fusion protein, but not by GST. The binding shift to the biotin-labeled target sequence was dramatically reduced when unlabeled competitor target DNA sequence was added, and no binding was detected when the mutated biotin-labeledGmDHNprobe was added (Fig. 10B). Thus,GmDREB5 bound specifically to the DRE element of theGmDHNpromoter.

A LUC expression system and EMSA were used to detect whether the binding activity of GmDREB5 was affected by GmPLATZ17. The pGreenII-0800 vector carrying a LUC reporter gene driven by theGmDHNpromoter was co-transformed into tobacco leaves with GmDREB5 or GmPLATZ17.Fluorescence signal was significantly reduced in plants co-transformed with GmDREB5, GmPLATZ17, andGmDHNpro::LUCcompared to plants co-transformed with just GmDREB5 andGmDHNpro::LUC(Fig. 10C). The ability of GmDREB5 to bind to theGmDHN-probe was reduced when GST-GmPLATZ17 was added (Fig. 10D). Thus,GmPLATZ17 reduced the promoter-binding activity of GmDREB5.

4. Discussion

Soybean is an important crop worldwide and is sensitive to drought. The changing climate has contributed to an increase in soil drought in recent years,posing a serious threat to soybean production.The biological functions of plant PLATZ genes are diverse,affecting growth and development [27-30] and participating in drought stress responses [25,26,31], but little was previously understood about their biological function in abiotic stress responses in soybean. In this present study, our results revealed thatGmPLATZ17mediated negatively drought response in plant(Figs. 5, 6).

Abiotic stresses affect the growth and development of plants due to accumulation of intracellular ROS [51-53]. Catalase (CAT)and peroxidase (POD) are ROS-scavenging enzymes and a component of an organism’s antioxidant capacity [54,55]. Severe stress conditions can induce high levels of ROS,which react directly with cellular lipids and DNA to cause oxidative damage and dysfunction[56]. High expression of CAT and POD genes can reduce ROSinduced cell damage, increasing plant stress tolerance [55].GmMYB84, a MYB transcription factor,depends on ROS to regulate primary root elongation, which contributes to drought tolerance[57]. Cold-regulated (COR) genes are considered to be markers for drought stress [58,59]. Late embryogenesis-abundant (LEA)protein and dehydrin (DHN) are two regulators of infiltration.When plants are exposed to drought conditions,cell osmotic pressure changes. This causes LEA and DHN proteins to accumulate to maintain osmotic homeostasis,which contributes to drought tolerance [60-62]. To determine the regulatory mechanism ofGmPLATZ17in plants, we measured the expression of ROSscavenging and stress-responsive genes inGmPLATZ17-OE andGmPLATZ17-RNAi plants, includingGmCAT1,GmPOD,GmMYB84,GmCOR47,GmLEA3, andGmDHN. These genes were repressed inGmPLATZ17-OE plants compared with the EV-control, and the opposite phenomenon was observed inGmPLATZ17-RNAi plants(Fig. 7). These results further suggest thatGmPLATZ17exerts an inhibitory effect on plant stress tolerance.

Fig.9. GmDREB5 function in soybean under drought stress. (A)Phenotypes of GmDREB5-RNAi,EV-control, and GmDREB5-OE under drought stress. (B)Survival rate of hairy root plants grown with or without drought treatment.(C-F)Expression levels of stress-response genes in GmDREB5-RNAi,EV-control,and GmDREB5-OE transgenic soybean.GmCYP2 was used as the internal control. Values are means ± standard deviation (n = 3). *, P ＜0.05; **, P ＜0.01 (ANOVA).

DREB factors are involved in the regulation of plant tolerance to drought and cold by binding specifically with the DRE element to regulate downstream stress-responsive gene expression [63,64].Overexpression ofDREB2Aincreased drought tolerance in transgenicArabidopsisplants [50]. Overexpression ofGmDREB1increased drought tolerance in transgenic wheat[65].In our study,overexpression ofGmDREB5increased the drought tolerance of transgenic soybean (Fig. 9), an effect contrary to that ofGmPLATZ17. We accordingly hypothesized that GmPLATZ17 mediates regulation of plant stress tolerance by interacting with and affecting the DRE-binding ability of GmDREB5. To verify our hypothesis, we analyzed the promoters of stress-responsive genes and found that the promoters ofGmDHNandGmLEA3contained the DREcis-element. EMSA showed that the binding ability between GmDREB5 and DRE was weakened in the presence of GmPLATZ17(Fig.10).This result confirmed that GmPLATZ17 could limit the function of GmDREB5.

We propose a model in which the stress-response regulatory network involves an interaction between GmPLATZ17 and GmDREB5 that interferes with the ability of GmDREB5 to bind its target genes, thereby tightly regulating drought stress response(Fig. 11).

Research[66-69]has revealed that several DREB TFs can inhibit plant growth and delay flowering, although overexpression of DREB genes can also increase plant drought tolerance. The functions of proteins such as GmPLATZ17 are thus of interest. We hypothesize thatGmPLATZ17alleviates the negative growth and developmental effects of DREBs in normal conditions via direct interaction, and that when plants experience stresses, expression ofGmPLATZ17decreases and GmDREB5 protein is released, allowing it to activate genes involved in drought response.

Fig.10. GmPLATZ17 interferes with the binding affinity of GmDREB5 to the GmDHN promoter.(A)Structure of the GmDHN promoter.The underlined sequences indicate the DRE elements in the GmDHN probe. (B) GmDREB5 binds directly to the DRE element. For the competitor test, the unlabeled target DNA fragments in 20-, 50- or 100-fold concentrations were added to assess binding specificity. (C) The interaction of GmDREB5 or GmPLATZ17 with the GmDHN promoter was detected using a dual luciferase reporter system.Empty vectors were used as negative controls.(D)GmPLATZ17 disrupted the binding affinity of GmDREB5 to the GmDHN promoter.For the competitor test,purified GST-GmPLATZ17 protein in 1-, 2- or 3-fold concentrations were added to the experiment.

Fig.11. A proposed working model of the GmPLATZ17-GmDREB5 interaction involved in drought-stress response in soybean.Under normal conditions,GmPLATZ17 interacts with GmDREB5,interfering with the binding of GmDREB5 to its target genes.Under stress conditions such as drought,GmPLATZ17 is down-regulated,freeing GmDREB5 and activating expression of stress-responsive genes such as GmDHN.

5. Conclusions

Thirty-one PLATZ genes were identified in the soybean genome.Overexpression ofGmPLATZ17reduces drought tolerance by interfering with the DRE-binding activity of GmDREB5. This study provides a theoretical basis for future detailed analysis of PLATZ functions in soybean.

CRediT authorship contribution statement

Juanying Zhao:Methodology, Investigation, Writing-original draft.Lei Zheng:Writing-review & Editing, Data curation.Jitong Wei:Methodology, Validation.Yixuan Wang:Investigation.Jun Chen:Methodology.Yongbin Zhou:Resources.Ming Chen:Methodology.Fengzhi Wang:Resources.Youzhi Ma:Supervision.Zhao-Shi Xu:Conceptualization, Funding acquisition, Project administration, Supervision.

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 work was supported by the National Natural Science Foundation of China (31871624), the Agricultural Science and Technology Innovation Program (CAAS-ZDRW202109 and CAAS-ZDRW202002), and the Central Public-interest Scientific Institution Basal Research Fund. We thank Profs. Lijuan Qiu, Shi Sun, and Dr. Taifei Yu of the Institute of Crop Sciences, CAAS for kindly providing soybean seeds and revising the manuscript.

Appendix A. Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2022.03.009.