Maize ZmHSP90 plays a role in acclimation to salt stress

Experimental design

At Shenyang Agricultural University’s maize breeding center, a total of 60 Zea mays L. plants were assessed for their resistance to salinity. The Seed Germination Rate-based salinity tolerance Index (GRI) was employed to evaluate the plants’ salinity tolerance, with the formula: GRI = germination rate of salt-treated seeds/germination rate of seeds under normal conditions. The germination process involved the use of different concentrations of saline culture medium for cultivation, with a control group (CK) established for comparison. The salt stress experiment utilized the same concentration gradient of salt stress as that used in Cao et al. (2020b). Next, maize lines with different salt tolerance grades were selected based on their GRI values. The growth status, as well as the concentrations of Na⁺ and K⁺, were measured as evaluation indexes for salt tolerance. This helped in determining the salt-tolerant and salt-sensitive maize materials. Subsequently, salt stress experiments were conducted on the selected materials. Maize plumule and seedling leaf tissues (third leaf stage) were collected for enzyme activity determination and transcriptome analysis. Based on the transcriptome data, candidate salt-tolerant genes were identified. To verify the function of these candidate genes, transgenic technology was employed using tobacco as the model organism. Figure 1 provides a schematic representation of the experimental design.

Part I Salt tolerance evaluation

Salt treatment and sampling during the seedling stage

Maize seedlings at the three-leaf stage and with uniform growth and size were subjected to different salt concentration gradients (0 mmol/L, 50 mmol/L, 100 mmol/L, and 150 mmol/L) under normal growth conditions, including temperature, light, and humidity.

Plant samples were collected at the seedling stage (three-leaf stage) from the flag-leaf under different salt treatment conditions after 24 h and 48 h of salt stress. The fresh samples were immediately frozen in liquid nitrogen and stored at −80 °C until salt resistance characteristics testing use.

Antioxidant enzyme activity determination

Firstly, leaf samples were collected and washed with a phosphate buffer solution (PBS) with a pH of 7.8. Then, the samples were homogenized using PBS with 4% polyvinylpyrrolidone (PVP) and 1 mM ethylene diamine tetraacetic acid (EDTA). The activities of SOD, CAT, and POD were determined using commercial assay kits. The T-SOD assay kit (hydroxylamine method, Cat. SCSP-503; Nanjing Jiancheng Corp., Nanjing, China), the Catalase (CAT) assay kit (Visible light, Cat. A007-1-1; Nanjing Jiancheng Corp., Nanjing, China), and the Peroxidase assay kit (colorimetry method, Cat. A084-3-1; Nanjing Jiancheng Corp., Nanjing, China) were used to determine the activities of SOD, CAT, and POD, respectively. Briefly, plant tissues were first homogenized using a plant tissue homogenizer (Laibei, Tianjin, China) and a refrigerated high-speed centrifuge (Eppendorf, Hamburg, Germany). After that, the SOD, CAT, and POD were measured using commercially available kits according to the manufacturer’s instructions. The activities of SOD, CAT, and POD were then calculated by dividing the total activity by the fresh weight. The experiment was conducted with three biological and three technical replicates.

Na⁺ and K⁺ ions determination

The method used for determining the concentration of Na+ and K+ in the leaf samples was described by Zhao et al. (2014). Firstly, clean leaf samples were dried at 60 °C for 48 h and then ground into a fine powder in a dry mortar. The powder was then subjected to 12-hour nitric acid digestion, and the resulting solution was centrifuged. The supernatant was diluted to 5 times its volume, and the concentration of Na+ and K+ in the acid-digested tissues was measured using a flame photometer (Model 410 Sherwood, UK).

Part II Transcriptome analysis

RNA isolation and RNA-seq libraries preparation

Total RNA was extracted from frozen tissues using TRIZOL reagent (Invitrogen, USA), then purified and concentrated using the RNeasy MinElute Cleanup kit (Qiagen, Germany). The RNA-seq libraries were constructed with an Illumina TruSeq RNA sample preparation kit and TruSeq index adaptors. Sequencing was done on an Illumina Hiseq 4000.

Transcriptome sequencing data analysis

The statistical power analysis for the given sample sizes was evaluated using the R package RNASeqPower version 1.18.0 (Hart et al., 2013). The transcriptome sequencing reads were mapped to the maize genome (https://www.maizegdb.org/genome/assembly/B73%20RefGen_v3) using TopHat2 (Kim et al., 2013), and transcripts were assembled and quantified using the Cufflinks pipeline (Trapnell et al., 2010). Differential gene expression analysis was performed using DESeq2 software with pairwise comparisons of identified genes with significant differences in expression between groups. Gene expression levels were quantified using reads per kilobase per million mapped reads (RPKM) values. Differentially expressed genes (DEGs) were screened based on the criteria of —log2(foldchange)—>1.5 and false discovery rate (FDR) adjusted p-value < 0.05. Up-regulated and down-regulated genes were defined as genes with higher and lower expression levels, respectively, in one group compared to another. Functional analysis was performed using the GO database (http://geneontology.org) and the KEGG database (http://www.genome.jp/kegg/), and Go terms and KEGG pathways were considered significantly enriched at an FDR < 0.05.

Part III Salt tolerance gene transgenic verification

Subcellular localization

“The candidate salt tolerance gene was analyzed for subcellular localization using a vector encoding a GFP-tagged fusion protein as markers. To create the fusion proteins, the salt tolerance gene was cloned into the 16318 h GFP vector with a CaMV 35S promoter: green fluorescent protein (35S: GFP) cassette. The recombinant vector was validated by Sanger sequencing and then transformed into protoplasts using the previously described polyethylene glycol (PEG)-mediated transformation approach (Bart et al., 2006). The fluorescence was observed under an Olympus FluoView FV1000 confocal laser scanning microscope (Olympus, Japan) 16 h after transformation.

Plasmid construction and Agrobacterium tumefaciens (A. tumefaciens)-mediated transformation of tobacco plants

The expression vector pCAMBIA3301, under the control of the CaMV 35S promoter (http://www.cambia.org), was used to clone the full-length cDNAs of the candidate salt tolerance gene. The recombinant vectors and a control pCAMBIA3301 were introduced into A. tumefaciens strains using the freeze-thaw method. The leaf disk transformation was performed based on a previous report (Curtis, Davey & Power, 1995). Briefly, the A. tumefaciens strain carrying the relevant plasmids was transformed into tobacco epidermal cells using the hand-infiltrating method. Three weeks later, the regenerated tobacco tissues were planted into tissue culture vessels containing medium. Positive transgenic plants were screened using qRT-PCR.

Evaluation of salt tolerance in transgenic tobacco

Transgenic and wild-type (WT) tobacco plants were subjected to salt stress treatments. Seeds were initially planted in MS medium for growth assessment, and after three weeks, plants of the same size were transferred to MS medium containing 0, 100, 150, and 200 mM NaCl for 7 days. The plant biomass was estimated as fresh weight, and the activities of SOD, CAT, and POD were evaluated. In addition, the ROS content, Na⁺, and K⁺ ion concentrations in the leaves were also measured. We conducted qRT-PCR analysis on the expression levels of key regulatory genes, namely NtSOS1, NtHKT1, and NtNHX1 investigate whether overexpression of ZmHSP90 has any impact on SOS pathway.

Statistical analysis

The data statistical analysis was conducted using GraphPad Prism 8 software. The T-test was used for comparisons between two groups, while one-way ANOVA with the Tukey test was used for comparisons between multiple groups. A p-value of less than 0.05 was considered statistically significant. For transcriptome analysis, statistical analysis was performed using the software’s established analysis pipeline.

Evaluation of salt tolerance

The study evaluated the GRI of 60 maize inbred lines, sorting the data from high to low. Results from Fig. 2 showed that under salt stress conditions, “975-12” and “DH3732” had the highest GRI values (95.54 each), while “GEMS64” had the lowest GRI value (52.22). Table S1 provided the original GRI data. These findings suggest that “975-12” and “DH3732” displayed salt-tolerant characteristics and were less affected by salt stress during germination, while “GEMS64” demonstrated salt-sensitive characteristics and was greatly affected by salt stress.

It is not comprehensive to judge the salt resistance of maize only by GRI. To further screen salt-tolerant and salt-sensitive materials, five plant materials with the highest, high, low, and the lowest GRI value were selected according to the GRI value: “975-12,” “DH3732,” “GML360,” “Jing2146” and “GEMS64”.

In addition, we conducted a phenotype analysis of these maize seedlings at different salt concentrations. The results revealed that salt stress significantly affected the growth of all five maize inbred lines, with greater damage observed at higher salt concentrations (Figs. 2B–2F). Based on the phenotype analysis, “975-12” (Fig. 2B) exhibited the least growth inhibition under salt stress, while “GEMS64” (Fig. 2F) was the most affected.

In order to investigate the salinity tolerance mechanisms of the selected maize lines, we measured the levels of plant Na⁺ and K⁺ content and calculated the Na⁺/K⁺ ratio. The results showed that seedlings exposed to salt stress had significantly higher Na⁺/K⁺ ratios, which increased as the NaCl concentration increased for all maize materials (Fig. 2G). Under 50 mmol/L salt stress, “Jing2146” and “DH3732” had the highest and lowest Na⁺/K⁺ ratio, respectively; under 100 and 150 mmol/L salt stress conditions, “GEMS64” and “975-12” had the highest and lowest Na⁺/K⁺ ratios, respectively. Original data for these results were provided in Table S2. Based on these findings, “975-12” was considered a salt-tolerant line, while “GEMS64” was a salt-sensitive line.

Effects of salt stress on germination characteristics of salt-tolerant and salt-sensitive lines

Under normal conditions, “975-12” and “GEMS64” had average germination rates of 99.00% and 96.33%, respectively. However, when exposed to salt stress of 50 mmol/L, their germination rates decreased to 95.67% and 62.33%, respectively. Further increasing the salt concentration to 100 mmol/L resulted in germination rates of 99.00% for “975-12” and 50.00% for “GEMS64”. At a salt concentration of 150 mmol/L, the germination rates for “975-12” and “GEMS64” were 90.00% and 46.33%, respectively (Fig. 3). All the germination original data were supplied in Table S2.

In the absence of salt stress, “975-12” and “GEMS64” exhibited comparable germination potentials of 97.67% and 93.33%, respectively. However, their germination potentials were adversely affected when subjected to salt stress. Specifically, at 50 mmol/L salt stress, “975-12” and “GEMS64” had germination potentials of 95.667% and 50.00%, respectively. At 100 mmol/L salt stress, their germination potentials were 95.667% and 41.00%, respectively. Finally, at 150 mmol/L salt stress, “975-12” and “GEMS64” had germination potentials of 88.00% and 42.33%, respectively (Fig. 3). The original data for these results can be found in Table S2 of the supplementary materials.

Effects of salt stress on antioxidant enzymes of candidate maize inbred lines

Under salt stress, the activity of POD in the stress group of maize was significantly higher than that in the CK group in both plumule and leaf tissues. The POD activity in plumule tissues exhibited an increasing trend with increasing salt stress concentrations (Fig. 4A). In leaf tissues, the POD activity showed a significant increase under 50 mmol/L salt stress, but there was no significant difference as the salt concentration continued to rise (Fig. 4B).

The study revealed that in the plumule tissues of “975-12”, SOD activity was significantly elevated at a salt concentration of 50mmol/L but did not exhibit significant changes with further increase in salt concentration. Conversely, in “GEMS64”, SOD activity showed a remarkable increase compared to the control group (CK) when the salt stress reached 150 mmol/L (Fig. 4C). Additionally, in leaf tissues, “975-12” demonstrated a significant change in SOD activity upon exposure to low-dose salt stress, while “GEMS64” did not exhibit any significant changes in SOD activity at different concentrations after 24 h of treatment. However, after being treated with a salt stress concentration of 150mmol/L for 48 h, “GEMS64” showed a significant increase in SOD activity compared to the CK (Fig. 4D).

In plumule tissues, “975-12” exhibited higher CAT activity than “GEMS64” at all salt stress concentrations, including non-salt stress conditions (Fig. 4E). In leaf tissues, CAT activity did not show a significant change after 24 h of salt stress in both maize inbred lines. However, after 48 h of salt stress, CAT activity in “975-12” continued to increase with increasing salt stress concentration, while in “GEMS64”, CAT activity reached its peak at 50mmol/L and did not exhibit significant changes under higher salt stress concentrations (Fig. 4F). All the POD, SOD and CAT original data were supplied in Table S2.

Effects of salt stress on transcriptome

We conducted RNA-seq analysis of plumule tissue from “975-12” and “GEMS64” to investigate the molecular mechanisms of the salt stress response across the genome. The two inbred lines were abbreviated as “975” and “G64,” respectively, and the control and salt stress groups were labeled as “CK” and “T,” respectively. We ensured that the quality of all sequencing data met the analysis requirements (Table S3). The experimental design had a statistical power (calculated by RNASeqPower) of 69.09%, 75.36%, 75.32%, 71.48%, 69.64%, 74.73%, 71.28%, and 79.26% for 975_T vs. G64_T, 975_CK vs. G64_CK, 975_T vs. 975_CK, G64_T vs. G64_CK, 975_48 h vs. 975_CK, 975_24 h vs. 975_CK, G64_48 h vs. G64_CK, and G64_24 h vs. G64_CK, respectively (Table S4).

In our initial analysis, we examined the RNA-seq data of plumule samples during the germination stage. By comparing G64_T vs. G64_CK, 975_T vs. 975_CK, 975_CK vs. G64_CK, and 975_T vs. G64_T, we identified 122, 6882, 7346, and 4262 DEGs, respectively (Fig. 5A). Our GO function and KEGG pathway enrichment analyses revealed that salt stress affected several metabolic pathways, such as defense response, abiotic stimulus–response, oxidation–reduction processes, photosynthesis, and plant hormone signal transduction, among others. Notably, the most enriched pathway was plant hormone signal transmission (zma04075), which includes crucial hormones and genes involved in stress response. Among the DEGs in plant hormone signal transduction, the ABA signal transduction and jasmonic acid pathways were the most affected. In ABA signal transduction pathway, ZmPYR/PYL and ZmPP2C responded positively to salt stress in “975-12” but not in “GEMS64”. More particularly in the jasmonic acid pathway, when exposed to salt stress, ZmJAZ did not show significant changes in “GEMS64”, but the expression level sharply decreased in “975-12” (Fig. 5B). Furthermore, we observed that the expression level of all ZmHSP90 transcripts, which generally considered to be responsible gene to heat stress, were significantly up-regulated in “975-12” after exposure to salt stress, while they were down-regulated in “GEMS64” (Fig. 5C). Among all the ZmHSP90 transcripts, Zm00001d020827 and Zm00001d052809 displayed contrasting expression patterns in response to salt stress across the two cultivars.

The analysis of RNA-seq data from seedling leaf samples indicated that salt stress induced changes in gene expression patterns, resulting in 5081, 7282, 1901, and 1417 differentially expressed genes (DEGs) in G64_24 h vs. G64_CK, G64_48 h vs. G64_CK, 975_24 h vs. 975_CK, and 975_48 h vs. 975_CK, respectively (Fig. 5D). Further investigating GO and KEGG function enrichment results revealed that salt stress influenced several metabolic pathways, such as those involved in abiotic stimulus–response, protein processing in the endoplasmic reticulum, and plant hormone signal transduction. Notably, under salt stress conditions, the expression levels of Zm00001eb172450 (ZmCCA1), Zm00001eb415770 (ZmLHY1), and Zm00001eb148420 (ZmHSP70) showed significant differences between the two varieties (Fig. 5E). The RPKM values can be found in Table S5 of the supplementary materials.

In addition, we utilized the STRING database (https://cn.string-db.org/) to examine the interaction network of the identified genes (which were chosen based on clustering analysis and gene expression), and then created an interaction network map using Cytoscape. A total of 80 differentially expressed genes (DEGs) were included in the analysis of the interaction network. Our results revealed that ZmHSP90 and ZmHSP70 exhibit strong interactions with other genes. Furthermore, ZmPR-P0 and ZmDRTS were found to have extensive interactions (as depicted in Fig. 5F). Taken together with the transcriptome findings, our results suggest that ZmHSPs play a crucial role in conferring salt stress tolerance in maize, thus necessitating further investigation and verification of their functions.

Transformation and regeneration of transgenic plants overexpressing ZmHSP90

As ZmHSP90 expression is induced by salt stress, transgenic tobacco plants were created to investigate the role of ZmHSP90 in salt stress tolerance. Nicotiana benthamiana was selected as the model plant due to its ease of cultivation, short lifespan (∼3 months), and well-established DNA delivery methods. Additionally, tobacco has been previously used to validate maize gene functions (Fang et al., 2021; Ju et al., 2021; Sun, Wang & Xia, 2019; Wang, Wang & Xia, 2019b; Xia et al., 2018). Based on subcellular localization tests, the recombinant plasmid and empty vector were expressed in the nucleus and chloroplast, respectively. Furthermore, the labeled protein fluorescence of the 16318h-ZmHSP90-GFP recombinant plasmid (Zhang et al., 2021) was found to be exclusively detectable in the nucleus and was high, indicating that the coding protein of ZmHSP90 is a nuclear localization protein (Fig. 6A). Two transgenic lines were confirmed by PCR using primers specific to ZmHSP90, and RT-qPCR analysis showed that ZmHSP90 mRNA was only present in the transgenic plants but not in the wild type (WT) (Fig. 6B). The two lines with higher levels of ZmHSP90 expression were then used in the stress tolerance test.

To investigate whether transgenic overexpression of ZmHSP90 was correlated with salt stress tolerance, the WT and transgenic lines were subjected to 100 mmol/L salt stress. Lines overexpressing ZmHSP90 demonstrated significant salt tolerance when the seedlings were grown in 100 mmol/L salt stress for 28 days (Fig. 6C). We examined the peroxidase activity of both transgenic and WT tobacco. The findings depicted that the POD increased significantly in both WT tobacco and transgenic tobacco after salt stress. The increase in POD caused by salt stress was significantly higher in ZmHSP90 overexpression tobacco than in WT tobacco (Fig. 6D). There was no significant difference in SOD activity between the two kinds of tobacco before and after salt stress (Fig. 6E). Regarding CAT, salt treatment induced more CAT in ZmHSP90 overexpression tobacco than in CK (Fig. 6F). Maintaining a balanced cytosolic Na⁺/K⁺ ratio has become a key salinity tolerance mechanism. We found that the balance of sodium and potassium ions was disrupted by salt stress. However, in tobacco plants overexpressing ZmHSP90, the adverse effects of salt stress on Na⁺/K⁺ were smaller than in WT tobacco plants (Fig. 6G). In addition, ZmHSP90 significantly enhanced the expression levels of NtSOS1, NtHKT1, and NtNHX1 as compared to those in the WT (Fig. 6H), suggesting that ZmHSP90 modulates Na⁺ transporters to maintain K⁺/Na⁺ balanced in tobacco. All the transgenic related original data were supplied in Table S6.