Pre-germination treatments of maize seeds were performed and growth conditions applied as described in our previous study (Sun et al., 2018). Germinated seeds were sown in 20 × 22 cm plastic pots containing 2.5 kg vermiculite. All pots were watered with a half-strength Hoagland nutrient solution once daily. At the 3-leaf seedling stage of maize, 24 pots were selected and randomly divided into two equal sets. One set was used as a control by watering with a half-strength Hoagland nutrient solution, and the other set was treated with 200 mmol/L NaCl or 10% polyethylene glycol (PEG6000) in half-strength Hoagland nutrient solution. Leaves were harvested at 0, 2, 6, and 12 h post-stress treatment for RNA extraction to detect the expression pattern of ZmSBP17 under abiotic stress. Three biological samples (three pots for one time in one set) were collected from each tissue.

A. thaliana Columbia ecotype was used in this study. Growth conditions in soil and stress treatment methods were applied as described by He et al., 2019, 2024. Four-week-old gene-overexpressing Arabidopsis lines (OE) and wild-type (WT) plants were treated with water (control) or 200 mmol/L NaCl and 10% PEG6000 for 36 h. Then, leaves were collected to analyze the expression levels of abiotic stress-related genes and determine the activities of key reactive oxygen species (ROS)-scavenging enzymes.

Total RNA was extracted from the samples using the RNAiso Reagent Kit (TaKaRa, Dalian, China). The FastQuant RT Kit (TIANGEN, China) was used to synthesize the first-strand cDNA, following the procedure outlined in our previous studies (Sun et al., 2012; 2018). ZmActin1 (NM_001155179) was used as the endogenous control to compare the expression level of ZmSBP17. ACT8 (AT1G49240) was utilized as the internal control to analyze the expression levels of abiotic stress-related genes in Arabidopsis. All primers used for qRT-PCR assay are listed in Supplementary File 1 . A standard SYBR^® Premix Ex Taq™ kit (TaKaRa, Dalian, China), Bio-Rad CFX96 Real-Time System (Bio-Rad, USA, Hercules, CA, USA), and CFX Manager System software version 2.0 were utilized per the manufacturer’s instructions. At the end of PCR, Ct values for each sample were obtained to analyze the transcript levels of each gene using the 2^−ΔΔCt method (Livak and Schmittgen, 2001).

Based on the sequence of ZmSBP17 available in the NCBI (http://www.ncbi.nlm.nih.gov/Blast.cgi) and maize genome (www.plantgdb.org/ZmGDB/cgi-bin/blastGDB.pl) databases, gene-specific primers for RT-PCR were designed using the NCBI primer designer. The primer sequences are listed in Supplementary File 1 . Following the procedure outlined in our previous study (Sun et al., 2012), the full-length coding sequence (CDS) of Hei ZmSBP17 was isolated from the first-strand cDNA, cloned into the pMD18-T cloning vector (TaKaRa, Dalian, China), and sequenced (BGI, China). NCBI’s Conserved Domain database (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) was used to analyze the sequences and identify the conserved regions. Phylogenetic trees of ZmSBP17 and sequences from homologous plants were then constructed using DNAMAN software.

The full-length CDS of ZmSBP17, lacking its termination codon, was inserted into the pCAMBIA2300-GFP vector using a combination of homologous recombination and restriction digestion (SacI and XbaI). This construct was then used to transform the Escherichia coli strain DH5α. Expression of ZmSBP17 was driven by a CaMV 35S promoter. The primer sequences used for amplification are listed in Supplementary File 1 . The validated recombinant pCAMBIA2300-ZmSBP17-GFP and pCAMBIA2300-GFP (control) vector were transformed into EHA105 Agrobacterium competent cells for injecting into 5-6 leaf old N. benthamiana leaves epidermal cells, respectively, as previously reported (Chao et al., 2023). The transformed tobacco plants were incubated at room temperature in the dark for 12-16 h, and then imaged by laser scanning confocal microscopy (Leica SP8) after cultured for 2-3 days at normal condition.

To detect the transcriptional activation of ZmSBP17, the full-length and different truncated versions, which were amplified (primers listed in Supplementary File 1 ) from the pMD18-T-ZmSBP17 plasmid into the pGBKT7 vector. These constructs, along with the pGBKT7 control vector, were used to transform the yeast strain Y2HGold according to the manufacturer’s protocol (Clontech, USA). The transformants were then cultured on SD/-Trp and SD/-Trp/-His media for 3–5 days at 30°C.

For DNA-binding assays, the full-length ZmSBP17 CDS was amplified and cloned into the pGAD7 vector. The sequences of the SBP17-F4 and R4 primers, which contain NdeI and EcoRI sites, are listed in Supplementary File 1 . The 5X GTAC cis-elements and the mutated GTACm version were inserted into the pHIS2 vector (Yu et al., 2017) and used to co-transform the Y2HGold cells. The transformants were streaked onto SD/-Trp and SD/-Trp/-His/-Ade media and incubated for 3–5 days at 30°C.

The pCAMBIA2300-ZmSBP17-GFP vector was introduced into the Agrobacterium tumefaciens strain EHA105, which was then used to transform A. thaliana via the floral dip method (Clough and Bent, 1998). Transgenic seeds (T₀) were selected on MS inorganic salts (Murashige and Skoog, 1962) medium supplemented with 50 mg·L⁻¹ kanamycin. At the 7–8 leaf seedling stage, the lowest leaves were used to identify PCR-positive plants using detection primers ( Supplementary File 1 ). Homozygous lines were selected on MS medium containing 50 mg·L⁻¹ kanamycin.

For osmotic stress seed germination assays, WT and T₃ homozygous OE seeds were sterilized and vernalized at 4°C for 3 days. They were then sown on half-strength MS medium supplemented with either 100 mM NaCl or 8% PEG. Germination rates were assessed after 7 days. All germination assays were performed in triplicate. To evaluate the sensitivity of adult plants to osmotic stress, 4-week-old WT and OE plants were irrigated with either 200 mM NaCl or 10% PEG at 3-day intervals for 7 days and subsequently photographed.

For histochemical analysis, 4-week-old WT and OE plants were treated with 200 mM NaCl or 10% PEG for 36 h. The entire plants were then used for staining. DAB and NBT staining were performed as described by Zhang et al. (2011) with slight modifications.

For detecting the activity of ROS-scavenging enzymes, 4-week-old WT and OE plants were irrigated with 200 mM NaCl or 10% PEG for 36 h, and then rosette leaves were collected. The activities of peroxidase (POD), superoxide dismutase (SOD), and catalase (CAT) were measured as described by Wang et al. (2019). Ascorbate peroxidase (APX) was extracted as described by Erinle et al. (2018) with slight modifications.

Following the expression analysis of SBPs under salt stress by qRT-PCR in our previous study, a putative salt stress-related SBP from the Hei maize inbred line was isolated and named ZmSBP17 (GenBank accession no. NM_001308478) according to blastp results ( Supplementary File 2 ). The open reading frame of ZmSBP17 is 1137 bp long, encoding a 379 amino acid polypeptide. Amino acid sequence alignment and phylogenetic analysis of ZmSBP17 homologs are shown in Figure 1 . The average identity homology of ZmSBP17 with those in rice and Arabidopsis was 12.9%. Multiple sequence alignment by DNAMAN revealed it to have C2HC and C3H domains at the N terminus ( Supplementary File 3 ). Phylogenetic analysis showed that ZmSBP17 was relatively close to OsSBP17, but ZmSBP17 was less identical to homologs other than OsSBP17, which suggested that ZmSBP17 markedly differs from other members. Analysis of the upstream 2000 bp promoter region of ZmSBP17 revealed the presence of multiple elements related to abiotic stress, including abscisic acid response element (ABRE), antioxidant response element (ARE), long terminal repeat (LTR), and MYB-binding sites (MBS) ( Supplementary File 4 ).

The expression pattern of ZmSBP17 under abiotic stress was investigated by subjecting the three-leaf stage maize seedlings to the indicated treatment. ZmSBP17 expression increased significantly and reached a peak at 12 and 2 h under salt and PEG stresses, respectively, indicating that it may participate in the osmotic stress responses in maize ( Figure 2A ; Supplementary File 5 ).

The subcellular localization of ZmSBP17 constructed transient expression vector pCAMBIA2300-ZmSBP17-GFP ( Supplementary File 2 ) and pCAMBIA2300- GFP vectors were transformed into the tobacco epidermal cells. The green fluorescence signals of expressing GFP gene was monitored by a confocal laser scanning microscope. As shown in Figure 2B , the pCAMBIA2300-ZmSBP17-GFP fusion protein was specifically detected in the nucleus compare to the control distributed throughout the cell.

The full-length ZmSBP17 was fused to the GAL4 (GAL4-BD) binding domain of pGBKT7 to confirm if ZmSBP17 has transcription activation ability. The constructed pGBKT7-ZmSBP17 vector is shown in Supplementary File 3 . A comparison of the growth phenotype of the pGBKT7-ZmSBP17 and pGBKT7 recombinant yeast revealed that they can survive on SD/-Trp selective medium with the same colony size, indicating the successful transfer of each vector into the Y2H cells. However, only those containing the pGBKT7-ZmSBP17 vector grew normally on SD/-Trp-His selective medium, suggesting that ZmSBP17 was able to activate the HIS reporter gene for survival. Thus, ZmSBP17 acts as a transcriptional activator in yeast ( Figure 3 ).

Protein sequence analysis revealed that the SBP superfamily domains are present in the ZmSBP17 protein (from 64 to 138 amino acid residues). To determine which domain in ZmSBP17 was the transcriptional activator, four deletion fragments were inserted into pGBKT7 and used to transform Y2H cells. All transformants grew well on the SD/-Trp medium. Contrary to those carrying the N-64 and N-138 fragments, the yeast cells carrying the C-64 fragment of ZmSBP17 could not grow on the selection medium. Meanwhile, yeast cells carrying the N-138 SBP domain fragment (amino acids 64–138) grew better than those carrying the N-64 fragment on the selection medium. These results indicate that the SBP domain enhanced the transcription activation ability of ZmSBP17, and the 1–64 amino acid fragment was crucial for this ability.

pCAMBIA2300-ZmSBP17 was used to transform the WT A. thaliana Columbia-0 plants to characterize whether ZmSBP17 is associated with the osmotic stress response. The transgenic plants were screened on kanamycin and identified by PCR ( Supplementary Figure 4 ). The T₃ plants were used for further analysis. No significant differences in seed germination rate were observed between transgenic and WT plants on MS medium. However, in the presence of 100 mM NaCl and 8% PEG, an obvious reduction in germination rate was detected in both transgenic and WT seeds but was significantly higher in the transgenic than WT seeds ( Figures 4A, B ; Supplementary File 6 ).

For further functional analysis of ZmSBP17, the adult stage ZmSBP17-OE lines were examined in soil. Under normal conditions, no significant phenotypic variations between the OE lines and WT were observed. However, post-treatment with 200 mM NaCl and 10% PEG for 36 h, the stems and leaves of the WT showed obvious wilting, with some leaves turning yellow and drying. However, the wilting of transgenic A. thaliana was significantly less severe than that of the WT under PEG stress. The stems and leaves of the transgenic plants remained relatively upright, and leaf yellowing was also less pronounced compared with the WT. However, under salt stress, there was no apparent difference between the WT and transgenic plants ( Figure 4C ). These results indicate that ZmSBP17 may improve the tolerance of Arabidopsis seeds at various stages to PEG.

The levels of H₂O₂ and O₂ ^¯ that are key ROS species were detected in ZmSBP17 OE line and WT plants using DAB and NBT staining, respectively to investigate whether ZmSBP17 is associated with the regulation of ROS scavenging under osmotic stress. After treatment with 200 mM NaCl and 10% PEG, the staining in the leaf tissues of the WT and OE lines increased significantly. However, the DAB staining was deeper in the WT than in the OE lines, and there was no significant difference in NBT intensity between the WT and OE lines under PEG stress. Under salt stress, the DAB and NBT staining was deeper in the WT than in the OE line ( Figures 5A, B ). Moreover, the activities of key ROS-scavenging enzymes POD, SOD, CAT, and APX were assessed in the OE and WT plants at 0 and 36 h after osmotic treatment. As shown in Figure 5C , the activities of all four increased significantly in the OE line compared to WT plants after stress treatment ( Supplementary File 6 ). These results indicate that ZmSBP17 might play a role in modulating the ROS-scavenging system by increasing the activities of protective enzymes under osmotic stress.

The expression levels of the abiotic stress-responsive genes—CSD1, AT1G08830.1; CSD2, AT2G28190.1; MSD1, AT3G10920.1; CAT1, AT1G20630.1; CAT2, AT4G35090.1; APX1, AT1G07890.3; DREB2A, AT5G05410; NHX2, AAT3G05030; MYB65, AT3G1140; and MYB53, AT5G65230—were analyzed in WT and transgenic plants by real-time qRT-PCR to better understand the mechanistic basis of the improved osmotic stress tolerance exhibited by the ZmSBP17-transgenic plants. As shown in Figure 6 , the expression levels of 10 genes increased significantly in transgenic plants compared to WT under salt or PEG stresses. For example, the expression levels of the genes encoding the ROS-scavenging enzymes, SODs (CSD1/2 and MSD1) and CATs (CAT1/2), were significantly elevated in the ZmSBP17-OE lines than that in the WT under PEG stress, while the expression levels of CAT1 and APX1 were significantly higher under salt stress. Moreover, the expression level of the drought-related gene (DREB2A) was significantly more than in WT under PEG stress. The expression levels of the Na⁺ or K⁺ transport-related gene (NHX1) were not only significantly lower in the WT than in transgenic plants under normal conditions but also significantly lower in the WT than in the transgenic plants after salt stress to maintain Na⁺ and K⁺ balance. Additionally, the expression level of the MYB TF family genes (MYB53/65) in the OE and WT plants was assessed. The expression level of MYB65 was significantly higher in the transgenic plants than in WT after PEG stress, while that of MYB53 was significantly enhanced in the transgenic plants than that in the WT after salt and PEG stress.

These results indicate that the overexpression of ZmSBP17 may aid in improving osmotic stress tolerance in plants by elevating the expression of ROS-scavenging enzymes encoding, drought-related, Na⁺ or K⁺ transport-related, and MYB TF family genes.

SBP TFs could bind to the GTAC-box ([C][C]GTAC[A/G]) cis-elements. Therefore, ZmSBP17 was cloned into pGADT7 while the GTAC core sequence and mutated elements were repeated five times and cloned separately into pHIS2. Growth assays revealed that the yeast co-transformed with GTAC and ZmSBP17 can grow better on the selection medium, while those containing the GTACm sequence displayed the same growth as the negative control pHis2 ( Figure 7B ). These results indicate that the core GTAC-box sequence is crucial for the ZmSBP17 protein to recognize its target genes.

The promoters (from −1500 to −1) of 10 putative target genes that were significantly upregulated by ZmSBP17 were analyzed to investigate the distribution of the core GTAC-box sequences. The promoters of all putative target genes contained more than four GTAC boxes. The schematic diagrams of the promoters of these genes are shown in Supplementary File 7 . The GTAC-box may be essential for ZmSBP17 to regulate its target genes.

The direct binding of ZmSBP17 to the promoters of the putative target genes containing GTAC-box core sites was verified. The promoter fragments of NHX2 and DREB2A containing the core GTAC-box sequences were isolated through genomic PCR and inserted into pHIS2 in yeast to identify the binding affinity to ZmSBP17 on the selection medium. The results show that all co-transformants exhibited binding affinity on the selection medium ( Figure 7B ), indicating that ZmSBP17 can directly interact with the NHX1 and DREB2A promoter fragments containing the core GTAC-box sequences.