Phenotypic observation of cotton seedlings under salt stress and re-watering
Cotton plants at the trefoil stage were treated with 200 mM NaCl solution. The growth of cotton plants under salt stress was observed at different time points (Fig.1). Under NaCl stress, the leaves and stems of the ST and SS plants gradually softened, and the cotyledons wilted severely at 48 h. Compared with those of the ST plants, the leaves and stems of the SS plants changed significantly, their cotyledons withered and fell at 48 h, and their true leaf margins were slightly charred. After 48 h of re-watering, ST growth was relieved, and the true SS leaves wilted.
Fig. 1 Growth of cotton plants under salt stress conditions at different time points.
Physiological Responses of Cotton Seedlings to Salt Stress and re-watering
Salt stress can disrupt the ion balance within plant cells. The generated ion toxins, osmotic pressure, and reactive oxygen subsequently affect the growth and development of cotton. As shown in Fig. 2, the malondialdehyde concentrations of the ST and SS plants significantly increased, and the chlorophyll content, superoxide dismutase activity and peroxidase activity decreased under NaCl treatment for 48 h, indicating that salt stress affected the normal growth and development of the cotton plants. The malondialdehyde concentrations in the ST and SS treatments increased by 36.07% and 64.61%, respectively, compared with that in the CK treatment, and after 48 h of re-watering, the malondialdehyde concentration in the ST treatment did not change, while the malondialdehyde concentration in the SS treatment increased by 11.54% compared with that in the CK treatment. This indicated that salt stress significantly affected the lipid peroxidation level and that re-watering had a greater mitigating effect on ST than on SS. The relative chlorophyll contents of the ST and SS plants were reduced by 10.29% and 13.89%, respectively, compared with those of the CK plants under salt stress and by 5.23% and 10.36%, respectively, compared with those of the CK plants after re-watering. This indicates that salt stress significantly inhibited chlorophyll synthesis in cotton and that re-watering also significantly improved chlorophyll synthesis in both ST and SS. Compared with those in the CK treatment, the superoxide dismutase activity and peroxidase activity in the ST treatment decreased by 3.06% and 12.98%, respectively, and after re-watering, they decreased by 0.36% and 12.51%, respectively. The superoxide dismutase activity and peroxidase activity in SS were significantly reduced by 27.5% and 13.89%, respectively, compared with those in CK and were reduced by 5.23% and 10.36%, respectively, compared with those in CK after re-watering under salt stress. Compared with those in the CK treatment, the stress in the CK treatment significantly decreased by 27.31% and 22.46%, respectively, and after re-watering, the stress significantly decreased by 39.79% and 36.67%, respectively. This indicated that the ability of ST to clear reactive oxygen species under salt stress was greater than that of SS and did not improve re-watering.
Fig.2 Physiological response of cotton seedlings to salt stress and re-watering.
Note: The same letters indicate no significant differences between treatments (P<0.05, LSD method).
Microstructures of Cotton Leaves under Salt Stress and re-watering
Salt stress not only affects the physiological and metabolic processes of plants but also causes changes in plant morphology, anatomical structure, and other aspects. According to the microstructure of the cotton leaves under salt stress and re-watering (Fig. 3), the ST and SS seedling slices were deeply stained, the microstructure was tightly arranged, the bulliform cells were shaped like single-layer irregular quadrilaterals and closely arranged under normal conditions, the bulliform cells were ovoid, the arrangement of palisade and sponge tissue cells was relatively loose, and the leaves were softened. After re-watering, the bulliform cells had a long egg shape, and the leaves were relatively flat, gradually returning to normal growth.
Analysis of the leaf epidermal structure revealed that under normal conditions, there was no significant difference in the epidermal structure of the leaves between the ST and SS cotton plants (Table S3). Under NaCl stress, the leaf thickness, palisade tissue thickness, and sponge tissue thickness of the ST and SS plants all decreased, and they relatively recovered after re-watering. After NaCl treatment, compared with those of the control, the leaf thickness, palisade tissue thickness, and sponge tissue thickness of the ST seedlings decreased by 16.17%, 13.92%, and 18.46%, respectively, without any significant difference; those of the SS decreased by 32.43%, 30.13%, and 41.10%, respectively. After re-watering, the leaf thickness, palisade tissue thickness, spongy tissue thickness and upper epidermis thickness of the ST cotton seedlings increased by 0.93%, 0.09%, 3.41% and 2.28%, respectively, and the leaf thickness, palisade tissue thickness and spongy tissue thickness of the SS plants significantly decreased by 28.60%, 28.71% and 32.53%, respectively, compared to those of the control plants.
Fig.3 Microstructure of cotton leaves under salt stress and re-watering.
Note: Upper, upper epidermis; Pt, palisade issue; St, spongy issue; Le, lower epidermis.
Transcriptome Sequencing Analysis of Cotton Seedlings under Salt Stress and re-watering
To explore key genes related to salt tolerance in cotton, this study sampled two cotton materials whose salt tolerance significantly differed under 200 mM NaCl stress at 0 h, 48 h, and 48 h of re-watering. Each treatment had three biological replicates. The 18 collected samples (3 time points × 2 varieties × 3 biological replicates) were used to construct 18 RNA-seq libraries, and Illumina HiSeq 4000 sequencing and analysis were performed. A total of 127.58 Gb of clean data were obtained, with a Q30 percentage of no less than 91.70% for all the samples. Sequence alignment was performed between the total reads of each sample and the reference genome (TM-1) of upland cotton. According to the alignment results, the alignment efficiency between the reads of each sample and the reference genome was between 94.06% and 96.60%, and the GC content of each sample exceeded 43.47%. These results demonstrated that the RNA-seq data in this study were of high quality and could be further analyzed (Table S4).
Based on the comparison results, 11,125 genes were discovered, of which 6,974 were functionally annotated. The normalized FPKM value was used to measure the expression level of each gene, and the Pearson correlation coefficient (PCC) was used to detect the correlation between all samples. The overall correlation between the three biological replicates of the two varieties at the same growth stage was high. Moreover, sample clustering exhibited good correlation between biological replicates in the same environment (Fig. S1A). To further confirm the correlation between ST and SS at different stages under salt stress and re-watering, principal component analysis was performed on the genes expressed as described above (Fig. S1B). These results indicated that the two cotton varieties with significant differences in salt tolerance have apparently diverse expression patterns in different environments. SS and ST exhibited consistency in three biological replicates in the same environment, indirectly verifying the reliability of the transcriptome data.
To confirm the reliability of the RNA-Seq results in this study, ten genes were randomly selected for qRT‒PCR (Table S2). The results were consistent with the expression trend of the transcriptome expression profile, proving that the transcriptome data were reliable for subsequent analysis (Fig. 4). In SS, 43,233, 40,970, and 42,529 genes were expressed in sequence, of which 36,890 genes were coexpressed. There were 43,593, 41,028, and 43,943 genes expressed in STs, respectively, of which 37,256 genes were coexpressed (Fig. 5).
Fig.4 qRT‒PCR validation of the RNA-Seq results.
Note: (A) qRT‒PCR results for ten genes. The relevant significance level (p value) has been added above each histogram. Ns indicates no significant impact at the p<0.05 level, while * indicates a significant impact at p<0.05. The black strip represents ST; the gray bar represents SS. (B) RNA-seq heatmap of ten genes.
Fig.5 (A) Venn diagram of gene expression in ST leaves; (B) Venn diagram of gene expression in SS leaves.
Analysis of DEG Expression between the Two Cotton Materials
To further analyze the differential mechanism between the two cotton materials, a total of 10,589 genes were screened, of which 5,960 were upregulated, including 136 under normal conditions, 2,151 under salt stress, and 3,673 under re-watering; 4,629 genes were downregulated, including 228 under normal conditions, 2,271 under salt stress, and 2,130 under re-watering (Fig. 6). As the growth period increased, the number of DEGs in the two cotton materials increased, indicating that these two cotton materials have distinct response mechanisms to salt stress and re-watering.
The DEGs between the two materials were compared against the GO database (Fig. 7). In terms of biological processes, DEGs were mainly enriched in metabolic processes, cellular processes, and single-organism processes. In terms of cell components, DEGs were enriched mainly in cell and cellular components, while in terms of molecular functions, DEGs were enriched mainly in binding and catalytic activities. Under normal conditions, a large number of genes were downregulated, while many genes were upregulated after re-watering. This fully demonstrates the adaptability of cotton to re-watering. Subsequently, KEGG enrichment analysis was performed on the DEGs. As shown in Fig. 8, DEGs of STs were significantly enriched in photosynthesis-antenna protein (ko00196), carbon fixation (ko00710) in photosynthetic organisms, and vitamin B6 metabolism (ko00750).
Fig. 6 DEGs between materials subjected to the same treatment.
(A) Histogram of DEGs between materials with the same treatment; (B) Venn diagram of DEGs between materials with the same treatment.
Fig.7 GO analysis of DEGs between materials subjected to the same treatment.
Fig.8KEGG analysis of DEGs between materials subjected to the same treatment.
(A) KEGG analysis of DEGs under normal conditions (ST-CK vs. SS-CK); (B) KEGG analysis of DEGs under salt stress (ST-NaCl vs. SS-NaCl); (C) KEGG analysis of DEGs under re-watering (ST-RW vs. SS-RW).
Expression analysis of DEGs under salt stress and re-watering
A fold change ≥ 2 and FDR<0.01 were adopted as screening criteria to obtain DEGs under salt stress and re-watering for further analysis of the differential mechanisms (Fig. 9). A total of 32,390 genes were screened. Under salt stress, 15,780 genes were screened in the ST, of which 7,996 DEGs were upregulated, and 7,784 DEGs were downregulated. Under re-watering, 2,454 genes were screened, including 659 upregulated genes and 1,795 downregulated genes. For SS under salt stress, 14,156 genes were screened. Among them, 6,925 DEGs were upregulated, and 7,231 DEGs were downregulated. Under re-watering, 9,390 genes were screened in SS, of which 4,666 DEGs were upregulated and 4,724 DEGs were downregulated. Under salt stress, the number of DEGs screened in ST was greater than that in SS. However, fewer DEGs were produced after re-watering in the ST treatment than in the SS treatment, which may be related to the greater recovery of the STs after re-watering.
The DEGs of the two materials under salt stress and re-watering were compared via the GO database (Fig. 10). In terms of biological processes, DEGs were mainly enriched in metabolic processes, cellular processes, and single organism processes. For cellular components, DEGs were mainly involved in cells, cellular components, and membranes. Regarding molecular functions, DEGs were mainly enriched in binding and catalytic activities.
DEGs annotated by KEGG were classified into multiple categories (Fig. 11). The response of cotton to salt stress and its adaptation to re-watering may be related to interactions such as photosynthesis (ko 00195), photosynthesis-antenna protein (ko 00196), plant hormone signal transduction (ko 04075), starch and sucrose metabolism (ko 00500), and porphyrin and chlorophyll metabolism (ko 00860). Additionally, STs respond to salt stress through pathways such as circadian rhythm-plant (ko 04712), carbon fixation (ko 00710) and carbon metabolism (ko 01200) pathways in photosynthetic organisms. The adaptability of STs to re-watering was regulated through pathways such as glyceride metabolism (ko 00561), the reciprocal transformation of pentose and glucuronate (ko 00040), circadian rhythm-plant (ko 04712), and phenylpropane biosynthesis (ko 00940). SSs resist salt stress through pathways such as fatty acid degradation (ko 00071); degradation of valine, leucine, and isoleucine (ko 00280); biosynthesis of brassinosteroid (ko 00905); and sphingolipid metabolism (ko 00600). The damage caused by salt stress was alleviated through plant‒pathogen interactions (ko 04626), the MAPK signaling pathway (ko 04016), and α-linolenic acid metabolism (ko 00592).
Fig. 9 DEGs between treatments.
(A) Histogram of DEGs between treatments; (B) Wayne plot of DEGs between treatments.
Fig. 10 GO analysis of DEGs between treatments.
Fig. 11 KEGG analysis of DEGs between treatments.
(A) KEGG analysis of DEGs in STs under salt stress (ST-CK vs. ST NaCl); (B) KEGG analysis of DEGs in STs under re-watering (ST-CK vs. ST-RW); (C) KEGG analysis of DEGs in SS under salt stress (SS-CK vs. SS-NaCl); (D) KEGG analysis of DEGs in SS under re-watering (SS-CK vs. SS-RW)
Effects of salt stress on genes related to photosynthesis and carbon metabolism in cotton
KEGG pathway analysis revealed enrichment of "carbon fixation in photosynthetic organisms" and "photosynthesis" pathways in cotton. The expression levels of genes encoding key enzymes involved in carbon fixation in STs and SS were suppressed under salt stress (Fig 12). Similarly, salt stress inhibited photosynthesis and reduced carbon assimilation in the ST and SS, which ultimately led to structural and functional damage to the cotton leaves. Moreover, the expression of most genes involved in photosynthesis and carbon metabolism was more affected by salt stress than by genotype, and the genes related to photosynthesis and carbon metabolism in STs and SS were suppressed under salt stress. After re-watering, more similar gene expression patterns were detected under salt stress and re-watering treatments. In particular, the expression of genes related to PSII, PSⅠ, FNR, Fd, LHCⅠ and LHCⅡ in the ST decreased in response to salt stress, and after re-watering, their expression levels increased to normal levels. After re-watering, the expression levels of related genes were still downregulated after SS re-watering.
Fig. 12 Expression of genes related to photosynthesis and carbon metabolism in cotton under salt stress and re-watering.
ST improved growth hormone, gibberellin and cytokinin signaling under salt stress and re-watering to promote cotton plant growth
Growth hormones, gibberellins and cytokinins control plant growth by promoting cell elongation or division, thereby increasing salt tolerance in plants under salt stress[51]. Furthermore, the growth hormone endocytic vector (AUX1) and gibberellin receptor (GID1) are required to cope with cell damage under salt stress, and the expression of cytokinin receptors and several Arabidopsis A-type response regulators is affected by salt treatment[39, 52]. In this study, under salt stress and re-watering treatment, the expression levels of 8 genes related to AUX1 in ST47 were greater than those in SS. There were 6 genes related to GID1 whose expression levels were elevated by salt stress, and these genes showed better recovery in STs with lower expression levels after re-watering, but SS still showed higher expression levels (Fig. 13). Four genes related to A-type response regulators in ST and SS showed low expression levels under normal conditions, and 2 genes were elevated under salt stress. After re-watering, the expression levels of these 2 genes in the STs decreased and recovered. However, these 2 genes of SS were still highly expressed. Our results suggest that ST coordinately regulates growth hormone, gibberellin and cytokinin signaling and promotes improved growth and recovery under salt stress.
Fig. 13Effects of salt stress and re-watering on the expression of growth hormone-, cytokinin- and gibberellin-related genes in cotton.
Response of plant circadian genes to salt stress and re-watering in cotton
Plant circadian rhythms and their response mechanisms to various abiotic stresses have been a hot topic in plant physiology and ecology[53]. There are many different models of biological clocks. A simple model of a plant biological clock consists of three circuits: CCA1, LHY and TOC1 (PRR1), which form the core circuit; PRR5/7/9, which forms the daytime circuit; and ELF3, ELF4, GI and LUX, which form the nighttime circuit[54]. In this study, KEGG pathway analysis revealed that ST cotton was enriched in the "plant circadian rhythm" pathway under salt stress and re-watering, and two TOC1 genes in ST and SS exhibited low expression levels under normal conditions, while the expression of ST-related genes increased under salt stress and decreased after re-watering, while that of SS decreased under both salt stress and re-watering (Fig 14). SS had lower gene expression levels in both the salt stress and re-watering treatments. Similarly, the expression of one GI and three PRR7 genes in both materials increased under salt stress and decreased after re-watering. There were 2 PRR9 genes in the ST, whose expression increased under salt stress and decreased after re-watering, and in the SS, whose expression decreased under both salt stress and re-watering.
MAPK signaling in response to salt stress
The MAPK cascade pathway plays an important role in environmental stress signaling via the MAPKKK-MAPKK-MAPK-mediated triple phosphorylation cascade to induce tolerance responses[55]. In this study, the majority of MAPK-related genes in ST and SS were altered by salt stress, and most of the genes in ST were restored to normal levels after re-watering, while most of the genes in SS were still activated to relatively high or low levels (Fig 15). The expression of the MEKK1, MEKK2, and MEK4/6 genes in both STs and SSs was induced by salt stress, and their expression was elevated. Interestingly, after re-watering, the expression of all genes related to ST decreased, and the expression of some genes related to SS decreased, while others maintained high expression.
Fig. 14Effects of salt stress and re-watering on the expression of genes related to biological clocks in cotton.
Fig. 15 Effects of salt stress and re-watering on MAPK signal transduction-related genes in cotton.
WGCNA of DEGs under Salt Stress and re-watering
To understand gene expression and re-watering and identify candidate genes for salt tolerance, WGCNA was used to construct a coexpression network for DEGs under salt stress and re-watering (Fig. 16). The expression matrix was imported. The dist function and hclust clustering were used to calculate the sample algorithm. The graph shows no outliers among the 18 samples. There were significant differences between ST and SS under the different treatments. ST and SS can be clearly distinguished under re-watering, which implies that the expression matrix has good specificity and can be used to effectively distinguish between ST and SS under different treatments. This study selected the optimal soft threshold of 26 to construct an infinitely close scale-free network of gene coexpression. Genes were divided into six modules. The gray module had the greatest correlation with SS re-watering (R=0.86 and p<0.0001).
By using the Cytohubba plugin in Cytoscape, the top five genes with the highest correlation were selected as the hub genes for the modules (Fig. 17). Five key hubs, namely, GH_A01G1528, GH_A08G2688, GH_D08G2683, GH_D01G1620, and GH_A10G0617, were screened from among the hub genes in the MEgray module (Table 1).
Fig. 16WGCNA of DEGs under salt stress and re-watering conditions.
Fig. 17 (A) Connection results related to Cytoscape imported by the gray module. (B) Filtering results for the Cytohubba plug-in.
Table 1. IDs and gene annotations of five key genes obtained from WGCNA.
Hub gene ID
|
Homologous gene ID in A.thaliana
|
Functional Annotation
|
Gene Name
|
GH_A01G1528
|
AT5G59310
|
Nonspecific lipid-transfer protein 4
|
LTP4
|
GH_A08G2688
|
AT5G57800
|
Protein ECERIFERUM 3
|
CER3
|
GH_D01G1620
|
AT4G38660
|
Nonspecific lipid-transfer protein 1
|
LTP1
|
GH_A10G0617
|
AT1g17710
|
Inorganic pyrophosphatase 2
|
PEPC1
|
GH_D08G2683
|
AT5G57800
|
Protein ECERIFERUM 3
|
CER3
|