Foxtail millet (S. italica; Jigu 11) seeds were germinated on moist filter paper for 24 h at 28°C and then planted in a soil mixture (nutrient soil: Vermiculite, 1:1, v/v) in a growth chamber (16-h light/8-h dark, 26–28°C, 60% relative humidity). After 2 weeks, the soil mixture on the seedling roots were washed away, and the seedlings were transferred to 1/4 Hoagland solution for 3 days. They were then subjected to treatment with 100 mM NaCl, 20% PEG6000 or 10 μM ABA and were harvested at the indicated times. Seedlings cultured in 1/4 Hoagland were used as controls. For tissue expression analyses, the leaves, stems and roots of 2-week-old seedlings and mature seeds were harvested. All samples were stored at −80°C after being frozen in liquid nitrogen.

Seeds of Arabidopsis thaliana (Col-0) were surface-sterilized and plated on MS medium containing 2% sucrose and 0.8% agar and incubated for 72 h at 4°C before being transferred to 22°C and a 16-h light/8-h dark photoperiod for germination. After 5 days, the seedlings were planted in a soil mixture (nutrient soil: Vermiculite, 1:1, v/v) and grown in the same conditions. Nicotiana benthamiana seeds were planted in a potting soil mixture (nutrient soil: Vermiculite, 1:1, v/v) and grown in a growth chamber under a 16-h light/8-h dark photoperiod for germination at 22–23°C.

Total RNA was extracted from foxtail millet and Arabidopsis using the TRIzol reagent (Invitrogen, USA). After digestion with DNaseI (Takara, Japan), 3–5 μg of total RNA was prepared, and cDNA was synthesized via reverse transcription using M-MLV Reverse Transcriptase (Promega, USA). Semi-quantitative RT-PCR was performed using 2 × Taq PCR StarMix with Loading Dye (GenStar, China). The PCR conditions were 95°C for 5 min, followed by 25 cycles of 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s, with a final step at 72°C for 10 min. Quantitative RT-PCR (qRT-PCR) was performed using 2 × Ultra SYBR Mixture (CWBIO, China) on a qTower 2.2 Real-Time PCR System (AnalytikJena, Germany). The PCR conditions were 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, and 60°C for 1 min. Relative gene expression levels were calculated using the 2^−ΔΔCT method (Livak and Schmittgen, 2001).

For GUS staining, fresh plant samples were immersed in GUS staining buffer containing a 1 mM 5-bromo-4-chloro-3-indolyl glucuronide (X-Gluc) solution in 100 mM sodium phosphate buffer (pH 7.0) with 0.5 M EDTA, 5 mM FeK₃(CN)₆, 5 mM FeK₄(CN)₆, and 0.1% Triton X-100. Following vacuum infiltration, the samples were stained at 37°C overnight, and the GUS staining solution was then replaced with 70% ethanol for decolorization. Finally, the samples were photographed under a dissecting microscope (Olympus SEX16, Japan).

To generate a GFP-fusion protein, the coding sequence of SiASR4 without the stop codon was amplified and inserted into the BamHI and SmaI sites of the pROK219-GFP plasmid and the XbaI/KpnI sites of the pSuper1300-GFP plasmid. The pROK219-GFP and pROK219-SiASR4-GFP plasmids were used for transformation. Foxtail millet seeds were germinated on moist filter paper for 24 h at 28°C and then grown in pots filled with nutrient soil and vermiculite mixed 1:1 (v/v) under 12 h-light/12 h-dark conditions at 28°C for 3 days, after which they were moved to dark conditions for 4–6 days. Stem and leaf tissues were cut into 0.5–1-mm-wide strips. Protoplasts were prepared and transformed according to the procedure described by Zhang et al. (2011). To prepare protoplasts, the tissues were transferred into 0.6 M mannitol and incubated for 10 min to plasmolyse, and then enzyme solution was used to digest tissues in the dark for 4–5 h. The protoplasts were collected through 74 μm nylon meshes. For transformation, 200 μL protoplasts (1 × 10⁶ cells mL⁻¹) were mixed with 20 μg plasmid DNA and 220 μL 40% PEG, and the samples were cultured for 16–20 h. Fluorescence was observed using a confocal laser microscope (Leica HQ).

The pSuper1300-GFP and pSuper1300-SiASR4-GFP plasmids were transformed into Agrobacterium strain EHA105. The Agrobacterium lines were then resuspended in transformation buffer and infiltrated into N. benthamiana leaves. These plants were grown at 22–23°C for 3 days, and fluorescence in the leaves was imaged using a confocal laser microscope (Leica HQ).

Transcriptional activation of SiASR4 was performed in the yeast strain YRG-2. The full-length SiASR4 sequence was inserted into the EcoRI and SmaI sites of the pBD-GAL4 plasmid (Stratagene, USA), driven by the yeast alcohol dehydrogenase1 (ADH1) promoter. The pBD-GAL4 and pGAL4 plasmid were used as negative and positive controls, respectively. The plasmids were transformed into yeast cells independently. These transformed yeast cells were grown on SD/Trp- or SD/Trp-/His- medium for 3 days at 30°C.

For SiASR4 overexpression in Arabidopsis, the full-length SiASR4 sequence without the stop codon fused to a flag tag was amplified via RT-PCR and cloned into the modified binary vector pCAMBIA2300 at the KpnI and SacI sites, controlled by the CaMV 35S promoter. For the SiASR4 promoter assay, the putative SiASR4 promoter was amplified and inserted into the pCAMBIA1391 vector at the HindIII and EcoRI sites to drive the GUS gene. The plasmids were transformed into Agrobacterium strain GV3101. Infection of Arabidopsis Col-0 was performed using the floral dip method (Zhang et al., 2006). Developing floral tissues were dipped into Agrobacterium solution containing 5% sucrose and Silwet-77 (500 μL L⁻¹). After inoculation, the plants were placed under a plastic dome for 12–24 h and then grown under a 16-h light/8-h dark photoperiod at 22°C. Seeds were obtained following self-pollination. The SiASR4 and proSiASR4 transgenic Arabidopsis seedlings were selected on MS medium containing kanamycin (50 μg/mL) or hygromycin (15 μg/mL) respectively and then confirmed by PCR. The expression of SiASR4 in transgenic Arabidopsis was determined by semi-qRT-PCR using SiASR4 (qRT)-F/R primer pairs (Table S1). Three independent homozygous T3 lines with kanamycin resistance and high expression levels of SiASR4 were chosen for further studies.

To construct the SiASR4-overexpressing vector, the coding region fused to a flag tag at the C-terminus was amplified and inserted into the BamHI and SacI sites of the modified binary vector pCOU under the control of the ubiquitin promoter. To generate SiASR4-RNAi constructs, the SiASR4 sequence with KpnI and SacI sites was amplified and inserted between the GUS gene and the NOS terminator, and the SiASR4 sequence with SpeI and BamHI sites was amplified and inserted in reverse orientation between the ubiquitin promoter and GUS gene of the modified binary vector pCOU. The SiASR4-overexpressing and SiASR4-RNAi constructs were first introduced into Agrobacterium tumefaciens strain LBA4404. The transformation of foxtail millet was performed according to a previously described method (Wang et al., 2011). 0.5–1-cm-long immature inflorescences of the greenhouse-grown Jigu 11 foxtail millet were used as explants. They were cultured on callus induction medium for 30 days to induce embryogenic calli. For transformation, the embryogenic calli were soaked in Agrobacterium infection medium (OD₆₀₀ ≈ 0.4–0.6) and inoculated for 20 min. Co-cultivation was performed in the dark at 22°C for 3 days on co-cultivation medium. After co-cultivation, the explants were incubated in the dark at 26°C on resting medium (150 mg/L timentin) for recovery. After 7 days, calli were cultured on selection medium (5 mg/L hygromycin and 150 mg/L timentin) for 2 weeks and then subcultured on selection medium (10 mg/L hygromycin and 150 mg/L timentin) for further 2 weeks, after which they were transferred to regeneration medium (150 mg/L timentin) under a 16-h light/8-h dark condition at 26°C and subcultured every 2 weeks. The regenerated plants were then rooted on rooting medium and further grown in the greenhouse. The SiASR4-overexpressing and SiASR4-RNAi foxtail millet were determined by PCR using a primer pairs of SiASR4-OE-F/R or SiASR4-RNAi-F/R, respectively (Table S1). The expression of SiASR4 in transgenic foxtail millet was examined by qRT-PCR using SiASR4 (qRT)-F/R primer pairs (Table S1). Two independent lines of T2 generation were chosen for further studies.

For the germination-stage salt stress analysis, Col-0 and SiASR4 transgenic Arabidopsis seeds were sown on MS medium containing 0, 75, or 100 mM NaCl for 10 days in a growth chamber under 16 h-light/8 h-dark conditions at 22°C. Subsequently, they were photographed, and the fresh weights of Col-0 and transgenic Arabidopsis were measured. For the early growth assay under salt stress, Col-0 and transgenic Arabidopsis were grown vertically on MS medium for 5 days in a growth chamber under 16 h-light/8 h-dark conditions at 22°C and then transferred to MS medium containing 0, 150, or 200 mM NaCl for 7 days. The seedlings were then subjected to measurement of chlorophyll contents, and the number of lateral roots was counted, after which the plants were photographed. For salt stress in soil, 5-day-old Col-0 and transgenic Arabidopsis seedlings grown on MS medium were transferred to soil for an additional 2 weeks. Subsequently, the soil was irrigated with water supplemented with 400 mM NaCl every 7 days for 14 days, and survival rates and chlorophyll contents were analyzed.

For ABA treatment during the germination stage, Col-0 and transgenic Arabidopsis seeds were sown on MS medium containing 0, 0.5, or 1 μM ABA for 10 days in a growth chamber under 16 h-light/8 h-dark conditions at 22°C. For the early growth assay, Col-0 and transgenic Arabidopsis were grown vertically on MS medium for 5 days at 22°C and then transferred to MS medium containing 0 and 30 μM ABA for 7 days. The resultant phenotypes were then observed.

For dehydration treatment, Col-0 and transgenic Arabidopsis seeds were sown on MS medium containing 0, 100, or 200 mM mannitol for 10 days in a growth chamber under 16 h-light/8 h-dark conditions at 22°C. For the early growth assay, Col-0 and transgenic Arabidopsis were grown vertically on MS medium for 5 days at 22°C and then transferred to MS medium containing 0 and 300 mM mannitol for 7 days, after which their phenotypes were observed. For the growth assay in soil, 5-day-old Col-0 and transgenic Arabidopsis seedlings were grown on MS medium and then transferred to soil for an additional 2 weeks. Irrigation was stopped for 18 days and then resumed for 5 days. Survival rates were calculated, and the chlorophyll content was measured.

For the germination-stage analysis, seeds of WT, SiASR4-overexpressing and SiASR4-RNAi transgenic foxtail millet were surface-sterilized and germinated on filter paper containing H₂O, 100 mM NaCl, 5 μM ABA, and 15% PEG in a growth chamber under 16 h-light/8 h-dark conditions at 28°C. The number of germinated seeds was subsequently counted at 2, 4, 6, and 8 days, and shoot and root lengths were measured on the 6th day. The germination stress index was determined using the following formula: Germination stress index = (germination index under stress treatment/germination index under control treatment) × 100; germination index = [(d2+0.75d4+0.5d6+0.25d8)/total seed number] × 100 (where d2, d4, d6, and d8 represent the number of germinated seeds at 2, 4, 6, and 8 days, respectively).

For the drought stress treatment in soil, 3-week-old WT, SiASR4-overexpressing and SiASR4-RNAi transgenic foxtail millet seedlings were deprived of water for 13 days, and then watering was resumed for 5 days. For the salt stress treatment in soil, 3-week-old WT, SiASR4-overexpressing and SiASR4-RNAi foxtail millet seedlings were irrigated with water supplemented with 400 mM NaCl every 7 days for 20 days. Survival rates were calculated, and the chlorophyll content was measured.

The fresh weights of WT and transgenic plant seedlings were determined, and the seedlings were placed in 1.5-mL Eppendorf tubes. Total chlorophyll was extracted with 80% acetone through incubation in the dark for 12 h at room temperature. The samples were then centrifuged at 12,000 rpm for 10 min, and the absorbance was measured at 663 and 645 nm using a microplate reader (BioTek, USA). The chlorophyll content was calculated according to Li et al. (2015): Chlorophyll a = 12.7A663−2.69A645; chlorophyll b = 22.9A645−4.68A663; total chlorophyll = chlorophyll a + chlorophyll b.

Total RNA from Col-0 Arabidopsis and SiASR4 transgenic Arabidopsis was used to analyze the expression levels of AtSOD1 (AT1G08830), AtCAT3 (AT1G20620), AtSOS1 (AT2G01980), AtLTP3 (AT5G59320), and Rd29B (AT5G52300) via qRT-PCR. Total RNA from Jigu 11 foxtail millet and SiASR4-overexpressing foxtail millet was used to analyze the expression levels of SiSOD (Seita.9G403600), SiCAT (Seita.4G286700), SiSOS1 (Seita.3G409000), and SiLTP (Seita.8G013500) through qRT-PCR. The primers employed for these assays are shown in Table S1.

Three-week-old Col-0 and transgenic Arabidopsis grown in pots were not watered and were subjected to salt treatment with 400 mM NaCl for 6 days, respectively, after which the fourth and fifth rosette leaves were harvested. H₂O₂ accumulation was detected with diaminobenzidine tetrahydrochloride (DAB). The leaf samples were immersed in DAB solution (1 mg ml⁻¹ in water, pH 3.8) and incubated for 4–5 h in the dark with gentle shaking. Following staining, the DAB solution was replaced with bleaching solution (ethanol: Acetic acid: Glycerol = 3:1:1), and the samples were boiled in water at 95°C for 15 min.

The bait sequence and mutant bait sequence were synthesized and cloned into the yeast vector pAbAi at the HindIII and XhoI sites (Table S1). According to the protocol of the Yeastmaker™ Yeast Transformation SystemII (Clontech, USA), the bait vectors were linearized through BstBI digestion and transformed into the yeast strain Y1HGold. Positive clones were screened on SD/Ura- medium containing different concentrations of Aureobasidin A (AbA). The SiARDP gene was inserted into the NdeI and XhoI sites of the pGADT7-AD vector according to Li et al. (2014). The resulting plasmid was then transformed into the bait and mutant bait strains and grown on SD/Leu- medium containing 800 ng/ml AbA at 30°C for 3 days.

SiARDP was cloned into the pGEX-TEV vector which contained a GST tag, by Li et al. (2014). This fusion protein was expressed in E. coli BL21 cells and purified using Glutathione Sepharose 4B (GE, USA). Oligonucleotides and their reverse complementary oligonucleotides were synthesized. These sequences are shown in Figure 10. The probe sequence was labeled with 5′ biotin. Double-stranded probe, competitor probes and mutant probes were obtained through annealing. The reaction mixtures were incubated at room temperature for 30 min and then analyzed via polyacrylamide gel electrophoresis. The electrophoretic mobility shift assay (EMSA) was performed using the LightShift® Chemiluminescent EMSA kit (Thermo, USA).

A full-length ASR gene (Seita.5G463800) was isolated from foxtail millet and designated SiASR4. Sequence analysis revealed that the full-length SiASR4 cDNA has a 309-bp opening reading frame (ORF), encoding 102 amino acids, with a predicted molecular mass of 11.5 kDa and an isoelectric point of 10.57. BLASTX analysis showed that SiASR4 has two highly conserved regions: A small N-terminal consensus of 12 amino acids containing a stretch of four His residues, which is typical of zinc-binding regions, and a longer C-terminal region of 80 amino acids with an ABA/WDS domain and a putative signal for nuclear targeting (Figure 1A). Phylogenetic analysis indicated that the closest homolog to SiASR4 in grass species is the green grass ASR protein, followed by rice and switch grass ASR clones. Here, we chose one of the closest homologs to SiASR4 from each grass species as a representative. As shown in Figure 1B, the results suggested that the SiASR4 sequence obtained in this study is a member of the ASR family of foxtail millet.

We examined the expression levels of SiASR4 in different foxtail millet tissues via qRT-PCR. SiASR4 was expressed in the leaves, stems, roots, and mature seeds, and the transcript levels were highest in mature seeds, followed by the roots, while the lowest level was detected in the leaves and stems (Figure 2A). We also isolated the putative SiASR4 promoter region and confirmed it by sequencing. To characterize the temporal and spatial expression patterns of SiASR4, the GUS gene was driven by the SiASR4 promoter and transformed into Arabidopsis (Figure 2B). GUS staining of transgenic Arabidopsis revealed GUS activity in the cotyledons, hypocotyls, stems, leaves, roots, flowers, and seeds, similar to foxtail millet, which indicated that SiASR4 is a constitutively expressed gene (Figure 2C).

To determine the response of SiASR4 to abiotic stress, PEG, NaCl, and ABA treatments were applied. Then, qRT-PCR was performed, and the results showed that the expression levels of SiASR4 increased in response to 20% PEG6000, 100 mM NaCl and 10 μM ABA treatment (Figure 3). In response to NaCl and ABA treatment, SiASR4 transcription increased after 1 h and peaked at 9 h (Figures 3A,C). In response to PEG treatment, SiASR4 expression was induced to peak levels at 12 h and was maintained at similarly high levels at 9 h (Figure 3B). The level of SiASR4 transcription decreased dramatically after 24 h of treatment. These results indicated that SiASR4 may be involved in salt and drought stress responses via an ABA-dependent signaling pathway.

To investigate the subcellular localization of SiASR4, a transient expression assay was performed with the SiASR4-GFP fusion protein and GFP alone (as a control) in N. benthamiana leaves. The fluorescence of SiASR4-GFP was distributed throughout whole cells in N. benthamiana leaves (Figures 4A,B), as was that of GFP. To confirm this result, the GFP and SiASR4-GFP fusion genes driven by the CaMV 35S promoter were transiently expressed in foxtail millet protoplasts. The results obtained in protoplasts showed that SiASR4 localized to the cell nucleus, cytoplasm and cytomembrane (Figures 4C,D).

To investigate the transcriptional activity of SiASR4, the yeast transcriptional activation system was used. Yeast cells were transformed with the pBD-GAL4-SiASR4 plasmid. All transformants grew well on SD/Trp-medium, but none of them grew on SD/Trp-/His-medium (Figure S1). These results indicated that SiASR4 lacked transcriptional activity in yeast cells.

To assess the function of SiASR4 in stress tolerance, transgenic Arabidopsis plants overexpressing SiASR4 under the control of the CaMV 35S promoter were generated. SiASR4 transgenic lines were obtained using a floral dip method. Transgenic T1 seedlings were selected on MS medium supplemented with 50 μg/ml kanamycin and then confirmed by PCR. Finally, 18 SiASR4-overexpressing transgenic Arabidopsis plants were obtained and three independent homozygous SiASR4 T3 lines (L11, L12, and L13) with high expression levels were chosen for further functional analyses (Figure 5A).

Under normal conditions, there were no obvious differences in fresh weight between transgenic and wild-type (WT) Arabidopsis in either the seed germination or growth stage. However, the transgenic plants grew better than wild-type plants on MS medium supplemented with 75 or 100 mM NaCl (Figure 5B). They displayed larger cotyledons than WT plants, and the fresh weights of the transgenic lines were significantly greater than those of WT plants (Figures 5B,D). However, when grown on MS medium containing different concentrations of mannitol or ABA, the WT and transgenic lines did not exhibit an obvious difference (Figures S2A,B). These results indicated that SiASR4 can enhance salt stress tolerance in transgenic Arabidopsis in both the seed germination and growth stages.

To evaluate the effect of abiotic stress during the seedling stage, 5-day-old Arabidopsis grown in normal medium were transferred to MS medium containing different concentrations of NaCl, mannitol and ABA for 7 days. There were no obvious differences between the transgenic and WT plants under normal conditions or in response to mannitol and ABA (Figures S3A,B). When grown on the MS medium containing 150 or 200 mM NaCl, the transgenic plants showed more robust growth than WT plants (Figure 5C), which was characterized by a significant increase in the number of lateral roots and chlorophyll content (Figures 5E,F). These results showed that WT plants were more sensitive than transgenic plants to salt stress during the seedling stage.

To further assess the function of SiASR4 during salt and drought tolerance, the corresponding treatments were applied to 3-week-old transgenic and WT plants grown under normal conditions. After irrigation with water supplemented with 400 mM NaCl for 14 days, more WT than transgenic plants were bleached (Figure 6A). Approximately 38% of the WT plants survived, while the survival rates of the transgenic lines were greater than 80% (Figure 6B). The chlorophyll content was increased in the transgenic lines compared with WT plants during salt stress treatment (Figure 6C). For the drought stress treatment, the plants were not irrigated for 18 days. When the plants were rewatered and grown for 5 days, almost all of the WT seedlings died (Figure 6D), while approximately 90% of transgenic plants survived (Figure 6E). The chlorophyll content was decreased by at least 50% in WT plants compared with the transgenic lines (Figure 6F). These results indicated that SiASR4 could enhance drought and salt tolerance in soil conditions. Considering the obtained results together, we concluded that SiASR4 functions in the stress response of Arabidopsis.

To further investigate the role of SiASR4, we obtained SiASR4-overexpressing and RNAi transgenic foxtail millet lines confirmed by PCR. Based on the qRT-PCR results, two overexpressing lines (OE13 and OE24) and two RNAi transgenic lines (RNAi53 and RNAi62) were chosen for further analysis (Figures 7A,B).

First, we examined abiotic stress tolerance in the transgenic lines during the germination stage. Transgenic and WT foxtail millet seeds were germinated under different treatments. There were no apparent differences in shoot or root length between the transgenic and WT lines exposed to water for 6 days. When germinated in 5 μM ABA, 100 mM NaCl, or 15% PEG for 6 days, the growth of the SiASR4-overexpressing lines was significantly better than that of WT plants, and the shoot and root lengths of SiASR4-overexpressing lines were dramatically longer compared with WT. However, there was no obvious suppression of the SiASR4-RNAi lines compared with WT, except with respect to shoot length after treatment with 100 mM NaCl (Figures 7C–E). We further analyzed the germination stress index after 8 days. The germination stress index was higher in the SiASR4-overexpressing lines compared with WT, but no differences were observed between SiASR4-RNAi and WT (Figure 7F). These results implied that overexpression of SiASR4 in foxtail millet enhanced resistance to drought and salt, and these plants were insensitive to ABA during the germination stage.

To analyze the performance of transgenic lines in response to drought and salt treatment in soil, 3-week-old SiASR4-overexpressing, SiASR4-RNAi and WT seedlings were subjected to the corresponding treatments. For drought stress treatment, seedlings were grown under water stress conditions for 13 days and then rewatered and grown under normal conditions for 5 days (Figure 8A). More than 80% of the overexpressing lines survived, while the survival rate of the WT lines was ~10% (Figure 8C). For salt stress treatment, seedlings were irrigated with water supplemented with 400 mM NaCl for 20 days (Figure 8B). Approximately 32% of the WT plants survived, while the survival rates of the overexpressing lines were greater than 68% (Figure 8E). In addition, the overexpressing lines displayed higher chlorophyll content than WT plants. However, there were no obvious differences between the RNAi and WT lines (Figures 8D,F).

Abiotic stresses, such as drought and high-salinity, can induce the production of ROS (Foreman et al., 2003). It has been reported that TaASR1 and SbASR1 induce the transcription of ROS scavenger-associated genes and reduce the accumulation of ROS (Hu et al., 2013; Tiwari et al., 2015). To analyze the role of SiASR4 in ROS scavenging in transgenic Arabidopsis, 3-week-old Col-0 and transgenic Arabidopsis grown in pots were not watered and were treated with salt for 6 days, respectively. The fourth and fifth rosette leaves were then harvested and stained with DAB to detect H₂O₂ production. The results showed that the levels of DAB staining in H₂O₂ in rosette leaves were similar among the Col-0 and transgenic lines under normal conditions. After drought and salt treatment, SiASR4 transgenic Arabidopsis accumulated less H₂O₂ than Col-0 Arabidopsis (Figure 9A), indicating that overexpression of SiASR4 reduces ROS accumulation in transgenic Arabidopsis.

To analyze the expression of ROS scavenger-associated genes in transgenic Arabidopsis, the AtCAT3 and AtSOD1 genes, which are involved in ROS detoxification, were selected. Without stress treatment, the expression of AtCAT3 and AtSOD1 was increased significantly in transgenic compared with WT Arabidopsis (Figure 9B). We further analyzed the expression levels of SiCAT and SiSOD in SiASR4-overexpressing foxtail millet. The results showed that the ROS scavenger-associated genes were upregulated in SiASR4-overexpressing foxtail millet (Figure 9C). Furthermore, we analyzed expression of stress-responsive genes, the SiSOS1 and SiLTP genes of foxtail millet, and the AtSOS1, AtLTP3, and Rd29B genes of Arabidopsis by qRT-PCR. The results showed that the expression of these genes was increased markedly in SiASR4-overexpressing foxtail millet and Arabidopsis under normal conditions (Figures 9D,E). These results indicated that overexpression of SiASR4 in foxtail millet and Arabidopsis enhances the expression of ROS scavenger-associated and stress-responsive genes.

Because SiASR4 responded to drought and salt stress and ABA treatment, we analyzed the SiASR4 promoter and identified five DRE motifs. SiARDP, which we had previously cloned from foxtail millet, is a transcription factor that can bind to DRE elements (Li et al., 2014). To assess whether SiARDP is involved in the regulation of SiASR4 in vivo and in vitro, yeast one-hybrid assays and EMSAs were performed. One bait sequence in the SiASR4 promoter containing the DRE element (GCCGAC) was inserted into a bait plasmid. Simultaneously, a mutated DRE element (AAAAAA) sequence was also prepared (Table S1) and transferred plasmids to yeast cells; 800 ng/mL AbA suppressed the basal expression. Additionally, SiARDP was inserted into the pGADT7-AD plasmid, and the resulting plasmid was then transformed into the bait strain or mutated strain. The bait strain yeast cells transformed with pGADT7-SiARDP could grow on medium lacking Leu and containing 800 ng/mL AbA, but the mutant strain yeast cells with pGADT7-SiARDP did not grow (Figure 10A).

We also conducted an EMSA experiment. The sequence containing the DRE element was used as a probe (P). Full-length SiARDP was expressed as a glutathione S-transferase (GST) fusion protein in E. coli. The results showed that SiARDP could bind to the probe. The addition of different concentrations of unlabeled probes reduced the binding signal to various degrees, but the addition of mutant probes did not reduce the binding signal (Figures 10B,D). These results suggested that SiARDP specifically bound to the DRE core element in the SiASR4 promoter.

To further analyze whether SiASR4 is regulated by SiARDP in vivo, we examined the expression level of SiASR4 in SiARDP-overexpressing foxtail millet. Two independent homozygous SiARDP-overexpressing T3 lines were chosen for the analysis. qRT-PCR showed that the expression of SiASR4 was upregulated at least 6-fold in SiARDP transgenic foxtail millet (Figure 10C). Based on these results, SiASR4 was likely regulated by SiARDP in foxtail millet.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer PS and handling Editor declared their shared affiliation, and the handling Editor states that the process nevertheless met the standards of a fair and objective review.