Integrated transcriptomic and metabolomic analyses elucidate the effects of exogenous spermidine priming on seed germination and seedlings growth of Elymus sibiricus under salt stress

Abstract

Background

Seed priming is an essential technical tool in agricultural production to solve the problem of slow seedling emergence and irregular seedling emergence. Salt stress significantly inhibits plant growth and development. Although Siberian wildrye (Elymus sibiricus L.) exhibits strong salt tolerance, the molecular mechanisms underlying its adaptation to high-salt stress require further investigation. This study aimed to elucidate how spermidine seed priming improves germination and seedling growth of Elymus sibiricus under salt stress by integrating physiological traits with transcriptomic and metabolomic profiles.

Methods

In this study, the E. sibiricus cv. ‘Lanyu No.2’ was utilized as experimental material. Seeds were soaked in four different concentrations of spermidine solution and germinated for 11 days under 200 mM NaCl to investigate the effects of spermidine priming on the germination stage. Additionally, seedlings grown normally to 60 days were subjected to 14 days of salt stress treatment to examine the effects of spermidine priming on the seedling stage.

Results

The results of this study indicate that under salt stress, 0.50 mM spermidine priming increased seed germination rates and bioaccumulation levels. To reveal the mechanism of spermidine priming alleviating the salt stress, the chlorophyll and carotenoid content, antioxidant enzyme activity, and osmotic regulatory substance content were detected. Our results demonstrated that seedlings from 0.50 mM spermidine-primed seeds showed better photosynthetic efficiency and a stronger ability to scavenge reactive oxygen species. Besides, transcription sequencing analysis, it was found that almost all the photosynthesis-related genes in the spermidine-primed seedlings were up-regulated after 14 days of salt stress treatment. It might be a way to enhance the photosynthetic rate to adapt to salt stress in E. sibiricus. Weighted gene co-expression network analysis (WGCNA) identified three hub genes that may be involved in E. sibiricus salt response. Additionally, integrated transcriptomes and metabolomes analyses revealed that spermidine may enhance salt tolerance in E. sibiricus seedlings through the phenylpropanoid biosynthesis and glyoxylate and dicarboxylate metabolism pathways.

Conclusions

Our findings provide candidate pathways and markers that could be exploited in breeding and seed technology for salt-tolerant forage grasses.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-08358-4.

Background

Elymus sibiricus has resilience to abiotic stress, possesses high biomass production, and has excellent adaptability [1, 2], it is essential to establishing artificial grasslands and restoring natural grasslands that have been damaged [3]. Additionally, E. sibiricus is becoming more and more acknowledged as a potentially useful energy crop for cultivation on marginal areas because of its exceptional fiber content, high production of biomass, and effective fermentation characteristic [3]. However, significant seed production problems brought on by recent industrial expansion include accelerated senescence, decreased seed germination rates, seedling growth is inhibited, and plants may even die [4]. Specifically, one of the most significant environmental issues is salinization, which is an abiotic stressor that modifies most plants’ physiological, biochemical, and metabolic functions [5, 6], causing oxidative stress and compromising the metabolism of plants, performance, and production [7], severely limiting the growth of plants, development and productivity, and this is among the most significant issues facing agricultural productivity today [8]. E. sibiricus has a stronger tolerance to salt than other forages; however, soils with excessive salt concentrations have impacted its growth and productivity. The main negative effects of salt stress on plants involve reactive oxygen species (ROS), osmotic stress, nutritional imbalance, and specific ion toxicity. Osmotic stress is an immediate occurrence brought on by the concentration of salt surrounding the roots during the initial stages of salt stress [9, 10]. Subsequently, Na⁺ accumulation results in nutrient imbalance, which in turn leads to specific ion toxicity [11]. In some instances, the plant reduces Na⁺ toxicity by encouraging compartmentation and ion transport, and it builds adaptive mechanisms by accumulating antioxidants, enzyme activities, and osmoregulatory substances [12]. In agricultural production, seed germination, seedling emergence, and seedling stages are more vulnerable to NaCl stress than later stages, which would result in more significant issues during E. sibiricus production of sibiricus and must be resolved immediately.

To solve the problem of inadequate germination in extremely salinized soil, various seed treatments lessen the negative effects of NaCl stress on the process by which seeds germinate, characteristics related to the emergence of seedlings, and its beneficial effects on subsequent growth stages for a higher yield. Additionally, a variety of techniques and technologies have been used on plants to increase resistance to salt stress [13]. Seed priming increases seed tolerance in the face of adversity and is an economically viable and environmentally beneficial type of seed pretreatment due to its simplicity and convenience of usage [14]. Seed priming therapy increases seed germination vigor and activates physiological and biochemical metabolisms associated with seed germination, strengthening seed tolerance to adversity and increasing yield [15]. Natural substances called polyamines (PAs) are made up of two or more amino groups. Spermidine (Spd) is one of the numerous common PAs in plants that serve a range of physiological functions [16, 17]. Spd is a signaling molecule and a growth regulator that aids in plant development and responses to biotic and abiotic stresses [18]. Drought, salinity, heavy metals, and similar factors are classified as abiotic stressors, and each stress type can induce distinct morphological, physiological, and genotoxic responses in plants [19, 20]. According to studies, Spd efficiently reduces damage from salt stress and has a protective impact on plants that is especially noticeable under stressful circumstances [21], stress caused by heavy metals in rice, and stressful situations [22]. Furthermore, Spd also scavenges ROS, protects photosynthetic organs, encourages root and bud growth, regulates plant hormones, and preserves redox balance [23]. Yang et al. [24] used salt-tolerant and salt-sensitive rice cultivars as experimental materials, demonstrating that exogenous Spd seed priming reduces malondialdehyde accumulation, enhances osmotic adjustment capacity, increases photosynthetic pigment content, alleviates seedling electrolyte leakage damage, and maintains ion homeostasis. Cheng et al. [25] demonstrated that under salt stress, Spd seed priming could enhance seed germination vigor, germination rate, and seed vigor index of the leguminous forage Trifolium repens, and could also facilitate starch hydrolysis by activating β-amylase activity, thus maintaining Na⁺/K⁺ homeostasis. Furthermore, Spd priming mitigates osmotic stress and oxidative damage, maintains cell membrane integrity, and enhances seed vigor and germination capacity of Leymus chinensis under salt-alkali stress [26].

Gene mining for significant agronomic traits now has strong technical support thanks to the development of omics methods [27]. Numerous investigations have employed histological research methods to examine how genes, proteins, and metabolites are expressed in abiotically stressed plants. Plant processes in reaction to NaCl stress are intricate and multifaceted. Complex biological processes and the regulation of biological networks are difficult to be fully described by a single type of omics data. Understanding the biological process that regulates the mechanism of plants reacting to abiotic stress is better achieved by the joint analysis of multi-omics data [28].

Multiomics joint analysis has been applied extensively to a number of plants, including Nicotiana tabacum [29], Zea mays [30], Oryza sativa [31], and sorghum [32]. Although Spd-based seed priming under salinity has been extensively studied in many model species, a comprehensive investigation combining “seed priming-germination-seedling omics” has not been performed in E. sibiricus. Furthermore, this study identified three core hub genes (EsIRE4, EsTPKA, and EsSRG1), which may hold substantial potential for the mechanistic elucidation of Spd seed priming-enhanced salt tolerance in E. sibiricus and provide well-defined candidate genes for subsequent functional validation. This study assumes that Spd priming may regulate photosynthesis, enhance the reactive oxygen species (ROS) scavenging system and osmotic adjustment capacity, and coordinate the phenylpropanoid biosynthesis as well as glyoxylate and dicarboxylate metabolism pathways to improve salt tolerance in E. sibiricus seedlings. The objective is to elucidate the mechanism by which Spd priming improves E. sibiricus seed germination and growth when exposed to salt stress, consequently offering a theoretical foundation for its use in the production of E. sibiricus.

Materials and methods

Materials for plant and seed priming

Seeds of E. sibiricus cv. ‘Lanyu No.2’ were supplied by the State Key Laboratory of Herbage Improvement and Grassland Agro-Ecosystems, Lanzhou University, Lanzhou. After disinfecting the seeds’ surfaces with 5% sodium hypochlorite, absorbent paper was used to carefully remove any remaining surface moisture. Seed priming was conducted using varying amounts of Spd solution (0.01 mM, 0.25 mM, 0.50 mM, and 1.00 mM) and control (CK) seeds were those that had not been primed. Based on prior investigations and preliminary tests, the Spd solution concentrations were selected. Using occasional manual agitation to preserve solution homogeneity and prevent mechanical seed injury by homogeneity and excessive shaking, the priming procedure was carried out for 12 h at 20 ± 1 °C in dark conditions. The purpose of this carefully regulated shaking procedure was to maintain the integrity of the seed coat while balancing solution-seed contact. Seeds primed with different concentrations of Spd were placed in an incubator at 20 ℃ and dried back for 24 h to the original weight of the seeds and then immediately subjected to a germination test.

Germination test and salt stress treatment

Using the filter paper germination method, which involves evenly distributing 50 seeds (these 50 seeds were completely randomly selected) per 15 cm Petri dish lined with double-layer filter paper, primed and unprimed seeds were simultaneously exposed to a salt stress germination test. Three different biological replicates were created (a Petri dish containing 50 seeds to represent one biological replicate) and processed under the same experimental circumstances in order to guarantee statistical robustness. Each petri dish was exposed to either 15 mL of 200 mM NaCl solution (salt stress) or deionized water (normal condition) for 11 days under growth conditions (25 ℃/ 20 ℃ (day/ night), 16 h of light and 8 h of darkness, 10,000 LX of light, and 70% relative humidity). NaCl solution, H₂O, and filter papers were replaced daily. To prevent position effect, every petri plate in the growth chamber was moved every day. Over the course of the 11-day germination period, germination features were noted. The experiment considered root length or bud length equal to seed length as germination, and the number of sprouts was counted daily and moldy seeds were promptly treated [33].

Salt stress treatment of seedlings

Seedling salt resistance was identified by the potting method. After different Spd-primed seeds of E. sibiricus were germinated under normal conditions for 7 d (with distilled water), selected seeds with consistent germination were sown in a soil mixture (humus: vermiculite = 3:1, PH = 6.5–7.2, EC = 0.35–0.52 mS/cm, major nutrient contents: nitrogen, phosphorus, potassium, and organic matter) that had been saturated with absorbed water and incubated for 60 d under sufficient moisture, 25 ℃, and 16 h of light/8 h of darkness. During this period, the seedlings were supplied with sufficient water, with the soil water content maintained at 80% of the field capacity. The gravimetric method was adopted for soil moisture control of potted plants. Each pot was weighed daily, and the irrigation volume per pot was calculated according to the target soil field capacity before irrigation was conducted. Subsequently, salt stress was applied to the seedlings. In this experiment, 200 mM NaCl solution, this concentration mimic the average saline level in salinized land in China [34]. The salt stress was induced by 200 mM NaCl solution with an electrical conductivity (EC) of 22.5 mS/cm at 25 ℃. Seedlings were designed into six treatments, including seedlings without priming and salt stress (CK), seedlings without priming under salt stress (CK + ST), seedlings from seeds primed with 0.01 mM Spd under salt stress (S0.01 + ST), seedlings from seeds primed with 0.25 mM Spd under salt stress (S0.25 + ST), seedlings from seeds primed with 0.50 mM Spd under salt stress (S0.50 + ST), and seedlings from seeds primed with 1.00 mM Spd under salt stress (S1.00 + ST). Since the samples for subsequent metabolomics sequencing were leaves from this step, four biological replicates were set up, a total of 24 pots.

Assessment of biomass and morphological characteristics

After two weeks of salt stress treatment, biomass (fresh weight and dry weight) and plant height were measured. Five seedlings were randomly selected from each pot, and the mean of the measured values from these five seedlings was calculated to obtain a pot mean value, with each pot regarded as one biological replicate. Setting up three biological replicates.

Quantification of leaf chlorophyll and carotenoid contents

The total chlorophyll content was determined using the Chlorophyll Assay Kit (Cat. No.: BC0995) from Solarbio Science & Technology Co., Ltd. (https://www.solarbio.com/). The Kirk and Allen (1965) method [35] was used to measure the carotenoid levels. Photosynthetic pigment extraction was carried out in an 80% acetone solution.

Measurement of ROS, MDA content, and antioxidant enzyme activities

The contents of ROS and MDA, as well as the activity of superoxide (SOD), peroxidase (POD), catalase (CAT), and ascorbate peroxidase (APX), were measured in the leaves of seedlings following a 14-day salt treatment. Following the grinding of the leaf material (0.5 g) with liquid N₂, ten milliliters of buffer (potassium phosphate) were added. Following that, centrifugation was performed for 20 min at 4 °C with a centrifugation force of 10,000 × g.

Nitrogen blue tetrazolium was used to measure the SOD activity [36]. Guaiacol was used to determine the POD activity [37]. The CAT activity was assessed using the previously mentioned methodology [38]. Using Nakano and Asada’s (1980) time scan approach, the APX activity from buffer-extracted leaf samples was determined, and the absorbance was measured at 290 nm [39]. The content of H₂O₂ was determined using the method of Song et al. [40]. The O₂⁻ content was determined using the method of Lin et al. [41]. The thiobarbituric acid method was used to measure the amount of MDA [42].

Determination of soluble sugar, soluble protein and proline contents

The anthrone method (glucose standard: 0.1 mg/ml of distilled water) was used to measure the content of soluble sugar in fully grown leaves at seedlings at various treatment levels [43]. Using the absorbance measurement optical density (OD) at 620 nm, the sugar content was expressed as mg glucose g⁻¹FW. The soluble protein content was determined by the Bradford method [44] using bovine serum albumin as the protein standard. The proline content was determined by the Ninhydrin colorimetric method [45].

Transcriptome analysis

Total RNA extraction, library preparation, and sequencing were performed by Personalbio (https://personalbio.bioon.com.cn). Sequencing the samples produced FASTQ raw data. To filter some reads with connections and bad quality in the raw data, HISAT2 (https://ccb.jhu.edu/software/hisat2/index.shtml), an improved version of TopHat2, was utilized [46]. Clean data, or high-quality sequences, were matched to this species’ reference genome, the reference genome is E. sibiricus cultivar ‘Chuancao No.2’ [47]. HTSeq was used to statistically align each gene’s read count, and the outcome served as the gene’s initial expression level [48]. FPKM (fragments per kilobase per million fragments) was used to normalize the expression level, which was then used as the gene’s expression level. Differential expression analysis between the control and treatment groups was performed using the DESeq R package (version 1.10.1). Genes with a false discovery rate (FDR) < 0.01, P value < 0.01, and a fold change ≥ 1.5 were identified as differentially expressed genes (DEGs) [2, 49]. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis was performed using ClusterProfiler. The differentially expressed genes (DEGs) identified by KEGG pathway analysis were used in the analysis to determine the gene list and the number of genes for each pathway. The P value (P value < 0.01 was considered to be an enrichment criterion) was then ascertained using the hypergeometric distribution approach. The primary biological roles of the DEGs were ascertained by comparing the KEGG pathways that were enriched in the DEGs with the whole-genome background. Three biological replicates were set up for each treatment in transcriptomic sequencing.

Weighted gene co-expression networks (WGCNA) analysis

WGCNA (weighted gene co-expression networks analysis) utilizes an R package to analyze gene expression data. It creates gene modules, groups genes with similar expression patterns according to their correlation coefficients, investigates connections between these modules and phenotypic features (P-value < 0.05 was considered statistically significant), and finds genes linked to the target trait [50]. After first identifying differentially expressed genes (DEGs) based on the criteria of FDR < 0.01 and fold change > 1.5, we selected those DEGs with FPKM > 1 as the expression dataset for WGCNA analysis. Trait data primarily included SS, SP, Pro, Car, Chl, SOD, CAT, APX, MDA, ROS. Visualization of the co-expression network was generated using Cytoscape software v3.3.0 [51].

Metabolome analysis

At 4 °C, the samples were extracted. The supernatant was examined via ultrahigh-performance liquid chromatography-tandem time-of-flight mass spectrometry (UHPLC-Q-TOF MS) following being vacuum-dried and redissolved in 100 µL of acetonitrile aqueous solution. The chromatographic column employed for analysis was an ACQUITY UPLC HSS T3 column (100 Å, 1.8 μm, 2.1 mm × 100 mm), with a flow rate of 0.4 mL/min, column temperature maintained at 40 °C, autosampler temperature at 8 °C, and injection volume of 2 µL. The gradient elution program was set as: 0 min, 5% B; 1 min, 5% B; 4.7 min, 95% B; 6 min, 95% B; 6.1 min, 95% B; 8.5 min, 5% B. Mass spectrometry data were acquired in both positive and negative ion modes using data-dependent acquisition (DDA) on a Thermo Orbitrap Exploris 120 mass spectrometer, which was controlled by Xcalibur software (Version 4.7, Thermo Fisher Scientific). XCMS software was employed for peak alignment, peak area extraction, and retention time correction. Utilizing liquid chromatography tandem mass spectrometry (LC-MS/MS), energy-metabolizing compounds were assessed. A 70% methanol-water solution was used to extract the material in 50 ± 2.5 mg. After being stored at 20 °C for half an hour, the supernatant was centrifuged at 4 °C. A protein precipitation plate was used to run the supernatant for additional measurement. The quality of the data was evaluated using the following methods: the population sample’s principal component analysis (PCA), the QC sample’s total ion chromatogram, the QC sample’s correlation, the population sample’s Hotelling’s T2 test, the QC sample’s multivariate control chart, and the QC samples’ relative standard deviation. Differentially abundant metabolites (DAMs) were identified based on a false discovery rate (FDR) < 0.05 calculated using the Benjamini-Hochberg (BH) procedure, with metabolites considered significant if they had a variable importance in projection (VIP) value > 1 and P value < 0.05. Multiple statistical analyses were conducted, including differentially abundant metabolite screening, KEGG pathway analysis, univariate statistical analysis, multidimensional statistical analysis, and correlation analysis of differentially abundant metabolites. Metabolome sequencing with four replicates per treatment.

Quantitative real-time PCR (qRT-PCR) analysis

Eight DEGs were selected in order to use qRT-PCR to corroborate the RNA-seq data. The Promega kit was used to extract the RNA from the E. sibiricus samples, and then the reverse transcription of the RNA to cDNA was performed. The CFX96 System was used to perform the qRT-PCR process using Vazyme’s SYBR Premix reagent. Primers unique to each gene have been generated using the NCBI Primer-BLAST program (Table 1). The total volume of the RT-qPCR reaction system was 10 µL, consisting of 5 µL of 2×TSINGKE Master qPCR Mix, 1 µL of template cDNA, 0.4 µL each of forward and reverse primers (10 µmol/L), 0.2 µL of 50×ROX Reference Dye Ⅱ, and 3 µL of ddH₂O. The qPCR reaction procedure was as follows: initial denaturation at 95 °C for 1 min, followed by 40 cycles of denaturation at 95 °C for 10 s and annealing-extension at 60 °C for 30 s; a melt curve analysis was subsequently performed under the conditions of 95 °C for 15 s, 60 °C for 1 min, and 95 °C for 15 s. Three technical replicates were set for each sample. All detected transcripts were normalized against the PP2A (Protein phosphatase 2 A) gene [52] as an internal control, PP2A is a highly conserved housekeeping gene in plants, which is involved in fundamental metabolism, with a stable basal expression level and no tissue specificity [53], preliminary experiments indicate that the expression of this gene remains stable across the six different treatment groups in this study. The gene expression levels were calculated using the 2^−ΔΔCT method [54].

Table 1.

The primer information for qRT-PCR

Gene name	Forward primer	Reverse primer
EsiS02g0063320 (TPS7)	TAGAGATTGAGGATGAACT	GCTTGGTATGTAACTGTT
EsiS04g0040270 (CYP96T1)	TTGAGGTGTAGTAGTAGTT	ATGGAAGTTATGGAGATTG
EsiH07g0039510	GTCGTCTGCAACAACAAC	GGTTTCTGTCCAATGATATTCC
EsiS06g0029390 (LCY1)	TCGAAGGTGGAGATCATG	TATTCCTGTCCTGTATCAAGT
EsiO0023510 (RPPL1)	TGGAGCCTGAGTTCTATGG	GCACTACGGTCAATTCATCTT
EsiS03g0048480 (CYP71D55)	TGAACCTGAGCGTGATGT	CGTTCTTCCGCTTCTTCC
EsiH07g0022170	CATCGTCACCGAGTTCGTGG	CGTCGTCGTCCTCTTGGTTA
L_transcript_20878 (EsPP2A)	AAGAGGAAACCACGAGAGCC	CAACCAATGCTGTCAACGGT

Statistical analysis

Experimental data were collated and preliminarily calculated using Microsoft Excel 2010 software. Prior to parametric statistical analysis, homogeneity of variances was verified for all datasets to ensure that they met the prerequisites for analysis of variance (ANOVA). Statistical analyses were performed using SPSS 22.0 software (SPSS, Inc., Chicago, IL, USA). In the present study, exogenous Spd concentration was set as the sole fixed factor, including different concentration gradients and a spermidine-free control treatment. Given that the study focused on the effects of exogenous spermidine on the traits of E. sibiricus under 200 mM NaCl stress, one-way analysis of variance (one-way ANOVA) was employed as the core linear model for all datasets, followed by post hoc multiple comparison tests. When significant differences among treatments were detected by one-way ANOVA (P < 0.05), Duncan’s multiple range test was further conducted for pairwise comparisons to determine the level of significant differences between treatment groups. Germination indices and physiological parameters are expressed as mean ± standard deviation (SD), representing the average of three biological replicates. Graphs were plotted using SigmaPlot 12.5 (Systat Software, USA).

Results

Effects of Spd priming on morphological traits of E. sibiricus under salt stress

On the 11th day after seed germination, the growth status of seedlings under different treatment conditions was compared. It was found that the growth status of E. sibiricus seedlings under salt stress was worse than that of the control group (CK), while seedlings treated with Spd priming performed better than unprimed ones (Fig. 1A). Among all groups, the CK had the highest germination rate of 69.33%, followed by the S0.50 + ST group with a germination rate of 38.00%, the situation of S0.01 + ST (32.67%) was similar to S0.25 + ST (31.33%), thereafter was the S1.00 + ST (29.33%), and CK + ST (26.67%) had the lowest germination rate (Fig. 1B). By measuring bud length and root length, it turned out that the bud length of CK (72.74 mm) was greater than that of bud length under salt stress. The bud length of S0.50 + ST (29.36 mm) and S0.25 + ST (28.04 mm) were the best among Spd-primed, in addition, the bud length of the S0.50 + ST group was 23.09%, 34.08%, and 71.74% higher than that of the S1.00 + ST, S0.01 + ST, and CK + ST groups, respectively (Fig. 1C). In addition, the root length of the CK group (44.61 mm) was significantly greater than that of the salt-stressed groups, and the root length of the S0.50 + ST treatment was 165.50% higher than that of the CK + ST group (Fig. 1D).

Fig. 1.

Phenotypes of seedlings under salt treatment (ST) with different concentrations (0.01 mM, 0.25 mM, 0.50 mM, 1.00 mM) of Spd (S0.01 S0.25, S0.50, S1.00) and control (CK). A represents each treatment’s growth state; B, C, and D compare the germination rate, bud length, and root length, respectively. Data are presented as bars representing mean ± SD of three biological replicates. Distinct letters indicate significant variations at P < 0.05, following Duncan’s multiple range test

Effects of Spd priming on morphological characteristics and seedling biomass during salt stress

On the 60th day following seedlings, the growth status of seedlings under various treatments is assessed. The findings demonstrated that the E. sibiricus seedlings primed with varying concentrations of Spd and CK under 200 mM NaCl stress did not significantly differ from one another, and their growth conditions were superior to those of the unprimed seedlings under salt treatment (CK + ST) (Fig. 2A). Shoot height measurement results showed that the shoot heights of the CK group and the S0.50 + ST group were higher than those of the other treatment groups, and the shoot height of the S0.50 + ST group was 40.99%, 20.73%, 23.67%, and 10.98% higher than that of the CK + ST, S0.01 + ST, S0.25 + ST, and S1.00 + ST groups, respectively (Fig. 2B).

Fig. 2.

The phenotypic and biomass of seedlings under salt treatment (ST) with different concentrations (0.01 mM, 0.25 mM, 0.50 mM, 1.00 mM) of Spd (S0.01 S0.25, S0.50, S1.00) and control (CK). A is the phenotype of various treatments; B is the comparison of shoot height. C is fresh weight, and D is dry weight. Data are presented as bars representing mean ± SD of three biological replicates. Distinct letters indicate significant variations at P < 0.05, following Duncan’s multiple range test

The fresh weight of the shoot in CK was higher than others except S0.50 + ST and S1.00 + ST (Fig. 2C). After salt stress, the fresh weight of the shoot of S0.50 + ST increased by 28.15%, 18.52%, and 21.63% over CK + ST, S0.01 + ST, and S0.25 + ST, respectively. Compared to others, the dry weight of the shoot in CK was higher, while under salt stress, S0.50 + ST manifested the highest dry weight of shoot (1.68 g), increased by 41.49%, 28.16%, 25.97%, and 12.84% over CK + ST, S0.01 + ST, S0.25 + ST, and S1.00 + ST, respectively (Fig. 2D).

Spd priming regulates the content of chlorophyll and carotenoid under salt stress

The content of chlorophyll was highest in CK and S0.50 + ST at 112.11 µg/g and 109.39 µg/g, respectively, with no significant difference between the two. CK + ST had the lowest content at 87.73 µg/g, followed by S1.00 + ST (99.40 µg/g), S0.01 + ST at 102.59 µg/g, and 105.52 µg/g for S0.25 + ST, among these, the chlorophyll content of S0.50 + ST was 24.69% higher than that of CK + ST (Fig. 3A). The carotenoid content of CK and CK + ST did not differ significantly, which had the lowest levels at 69.89 µg/g and 71.73 µg/g, respectively. Under salt stress, the carotenoid content in the leaves of E. sibiricus seedlings primed with different concentrations of Spd was significantly higher than that in the unprimed group (CK + ST). Specifically, the carotenoid content of the S0.01 + ST, S0.25 + ST, S0.50 + ST, and S1.00 + ST groups was 8.96%, 8.37%, 15.84%, and 12.35% higher than that of the CK + ST group, respectively. It can be concluded that the leaves of E. sibiricus seedlings primed with 0.50 mM Spd had the highest carotenoid content (Fig. 3B).

Fig. 3.

Chlorophyll (A) and carotenoid (B) content in control (CK) and different concentrations (0.01 mM, 0.25 mM, 0.50 mM, 1.00 mM) of Spd (S0.01, S0.25, S0.50, S1.00) under salt treatment (ST). Data are presented as bars representing mean ± SD of three biological replicates. Distinct letters indicate significant variations at P < 0.05, following Duncan’s multiple range test

Spd priming decreases ROS and MDA levels during salt stress and stimulates the activity of antioxidant enzymes, soluble sugar, soluble protein, and proline

Based on the data gathered, the lowest levels of SOD, POD, CAT, and APX activities were identified in CK, with respective values of 2.85 IU/mL, 1.7 × 10^− 5 IU/mL, 7.24 IU/mL, and 0.19 IU/mL (Fig. 4). Of all the treatments, S0.50 + ST had the highest antioxidant enzyme activity, rising by 50.84% (SOD), 60.04% (POD), 41.38% (CAT), and 38.17% (APX), in that order. For SOD activity, the corresponding values for CK + ST, S0.01 + ST, S0.25 + ST, and S1.00 + ST were 25.12%, 26.28%, 14.31%, and 39.10% lower than S0.50 + ST (Fig. 4A). For POD activity, S0.50 + ST was 31.82%,11.00%,17.80% and 27.24% higher than CK + ST, S0.01 + ST, S0.25 + ST and S1.00 + ST, respectively (Fig. 4B). For CAT activity, S0.50 + ST was 23.75%, 17.93% and 31.39% higher than CK + ST, S0.01 + ST and S1.00 + ST, respectively, and was not significantly different from S0.25 + ST (Fig. 4C). For APX activity, S0.50 + ST was 23.05%, 6.83%, 14.10% and 11.88% higher than CK + ST, S0.01 + ST, S0.25 + ST and S1.00 + ST respectively (Fig. 4D). These findings suggest that priming E. sibiricus seedlings with 0.50 mM Spd can increase the activities of superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), and ascorbate peroxidase (APX) under salt stress conditions.

Fig. 4.

The activity of four antioxidative enzymes including (A) SOD, (B) POD, (C) CAT, (D) APX and the content of (E) MDA, (F) ROS, (G) soluble sugar, (H) soluble protein, (I) proline in CK and different concentrations (0.01 mM, 0.25 mM, 0.50 mM, 1.00 mM) of Spd (S0.01 S0.25, S0.50, S1.00) priming under salt treatment (ST). Data are presented as bars representing mean ± SD of three biological replicates. Distinct letters indicate significant variations at P < 0.05, following Duncan’s multiple range test

Additionally, the results showed that each of the treatments during NaCl stress exhibited greater MDA contents than the control. There was a difference between the two, with CK + ST demonstrating a substantial increase (57.89%) and S0.01 + ST showing the least increase (21.66%) and S0.50 + ST exhibiting the least increase (13.21%) (Fig. 4E). The content of ROS in the S0.50 + ST group was the lowest, which was 19.79% lower than that in the CK + ST group (Fig. 4F).

The lowest soluble sugar content was found within CK (9.76 mg/g) and CK + ST (9.73 mg/g), under salt stress, the soluble sugar contents of the S0.01 + ST, S0.25 + ST, S0.50 + ST and S1.00 + ST groups increased by 14.14%, 18.14%, 14.57%, and 7.70%, respectively, compared with the CK + ST group (Fig. 4G). For soluble protein content, CK was lowest, which is 11.56 mg/g. The S0.50 + ST had the highest soluble protein content of 14.35 mg/g. CK + ST, S0.01 + ST, S0.25 + ST and S1.00 + ST showed no significant difference in soluble protein, which was 17.72%, 13.46%, 17.84% and 13.31% lower than that of S0.50 + ST, respectively (Fig. 4H). There was no discernible difference between the maximal proline contents of S0.50 + ST and S1.00 + ST, which were 0.67 µg/g and 0.61 µg/g, respectively. Compared to CK, CK + ST, S0.01 + ST, and S0.25 + ST, the proline content of S0.50 + ST was 352.74%, 182.37%, 87.45%, and 69.70% higher, respectively (Fig. 4I). A summary table of key physiological parameters is shown as Table 2.

Table 2.

Summary table of key physiological parameters

Physiological parameter	% Increase/decrease	Physiological parameter	% Increase/decrease
Germination rate	42.50%	POD	39.75%
Bud length	71.74%	CAT	23.75%
Root length	165.50%	APX	23.04%
Shoot height	66.93%	MDA	-28.30%
Fresh weight	39.18%	ROS	-19.79%
Dry weight	70.92%	Soluble sugar	14.57%
Chl	24.69%	Soluble protein	16.90%
Car	15.84%	Proline	182.33%
SOD	31.60%	-	-

% Increase/decrease refers to the percentage of increase or decrease in S0.50 + ST compared with CK + ST

Analysis of transcriptome sequencing, screening of DEGs, go enrichment and KEGG pathway analysis

According to the experimental findings, E. sibiricus seedlings under salt stress responded best to priming with 0.50 mM Spd. Therefore, to demonstrate how Spd under salt stress alters the gene regulatory network, sequencing the transcriptome of the leaf sample from CK, CK + ST, S0.50, and S0.50 + ST after 14 d of treatment was performed. RNA-Seq sequencing produced 84.56 Gb of clean data in total. All of the samples had base percentages (Q30) higher than 96.23%. Over 90% of the readings were compared to the reference genome of E. sibiricus cv. ‘Chuancao No.2’ (Table S1). These results indicate that filtered data obtained through transcriptome sequencing exhibits high quality, with favorable alignment outcomes and elevated sequence coverage, rendering it suitable for subsequent data analysis.

Analysis of DEGs among different treatments in E. sibiricus revealed that under Spd-primed conditions, 10,340 DEGs were shared between E. sibiricus (S0.50 + ST) and S0.50, comprising 4,914 up-regulated genes and 5,426 down-regulated genes; under non-primed conditions, salt stress (CK + ST) versus non-stressed (CK) conditions yielded 8,423 DEGs, comprising 4,086 up-regulated genes and 4,337 down-regulated genes; under non-stress conditions, 4,562 DEGs were identified between Spd primed (S0.50) and non-primed (CK), comprising 1,967 up-regulated genes and 2,595 down-regulated genes; under salt stress, 578 DEGs were identified between E. sibiricus (S0.50 + ST) and (CK + ST), comprising 279 up-regulated genes and 299 down-regulated genes (Fig. 5A). Among these DEGs, a total of 25 genes were present in all four comparison groups (Fig. 5B).

Fig. 5.

Analysis of DEGs for (A) S0.50 + ST vs. S0.50, CK + ST vs. CK, S0.50 vs. CK, S0.50 + ST vs. CK + ST. (B) Venn displays the DEGs for each of the four treatments. DEGs enrichment in the KEGG pathway across the four comparator groups. (C) S0.50 + ST vs. S0.50, (D) CK + ST vs. CK, (E) S0.50 vs. CK, and (F) S0.50 + ST vs. CK + ST are the top 20 enriched KEGG pathways of DEGs. The enrichment increases with the size of the rich factor

The KEGG pathway enrichment analyses were conducted on the identified DEGs to further elucidate the molecular mechanisms by which Spd alleviates salt stress in E. sibiricus. DEGs were associated with 136, 134, 129, and 84 KEGG pathways in S0.50 + ST vs. S0.50, CK + ST vs. CK, S0.50 vs. CK, and S0.50 + ST vs. CK + ST, respectively, according to the KEGG pathway enrichment analysis. In S0.50 + ST vs. S0.50, the DEGs were predominantly enriched in photosynthesis-antenna proteins, porphyrin and chlorophyll metabolism, one carbon pool by folate, vitamin B6 metabolism, and carbon fixation in photosynthesis organisms (Fig. 5C). In CK + ST vs. CK, the DEGs were predominantly enriched in photosynthesis-antenna proteins, vitamin B6 metabolism, thiamine metabolism, linoleic acid metabolism, and carbon fixation in photosynthesis organisms (Fig. 5D). In S0.50 vs. CK, the DEGs were predominantly enriched in photosynthesis-antenna proteins, beta-alanine, linoleic acid metabolism, flavone and flavonol biosynthesis, and isoflavonoid biosynthesis, and fatty acid degradation (Fig. 5E). In S0.50 + ST vs. CK + ST, the DEGs were predominantly enriched in monobactam biosynthesis, steroid biosynthesis, linoleic acid metabolism, carbon fixation in photosynthesis organisms, carotenoid biosynthesis, and lysine biosynthesis (Fig. 5F).

Spd priming regulates gene expression in photosynthesis pathways under salt stress

KEGG pathway enrichment analysis revealed that the expression of photosynthesis-antenna proteins was upregulated in S0.50 + ST vs. S0.50 and CK + ST vs. CK but downregulated in S0.50 vs. CK (Fig. 6A, B). DEGs involved in photosynthesis and photosynthetic antenna proteins were screened using criteria of Fold change ≥ 1.5 and P value < 0.01. In this study, we identified 154 DEGs related to photosynthesis, including 19 photosystem I (PS I), 82 photosystem II (PS II); 9 Cytochrome (Cyt); 16 Ferredoxin (Fd); 6 Oxygen-evolving enhancer protein (OEE); 7 Photosynthetic NDH subunit of lumenal location (PsbNDH-L); 5 F-box protein (FBX); 5 PsbP-like protein (PsbP); 5 Copper binding proteins (CuBP). Results revealed distinct patterns of photosynthesis-related DEGs among different treatment conditions in E. sibiricus seedlings. Most DEGs were upregulated in S0.50 vs. CK and S0.50 + ST vs. CK + ST, while downregulation occurred in CK + ST vs. CK and S0.50 + ST vs. S0.50, indicating that salt stress induced more downregulated genes, whereas Spd treatment triggered more upregulated genes (Fig. 6C, D). 33 genes that regulate photosynthesis-antenna proteins were identified. The expression patterns of these genes are consistent with those of the genes regulating photosynthesis (Fig. 6E).

Fig. 6.

The influence of Spd (0.50 mM) on the salt-stressed photosynthesis pathway. (A) photosynthesis (ko00195), (B) photosynthesis-antenna proteins (ko00196). The figures display a heatmap of DEGs associated with (C, D) photosynthesis and (E) photosynthesis-antenna protein for E. sibiricus S0.50 + ST vs. S0.50, CK + ST vs. CK, S0.50 vs. CK, and S0.50 + ST vs. CK + ST. Different colors in the heatmap represent log₂foldchange values, red indicates up-regulated genes (log₂FC > 0), blue indicates down-regulated genes (log₂FC < 0), white/light gray indicates no significant expression change, with P values < 0.01 indicating significant expression differences. PS Ⅰ: Photosystem I; PS II: Photosystem II; Cyt: Cytochrome; Fd: Ferredoxin; OEE: Oxygen-evolving enhancer protein; PsbNDH-L: Photosynthetic NDH subunit of lumenal location; FBX: F-box protein; PsbP: PsbP-like protein; CuBP: Copper binding proteins

Spd priming regulates generative expression in phytohormone pathways when exposed to salt stress

It was shown that the phytohormone signaling pathway produces a total of 295 DEGs, which contain genes related to brassinosteroid (BR), auxin, gibberellin (GA), cytokinine (CTK), ethylene (ETH), and abscisic acid (ABA) (Fig. 7 and Table S2). In the auxin pathway, 67 DEGs were identified across four distinct treatment comparison groups. Among these were six auxin-responsive GH3 (GH3) genes, two auxin influx carrier (AUX1) genes, five transport inhibitor response 1 (TIR1) genes, forty-one auxin response factor (ARF) genes, seven AUX1/IAA genes, and six small auxin-up RNA (SAUR) genes (Fig. 7A). Regarding the cytokinin pathway, a total of 67 DEGs were identified across the four distinct treatment groups. Among these were seven cytokinin response 1 (CRE1), five auxin-responsive (AHP), thirty-four type-B authentic response regulators (B-ARR), and eleven type-A authentic response regulators (A-ARR) (Fig. 7B). Regarding the brassinosteroid pathway, 19 DEGs were identified across four distinct treatment groups. Among these were nine BRI1-associated receptor kinase 1 (BAK1/BRI1), eight brassinosteroid-signaling kinase (BSK), and two Brassinazole-Resistant 1/2 (BZR1/2) genes (Fig. 7C). Regarding the gibberellin pathway, a total of 63 DEGs were identified across four distinct treatment groups. Among these were ten Gibberellin insensitive dwarf1 (GID1), forty-three DELLA protein (DELLA), and ten transcription factor (TF) genes (Fig. 7D). Regarding the abscisic acid pathway, 71 DEGs were identified across different treatment groups. Among these were eight PYRABACTIN RESISTANCE1/PYR1-LIKE (PYR/PYL), eighteen Protein Phosphatase 2 C (PP2C), twenty-one snf1-related protein kinase SRK2A (SnRK2), and fifteen ABA-responsive element binding factor (ABF) (Fig. 7E). In the ethylene expression pathway, seventeen DEGs were identified across four distinct treatment groups. Among these were five SIMK kinase (SIMKK), two Ethylene Insensitive 2 (EIN2), eight Ethylene Insensitive 3 (EIN3), and two Ethylene Response Factor 1/2 (ERF1/2) genes (Fig. 7F).

Fig. 7.

Effects of salt stress on hormone signal transduction induced by Spd (S0.50). DEGs representation within the plant hormone signal transduction pathway. (A) Auxin, (B) Cytokinine, (C) Brassinosteroid, (D) Gibberellin, (E) Abscisic acid, (F) Ethylene. The different colors in the figure represent log₂FoldChange, red indicates up-regulated genes (log₂FC > 0), blue indicates down-regulated genes (log₂FC < 0), white/light gray indicates no significant expression change, with P values < 0.01 indicating significant expression differences

Co-expression network analysis of salt-resistant candidate genes in E. sibiricus

To further identify salt-tolerant genes and elucidate the salt tolerance mechanism in E. sibiricus, we constructed co-expression networks for 12 E. sibiricus samples. After screening for low-expression genes, WGCNA retained 7,515 differentially expressed genes (DEGs). Using the Dynamic Tree Cut algorithm with a minimum module size of 30 genes and a maximum module distance of 0.25, gene modules were generated and subsequently merged based on high similarity (Fig. 8A). A gene clustering tree was constructed based on correlation coefficients among differentially expressed transcripts. The network was partitioned into 10 modules-lightgreen, blue, royalblue, skyblue, orange, purple, darkorange, greenyellow, darkred, and turquoise according to inter-gene clustering relationships. Analysis revealed strong negative correlations between SP, Pro, CAT, APX, and MDA, which showed strong negative correlations with the blue module, while SS, SP, Pro, SOD, CAT, and APX exhibited strong positive correlations with the greenyellow module (Fig. 8B).

Fig. 8.

WGCNA related to salt response. (A) WGCNA gene clustering tree; (B) Correlation heatmap between WGCNA modules and traits; (C) KEGG pathway enrichment analysis of DEGs in the greenyellow module; (D) KEGG pathway enrichment analysis of DEGs in the blue module; (E) Co-expression network analysis of the greenyellow module; (F) Co-expression network analysis of the blue module

KEGG enrichment analysis was performed on the greenyellow and blue modules. We found that pyrimidine metabolism, alpha-linolenic acid metabolism, circadian rhythm-plant, glycerophospholipid metabolism, and starch and sucrose metabolism were the top five significantly enriched pathways in the greenyellow module (Fig. 8C). Photosynthesis-antenna proteins, plant-pathogen interaction, starch and sucrose metabolism, MAPK signaling pathway-plant, and phenylpropanoid biosynthesis were the top five significantly enriched pathways in the blue module (Fig. 8D). Furthermore, we visualized the co-expression networks of the top 50 differentially expressed genes in the greenyellow and blue modules to identify hub genes within these modules. As shown in Fig. 8E, the key genes in the greenyellow module were EsiH03g0043070 (EsIRE4) and EsiH05g0016010 (EsTPKA), while the key gene in the blue module was EsiS07g0045150 (EsSRG1) (Fig. 8F).

Metabolome sequencing analysis and differential metabolite screening

Metabolomic analysis was conducted for E. sibiricus seedlings at various treatments using LC-MS/MS in both positive and negative ion modes to enhance the detection range and provide more thorough phytochemical information on E. sibiricus seedlings under salt stress. Principal component analysis and Pearson correlation coefficients indicate strong correlations among biological replicates within each treatment, enabling subsequent personalized functional analysis. To understand the metabolites involved in the salt stress response of E. sibiricus, the identified metabolites were classified and statistically analyzed based on chemical classification information. All samples of E. sibiricus seedlings contained a total of 1288 metabolites (Table S3). The classification of various metabolites is shown in Fig. 9A, lipids and lipid-like molecules accounted for 25.4% of total metabolites, organic acids and derivatives constituted 23.1% of total metabolites, organoheterocyclic compounds accounted for 15.2%, phenylpropanoids and polyketides represented 10.8%, and benzenoids comprised 10.2% of total metabolites.

Fig. 9.

Differences of 1288 different metabolites in all E. sibiricus seedlings samples. (A) Classification of the metabolites in E. sibiricus seedlings; (B) Venn displays the DEGs for each of the four treatments; (C-F) KEGG pathways of the differentially expressed metabolites in S0.50 + ST vs. S0.50, CK + ST vs. CK, S0.50 vs. CK, and S0.50 + ST vs. CK + ST

Screening for DAMs in E. sibiricus seedlings under different treatments was conducted using the criteria P < 0.05 and VIP > 1. DAMs varied among the four treatments. A total of 577 DAMs were detected in S0.50 + ST vs. S0.50, 470 DAMs in CK + ST vs. CK, 667 DAMs in S0.50 vs. CK, and 321 DAMs in S0.50 + ST vs. CK + ST. and 42 DAMs were detected across the four comparison groups (Fig. 9B). Among these 42 differentially expressed metabolites, 12 are lipids and lipid-like molecules, 8 are organic acids and derivatives, 7 are phenylpropanoids and polyketides, etc. Perform KEGG enrichment analysis on all filtered DAMs, and display the top 20 pathways with the smallest P-values, indicating the most significant enrichment. In S0.50 + ST vs. S0.50, the top five significantly enriched pathways are ABC transporters, aminoacyl-tRNA biosynthesis, biosynthesis of amino acids, alanine, aspartate and glutamate metabolism, galactose metabolism (Fig. 9C). In CK + ST vs. CK, the top five significantly enriched pathways are biosynthesis of amino acids, linoleic acid metabolism, biosynthesis of cofactors, glycine, serine and threonine metabolism, and 2-oxocarboxylic acid metabolism (Fig. 9D). In S0.50 vs. CK, the top five significantly enriched pathways are ABC transporters, biosynthesis of amino acids, biosynthesis of cofactors, 2-oxocarboxylic acid metabolism, and metabolic pathways (Fig. 9E). In S0.50 + ST vs. CK + ST, the top five significantly enriched pathways are ABC transporters, pyruvate metabolism, phenylalanine, tyrosine and tryptophan biosynthesis, and starch and sucrose metabolism, and galactose metabolism (Fig. 9F).

Analysis of DEGs and dams associated with phenylpropanoid biosynthesis and glyoxylate and dicarboxylate metabolism pathways

KEGG pathways enriched in both omics types were used to create bar graphs that demonstrate the number of DEGs and differentially abundant metabolites enriched in a given pathway. The phenylpropanoid biosynthesis (ko00940), glyoxylate and dicarboxylate metabolism (ko00630), ABC transporters (ko02010), and pyruvate metabolism (ko00620) are significantly enriched in all four groups (Fig. S1). The metabolome and transcriptome KEGG enrichment revealed that the phenylpropanoid biosynthesis and dicarboxylate metabolism pathways were significantly enriched after Spd priming during NaCl treatment. To predict the molecular mechanism of phenylpropanoid biosynthesis pathways in E. sibiricus seedlings during NaCl stress, we reconstructed the phenylpropanoid biosynthesis (Fig. 10A). The primary enzyme in the production of lignin is cinnamyl alcohol dehydrogenase (CAD). Among the four comparison groups, a total of 9 CAD genes were identified. In S0.50 + ST vs. S0.50, 7 genes showed upregulation and 2 genes showed downregulation; in CK + ST vs. CK, 3 genes were upregulated while 6 genes were downregulated or unchanged; in S0.50 vs. CK, 2 genes were upregulated and 7 genes were downregulated. In S0.50 + ST vs. CK + ST, 5 genes were up-regulated, while 4 genes were down-regulated or unchanged. Furthermore, the 4-Coumarate-CoA ligase (4CL), cytochrome P450 98 A (CYP98A), and phenylalanine ammonia-lyase (PAL) genes were almost entirely up-regulated in S0.50 + ST vs. CK + ST, while the phenylalanine/tyrosine ammonia-lyase (PTAL), caffeic acid O-methyltransferase (COMT), and Cinnamate 4-hydroxylase (CYP73A) genes were almost entirely down-regulated. Among the genes responsible for CSE regulation, EsiH02g0021450 showed a significant positive correlation with caffeic acid (Fig. S2 A). Another pathway showing significant enrichment is the glyoxylate and dicarboxylate metabolism pathway. In S0.50 + ST vs. CK + ST, acetyl-CoA acetyltransferase (ACAT), D-glycerate 3-kinase (GLYK), catalase E (KatE), ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit (rbcL), isocitrate lyase (ICL), glycine/serine hydroxymethyl transferase (glyA), and glycine decarboxylase (GLDC) are upregulation, while in the downregulation of CK + ST vs. CK, the expression levels of the DAMs involved in this pathway (isocitrate, glycolate, and 2-oxoglutarate) align with the expression trends of DEGs (Fig. 10B). In this pathway, EsiS06g0041680 and EsiS07g0031480, which are genes involved in the regulation of katE, exhibited a significant positive correlation with glycolate. Among genes that mediate rbcL regulation, EsiS02g0056340, EsiS02g0056330 and EsiH02g0051120 showed a highly significant positive correlation with glycolate, whereas EsiH05g0038250, EsiH05g0038270 and EsiH02g0009830 displayed a significant positive correlation with glycolate. Additionally, EsiO0018190, a gene that regulates glnA, EsiO0020190 and EsiH06g0012740, genes that regulate GLDC, as well as EsiH06g0034810, a gene that regulates glyA, were significantly and positively correlated with glycolate. In contrast, EsiS03g0037140 and EsiH03g0035140, two genes that are responsible for ACO regulation, showed a significant negative correlation with glycolate (Fig. S2 B).

Fig. 10.

Analysis of gene and metabolite associations. For the (A) phenylpropanoid biosynthesis and (B) glyoxylate and dicarboxylate metabolism pathway gene and metabolite association analysis was performed for S0.50 + ST vs. S0.50, CK + ST vs. CK, S0.50 vs. CK, and S0.50 + ST vs. CK + ST. The pathways were constructed based on the literature and the KEGG pathway database

qRT-PCR verification

To validate the transcriptomic sequencing results, nine DEGs were randomly selected in the CK, CK + ST, S0.50, and S0.50 + ST groups to ensure the accuracy of the RNA sequencing (Fig. 11). Among these 9 DEGs, RPPL1 participates in signal transduction mechanisms; CYP71D55 and CYP96T1 participate in secondary metabolite biosynthesis, transport, and catabolism; TPS7, EsiH07g0039510 and LCY1 have catalytic activity; LOX1.1 is involved in the amino acid transport and metabolism; EsiH07g0022170 has biological regulatory functions; and MADS51 is a transcription factor. RPPL1 and CYP96T1 were highly expressed in CK + ST compared to CK, S0.50, and S0.50 + ST. CYP71D55 and LOX1.1 were highly expressed in CK + ST and S0.50 + ST compared to CK, while EsiH07g0039510, LCY1, and MADS51 were expressed at low levels. TPS7 is most highly expressed in S0.50 + ST. Based on the results of qRT-PCR, the expression levels of these nine DEGs were consistent with transcriptomic sequencing data, indicating that the DEGs identified through transcriptomic sequencing possess accuracy and reproducibility.

Fig. 11.

Quantitative real-time PCR (qRT-PCR) is used to validate the expression level of particular genes. The FPKM values, which correlate to the Y-axis on the right, are displayed as red lines that indicate the relative expression levels of DEGs. The Y-axis on the left side of the bar chart was used to do the qRT-PCR validation

Discussion

Traditionally, superior seeds have been essential to crop production [55]. Additionally, the development of seedling morphogenesis is a prerequisite for field performance [56]. However, agricultural growth is being hampered by the rise of global industrialization and problems like global warming, pollution, and pressures like drought and salt [56]. Enhancing the quality of seeds is therefore essential since it can promote earlier and more consistent plant growth, which will increase crop yields, in addition to helping seeds germinate under unfavorable environmental conditions. The practice of seed priming has been used in a variety of crops and has the potential to greatly aid the early growth stage of crops in establishing morphogenesis and overcoming adverse conditions by controlling routes for signaling, expression of gene networks, and physiological responses [57, 58]. It was mentioned that investigations on seed priming to increase salt tolerance have been conducted [59–61]. The present findings showed that Spd-primed seedlings performed better under salt stress than seedlings without priming materials, suggesting that seed priming could improve E. sibiricus’s tolerance to salt.

Spd has been shown to effectively adjust resilience to several stressors [62]. Prior research investigations have demonstrated that the application of Spd improves the ability of yarrow [63] plants to tolerate high salt levels by adjusting the antioxidant enzyme activity and reducing the electrolyte leakage, hence reducing oxidative damage. Islam et al. [64] demonstrated that rice (Oryza sativa) roots and leaves had higher endogenous Spd concentrations during salt stress, which was an essential adaptive strategy to deal with salt stress.

The effects of Spd priming on the germination and seedling salt tolerance of E. sibiricus

Salt stress dramatically reduced seed germination, reduced seedling bioaccumulation, postponed water absorption, disrupted cellular structures, and extended the average germination time [65]. In addition to having a detrimental effect on sunflower and canola plant development and biomass, salt stress caused a notable decrease in the plants’ length of shoot, length of root, fresh weight of stem, and dry weight of stem-root [66]. In our study, under salt stress, parameters linked to plant growth were suppressed, demonstrating that the extreme salt stress had an inhibitory impact on plant growth. It was found that 0.50 mM Spd treatment had potential for promoting plant growth by reducing salt stress effects. The seed germination and seedling bioaccumulation of E. sibiricus under high salt stress were increased after priming at a concentration of 0.50 mM Spd. Exogenous regulators have been shown to be an effective way to increase the root growth of Sophora alopecuroides seeds and encourage seed germination during NaCl treatment [67]. Spd seed priming has been shown to mitigate the reduction in rice biomass during salt stress, according to Yang et al. [24]. These results are in agreement with the study’s conclusion.

The effects of Spd priming on photosynthesis

The present investigation demonstrated that exposure to salinity led to a decline in the levels of chlorophyll and carotenoid in the E. sibiricus. Nevertheless, the use of external 0.50 mM Spd could potentially mitigate the reduction in photosynthetic pigments caused by salinity. The content of carotenoids increased after priming by 0.50 mM Spd. The expression trends of DEGs involved in photosynthetic regulation in this study were consistent with photosynthetic physiological indicators. This may be attributed to the enhancement of photosynthesis induced by Spd treatment, which alleviated the degradation of chlorophyll and carotenoids while boosting their synthesis capacity. This finding aligns with the observation that Spd treatment increases photosynthetic pigment content in rice seedlings under salt stress [24].

Changes in plant hormone signaling pathways by Spd priming

Exogenous Spd may regulate phytohormone-related genes, thereby modulating content and promoting plant growth under salt stress. The phytohormone signaling pathway also plays a key role in the regulation of NaCl stress [68]. Particularly important regulators in plants under salt stress include phytohormones such as auxin [69], gibberellin acids [70], abscisic acid [71], cytokinins [68], ethylene [72], and salicylic acid [73]. In this study, it was found that in plant hormone signaling pathways, Spd primed alterations in the expression of several key genes. In the S0.50 + ST vs. CK + ST treatment, some genes involved in six plant hormone signal transduction pathways were upregulated, while the majority were downregulated. Genes involved in brassinosteroid signal transduction were all downregulated. In auxin signal transduction, Spd treatment upregulates the expression of SAUR genes. Shen et al. [74] subjected rice roots to salt stress and Spd treatment, and transcriptome sequencing revealed that Spd enhanced salt tolerance by regulating phytohormone signal transduction. Napieraj et al. [75] demonstrated that exogenous Spd reduced abscisic acid (ABA) content in drought-tolerant plants under osmotic stress. Additionally, among all phytohormones, the crosstalk between Spd and ethylene under salt stress appears to be the most common [75]. Zapata et al. [76] indicated that the contents of ethylene ETH and Spd increased in plant species with different salt sensitivities under saline-alkali stress. However, as the hormone levels within the leaves of E. sibiricus seedlings were not measured, this study on plant hormones has certain limitations.

Spd priming regulates antioxidant enzyme activity in response to salt stress

Plants are able to respond to stressors by enhancing the synthesis of antioxidant metabolites [56]. In the present study, exogenous Spd was applied as seed priming solution, which enhanced POD, SOD, APX, CAT activities and decreased the accumulation of ROS and MDA in order to avoid salt damage. Furthermore, S0.50 + ST had the lowest ROS and MDA concentration and the best antioxidant enzyme activity of any treatment. In addition, both POD and CAT related genes were up-regulated after pretreatment with Spd during NaCl stress. It has been suggested that Spd may enhance the activity of antioxidant enzymes in E. sibiricus seedlings by promoting the expression of genes encoding antioxidant enzymes, thus helping them to cope with salt stress. According to Qiang and Bo [77], using Spd may enhance POD and CAT activity, reducing salt toxicity in soybean (Glycine max) seedlings. Xing et al. [78] showed that Spd application improved germination and vigor of aged seeds by enhancing antioxidant enzyme activities and inhibiting reactive oxygen species. Cheng et al. [25] found that Spd pretreatment enhanced accumulation of metabolites associated with improved osmoregulation and antioxidant capacity during germination under salt stress conditions.

The effects of Spd priming on osmoregulatory substances

One of the primary challenges that plants encounter in saline soil is osmotic stress [79]. Soluble sugar, soluble protein, and proline are common osmoregulatory substances, which are crucial to the regulation of osmotic balance and the protection of cellular structure [80]. In this study, the content of soluble sugars, proline and soluble protein in E. sibiricus seedlings during NaCl treatment after priming at a concentration of 0.50 mM Spd was greater than that of the salt-stressed control group under salt stress. In agreement with our study, according to Alijani et al. [63], Spd synergistically enhanced proline accumulation under salinity stress, with a maximal 14.4% increase detected at 90 mM NaCl supplemented with 3 µM Spd. The results of the Yang et al. [81] study showed that 1 mM Spd promoted the buildup of beneficial compounds like proline, free amino acids, and sugars.

WGCNA-selected salt tolerance hub genes

In constructing this study module, a total of 10 modules were identified. Among them, the greenyellow and blue modules showed strong correlations with salt tolerance traits in E. sibiricus. Visualization of the co-expression network for these two modules revealed three hub genes. Serine/threonine protein kinase IRE4 is a key component of stress signal pathways; its homologous gene IRE1 mediates salt stress response by regulating unfolded protein response in Arabidopsis [82], and similar kinases in wheat enhance salt tolerance through ROS scavenging [83]. TPK maintains intracellular K⁺/Na⁺ balance by regulating vacuolar K⁺ transport; overexpression of PeTPK1 in poplar or OsTPKb in rice significantly improves salt tolerance by reducing Na⁺ accumulation [84, 85]. SRG1 homologous gene OsGASR1 in rice is induced by salt and ABA, and its overexpression reduces ROS content to enhance salt tolerance [86], while AtSR1 regulates salt-responsive gene expression to mediate salt tolerance [82]. These results suggest that Spd may upregulate these hub genes to coordinate signal transduction, ion homeostasis, and oxidative stress scavenging, offering new insights into how Spd priming mechanisms enhance salt tolerance in E. sibiricus.

Integrated analysis of the transcriptome and metabolomics

The combined transcriptomic and metabolomic analysis revealed that phenylpropanoid biosynthesis and glyoxylate and dicarboxylate metabolism are key pathways mediating Spd-enhanced salt tolerance in E. sibiricus seedlings. These two pathways form an interconnected regulatory network through metabolite crosstalk and synergistic anti-stress functions, where DEGs directly or indirectly modulate the accumulation of DAMs, collectively improving plant adaptability to salt stress.

Phenylpropanoid metabolism is a key secondary metabolic pathway for plant stress resistance, with its metabolites (lignin, flavonoids, and phenols) playing critical roles in cell structure reinforcement and oxidative stress alleviation [87]. Our results showed that in the Spd-primed salt stress group (S0.50 + ST vs. CK + ST), key enzyme genes including PAL, 4CL, and CAD were upregulated, while PTAL, COMT, and CYP73A were downregulated. PAL, as the rate-limiting enzyme initiating phenylpropanoid biosynthesis, catalyzes the deamination of phenylalanine to form cinnamic acid [88]. The upregulation of PAL genes in our study likely promoted the synthesis of downstream 4-coumaroyl-CoA and ultimately enhanced lignin accumulation. CAD, the terminal enzyme in lignin monomer synthesis, which would strengthen cell wall lignification to reduce Na⁺ influx and prevent cell turgor loss [89]. Consistent with our findings, overexpression of ZmPAL5 in maize significantly increased lignin content and improved drought tolerance [90], while CAD gene upregulation was associated with enhanced salt tolerance in rice by reinforcing cell wall integrity [91]. The downregulation of PTAL and COMT may reflect metabolic flux redirection towards the synthesis of stress-responsive phenols rather than other phenylpropanoid derivatives, a regulatory strategy also observed in rice under abiotic stress [92].

Glyoxylate and dicarboxylate metabolism serve as a critical link between carbon metabolism and stress adaptation, optimizing energy supply and metabolite homeostasis under adverse conditions [93]. In S0.50 + ST vs. CK + ST, DEGs encoding ACAT, ICL, GLYK, and KatE were upregulated, and their expression trends were consistent with the accumulation of DAMs such as isocitrate, glycerate, and 2-oxoglutarate. ICL, as a key rate-limiting enzyme in the glyoxylate cycle, catalyzes the cleavage of isocitrate into glyoxylate and succinate, which bypasses the tricarboxylic acid (TCA) cycle and conserves carbon sources for plant stress adaptation [94]. The upregulation of ICL genes in our study directly accelerated the catabolism of isocitrate, while the concurrent upregulation of GLYK genes promoted the phosphorylation of glycerate, providing carbon skeletons for amino acid biosynthesis and cellular osmotic adjustment. Notably, the upregulation of KatE directly scavenges hydrogen peroxide (H₂O₂) accumulated under salt stress, thereby mitigating oxidative damage to cellular components. This finding is consistent with previous research that the upregulation of key genes in the glyoxylate pathway enhances plant stress tolerance by maintaining intracellular metabolite balance and reducing ROS accumulation [95]. In addition, the upregulation of rbcL might improve the efficiency of photosynthetic carbon fixation, providing sufficient acetyl-CoA for downstream metabolic processes via the catalytic action of ACAT [96].

The phenylpropanoid biosynthesis pathway and the glyoxylate and dicarboxylate metabolism pathway form a synergistic regulatory network through two core regulatory links: carbon source supply and oxidative stress mitigation. First, the glyoxylate and dicarboxylate metabolism pathway provides key carbon precursors for the phenylpropanoid biosynthesis pathway. Acetyl-CoA, produced under the catalysis of ACAT in the glyoxylate pathway, acts as a direct carbon source for the synthesis of phenylpropanoid precursors. Under Spd priming, the upregulation of ACAT genes enhances the supply of acetyl-CoA, while the activation of PAL and 4CL genes in the phenylpropanoid pathway efficiently utilizes this precursor to synthesize stress-responsive metabolites [97]. This coordinated carbon flux distribution ensures the efficient allocation of limited carbon resources under salt stress, which is a conserved regulatory mechanism also reported in the salt stress response of maize [98]. Second, these two pathways synergistically regulate the homeostasis of oxidative stress in plants. KatE derived from the glyoxylate pathway directly scavenges intracellular H₂O₂, while metabolites of the phenylpropanoid pathway function as non-enzymatic antioxidants to quench excess ROS. Spd priming simultaneously activates the expression of KatE genes and promotes the phenylpropanoid biosynthesis pathway, thereby forming an “enzymatic + non-enzymatic” antioxidant defense network that comprehensively alleviates oxidative damage in plant cells. In our study, Spd priming induced the coordinated upregulation of key functional genes in both pathways, suggesting that Spd may integrate environmental stress signals to activate this synergistic regulatory network, thereby enhancing the salt tolerance of E. sibiricus seedlings.

Unfortunately, our transcriptomics findings did not show the expected strong correlation with metabolomics results. This discrepancy may stem from the fact that metabolic changes often occur more rapidly, while gene expression alterations may require a longer time to manifest, leading to temporal inconsistencies between metabolomics and transcriptomics data. Another possible reason lies in the differing regulatory mechanisms governing metabolites and gene expression, which may result in inconsistencies in their regulatory pathways. For instance, changes in metabolites may be influenced by multiple biochemical pathways, such as enzyme activity and metabolic pathways, whereas gene expression is regulated by factors like transcription factors. These differing regulatory mechanisms may result in less pronounced correlations between metabolites and gene expression.

Research limitations and future prospects

Several clear future research directions are proposed to extend this study on Spd-primed salt tolerance in E. sibiricus. First, endogenous hormone levels will be quantified to clarify their regulatory crosstalk with Spd in mediating salt adaptation. Second, longer-term greenhouse and field experiments are needed to verify the stability and practical applicability of the priming effect under natural conditions. Third, multiple NaCl concentration gradients will be tested to define the salt tolerance threshold of primed plants. Fourth, diverse Elymus genotypes will be incorporated to assess genotypic differences in response to Spd priming, laying a foundation for forage germplasm screening.

Conclusions

Under salt stress, Spd priming promotes seed germination and subsequent seedling growth. We propose a simplified model to describe the role of exogenous Spd in salt tolerance (Fig. 12). This occurs primarily through the following mechanisms: (1) Spd priming regulated photosynthesis, modulating the expression of photosynthesis-related genes. (2) The mechanism for scavenging ROS was improved, and the activity of antioxidant enzymes was increased. (3) Under salt stress, exogenous Spd increased proline, soluble protein, and total soluble sugar accumulation. (4) Spd seed priming promotes E. sibiricus to alleviate salt stress by modifying the phenylpropanoid biosynthesis and glyoxylate and dicarboxylate metabolism pathways. The results of our study have advanced knowledge of the complex regulatory processes controlling grass adaptation to saline-alkaline conditions, offering a basis of information for molecular breeding and aiding in the creation of perennial forage grasses that can withstand salt.

Fig. 12.

Schematic model of exogenous Spd regulating different mechanisms for salt tolerance in E. sibiricus seedlings. Up-pointing red arrows: enhance; down-pointing green arrows: diminish

Abbreviations

E. sibiricus

Elymus sibiricus

MDA

Malondialdehyde

ROS

Reactive oxygen species

Spd

Spermidine

ABA

Abscisic acid

GAs

Gibberellic acids

FW

Fresh weight

DW

Dry weight

SOD

Superoxide

POD

Peroxidase

CAT

Catalase

APX

Ascorbate peroxidase

KEGG

Kyoto encyclopedia of genes and genomes

DEGs

Differentially expressed genes

DAMs

Differential accumulated metabolites

LC-MS/MS

Liquid chromatography tandem mass spectrometry

PCA

Principal component analysis

qRT-PCR

Quantitative real-time PCR

BR

Brassinosteroid

CTK

Cytokinine

ETH

Ethylene

ABA

Abscisic acid

ARF

Auxin response factor

SAUR

Small auxin-up RNA

AUX1

Auxin influx carrier

TIR1

Transport inhibitor response 1

GH3

Auxin-responsive GH3

SAUR

Auxin-up RNA

CRE1

Cytokinin response 1

AHP

Auxin-responsive

B-ARR

Type-B authentic response regulators

A-ARR

Type-A authentic response regulators

BAK1/BRI1

BRI1-associated receptor kinase 1

BSK

Brassinosteroid-signaling kinase

BZR1/2

Brassinazole-resistant 1/2

GID1

Gibberellin insensitive dwarf1

TF

Transcription factor

PYR/PYL

PYRABACTIN RESISTANCE1/PYR1-LIKE

PP2C

Protein phosphatase 2 C

SnRK2

Snf1-related protein kinase SRK2A

ABF

ABA-responsive element binding factor

SIMKK

SIMK kinase

EIN2

Ethylene insensitive 2

EIN3

Ethylene insensitive 3

ERF1/2

Ethylene response factor 1/2

ACAT

Acetyl-CoA C-acetyltransferase

rbcL

Ribulose-bisphosphate carboxylase large chain

PAL

Phenylalanine ammonia-lyase

4CL

4-Coumarate-CoA ligase

CCR

Cinnamoyl-CoA reductase

CH4

Trans-cinnamate 4-monooxygenase

PTAL

Phenylalanine/tyrosine ammonia-lyase

CSE

Caffeoyl shikimate esterase

CAD

Cinnamyl alcohol dehydrogenase

F5H

Ferulate-5-hydroxylase

Authors’ contributions

YC. Z.: Writing-original draft, Software, Methodology, Formal analysis, Data curation, Conceptualization. JY. Z.: Validation, Resources, Methodology, Data curation. LB. T: Visualization, Validation, Software. HH. L.: Methodology, Data curation. F. Z.: Writing-review & editing, Formal analysis. WG. X.: Writing-review & editing, Funding acquisition, Conceptualization.

Funding

This work was supported by earmarked fund for CARS (CARS-34), Leading Scientist Project of Gansu Province (23ZDKA013), the Leading Scientist Project of Qinghai Province (2023-NK-147), and Gansu Provincial Science and Technology Major Projects (22ZD6NA007).

Data availability

The datasets generated during the current study are available in the China National Center for Bioinformation (accession number: PRJCA044790 (https://ngdc.cncb.ac.cn/gsub/submit/bioproject/list).

Declarations

Ethics approval and consent to participate

All methods were performed in accordance with the relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.