Seeds of cucumber cultivar “Chunxiaqiuwang” were purchased from the Ningyang County Lu-Ming Seed Co., Ltd. (Shandong province). Seed germination and growth conditions followed those we previously described (Jiang et al., 2019). The seedlings of cucumber were maintained in a controlled growth chamber with a light/dark regime of 14/10 h, relative humidity of 70%, temperature of 25℃ and a photosynthetically active radiation (PAR) of 800 μmol.m⁻².s⁻¹. When the first true leaf emerged, uniform and healthy seedlings were selected and divided into three groups for the following treatments: (i) Hoagland’s nutrient solution (as control, C); (ii) Hoagland’s nutrient solution +200 mM NaCl (as salt treatment, S); (iii) Hoagland’s nutrient solution + 200 mM NaCl +15 µM NaHS (as H₂S treatment, H₂S). In our previously studies, 5, 10, 15, and 20 µM NaHS was tested for alleviation of NaCl-induced damage, and morphological and physiological assessments showed that exogenous supplementation with 15 µM NaHS was the most effective in boosting cucumber tolerance to NaCl stress (Jiang et al., 2019). Thus, the control (C), 200 mM NaCl (S), and 15 µM NaHS (H₂S) treatments were chosen to study the transcriptomic and proteomic responses to H₂S alleviation of salt-stress in cucumber leaves. NaHS was purchased from Sigma (St. Louis, MO, United States) and used as H₂S donor. NaHS is commonly used as an H2S donor since it dissociates to water produce HS⁻ and Na⁺, and then, combination of HS⁻ with H⁺ produces H₂S (Lin et al., 2012). NaHS is responsible for induction of stress tolerance, but not Na₂S, Na₂SO₄, NaHSO₄, Na₂SO₃, NaHSO₃, or CH₃COONa (Deng et al., 2016). Three independent biological replicates were conducted per treatment (8~10 plants per treatment) under the same experimental conditions. A 10-ml fresh treatment solution was irrigated every other day at 9:00 AM during the whole culture process, and the remaining treatment solution, which was not fully absorbed by plants, was removed. After 7 days of treatment, the leaves from at least eight seedlings of each treatment sample were harvested as three replicate, immediately frozen in liquid nitrogen and then stored at −80°C until RNA and protein extraction.

Total RNA was isolated from each treatment using the RNAplant Plus Reagent DP437 (Tiangen, Beijing, China) according to the manufacturer’s instructions. The concentration and quality of RNA were determined using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Then, the RNA was sent to the Shanghai Personal Biotechnology Cp. Ltd. for RNA-Seq analysis (Personalbio, Shanghai, China).

A total of 20-mg RNA was equally isolated from the each sample for cDNA library construction. Beads with oligo (dT) were used to enrich poly(A) mRNA, and then, fragmentation buffer was added for interrupting mRNA to short fragment (200~300 nt) which were used as templates to synthesize the first-strand cDNA using random hexamer primer, and the second-strand cDNA was synthesized by adding buffer, dNTPs, RNaseH, and DNA polymerase I, respectively. The double-strand cDNA was purified with QiaQuick PCR Purification Kit (Qiagen, Germany) and washed with EB buffer for end repair and single nucleotide. Finally, the cDNA fragments were connected with sequencing adaptors and gel-electrophoresis was used to select cDNA fragments of 200 bp in size as the templates for amplification with PCR. The library products were ready for sequencing analysis via Illumina HiSeq™ 2000 (Illumina, San Diego, CA, USA).

The RNA-Seq raw sequencing data was filtered using the Cutadapt program (http://cutadapt.readthedocs.io/en/stable/). The filtered clean reads were further mapped to the reference genome (Cucumis_sativus.ASM407v2.dna.toplevel.fa, v 2.39), downloaded from the Ensembl database (http://www.ensembl.org/) using Tophat2 (http://tophat.cbcb.umd.edu/), with up to two mismatches allowed. The transcripts were screened for significant changes in abundance between C, S, and H₂S samples using DESeq (http://www.bioconductor.org/packages/release/bioc/html/DESeq.html) with a condition of |log₂ ^fold-change|> 1.0 and p-value < 0.05. Principal Coordinate Analysis (PCA) was performed on distance matrices, and coordinates were used to draw 2D graphical outputs (Lozupone et al., 2007). All DEGs were searched against the Kyoto Encyclopedia of Genes and Genomes database (KEGG, http://www.genome.jp/kegg/pathway.html) to identify the main metabolic pathways and signal transduction pathways of DEGs using Blastall software (Kanehisa and Goto, 2000).

Protein extraction was performed using a modified version of the trichloroacetic acid (TCA)-acetone precipitation method described by Wu et al. (2011) (Wu et al., 2011). Protein concentrations were quantified according to the Bradford method (Bradford, 1976). Isoelectric focusing (IEF) was carried out on 18-cm immobilized pH gradient (IPG) linear gradient strips at pH 4–7. IPG strips in a rehydration tray were loaded with 350 μl of protein sample containing 500-μg protein for 12–16 h at room temperature. The IPG strips were then run on an Ettan IPGphor 3 (GE Healthcare, USA) using the following voltage settings: 50 V for 1 h, 100 V for 2 h, 500 V for 1 h, 1,000 V for 2 h, and 2,000 V for 1 h, followed by a gradient of 5,000 V for 1 h and 10,000 V for 2 h, with a subsequent 10,000 V rapid focusing to reach 65,000 V.h. The temperature was maintained at 20°C, with a maximum current of 50 μA per strip. After running the first dimension, IEF strips were incubated for 15 min in 10-ml equilibration buffer containing 6 M urea, 30% (v/v) glycerol, 2% (w/v) SDS, 1.5-ml Tris-HCl (1.5 M, pH 8.8), 0.002% (w/v) bromophenol blue, and 1% (w/v) DTT and then equilibrated with fresh equilibration buffer containing 2.5% iodoacetamide (IAM) for another 15 min.

The equilibrated strips were placed directly onto 12.5% polyacrylamide-SDS slab gels and sealed with 0.5% agarose solution containing bromophenol blue dye. The second dimensional SDS–polyacrylamide gel electrophoresis (SDS-PAGE) was conducted using the Ettan DALTSix electrophoresis system (GE Healthcare, USA). Electrophoresis was carried out at 15 mA per gel for 30 min, and then, 38 mA per gel until the bromophenol blue dye front was approximately 1 cm from the bottom of the gel. After 2-DE, protein spots were visualized using Coomassie Brilliant Blue (CBB) R-250.

The 2-DE gels were scanned at 300 dpi using an Image Scanner III (GE Healthcare). Spot detection, quantification, and matching were conducted using ImageMaster™ 2D Platinum version 7.0 analysis software (GE Healthcare). The optimized parameters were: smooth 2.0, min area 5.0, and saliency 3.0. The abundance of each spot was estimated by the percentage volume (%VOL), which was normalized as a percentage of the total volume of all the spots present in the gel to correct for variability caused by loading, gel staining, and destaining. Only spots with significant and biologically reproducible changes (abundance variation ≥1.3-fold or ≤0.67-fold, one way ANOVA, p ≤ 0.05) were considered to be differentially expressed proteins (DEPs).

The selected DEPs were manually excised from the 2-DE gels, and gel pieces were digested overnight in trypsin at 37°C. The hydrolysates were then transferred to new tubes and disrupted by sonication for 15 min in a buffer containing 100-μl 60% acetonitrile (ACN)/0.1% trifluoroacetic acid (TFA) and desalinated in a ziptip (Millipore). Peptides in the supernatant were collected by centrifugation. The peptide pellets were dissolved with 1-μl matrix solution. After drying completely at room temperature, 0.6-μl over-saturated CHCA (50% ACN/0.1% TFA) was added. MS and MS/MS data for protein identification were obtained by using a MALDI-TOF-TOF instrument with an AB SCIEX MALDI TOF-TOF™ 5800 Analyzer (AB SCIEX, Foster City, CA, USA) equipped with a neodymium (Nd): yttrium-aluminum-garnet (YAG) laser (laser wavelength: 355 nm). Instrument parameters were set using the 4000 Series Explorer software (Applied Biosystems). The MS spectra were recorded in reflector mode in a mass range from 800 to 4000 with a focus mass of 2000. The TOF/TOF calibration mixtures (AB SCIEX) were used to calibrate the spectrum to a mass tolerance within 10 ppm. The MS spectra were processed using TOF-TOF Series Explorer software (V4.0, AB SCIEX). At least 1,000 laser shots were typically accumulated with a laser pulse rate of 400 Hz in the MS mode, whereas in the MS/MS mode spectra, up to 2,000 laser shots were acquired and averaged with a pulse rate of 1,000 Hz. For MS calibration, autolysis peaks of trypsin ([M⁺H]⁺ 842.5100 and 2, 211.1046) were used as internal calibrates, and the most intense ion signals (up to 10) were selected as precursors for MS/MS acquisition, excluding the trypsin autolysis peaks and the matrix ion signals. Peptide mass finger printing (PMF) and MS/MS queries were performed by using the MASCOT search engine 2.2 (Matrix Science, Ltd. UK) embedded into GPS-Explorer Software 3.6 (Applied Biosystems) on the database of uniprot_Eimeria acervulina with the following parameter settings: 100 ppm mass accuracy, trypsin cleavage one missed cleavage allowed, carbamidomethylation set as fixed modification, oxidation of methionine was allowed as variable modification, MS/MS fragment tolerance was set to 0.4 Da. All proteins with a confidence interval (CI) > 95% were considered to be identified successfully.

The DEPs were classified according to the biological processes in which they are involved based on UniProtKB (http://www.uniprot.org/) and NCBI (https://www.ncbi.nlm.nih.gov/) databases. Hierarchical clustering of protein expression patterns was performed using MultiExperiment Viewer (MeV) software version 4.9. The relative ratios of DEPs were subjected to log² transformation, the Euclidean distance similarity metric was used to define the similarity, and hierarchical clusters were assembled using the complete linkage clustering method. The clustering result was visualized through MeV software.

Data for RNA data and protein expression levels was analyzed using SPSS software (version 21.0, SPSS Inc., USA). Differences in proteins expression levels were analyzed by Duncan’s multiple range test at P ≤ 0.05 level of significance.

Raw transcriptome data from this manuscript can be found in the NCBI BioProject database under accession numbers: SRP220380.

Total RNA from three replicates each of the control (C), 200 mM NaCl (S), and 15 µM NaHS (H₂S) treatment samples was extracted for RNA sequencing by Illumina Hiseq 2500. Nine libraries was constructed and searched for DEGs associated with H₂S-regulated salt-stress tolerance. After filtering raw reads, a total of 382,202, 188 high-quality reads were obtained from the nine samples. The clean reads of the sequencing results for each cDNA library accounted for >97.42% of the raw reads (Q₃₀ > 86.22%). The standard deviation of clean reads among the three replicates for each treatment was ~0.2% ( Table S1 ). Among all nine samples, 85.75~90.78% of the reads could be mapped to the genome of cucumber. Of these mapped reads, 84.79~89.90% matched uniquely.

The similarity between the expression of genes between the cucumber leaf samples following C, S, and H₂S treatment was determined by PCA. As shown in Figure 1 , the genes identified in the C, S, and H₂S samples were clearly separated from each other, and the difference was statistically significant. The expressed genes in the control and H₂S samples clustered more closely to each other than those from the salt treatment, indicating that salt stress remarkably changes gene expression and that H₂S may alleviate some of these changes at the transcript level. The principal coordinates PC1 and PC2 represented 71% and 25% of the variance, respectively, and the contribution of the cumulative variance of the two principal coordinates (PC1 and PC2) accounted for 96% ( Figure 1 ). The transcripts were compared between C and S, between H₂S and S, and between C and H₂S by DESeq. Transcripts with a P < 0.05, and |log₂ ^{fold change}| > 1 were defined as DEGs. All identified DEGs are shown in Table S2 . A total of 11,761 (5,417 up-regulated and 6,344 down-regulated) DEGs were detected from the C, S, and H₂S samples ( Figure 1 ). Among these DEGs, 4,942 DEGs (2,629 up-regulated and 2,313 down-regulated) in S vs. C, 2,351 DEGs (741 up-regulated and 1,610 down-regulated) in H₂S vs. S, and 4,468 DEGs (2, 047 up-regulated and 2, 421 down-regulated) in H₂S vs. C were detected ( Figure 1 ). A Venn diagram of the genes show how many genes are specific to one treatment and how many genes transcriptionally respond to multiple treatments ( Figure 1 ). Of the 11,761 DEGs, 1,124 DEGs were unique to untreated plants, 366 genes were unique to salt treatment, and 734 were unique to H₂S and S treatment ( Figure 1 ). Additionally, the C, S, and H₂S treatment samples shared 519 DEGs as shown in Figure 1 .

After gene annotation, GO enrichment analysis of differentially expressed genes was carried out. As shown in Figure 2 , photosystem took the first place in cellular components in the S vs. C and H₂S vs. S, and intrinsic component of membrane took the first place in cellular components in the H₂S vs. C. Catalytic activity was the largest category in molecular function in the S vs. C and H₂S vs. C; however, tetrapyrrole binding was the largest category in molecular function in the H₂S vs. S ( Figure 2 ). Oxidation-reduction process occupied the largest proportion of biological processes in the S vs. C and H₂S vs. S, and photosynthesis, light harvesting occupied the largest proportion of biological processes in the H₂S vs. C ( Figure 2 ).

The DEGs were compared to known metabolic and signaling pathways to identify those that were more responsive to salt stress and H₂S amelioration. In the transcriptome, there were 118 KEGG pathways differentially regulated in the S vs. C samples, 111 in the H₂S vs. C, and 88 in the H₂S vs. S ( Table S3 ). The 20 most enriched pathways, based on the number of DEGs (including up- and down-regulated genes), are shown in Figure 2 . The signal transduction, carbohydrate metabolism, energy metabolism, amino acid metabolism, environmental adaptation, folding, sorting and degradation, and secondary metabolites biosynthesis pathways are involved in the three pairwise comparisons (S vs. C, H₂S vs. C, and H₂S vs. S), and the detailed metabolic pathway are as follows: plant hormone signal transduction, MAPK signaling pathway, glycolysis/gluconeogenesis, amino sugar and nucleotide sugar metabolism, pentose and glucuronate interconversions, photosynthesis, tyrosine metabolism, glutathione metabolism, plant-pathogen interaction protein processing in endoplasmic reticulum, and phenylpropanoid biosynthesis ( Figure 3 ). Salt stress significantly resulted in the changes of basal metabolism such as carbohydrate metabolism (starch and sucrose metabolism and glyoxylate and dicarboxylate metabolism), energy metabolism (carbon fixation in photosynthetic organisms), amino acid metabolism (arginine biosynthesis, cysteine and methionine metabolism, alanine, aspartate and glutamate metabolism and phenylalanine metabolism), and transport and catabolism (peroxisome) in S vs. C ( Figure 3 ). However, the treatment of H₂S regulated the lipid metabolism (alpha-Linolenic acid metabolism and glycerophospholipid metabolism), and energy metabolism (photosynthesis-antenna proteins), as well as amino acid metabolism in leaves of cucumber under salt stress (H₂S vs. S) ( Figure 3 ). In addition, the treatment of H₂S alone regulated the genes involved in the amino acid metabolism, carbohydrate metabolism, and transport and catabolism (H₂S vs. C) ( Figure 3 ).

Proteomics analysis using 2-DE for protein separation was performed to examine changes in protein levels in cucumber leaves and representative gel maps are shown in Figure S1 . The total numbers of protein spots reproducibly detected using ImageMaster™ 2D Platinum software were 875 ± 93, 986 ± 11, and 776 ± 47 in the S vs. C, H₂S vs. S, and H₂S vs. C, respectively. Quantitative image analyses revealed a total of 69 protein spots showing significant changes in their abundance in pairwise comparisons (P < 0.05). Among the 69 DEPs observed in the gels, there were 29 DEPs in S vs. C, 25 DEPs in H₂S vs. S, and 15 DEPs in H₂S vs. C. Finally, 61 DEPs were chosen for further identification using MALDI-TOF/TOF MS ( Figure S1A ) after removing eight very low abundance proteins. Of these 61 DEPs, 42 proteins were up-regulated and 19 proteins were down-regulated, in pairwise comparisons ( Figure S1B ).

Altogether, 49 DEPs were successfully identified by MALDI-TOF/TOF-MS. To investigate the functional information of these proteins, the Viridiplantae (taxonomy, Green plant) data in NCBInr and Uniprot Swissprot were searched using the Mascot search engine. The basic identification information for each of these DEPs was listed in Table S4 and summarized in Table 1 . It was interesting to note that a chloroplast stem-loop binding protein of 41 kDa (spots 14 and 17) and an acetylornithine deacetylase-like protein (spots 31, 32, and 51) were identified in more than one spot on the same gel. This multiple observation of the same protein could result from the presence of different protein isoforms, post-translational modification, or degradation. Forty-seven proteins out of the 49 identified MS-sequenced proteins (except spots 1 and 4) were matched with proteins in Cucumis sativus L.

Salt stress resulted in 22 DEPs, 17 of which were up-regulated and 5 of which were down-regulated in the S vs. C sample ( Figure S2A ). However, H₂S treatment of cucumber induced only 8 DEPs, 7 of which were up-regulated and only one of which was down-regulated in the H₂S vs. C sample, and 19 proteins were significantly altered (P < 0.05) in the H₂S vs. S sample, with 9 proteins showing up-regulation and 10 proteins showing down-regulation ( Figure S2A ). Functional categorization of the DEPs in S vs. C showed that 31.82% of the DEPs participated in photosynthesis and carbohydrate and energy metabolism, followed by protein folding, assembly and degradation (22.73%), amino acid biosynthesis (9.09%), and secondary metabolite biosynthesis (4.08%) ( Figure S2B ). In the H₂S vs. S, 21.05% of the DEPs belonged to stress and defense-related protein and another 21.05% to photosynthesis, followed by protein folding, assembly and degradation (15.79%), carbohydrate and energy metabolism (15.79%), and amino acid biosynthesis (10.53%). In the H₂S vs. C, 25% of the DEPs participated in amino acid biosynthesis and protein folding, assembly and degradation, followed by photosynthesis, secondary metabolite biosynthesis, and carbohydrate and energy metabolism ( Figure S2B ). Based on their putative physiological and biochemical functions, the majority of these DEPs belonged to one of seven categories, namely, photosynthesis (12, 24.49%), carbohydrate and energy metabolism (11, 22.45%), protein folding, assembly and degradation (10, 20.41%), amino acid biosynthesis (6, 12.24%), secondary metabolite biosynthesis (2, 4.08%), stress and defense (4, 8.16%), and other and unknown proteins (4, 8.16%) in all treatment samples ( Figure S2B ).

There were 49 KEGG metabolic pathways that were enriched for DEPs responding to S and H₂S treatments ( Table S5-a ), and most of DEPs were enriched in 13 pathways ( Table S5-b , Figure S3 ). The primary metabolism-related pathways, such as photosynthesis (ko00710, 7), energy metabolism (ko00010, 6), and amino acid biosynthesis (ko00940, 4) were enriched in H₂S vs. S. It is noteworthy that several DEPs were enriched in the sulfur and cysteine metabolism pathways (ko00270, 3). These pathways are responsible for the synthesis of endogenous H₂S, suggesting that exogenous H₂S treatment might cause changes in endogenous H₂S production along with altered regulation of other signaling pathways. However, very few other DEPs were enriched in other primary metabolism-related pathways, only nitrogen metabolism (ko00910, 1 DEP), lipid metabolism (ko00603, 1), and purine metabolism (ko00230, 1). Additionally, some DEPs were enriched in defense-related pathways, such as glutathione metabolism (ko00480, 1), inositol phosphate metabolism (ko00562, 1), protein processing in ER (endoplasmic reticulum) (ko04141, 1), proteasome (ko03050, 2), and RNA transport (ko03013, 1).

To correlate transcript and protein expression profiles, accession numbers from the proteome were extracted and compared with the annotated RNA-Seq libraries ( Table S6 ). The expression levels of 17 out of the 47 proteins were consistent with the mRNA levels, indicating that only a few proteins were regulated directly at the transcriptional level. Cluster analyses of the 47 differentially expressed genes and proteins were conducted ( Figures 4A ). Then, comparative analyses of 12 genes (27.66%) that showed similar regulation at both the transcript and protein levels and 4 genes (8.51%) that showed opposite at the transcript and protein levels was conducted ( Figures 4C ). The Pearson Correlations (r) were calculated to be 0.399 ( Figure 4 ), 0.839 ( Figure 4 ), and −0.872 ( Figure 3 ).

Forty DEG/DEPs could be mapped to just two metabolic pathways within cucumber ( Figure 5 ). Among them, 13 DEGs (6 up-regulated genes and 7 down-regulated genes) and 7 DEPs (7 up-regulated proteins) were mapped onto carbon fixation in photosynthetic metabolism using the KEGG database ( Figure 5 ). Nine of the transcripts, Phosphoenolpyruvate carboxylase (PPC), malate dehydrogenase, glutamic oxaloacetic transaminase (GOT1), alanine transaminase (ALT), transketolase, malate dehydrogenase (oxaloacetate-decarboxylating)(NADP⁺) (MAEB), phosphoenolpyruvate carboxykinase (ATP) (PCKA), glyceraldehyde-3-phosphate dehydrogenase (NADP⁺) (GAPA), and fructose-1,6-bisphosphatase I (FBP), are mainly related to carbon fixation ( Figure 5A ). Meanwhile, malate dehydrogenase, ribulose-bisphosphate carboxylase large chain, ribulose-phosphate 3-epimerase, phosphoglycerate kinase, and sedoheptulose-1,7-bisphosphatase showed significant up-regulated at the proteome level. Among these, malate dehydrogenase showed the same up-regulated pattern at both the transcriptome and proteome levels ( Figure 5 ). However, ribulose-bisphosphate carboxylase large chain, ribulose-phosphate 3-epimerase, phosphoglycerate kinase, and sedoheptulose-1,7-bisphosphatase showed a down-regulation in the transcriptome and an up-regulation in the proteome ( Figure 5 ).

Twenty of the DEGs could be mapped to the sulfur, cysteine and methionine metabolism pathways, all of which play important roles in H₂S regulation ( Figure 5 ). Eighteen DEGs (14 up-regulated genes and 4 down-regulated genes) and 2 DEPs (2 up-regulated proteins) were mapped onto sulfur, cysteine, and methionine metabolism using the KEGG database. Cysteine synthase, S-adenosylmethionine decarboxylase, spermidine synthase (speE), methylthioribulose 1-phosphate dehydratase (DEP1), tyrosine aminotransferase, 5’-methylthioadenosine nucleosidase (MTN), adenylyl-sulfate reductase (APR), 5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase (metE), alanine-glyoxylate transaminase (AGXT2), methylthioribose-1-phosphate isomerase (mtnA), and aspartate aminotransferase (ASP5) were identified as mainly related to sulfur, cysteine, and methionine metabolism at the transcriptome level. Additionally, cysteine synthase and malate dehydrogenase (MDH1) showed the same up-regulated pattern at both the transcriptomic and proteomic levels ( Figure 5 ). The majority of the DEGs were up-regulated at the transcriptome level, except adenylyl-sulfate reductase (APR), homocysteine S-methyltransferase (mmuM), DNA (cytosine-5)-methyltransferase1 (DNMT1), and aminocyclopropanecarboxylate oxidase (E1.14.17.4) ( Figure 5 ).

Carbohydrate and energy metabolism are necessary for both plant growth and stress response. In this work, salt stress caused significant changes at the transcriptomic level in carbon metabolism (62 DEGs), biosynthesis of amino acids (59 DEGs), carbon fixation in photosynthesis (25 DEGs), nitrogen metabolism (16 DEGs), and fatty acid degradation (15 DEGs) (S vs. C, Figure 2 ). However, addition of H₂S to the salt treatment resulted in the disappearance of these pathway compared to salt stress alone (H₂S vs. S, Figure 3 ). The balance between protein synthesis and degradation also plays an essential role in the regulation of cellular processes in response to environmental cues Reinbothe et al., 2010; Rothman, 2010). Salt stress also caused significant changes in the levels of protein involved in photosynthesis (7 DEPs), carbohydrate and energy metabolism (7 DEPs), and protein folding, assembly, and degradation (5 DEPs) (S vs. C, Figure S2B ). Addition of H₂S to the salt treatment caused a decline in the number of DEPs compared to the number seen under salt stress alone (H₂S vs. S, Figure S2B ). This data is consistent with a previous report in sesame showing that salinity stress changed the abundance of many proteins involved in the biosynthesis and degradation of amino acids such as alanine, aspartate, valine, leucine, isoleucine, and glutamate (Zhang et al., 2019). These results indicated that salt stress caused significant changes in primary and energy metabolism, which might lead to plant growth inhibition. However, H₂S could regulate these pathways to recover to levels similar to the control sample. Plants need energy to regulate a variety of processes, including ROS scavenging and protective substance synthesis, to prevent salt stress-induced damage, and adequate energy is crucial for cucumber roots to resist salt stress (Du et al., 2018).

Under salt stress, photosynthesis is impaired due to stomatal closure and limited carbon dioxide intake. Plants with more active resistance to stress always have more energy demands (Zhu, 2001; Kosova et al., 2011). In this study, a large number of DEGs or DEPs related to photosynthesis, such as ribulose-phosphate 3-epimerase, chloroplastic (Csa_1G038360; XP_004137418.1), ribulose bisphosphate carboxylase small chain (Csa_5G609710; O65194), photosystem I reaction center subunit N (Csa_6G483300), chlorophyll a-b binding protein 6A (Csa_5G490820), photosynthetic NDH subunit of lumenal location 3(Csa_3G414060), carbonic anhydrase 2 isoform X2 (XP_004141091.2), chloroplast stem-loop binding protein of 41 kDa a (XP_004140080.1), and rubisco accumulation factor 1 (XP_004142574.1), were detected at the transcriptomic or proteomic level during H₂S regulation of salt-stress tolerance in cucumber ( Tables 1 and S2 ). In agreement with this observation, Jiang et al. (2019) found significant declines in the photosynthetic parameters P_n, T_r, and F_v/F_m and in the chlorophyll content (C_a, C_b, C_a+b, and C_x+c) under salt stress, and that H₂S alleviated these declines (Jiang et al., 2019).

Carbohydrate and nitrogen metabolism are complex phenomena driven by the coordinate expression of numerous genes (Zhao et al., 2019). Our transcriptome data showed that numerous DEGs involved in carbohydrate metabolism, such as malate synthase, glyoxysomal (Csa_1G050360), NADP-dependent glyceraldehyde-3-phosphate dehydrogenase (Csa_2G009470), citrate synthase (Csa_2G012660), formate dehydrogenase (Csa_3G836500), acyl-CoA dehydrogenase family member 10 (Csa_3G164520), and fructose-bisphosphate aldolase cytoplasmic isozyme (Csa_3G750920), and some DEGs involved in nitrogen metabolism, such as nitrate reductase [NADH]-like (Csa_4G377160) and cyanate hydratase (Csa_3G134790), responded to H₂S regulation of salt-stress tolerance in cucumber ( Table S2 ). In the proteomic data, metabolism-related proteins likely corresponding to these DEGs, namely, 6-phosphogluconolactonase 4 (XP_004145948.1), ribulose bisphosphate carboxylase (XP_004138462.1), ketol-acid reductoisomerase (XP_004136850.1), acetylornithine deacetylase (XP_004148484.1), malate dehydrogenase (XP_004147146.1), pyruvate dehydrogenase E1 component subunit beta-1 (XP_011656794.1), glyceraldehyde-3-phosphate dehydrogenase (XP_004152516.1), and sedoheptulose-1,7-bisphosphatase (NP_001267658.1), also responded to H₂S regulation of salt-stress tolerance in cucumber ( Table 1 ). In sesame, proteins related to energy metabolism and amino acid biosynthesis and metabolism were also salt-responsive (Zhang et al., 2019).

In this study, V-type proton ATPase subunit a2 (Csa_2G364570) and cation/H⁺ antiporter15/18 (Csa_3G563310; Csa_4G050260) were both detected in the transcriptomic data, indicating that H₂S might regulate proton and ion transport under salt-stress ( Table S2 ). Removing cytosolic Na⁺ from plant cell generally through the activities of Na⁺/H⁺ antiporters is critically important for plants to survive under salt stress. Additionally, previous research has shown that the shaker AKT1-like channels are likely involved in both high- and low-affinity K⁺ uptake and correlate with salt tolerance in plants (Duan et al., 2015). Our transcriptome data showed that potassium channel AKT1 (Csa_1G303700) was involved in H₂S regulation of salt-stress tolerance in cucumber ( Table S2 ).

Sulfur-containing compounds, including cysteine, methionine, glutathione, thioredoxin, and H₂S, are crucial for the survival of plants under biotic and abiotic stress. Cysteine exists at the cross pathway of primary metabolism, protein synthesis, and the formation of low M_r sulfur-containing defense compounds (Rausch et al., 2005). Our data revealed that H₂S treatment resulted in significant changes in cysteine and methionine metabolism (21 DEGs) and glutathione metabolism (20 DEGs) at the transcriptomic level (H₂S vs. C, Figure 3 ). Sulfur-containing compounds impact plant defense, and their formation saturates at a higher S-supply than does plant growth (Bloem et al., 2005). Our transcriptome analysis showed that numerous DEGs related to metabolism of sulfur-containing compounds, including probable sulfate transporter 3.5 (Csa_1G378520), thioredoxin H-type/M-typ (Csa_3G865340; Csa_6G343710), thioredoxin-like 3-2 (Csa_5G622670), cysteine-rich repeat secretory protein 3 (Csa_2G171840), and methionine S-methyltransferase (Csa_3G113310), responded to the addition of H₂S during salt stress in cucumber ( Table S2 ). In addition, H₂S homeostasis in plant cells is closely related to cysteine metabolism. Similarly, application of H₂S also caused significant changes at proteome level in sulfur, cysteine and methionine metabolic pathways (3 DEPs) and glutathione metabolism (1 DEP) ( Figure S3 ). Cysteine synthase 1 (XP_004138495.1), glutathione S-transferase U25-like (XP_004147762.1), and protein disulfide-isomerase (XP_004149798.1) were observed through proteomic analysis ( Table 1 ). H₂S homeostasis is closely regulated by L-/D-cysteine desulfhydrase and L/D-cystiene desulphhydrase can catalyze the degradation of L-/D-cysteine to produce H₂S, amine and pyruvate. It is surprising that L/D-cystiene desulphhydrase is not detected through transcriptome and proteome sequencing. However, Jiang et al. (2019) demonstrated that the activities of D/L-cysteine desulfhydrase significantly increased under excess NaCl and the endogenous H₂S level also increased (Jiang et al., 2019). The metabolism related to sulfur-containing compounds, especially those that control the endogenous H₂S level by regulating H₂S synthetic and decomposition enzymes, plays a very important role in cucumber during H₂S regulation of salt-stress tolerance.

Perception and transmission of signals would be the first steps of a response within cucumber leaves to H₂S regulation of salt-stress tolerance, and these interactions form a network of regulation leading to a variety of physiological responses (Sang et al., 2016). Here, several signal transduction-related pathways, such as plant hormone signal transduction (32 DEGs) and cAMP signaling pathway (6 DEGs), significantly responded at the transcriptomic level with application of H₂S (H₂S vs. S, Figure 3 ). Although no plant hormone signal transduction-related protein was found among the differentially accumulated proteins, numerous plant hormone signal transduction-related genes, such as auxin-responsive protein IAA13 (Csa_3G134550), auxin-binding protein ABP19a (Csa_7G450510), ethylene-responsive transcription factor ERF017 (Csa_1G597730), salicylic acid-binding protein 2 (Csa_6G490080), abscisic acid receptor PYR1-like (Csa_3G902310), and gibberellin-regulated protein (Csa_3G826660), were detected in the DEG data ( Table S2 ). Ethylene-, abscisic acid-, and salicylic acid-induced signal transduction pathways are frequently reported to be closely related to plant resistance to various stresses. Additionally, calcium signals are involved in plant responses to various stimuli, including abiotic and biotic stresses, and regulate a wide range of physiological processes (Harper, 2001). Our data showed that calcium sensing receptor (Csa_1G041560), calcium-dependent protein kinase 32 (Csa_4G172540), putative calcium-binding protein CML19 (Csa_1G002110), calcium-binding protein PBP1-like (Csa_1G071800), calcium uniporter protein 3, mitochondrial-like (Csa_6G487650), calcium permeable stress-gated cation channel 1 (Csa_2G416110), calcium-transporting ATPase4 (Csa_5G310810), and cation/calcium exchanger 2-like (Csa_3G822350) were detected in the DEG data during H₂S regulation of salt-stress tolerance in cucumber ( Table S2 ). H₂S has been reported to alleviate oxidative damage under excess nitrate stress through mitogen-activated protein kinase signaling in cucumber (Qi et al., 2019). As expected, mitogen-activated protein kinase 3-like (Csa_1G479630) and cyclic nucleotide-gated ion channel 1-like (Csa_3G607120) as well as a few transcription factor IBH1/BIM1/MYB86-like/RAX3 genes (Csa_1G294080; Csa_1G015080; Csa_1G046820; Csa_2G352410) were also detected in the transcriptomic data during H₂S regulation of salt-stress tolerance in cucumber ( Table S2 ). Moreover, the proteome data showed that inositol phosphate metabolism (Inositol-3-phosphate synthase-like; XP_004149141.2), an important signal transduction pathway, was involved in H₂S regulation of salt-stress tolerance in cucumber.

Plants suffering from NaCl toxicity often exhibit symptoms associated with oxidative stress and membrane lipid peroxidation, which can result in accumulation of reactive oxygen species and malondialdehyde (Hossain et al., 2005). Thus, under various biotic and abiotic stresses, stress- and defense-related proteins have been frequently reported (Liu et al., 2018). Under salt stress, the genes L-ascorbate peroxidase (Csa_6G507100), peroxidase P7 (Csa_1G586250), superoxide dismutase [Fe]/[Mn]/[Cu-Zn] (Csa_1G571850; Csa_1G025980; Csa_2G013250), and catalase isozyme 1 (Csa_4G658590) were significantly up-regulated and the L-ascorbate oxidase were significantly down-regulated with the transcriptome sequencing. However, the genes L-ascorbate peroxidase (Csa_6G507100), peroxidase P7 (Csa_1G586250), and superoxide dismutase [Cu-Zn] (Csa_2G013250) were significantly down-regulated with the transcriptome sequencing, when the exogenous H₂S was applied the cucumber under salt stress ( Table S2 ). Consistent with the transcriptome analysis, peroxidase (AAA33129.1) and glutathione S-transferase U25-like (XP_004147762.1) were also detected in the proteomic data ( Table 1 ). Peroxidase, which can also catalyze the reduction of H₂O₂, was induced by salt stress (Yan et al., 2005). Glutathione S-transferases (XP_004147762.1) represent a major sample of detoxification enzymes that play important roles in protecting plants from impairments caused by abiotic stresses (Liu et al., 2019). Moreover, heat shock 70 kDa protein (Csa_4G617390) was detected in the transcriptomic data during H₂S regulation of salt-stress tolerance in cucumber ( Table S2 ). Heat shock proteins are also associated with the stress response and play an important role as chaperones, including in protein folding and degradation under stress conditions (Carmo et al., 2019).