Phosphorylation of SiRAV1 at Ser31 regulates the SiCAT expression to enhance salt tolerance in Setaria italica

Ll Qiao-lu, Ll Zhi-yong, WANG Meng-meng, YAN Jing-wei, FANG Lin#

1 Guangdong Provincial Key Laboratory of Applied Botany/South China Botanical Garden, Chinese Academy of Sciences,Guangzhou 510650, P.R.China

2 Institute of Millet Crops, Hebei Academy of Agriculture and Forestry Sciences, Shijiazhuang 050000, P.R.China

3 Nanjing Institute of Environmental Science, Ministry of Ecology and Environment, Nanjing 210042, P.R.China

4 State Key Laboratory of Subtropical Silviculture, Zhejiang A&F University, Hangzhou 311300, P.R.China

Abstract Salinity severely affects plant growth and development.Thus, it is crucial to identify the genes functioning in salt stress response and unravel the mechanism by which plants against salt stress.This study used the phosphoproteomic assay and found that 123 of the 4 000 quantitative analyzed phosphopeptides were induced by salt stress.The functional annotation of the non-redundant protein database (NR) showed 23 differentially expressed transcription factors, including a phosphopeptide covering the Serine 31 in the RAV (related to ABI3/VP1) transcription factor (named SiRAV1).SiRAV1 was located in the nucleus.Phenotypic and physiological analysis showed that overexpressing SiRAV1 in foxtail millet enhanced salt tolerance and alleviated the salt-induced increases of H2O2 accumulation, malondialdehyde (MDA)content, and percent of electrolyte leakage.Further analysis showed that SiRAV1 positively regulated SiCAT expression to modulate the catalase (CAT) activity by directly binding to the SiCAT promoter in vivo and in vitro.Moreover, we found that phosphorylation of SiRAV1 at the Ser31 site positively regulated salt tolerance in foxtail millet via enhancing its binding ability to SiCAT promoter but did not affect its subcellular localization.Overall, our results define a mechanism for SiRAV1 function in salt response where salt-triggered phosphorylation of SiRAV1 at Ser31 enhances its binding ability to SiCAT promoter, and the increased SiCAT expression contributes to salt tolerance in foxtail millet.

Keywords: foxtail millet, salt stress, phosphoproteomic, SiRAV1, SiCAT

1.lntroduction

As an important crop in northern China and some Asian countries, foxtail millet has a long history of cultivation for about 7 000 years.Due to its medicinal properties and antioxidant ability, it is widely recommended in many nutritional and health foods (Adekunle and Fatunbi 2012;Liet al.2021).It strongly tolerates hostile environments,such as salt and drought stresses (Doustet al.2009;Sharma and Niranjan 2018; Jieet al.2021).However,the mechanisms underlying tolerance to these hostile environments are still poorly understood.

Specific proteins inside cells play a crucial role in alleviating stress by enhancing cellular processes that allow plants to survive adverse conditions.Before plants can synthesize nascent proteins in response to stress,proteins already present in cells can undergo a series of post-translational modifications, such as phosphorylation,acylation, glycosylation, nitration, and ubiquitination,resulting in a rapid response (Kwonet al.2006; Vuet al.2018).These activated proteins can, in turn, aid further stress responses.Among them, phosphorylation modification is one of the most common modification types at present.The phosphorylation process is catalyzed by kinases, the phosphate group of ATP is transferred to the amino acid side chain of the protein,and ATP becomes ADP.It is estimated that nearly onethird of all proteins in eukaryotes may be phosphorylated at any time (Damaris and Yang 2021).Thus, excavating the phosphorylated proteins in the field of plants and unraveling the mechanism of these proteins in foxtail millet is very important to increase agricultural productivity.

Phosphoproteomic analysis is widely used to investigate the role of phosphoproteins in plant growth and development, as well as abiotic and biotic stress responses.The phosphoproteomic approach could identify the phosphorylated proteins and obtain the phosphorylation sites and the phosphorylation levelsin vivo(Amagaiet al.2018; Piet al.2018; Xiet al.2021).For example, 27 phosphorylation sites from 20 different phosphorylated proteins in NaCl-treatedArabidopsiswere identified using the label-free quantitative analysis (Changet al.2012).In maize, Huet al.(2013) identified a total of 139 phosphorylation sites from 72 proteins after NaCl treatment.Among them, 37 phosphorylated proteins were newly identified as players in salt stress response (Huet al.2013).In addition, the phosphoproteomic approach was used to explore the mechanisms underlying salt tolerance in other plant species, such as soybean, rice,and wheat (Zhanget al.2014, 2015; Panget al.2018;Piet al.2018).However, phosphoproteomic analysis in foxtail millet exposed to salt stress had not been reported in the literature.

The transcription factor RAV (related to ABI3/VP1)belongs to the APETALA2/ethylene responsive factor(AP2/ERF) family, which widely exists in plants such asArabidopsis, maize, and foxtail millet (Lataet al.2014;Gaoet al.2020).The characteristic feature of the RAV superfamily is the presence of two highly conserved motifs,the AP2- and the B3-like domains.RAV can recognize the CAACA and CACCTG motifs of its target genes’ promoters to directly modulate their expressions (Fenget al.2014).RAV proteins play vital roles in regulating the early stage of development and growth retardation, leaf angle, seed development, root development, anthocyanin biosynthesis,and leaf senescence (Wooet al.2010; Fenget al.2014;Shin and Nam 2018; Tianet al.2019; Gaoet al.2020;Liuet al.2021).Besides that, RAV proteins coordinate the response to biotic and abiotic stresses, including plant disease, drought stress, ABA response, BR response, salt stress, and cold stress (Duanet al.2016; Zhaoet al.2017;Amagaiet al.2018; Weiet al.2018; Gaoet al.2020).For instance, AtRAV1, AtRAV1L, and AtRAV2 modulate salt stress response inArabidopsis(Fuet al.2014).In maize,overexpression ofZmRAV1inArabidopsisenhances salt tolerance (Minet al.2014).In soybean, overexpressing ofGmRAV-03inArabidopsisenhances salt resistance (Zhaoet al.2017).InMedicagotruncatula, MtRAV3 positively regulates the salt responseviamodulating the expression of adversity-related genes (Wang Set al.2021).However,whether the foxtail millet RAV is involved in salt response is still unclear.

Here, we first took a phosphoproteomic approach to reveal key differentially phosphorylated proteins (e.g.,SiRAV1 and SiWRKY) that might be involved in salt response in foxtail millet.Further, we unraveled the mechanism of phosphorylated modification on SiRAV1 in salt response.

2.Materials and methods

2.1.Plant materials and growth conditions

Foxtail millet (Seteriaitalica) inbred line Yugu 1 was used in this study.Seeds were sown on pots containing soil mixture (soil:vermiculite, 1:2, v/v) in a growth chamber(25°C, 200 μmol L-1m-2s-1, 14 h light/10 h dark, 60%humidity).

2.2.lsolation of total RNA and real-time PCR analysis

Total RNA was isolated using the RNAprep Pure Plant Kit(TIANGEN, China) as described by Yanet al.(2023).The NanoDrop? 2000 spectrophotometer (Thermo Scientific,USA) was used to detect the quantity and purity of RNA;then the high-quality RNA was used to synthesize the cDNA using the HiScript?II 1st Strand cDNA Synthesis Kit (+gDNA wiper, Vazyme, China).Real-time PCR was carried out on a CFX96 real-time PCR system (Bio-Rad,USA) using ChamQ SYBR?qPCR Master Mix (Vazyme)according to the manufacturer’s instructions.The relative expression levels forSiRAV1andSiCATwere measured using themethod withSiActin7as described previously (Livak and Schmittgen 2001).

2.3.Foxtail millet transformation and regeneration

The full-lengthSiRAV1or its mutants were introduced into the plant express vector pCUN-NHF driven byubiquitinpromoter (Appendix A).The foxtail millet inbred line Yugu 1 was used as the plant receptor.Agrobacteriummediated foxtail millet shoot-tip transformation was used(Yanget al.2021).In brief, the tip of the coleoptile of etiolated seedlings (about 1 cm) was cut by a blade,then placed in the infection solution (800 μmol L-1acetosyringone, OD600=1.4) under 0.05 MPa vacuum pressure for 40 min.After infection, the foxtail millet was placed on a pot (nutrient soil:vermiculite=1:1, v/v) and continued to culture normally in the greenhouse at 28°C.Positive transformants were selected by spraying with 25 μg mL-1DL-phosphinothricin (Sigma-Aldrich, USA)and then further confirmed by PCR amplification.The T2seedlings showed resistant-to-susceptible segregation of 3:1, and resistant seedlings were transferred to pots for continuous cultivation until homozygous T3seeds from individual lines were obtained.

2.4.Subcellular localization

Agrobacteriumtumefaciensstrain GV3101 carrying the35S-SiRAV1-YFPor its mutants (35S-SiRAV1S31AYFPand35S-SiRAV1S31D-YFP) and nucleus marker(NLS-mCherry) was co-infiltrated into four-week-old tobacco (Nicotianabenthamiana) leaves as described by Grebenoket al.(1997).The signals were observed using a Leica laser-scanning confocal microscope (Leica, TCP SP8, Germany) (YFP, excitation at 513 nm and emission at 527 nm; mCherry, excitation at 587 nm and emission at 610 nm).

2.5.Phenotype and oxidative damage analysis

One-week-old seedlings (WT and transgenic plants)grown on pots were irrigated with or without 300 mmol L-1NaCl every three days for nine days.The fresh weight was measured.

For the measurement of malondialdehyde (MDA) and the percentage of electrolyte leakage, one-week-old seedlings(WT and transgenic plants) grown on pots were irrigated with or without 300 mmol L-1NaCl for three days, then the MDA and the percentage of electrolyte leakage were analyzed as described by (Huet al.2021; Wuet al.2022).

2.6.Detection of H2O2 content and CAT activity

One-week-old seedlings (WT and transgenic plants)grown on pots were irrigated with or without 300 mmol L-1NaCl for three days, then the samples were harvested.H2O2content was measured using the Hydrogen Peroxide Assay Kit (Beyotime) by following the manufacturer’s instructions.For measurement of the CAT activity,samples were homogenized in 0.6 mL of 50 mmol L-1potassium phosphate buffer (pH 7.0) containing 1 mmol L-1EDTA and 1% polyvinylpyrrolidone.The homogenate was centrifuged at 12 000×g for 30 min at 4°C, and then total CAT activity was assayed by measuring the rate of decomposition of H2O2at 240 nm.

2.7.Protein extraction, digestion, and enrichment

One-week-old seedlings of foxtail millet treated with or without 300 mmol L-1NaCl were sampled to extract proteins with the urea-extraction method.Briefly, 0.5 g sample was ground into fine powder after liquid nitrogen refrigeration, then added with 2 mL extraction buffer(10 mmol L-1dithiothreitol, 1% protease inhibitors, 1%phosphatase inhibitors, 50 μmol L-1PR-619, 3 μmol L-1trichostatin A, 50 mmol L-1nicotinamide, and 2 mmol L-1EDTA).The mixture was lysed by sonication and then centrifuged at 6 000×g for 10 min.The supernatant was precipitated overnight with ammonium acetate/methanol,washed with methanol and acetone, and then resolved in 8 mmol L-1urea to obtain total protein.The extracted protein was reduced with 5 mmol L-1dithiothreitol at 56°C for 30 min, alkylated with 11 mmol L-1iodoacetamide in the dark for 15 min, then digested with trypsin (v/v, 50:1) at 37°C overnight to obtain peptides.The digested peptides were resolved with binding buffer (80% acetonitrile, 5%trifluoroacetic acid, 1 mol L-1lac acid), then incubated with TiO2beads three times.The TiO2beads enriched with phosphopeptides were washed with elution buffer (40%acetonitrile, 15% NH3H2O) four times, and the eluates were finally dried to powder with speed vacuum and then submitted to Novogene Company (Beijing, China) for LCMS/MS analysis.

2.8.Phosphoproteomic analysis

Peptides were separated on a Michrom Magic C18AQ(200 μm×15 cm) capillary reverse-phase column at a flow rate of 300 nL min-1, and eluted with 5-30% acetonitrile gradient with an Easy-nLC1000 liquid chromatography system (Thermo Scientific, USA).Obtained phosphosites were filtered with a criterion of peptide Andromeda search engine score>60, phosphosite probability>0.5, and delta score>0.5, and phosphosite intensities were normalized by using the subtraction of the most frequent value(position of global maximum) function.A threshold of fold change>1.2 (or <0.83) andq-value<0.05 was used to screen the differentially expressed phosphorylated proteins.The selected proteins were annotated with the Gene Ontology (GO) database performed by TBtools Software (Chenet al.2020).The transcription factor prediction was carried out by blast against the Plant Transcription Factor Database (http://planttfdb.gao-lab.org/prediction.php).The subcellular localization was predicted with TargetP-2.0 online tools (https://services.healthtech.dtu.dk/service.php?TargetP-2.0).The interactions between differentially expressed proteins were predicted by STRING online tools (https://stringdb.org/), in which the confidence score was set to 0.9 to assure reliability.The results were further visualized by Cytoscape Software (Shannonet al.2003).

2.9.Luciferase assays

The promoter ofSiCATwas introduced into the 1381-LUC vector.The full-lengthSiRAV1or its mutants was introduced into the 1305-YFP vector.Agrobacterium tumefaciensstrain GV3101 carrying the above vectors,and 1305-REN vector were infiltrated into four-week-old tobacco leaves with various combinations as indicated.The LUC and REN activities were separately measured using Dual-Luciferase Reporter Gene Assay Kit (Vazyme,China), and the LUC/REN was also calculated after three days of infiltration.

2.10.Yeast one hybrid

The vector pAbAi containing the fragment ofSiCATpromoter was transformed into the Y1HGold strain to generate the bait strain.Then, the empty vector or pGADT7 AD containing the coding sequence ofSiRAV1or its mutants (SiRAV1S31AandSiRAV1S31D) was transformed into the Y1H bait strain and cultured on a synthetic dropout/-Leu/-Ura plate with 300 ng mL-1Aureobasidin A(AbA) (Clontech, USA) (Liet al.2021).

2.11.Western blot analysis

Proteins were extracted from plant samples as described previously (Yanet al.2022).The protein content was determined by Bradford assays.Total proteins were separated by 12% SDS-PAGE.The PAGE gels were transferred onto polyvinylidene fluoride (PVDF)membranes and then incubated overnight with primary antibodies at 4°C, including anti-Flag antibodies (1:1 500,produced in mouse, Beijing Solarbio Science &Technology Co., Ltd.) and anti-Actin antibodies (1:2 000,produced in mouse, Abmart).After washing with PBST(3 times), the membranes were then incubated with HRPconjugated secondary antibodies (1:2 000, produced in goat, Abmart) at room temperature for 2 h.The signals were analyzed using chemiluminescence imaging (Tanon 5200 Multi, Tanon Biomart).

For detecting phosphorylated protein, crude proteins were immunoprecipitated with anti-Flag antibodies bound to Protein A/G MagBeads and separated by 12%SDS-PAGE.After transferring onto PVDF membranes,phosphorylated SiRAV1 was detected with the anti-Phos-(Ser/Thr) antibodies (1:10 000, produced in rabbit, Abcam).The protein inputs were detected by immunoblotting with the anti-Flag antibodies.

3.Results

3.1.Phosphoproteomic analysis of salt-treated foxtail millet

As one of the most post-translational modifications,protein phosphorylation plays a vital role in response to salt stress.Thus, a quantitative phosphoproteomic assay was used to excavate the phosphorylated protein of foxtail millet in response to salt stress.In total, 4 064 phosphorylated sites from 2 236 phosphorylated proteins were quantitatively analyzed.A heatmap showed 293 of the phosphorylated proteins were obviously induced by salt stress (Fig.1-A; Appendix B).Then, a GO annotation was performed, and the function of the saltregulated differentially expressed phosphorylated proteins could be classified into three main categories: biological process, cellular component, and molecular function.The result indicated that the phosphorylated proteins mainly participated in cellular process, metabolic process, cellular anatomical entity, binding, and catalytic activity (Fig.1-B).

The functions of proteins are critically influenced by their subcellular localization, and plant cells control protein localization to regulate biological processes.As shown in Fig.1-C and Appendix C, most of the differentially expressed proteins were located in the nucleus (51.19%),followed by cytoplasm (15.02%), plasma membrane(13.99%), and chloroplast (12.29%).Only a small number of proteins were located in mitochondria (5.46%), Golgi apparatus (0.34%), endoplasmic reticulum (0.34%), or extracellular space (1.37%).Besides, 24 transcription factors were differentially phosphorylated, including 10 down-phosphorylated and 14 up-phosphorylated(Appendix D).The transcription factors consist of RAV,WRKY, etc.Interestingly, the phosphorylation peptide ALAAAAAsAGAPLER in SiRAV (mass-to-charge ratio=710.35, QvalityPEP=2E-4, and retention time=59.62 min) (Seita.5G280700.1.p) was only detected after salt treatment (normalized abundance=520 498.38), implying that the phosphorylation of SiRAV (Seita.5G280700.1.p)might play an important role in regulating salt response in foxtail millet.RAV proteins have been found to be involved in salt stress response inArabidopsis, maize,soybean, andMedicago(Wang Set al.2021).However,the contributions of RAV in foxtail millet are still unclear.

3.2.The sequence analysis and subcellular localization of SiRAV1

Since the function of the RAV transcription factor in foxtail millet has not been described, we focused on the role of SiRAV in response to salt stress.A phylogenetic analysis was first used to evaluate the evolutionary between SiRAV(XP_004969625.1) and the other RAV proteins.As shown in Fig.2-A, SiRAV (Seita.5G280700.1.p) displayed a close relationship with ZmRAV1 (Zeamays, ACG41637.1),PmRAV1 (Panicummiliaceum, RLN24973.1), AtRAV1(Arabidopsisthaliana, OAP16099.1), and NtRAV1(Nicotianatabacum, XP_016512131.1), thus was designated as SiRAV1.Sequence alignment indicated that SiRAV1 exhibited high identities (58.3-87.4%) with its close RAV proteins.SiRAV1 had a single typical AP3 domain and a B3 domain (Fig.2-B).

Fig.2 The sequence analysis and subcellular localization of SiRAV1.A, phylogenetic analysis of SiRAV1 and closely related RAV protein in other species using DNAMAN V6 Program.The accession numbers of these RAV proteins are as follows: SiRAV1 (Setaria italica, Seita.5G280700.1.p), ZmRAV1 (Zea mays, ACG41637.1), NtRAV1-like (Nicotiana tabacum, XP_016512131.1), PmRAV1(Panicum miliaceum, RLN24973.1), and AtRAV1 (Arabidopsis thaliana, OAP16099.1).Numbers in A indicate the bootstrap values after 1 000 replicates.B, alignment of the SiRAV1 with closely related RAV protein in other species using MEGA6.The putative AP2 domain is marked by a black arrow, and B3 domain is marked by a red arrow.C, the subcellular localization of SiRAV1.The 35S:YFP or 35S:YFP-SiRAV1 were transiently expressed in N.benthamiana leaf cells.Scale bar=20 μm.The experiment was performed at least three times with similar results.

The bioinformatics analysis predicted that SiRAV1 was located in the nucleus.To verify that, the transient expression ofSiRAV1fused with yellow fluorescence proteins (YFP) in tobacco cells was analyzed.As shown in Fig.2-C, SiRAV1-YFP protein was observed at the nucleus co-localized with established nucleus marker protein (NLS-mCherry) (Grebenoket al.1997), indicating that SiRAV1 is located in the nucleus.

3.3.Overexpression of SiRAV1 in foxtail millet enhances salt tolerance

To study the role ofSiRAV1in response to salt stress in foxtail millet, theSiRAV1overexpressors were obtained,and two independent lines (OE-SRAV1#1andOESiRAV1#2) confirmed by qRT-PCR and immunoblotting assays were used to perform salt hypersensitivity test(Fig.3-A and B).In the absence of NaCl, there was no significant morphological difference in the growth of foxtail millet seedlings between wild type (WT) andSiRAV1overexpressors.However, in the presence of NaCl,SiRAV1overexpressors showed less severe wilting and chlorosis than WT plants (Fig.3-C).Also,the shoot fresh weight ofSiRAV1overexpressors was greater than those of the WT under salt treatment(Fig.3-D).

Fig.3 Overexpressing SiRAV1 in foxtail millet enhances salt tolerance.The expression (A) and protein level (B) of SiRAV1 in wild type (WT) and SiRAV1 overexpressors plants.Expression of SiRAV1 in these lines was analyzed by qRT-PCR assay.The protein level of SiRAV1 was detected by immunoblotting analysis using an anti-Flag antibody.C, phenotype of the WT and SiRAV1 overexpressors under salt stress.One-week-old seedlings of WT and SiRAV1 overexpressors grown on pots were irrigated with or without 300 mmol L-1 NaCl for nine days, then the representative photographs were shown.D, shoot fresh weight of the WT and SiRAV1 overexpressors.E, electrolyte leakage rates.F, malondialdehyde (MDA) contents.G, H2O2 contents.H, catalase (CAT) activities in foxtail millet leaves of WT and SiRAV1 overexpressors under NaCl treatment.Oneweek-old seedlings of WT and SiRAV1 overexpressors grown on pots were irrigated with or without 300 mmol L-1 NaCl for three days, then the above index was analyzed.DW, dried weight.I-M, salt-responsive genes (SiSOS1, SiSOD, SiLTP, SiNHX4,and SiASR4) in foxtail millet leaves of WT and SiRAV1 overexpressors under NaCl treatment.One-week-old seedlings of these plants grown on pots were irrigated with or without 300 mmol L-1 NaCl for 12 h, and then the SiCAT expressions were analyzed by qRT-PCR.Error bars in A and D-M indicate SD (n=3).Different letters in A and D-M indicate significant differences at P<0.05 according to Duncan’s multiple range test.

Next, we analyzed several physiological indexes,such as ion leakage, malondialdehyde (MDA), and H2O2accumulation, which could typically reflect the degree of oxidative damage under salt stress (Wang Met al.2021; Wanget al.2023).In the absence of NaCl, there were no significant differences in electrolyte leakage, MDA content, and H2O2accumulation between WT andSiRAV1overexpressors.However, in the presence of NaCl, the above physiological index inSiRAV1overexpressors was much lower than those in the WT (Fig.3-E-G).Catalase (CAT) plays a vital role in response to salt stressviascavenging excess H2O2accumulation caused by salt stress (Zhouet al.2018).Hence, the CAT activity in WT andSiRAV1overexpressors in response to salt stress was measured.As expected, the CAT activity inSiRAV1overexpressors was higher than those in WT plants under NaCl treatment (Fig.3-H).Furthermore, several salt stress-responsive genes, such asSiSOS1,SiSOD,SiLTP,SiNHX1, andSiASR4, were also detected in WT andSiRAV1overexpressors exposed to salt stress (Liet al.2017; Xiaoet al.2023).As shown in Fig.3-I-M,the expressions ofSiSOS1,SiSOD,SiLTP, andSiASR4induced by salt stress inSiRAV1overexpressors were much higher compared to those in WT, while the expression ofSiNHX1was not affected.Taken together, these results suggest that SiRAV1 enhances salt tolerance by increasing the expressions of ROS scavenger-associated and stress-responsive genes to alleviate the salt-induced oxidative damage.

3.4.SiRAV1 positively regulates SiCAT expression by directly binding to the SiCAT promoter in vivo and in vitro

A previous study has shown that a foxtail millet geneSiCAT, encoding the CAT enzyme, functioned in response to salt stress (Liet al.2017).Thus, the effect of SiRAV1 on the expression ofSiCATwas first analyzed.Under the NaCl treatment, the expression ofSiCATinSiRAV1overexpressors was much higher than those in the WT (Fig.4-A).Next, a dual luciferase assay system in tobacco leaves was used to test the interaction between SiRAV1 andSiCATpromoter.As shown in Fig.4-B,SiRAV1 could obviously enhance the activity of LUC driven by theSiCATpromoter, which suggests that SiRAV1 might directly regulate theSiCATexpression.

Fig.4 SiRAV1 positively regulates SiCAT expression by directly binding to the SiCAT promoter in vivo and in vitro. A, the expression of SiRAV1 in foxtail millet leaves of wild type (WT) and SiRAV1 overexpressors under NaCl treatment.One-week-old seedlings of WT and SiRAV1 overexpressors were irrigated with or without 300 mmol L-1 NaCl for 12 h, and then the SiRAV1 expressions were analyzed by qRT-PCR.B, the effect of SiRAV1 on luciferase activity driven by SiCAT promoter in tobacco.The above is the schematic representation of the double-reporter and effector plasmids.LUC represents firefly luciferase and REN means Renilla luciferase.The LUC/REN ratios represent the SiCAT promoter activities.C, SiRAV1 binds the promoter of SiCAT in yeast.The fragment of SiCAT promoter (-30 to -100) was used in the Y1H assay.The number at the left represents the dilution times of an optical density at 600 nm.Values in A and C show mean±SD (n=3).The experiment in B was performed at least three times with similar results.Different letters in A and C indicate significant differences at P<0.05 according to Duncan’s multiple range test.

RAV proteins can bind the specific motif (CAACA and CACCTG) to modulate their target genes’ expression(Zhanget al.2020).The CAACA motif was found in theSiCATpromoter (-45 to -49).To confirm the interaction between SiRAV1 andSiCATpromoterin vitro, the yeast one hybrid (Y1H) test was used.As shown in Fig.4-C,SiRAV1 could directly bind to theSiCATpromoter in yeast.

3.5.Phosphorylation of SiRAV1 at the Ser31 site positively regulates salt tolerance in foxtail millet

Since SiRAV1 positively regulated the salt tolerance(Fig.3) and salt stress induced the phosphorylation of SiRAV1 at the Ser31 site (Appendix D), it is reasonable to assume that the phosphorylation site of SiRAV1, Ser31,plays a vital role in response to salt stress.SiRAVS31A,a phosphor-ablative mutant, was obtained by mutating Ser31 to Ala31, and SiRAVS31D, the phosphor-mimicking mutant was obtained by mutating Ser31 to Asp31 (Zouet al.2020).Two independentSiRAVS31Aoverexpressors and two independentSiRAVS31Doverexpressors were generated and confirmed by qRT-PCR and immunoblotting assays (Fig.5-A and B).

Then, the salt hypersensitivity test was performed to explore the role of phosphorylation of SiRAV1 at Ser31 in response to salt stress.In the absence of NaCl,there was no significant morphological difference in the growth of foxtail millet seedlings between WT,SiRAVS31Aoverexpressors, andSiRAVS31Doverexpressors.However,in the presence of NaCl,SiRAVS31Doverexpressors showed less severe wilting and chlorosis than WT andSiRAVS31Aoverexpressors (Fig.5-C).The shoot fresh weight ofSiRAVS31Doverexpressors was greater than those of the WT andSiRAVS31Aoverexpressors under salt treatment (Fig.5-D).Also, the malondialdehyde(MDA), ion leakage, and H2O2accumulation inSiRAVS31Doverexpressors were much lower than those in WT andSiRAVS31Aoverexpressors exposed to salt stress(Fig.5-E-G).Meanwhile, the CAT activity andSiCATexpression inSiRAVS31Doverexpressors were much higher than those in WT andSiRAVS31Aoverexpressors exposed to salt stress (Fig.5-H and I).To further investigate the effect of salt stress on phosphorylation of SiRAV1 at Ser31, SiRAV1 phosphorylation was examined in WT andSiRAVS31Aoverexpressors exposed to salt stress using anti-Phos-(Ser/Thr) antibody.As shown in Fig.5-J,SiRAV1 phosphorylation was only detected in NaCltreated WT plants, while it cannot be detected in normal plants.Importantly, SiRAV1 phosphorylation triggered by salt stress was lost when Ser31 was mutated to Ala31,indicating that salt stress could trigger the phosphorylation of SiRAV1 at Ser31.Taken together, these results suggest that phosphorylation of Ser31 in SiRAV1 plays a positive role in the salt tolerance of foxtail millet.

3.6.Phosphorylation of SiRAV1 at the Ser31 site enhances its binding ability to SiCAT promoter but does not affect its subcellular localization

To explore the effect of phosphorylation of SiRAV1 at Ser31 on SiRAV1 function, the subcellular localization of SiRAVS31Aand SiRAVS31Dwere analyzed.As shown in Fig.6-A, the mutation of SiRAV1 did not affect its subcellular localization.Due to the difference inSiCATexpression betweenSiRAVS31DandSiRAVS31Aoverexpressors, it was unclear whether SiRAV1’s phosphorylation at Ser31 affected its binding ability toSiCATpromoter.As expected, the enhancement of SiRAVS31Don the activity of LUC driven bySiCATpromoter in tobacco was much stronger (Fig.6-B).Next,we tested the binding ability of SiRAVS31Aand SiRAVS31Don theSiCATpromoter in yeast.The binding ability of SiRAVS31DtoSiCATpromoter was much higher than that of SiRAVS31AtoSiCATpromoter (Fig.6-C).All these results demonstrate that phosphorylation of SiRAV1 at the Ser31 site enhances its binding ability to theSiCATpromoter.

Fig.6 Phosphorylation of SiRAV1 at the Ser31 site enhances its binding ability to SiCAT promoter but does not affect its subcellular localization.A, the subcellular localization of SiRAV1S31A and SiRAV1S31D.The 35S:YFP-SiRAV1, 35S:YFP-SiRAV1S31A, and 35S:YFP-SiRAV1S31D were transiently expressed in Nicotiana benthamiana leaf cells.Scale bar=20 μm.B, phosphorylation of SiRAV1 at Ser31 site enhances its binding ability to SiCAT promoter in tobacco.The above is the schematic representation of the double-reporter and effector plasmids.LUC, firefly luciferase; REN, Renilla luciferase.The LUC/REN ratios represent the SiCAT promoter activities.C, phosphorylation of SiRAV1 at Ser31 site enhances its binding ability to SiCAT promoter in yeast.The experiments in A and C were performed at least three times with similar results.Values in B show mean±SD (n=3).Different letters in B indicate significant differences at P<0.05 according to Duncan’s multiple range test.D, a proposed model for SiRAV1 function in the salt response of foxtail millet.Salt-triggered phosphorylation of SiRAV1 at Ser31 enhances its binding ability to SiCAT promoter and the increased SiCAT expression, which causes an increase in catalase (CAT) enzyme activity to reduce reactive oxygen species accumulation.

4.Discussion

Protein phosphorylation/dephosphorylation is one of the important post-translational modifications and is critical in the rapid response of plants to salt stress (Panget al.2018; Piet al.2018; Andrásiet al.2019; Maet al.2019).Previous studies have successfully excavated a number of salt-responsive phosphoproteins and phosphorylated sites in many plant species (e.g., maize and soybean)using phosphoproteomic approaches (Huet al.2013;Piet al.2018).Surprisingly, the phosphoproteomic analysis of foxtail millet in response to salt stress is still lacking.Here, we used this approach to unravel the NaCl-responsive phosphoproteins in foxtail millet and detected 293 salt-regulated phosphorylated proteins(Fig.1; Appendix B).Notably, among the 293 genes encoding phosphorylated proteins mentioned above,only a few genes were also responsive to salt stress in previous transcriptome analysis and proteomic analysis(Veeranagamallaiahet al.2008; Puraniket al.2011),implying the specific roles of phosphorylation modification in the rapid response of foxtail millet to salt stress.

Many studies have shown that RAV proteins play pivotal roles in regulating plant growth and development process, as well as response to abiotic and biotic stresses(Amagaiet al.2018; Weiet al.2018; Gaoet al.2020; Liuet al.2021).However, there are few prior reports about the function of RAV in foxtail millet.Here, we excavated a RAV transcription factor in foxtail millet, SiRAV1, that could be phosphorylated in response to salt stress and further confirmed that the phosphorylated site Ser31 was crucial for salt tolerance (Fig.5; Appendix D).InArabidopsis,AtRAV1 is regulated by ABA at both transcriptional and posttranscriptional levels (Fenget al.2014).However,a previous study has shown thatSiRAV1could not be induced by salt stress (Lataet al.2014), and our study also shows that NaCl treatment did not up-regulate theSiRAV1expression (Appendix E), indicating that salt stress could largely activate SiRAV1 at the posttranscriptional level in foxtail millet.Until now, only Fenget al.(2014)reported that AtRAV1 could be phosphorylated in ABA signaling.However, this study did not further excavate the key phosphorylation site.Our study identified the key phosphorylation site in the RAV transcription factor (Fig.5-J;Appendix D).Phosphorylation modification might affect the localization, DNA binding ability, or transcriptional activation of transcription factors (Jianget al.2019).We found that phosphorylation of SiRAV1 at the Ser31 site enhanced its binding ability toSiCATpromoter but did not affect its subcellular localization (Fig.6).Thus, it is reasonable to speculate that the phosphorylation of SiRAV1 at Ser31 might determine its spatial structure.InArabidopsis,three SNF1-RELATED PROTEIN KINASE 2s (SnRK2s),SnRK2.2, SnRK2.3, SnRK2.6, could interact with AtRAV1 and directly phosphorylate it (Fenget al.2014).Except for SiRAV1, several kinases, including Seita.7G123400.1.p,Seita.9G554700.1.p, Seita.6G081700.1.p, and Seita.3G192600.1.p were also regulated by salt stress using the phosphoproteomic analysis (Appendix F).Proteinprotein interaction network analysis showed that the above kinases had a close relationship between them with SIRAV1(Appendix G).Future work needs to identify the upstream kinase of SiRAV1 and further unravel this pathway.

Previous studies have shown that the overexpressing ofZmRAV1orMtRAV3enhanced the tolerance of plants to salt stress (Minet al.2014; Wang Set al.2021).Consistent with these reports, our data illustrate that SiRAV1 is a positive regulator in modulating the salt response of foxtail millet based on the following reasons: first, overexpression ofSiRAV1in foxtail millet enhanced salt stress tolerance(Fig.3-A-C).Secondly, SiRAV1 alleviated oxidative damage in response to salt stress (Fig.3-D-G).RAV proteins can bind the specific motif (CAACA and CACCTG)to modulate their target genes’ expression (Fenget al.2014; Zhanget al.2020).For example, AtRAV1 binds to the CAACA motif in the promoter of the genesABI3,ABI4,andABI5to modulate their expressions, contributing to ABA response inArabidopsis(Fenget al.2014).In strawberry,FaRAV1 binds to the CAACA motif of theFaMYB10promoter to regulate anthocyanin accumulation (Zhanget al.2020).Our results showed that SiRAV1 positively regulatedSiCATexpression by directly binding to theSiCATpromoter (Fig.4).SiCATencodes the CAT enzyme, which functions in scavenging excess H2O2accumulation caused by salt stress (Zhouet al.2018).Our results found that overexpressing ofSiRAV1increased CAT enzyme activity to decrease the H2O2accumulation (Fig.3), which was consistent with the positive role of RAV in enhancing the cellular antioxidant defense system exposed to salt stress(Minet al.2014).It is worth noting that some other studies also indicate the negative role of RAV proteins in response to salt stress.For instance, knockout ofAtRAV1,AtRAV1L,orAtRAV2enhances the salt tolerance inArabidopsis(Fenget al.2014).Phylogenetic analysis showed that RAV genes derived from monocots (foxtail millet and maize) clustered separately from those of dicots (Arabidopsis).Notably, the phosphorylation site (Ser31) of SiRAV1 in salt response is only conserved orthologs in maize but does not exist inArabidopsis, implying that the regulatory mechanism of phosphorylation on SiRAV1 at Ser31 in salt response is highly specific.This might lead to the opposite role of RAV in various plant species in environmental stress response.More details structural and functional analyses of RAV protein in salt response in different species are needed in the future.

5.Conclusion

We present a model for SiRAV1 function in salt response where salt-triggered phosphorylation of SiRAV1 at Ser31 enhances its binding ability toSiCATpromoter, and the increasedSiCATexpression contributes to salt tolerance in foxtail millet.

Acknowledgements

This work was funded by the National Natural Science Foundation of China (31902062) and the South China Botanical Garden, Chinese Academy of Sciences (QNXM-02).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on https://doi.org/10.1016/j.jia.2023.04.034