Comparative Proteomic Analysis Reveals the Regulatory Effects of H2S on Salt Tolerance of Mangrove Plant Kandelia obovata

As a dominant mangrove species, Kandelia obovata is distributed in an intertidal marsh with an active H2S release. Whether H2S participates in the salt tolerance of mangrove plants is still ambiguous, although increasing evidence has demonstrated that H2S functions in plant responses to multiple abiotic stresses. In this study, NaHS was used as an H2S donor to investigate the regulatory mechanism of H2S on the salt tolerance of K. obovata seedlings by using a combined physiological and proteomic analysis. The results showed that the reduction in photosynthesis (Pn) caused by 400 mM of NaCl was recovered by the addition of NaHS (200 μM). Furthermore, the application of H2S enhanced the quantum efficiency of photosystem II (PSII) and the membrane lipid stability, implying that H2S is beneficial to the survival of K. obovata seedlings under high salinity. We further identified 37 differentially expressed proteins by proteomic approaches under salinity and NaHS treatments. Among them, the proteins that are related to photosynthesis, primary metabolism, stress response and hormone biosynthesis were primarily enriched. The physiological and proteomic results highlighted that exogenous H2S up-regulated photosynthesis and energy metabolism to help K. obovata to cope with high salinity. Specifically, H2S increased photosynthetic electron transfer, chlorophyll biosynthesis and carbon fixation in K. obovata leaves under salt stress. Furthermore, the abundances of other proteins related to the metabolic pathway, such as antioxidation (ascorbic acid peroxidase (APX), copper/zinc superoxide dismutase (CSD2), and pancreatic and duodenal homeobox 1 (PDX1)), protein synthesis (heat-shock protein (HSP), chaperonin family protein (Cpn) 20), nitrogen metabolism (glutamine synthetase 1 and 2 (GS2), GS1:1), glycolysis (phosphoglycerate kinase (PGK) and triosephosphate isomerase (TPI)), and the ascorbate–glutathione (AsA–GSH) cycle were increased by H2S under high salinity. These findings provide new insights into the roles of H2S in the adaptations of the K. obovata mangrove plant to high salinity environments.


Introduction
Mangroves comprise a complex and unique ecosystem that distributes along tropical and subtropical coastal tidal zone [1]. Mangroves are facultative halophytes and potential stress adaptors due to their special morphological, anatomical, physiological, and biochemical features [1]. Different mangrove species have different salinity preferences and achieve optimal growth at different salinity levels. Jayatissa et al. reported that there is an optimal growth for Sonneratia caseolaris with the

H2S Rather than other Derivatives from Nahs Alleviates NaCl-Induced Reduction in Chlorophyll Content
It should be noticed that the NaHS solution contained not only H2S/HSbut also other sulfurcontaining components. The research results showed that the above chemicals (Na2SO4, NaHSO3, NaHSO4, and CH3COONa), failed to rescue the NaCl-induced reduction of chlorophyll content (Supplementary Figure S2A). Meanwhile, compared with the experimental control group (CK), the endogenous H2S content kept a high level with the NaHS treatment but maintained a stable low level under other sulfur-containing compounds or sodium-containing compounds (Supplementary Figure  S2B). HSand/or H2S has an essential role in relieving the reduction in chlorophyll content in K. obovata seedlings under salt treatment but not other sulfur-containing compounds or sodiumcontaining compounds (Na2SO4, NaHSO3, NaHSO4, and CH3COONa).

Effects of H2S and NaCl on the Characteristics of Chlorophyll Fluorescence
The variable fluorescent/maximal fluorescent (Fv/Fm) ratio reflects the potential quantum efficiency of photosystem II (PSII), widely used as a sensitive indicator of stress-induced damage to the photosynthetic apparatus [18]. After seven days of treatment with NaCl, K. obovata seedlings' leaves were used to measure the chlorophyll fluorescence. The results showed that the NaCl treatment significantly decreased the ratio of Fv/Fm by 27.18% compared to the control. However, when exogenous H2S was applied to the NaCl-treated seedlings, the Fv/Fm was increased by 16.8% (Supplementary Figure S3).

H 2 S Rather than other Derivatives from Nahs Alleviates NaCl-Induced Reduction in Chlorophyll Content
It should be noticed that the NaHS solution contained not only H 2 S/HS − but also other sulfur-containing components. The research results showed that the above chemicals (Na 2 SO 4 , NaHSO 3 , NaHSO 4 , and CH 3 COONa), failed to rescue the NaCl-induced reduction of chlorophyll content (Supplementary Figure S2A). Meanwhile, compared with the experimental control group (CK), the endogenous H 2 S content kept a high level with the NaHS treatment but maintained a stable low level under other sulfur-containing compounds or sodium-containing compounds (Supplementary Figure S2B). HS − and/or H 2 S has an essential role in relieving the reduction in chlorophyll content in K. obovata seedlings under salt treatment but not other sulfur-containing compounds or sodium-containing compounds (Na 2 SO 4 , NaHSO 3 , NaHSO 4 , and CH 3 COONa).

Effects of H 2 S and NaCl on the Characteristics of Chlorophyll Fluorescence
The variable fluorescent/maximal fluorescent (F v /F m ) ratio reflects the potential quantum efficiency of photosystem II (PSII), widely used as a sensitive indicator of stress-induced damage to the photosynthetic apparatus [18]. After seven days of treatment with NaCl, K. obovata seedlings' leaves were used to measure the chlorophyll fluorescence. The results showed that the NaCl treatment significantly decreased the ratio of F v /F m by 27.18% compared to the control. However, when exogenous H 2 S was applied to the NaCl-treated seedlings, the F v /F m was increased by 16.8% (Supplementary Figure S3).

Effects of H 2 S and NaCl on Oxidative Stress and Activity of the Antioxidant System
Cytomembranes are damaged by reactive oxygen species (ROS) that were induced by salinity. The content of H 2 O 2 was analyzed to evaluate the effect of H 2 S on the oxidative burst of NaCl-treated K. obovata seedlings. When exogenous H 2 S was applied to NaCl-treated leaves, the content of H 2 O 2 was decreased, but it did not significantly differ from that of the NaCl treatment ( Figure 2A). Leaf electrolyte leakage percentage (REL) and the thiobarbituric acid reactive substances (TBARS) accumulation were measured to estimate the protective role of H 2 S on the stability of the membrane. The TBARS levels were decreased by the exogenous H 2 S addition, but these levels did not significantly differ from that of the NaCl treatment ( Figure 2B). Meanwhile, high salinity significantly increased the REL of K. obovata leaves, being 2.0-fold higher than that of the control. The exogenous H 2 S significantly alleviated the NaCl-induced increase in REL ( Figure 2C).
Additionally, the content of glutathione (GSH) significantly increased by around 4-fold after the NaCl and H 2 S treatment compared to the NaCl treatment ( Figure 2D). Plant cells have evolved the antioxidant enzymatic system, such as SOD and APX, to cope with stress-induced ROS generation. When exogenous H 2 S was applied to NaCl-treated seedlings, the activity of APX and SOD were increased but did not significantly differ from that of the NaCl treatment ( Figure 2E,F).

Effects of H2S and NaCl on Oxidative Stress and Activity of the Antioxidant System
Cytomembranes are damaged by reactive oxygen species (ROS) that were induced by salinity. The content of H2O2 was analyzed to evaluate the effect of H2S on the oxidative burst of NaCl-treated K. obovata seedlings. When exogenous H2S was applied to NaCl-treated leaves, the content of H2O2 was decreased, but it did not significantly differ from that of the NaCl treatment ( Figure 2A). Leaf electrolyte leakage percentage (REL) and the thiobarbituric acid reactive substances (TBARS) accumulation were measured to estimate the protective role of H2S on the stability of the membrane. The TBARS levels were decreased by the exogenous H2S addition, but these levels did not significantly differ from that of the NaCl treatment ( Figure 2B). Meanwhile, high salinity significantly increased the REL of K. obovata leaves, being 2.0-fold higher than that of the control. The exogenous H2S significantly alleviated the NaCl-induced increase in REL ( Figure 2C).
Additionally, the content of glutathione (GSH) significantly increased by around 4-fold after the NaCl and H2S treatment compared to the NaCl treatment ( Figure 2D). Plant cells have evolved the antioxidant enzymatic system, such as SOD and APX, to cope with stress-induced ROS generation. When exogenous H2S was applied to NaCl-treated seedlings, the activity of APX and SOD were increased but did not significantly differ from that of the NaCl treatment ( Figure 2E,F).

Figure 2.
Effects of NaHS on oxidative stress and antioxidant system activity of the leaves of K. obovata seedlings treated by high salinity. (A) H2O2 content, (B) thiobarbituric acid reactive substances (TBARS) accumulation, (C) electrolyte leakage percentage (REL), (D) glutathione (GSH) content, (E) ascorbate acid peroxidase (APX) activity and (F) superoxide dismutase (SOD) activity. CK stands for the control treated only by a 1/4 strength Hoagland's nutrient solution. NaCl stands for the 400 mM of the NaCl treatment. NaCl and H2S stands for the treatment with 400 mM of NaCl and 200 μM of NaHS. Error bars are SE (n = 3). The columns labeled with different letters are significantly different at p < 0.05 according to Duncan's multiple range test.

Identification and Classification of DEPs
The 2-DE proteomic approach was employed to analyze the protein expression profile to investigate the mechanism of the H2S-mediated salt tolerance of K. obovata. Three biological replicates were performed with similar results. According to the image analysis by the PDQuest software, approximately 450 protein spots were reproducibly resolved on each gel. The protein spots were detected in the isoelectric point (pI) range from 4 to 7, and the range of molecular weight (MW) was 14.4-66.2 kDa. A total of 37 spots with significant changes (higher than 2.0-fold, p < 0.05) were identified as differentially expressed proteins (DEPs) and presented in representative image ( Figure  3A). Close-up images from four protein spots were displayed in Figure 3B.
We provide the information for 37 DEPs in Table 1 and Supplementary Table S2. The difference between the two groups (NaCl vs. CK and NaCl and H2S vs. CK) was statistically significant. A

Identification and Classification of DEPs
The 2-DE proteomic approach was employed to analyze the protein expression profile to investigate the mechanism of the H 2 S-mediated salt tolerance of K. obovata. Three biological replicates were performed with similar results. According to the image analysis by the PDQuest software, approximately 450 protein spots were reproducibly resolved on each gel. The protein spots were detected in the isoelectric point (pI) range from 4 to 7, and the range of molecular weight (MW) was 14.4-66.2 kDa. A total of 37 spots with significant changes (higher than 2.0-fold, p < 0.05) were identified as differentially expressed proteins (DEPs) and presented in representative image ( Figure 3A). Close-up images from four protein spots were displayed in Figure 3B.
increased, and five proteins were decreased. There were six proteins that were affected by both the NaCl and NaCl and H2S treatments ( Figure 4A). Compared with the NaCl vs. CK group, there were more up-regulated DEPs in the NaCl and H2S vs. CK group. These up-regulated DEPs were mostly divided into carbohydrate metabolism, energy metabolism, and stress response proteins ( Figure 4B). There were less down-regulated DEPs divided into carbohydrate metabolism, energy metabolism, and photosynthesis under the NaCl and H2S treatment ( Figure 4B).  The numbers assigned to the protein spots correspond to those listed in Table 1. (A) Representative coomassie brilliant blue-R250 (CBB-R250) stained 2D gel of total proteins extracted from K. obovata leaves under the control condition. Proteins (1.5 mg) were loaded onto the pH 4-7 gradient immobilized pH gradient (IPG) strip for the first-dimensional isoelectric focusing (IEF) and then separated in the second dimension on a 15% SDS-PAGE gel. The isoelectric point (pI) and molecular weight (MW) in kilodaltons are indicated on the top and left of the gel, respectively. Arrows indicate 37 spots with at least 2.0-fold changes (p < 0.05) that were analyzed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI TOF MS). (B) The enlarged windows of the representative protein spots with different expressions under both the NaCl treatment and the NaCl and H 2 S treatment. NaCl stands for the 400 mM of NaCl treatment, NaCl and H 2 S stands for the treatment with 400 mM of NaCl and 200 µM NaHS, and CK stands for the control treated only by a 1/4 strength Hoagland solution. Table 1. Identification of differentially expressed proteins with greater than 2.0-fold change in the leaves of K. obovata seedlings under the salinity and NaHS treatments. We provide the information for 37 DEPs in Table 1 and Supplementary Table S2. The difference between the two groups (NaCl vs. CK and NaCl and H 2 S vs. CK) was statistically significant. A comprehensive overview of the expression patterns of 37 identified proteins is shown in Figure 4A. They were divided into five functional categories, including carbohydrate and energy metabolism (29.73%), photosynthesis (18.92%), stress response proteins and cell structure (18.92%), amino acid and protein metabolism (16.22%), and hormone biosynthesis and transcription factor (10.81%) by their biological functions, each of which is in the UniProt database ( Figure 4B). There were fifteen proteins that significantly changed under the NaCl treatment; specifically, the abundances of seven proteins were increased, and the abundances of eight proteins were decreased. Under the NaCl and H 2 S treatment, fourteen proteins were significantly changed, the abundances of nine proteins were increased, and five proteins were decreased. There were six proteins that were affected by both the NaCl and NaCl and H 2 S treatments ( Figure 4A). Compared with the NaCl vs. CK group, there were more up-regulated DEPs in the NaCl and H 2 S vs. CK group. These up-regulated DEPs were mostly divided into carbohydrate metabolism, energy metabolism, and stress response proteins ( Figure 4B). There were less down-regulated DEPs divided into carbohydrate metabolism, energy metabolism, and photosynthesis under the NaCl and H 2 S treatment ( Figure 4B).

STRING Interaction Networks among DEPs
To predict protein-protein interaction (PPI) networks with the Search Tool of the Retrieval of Interaction Genes/Proteins (STRING) database, an alignment of the amino acid sequence to find the corresponding homologous proteins from Arabidopsis thaliana was carried out. The PPI network visualized by Cytoscape is shown in Supplementary Figure S4. The PPI network of DEPs was found to be composed of 27 nodes and 129 edges. A total of two modules were identified in the PPI network, and these modules include nine and eight proteins ( Figure 5, Supplementary Table S3). Module A

STRING Interaction Networks among DEPs
To predict protein-protein interaction (PPI) networks with the Search Tool of the Retrieval of Interaction Genes/Proteins (STRING) database, an alignment of the amino acid sequence to find the corresponding homologous proteins from Arabidopsis thaliana was carried out. The PPI network visualized by Cytoscape is shown in Supplementary Figure S4. The PPI network of DEPs was found to be composed of 27 nodes and 129 edges. A total of two modules were identified in the PPI network, and these modules include nine and eight proteins ( Figure 5, Supplementary Table S3). Module A has nine nodes and 32 interactions. The key node in module A is glutamine synthetase 2 (GS2). Module B has eight nodes and 22 interactions. The key node in module B is ribulose-1,5-bisphosphate carboxylase large subunit (RBCL).
An acyclic graph was constructed and directed by using the Biological Network Gene Ontology tool (BiNGO) to depict the visual interactions of functions based on the enrichment levels of the GO terms. The present study identified eight significantly enriched GO terms in biological processes, and especially in terms of response to abiotic stimulus, response to stress, and the generation of precursor metabolites and energy (Supplementary Figure S5).

Comparisons of Expression Patterns between Protein and Transcript of Selected DEPs
To further verify the results of the proteome, we randomly selected six DEPs to further analyze their transcriptional expressions under different treatments by using quantitative real-time PCR. In particular, the consistency between the protein and mRNA levels of the following proteins was compared: transcription factor basic helix-loop-helix 145 (bHLH145, spot 13), superoxide dismutase 4 (SOD, spot 18), heat-shock protein (HSP, spot 19), oxygen-evolving enhancer protein 1 (OEE1, spot 14), abscisic acid stress ripening protein (Asr, spot 31) and glutamine synthetase 1;1 (GS1;1, spot 33). The detailed list of the corresponding primer pairs and the detailed PCR procedures used for the realtime PCR are summarized in Supplemental Table S4. As shown in Figure 6, four selected genes showed a similar pattern with their corresponding protein, except for bHLH 145 and GS1;1.

Comparisons of Expression Patterns between Protein and Transcript of Selected DEPs
To further verify the results of the proteome, we randomly selected six DEPs to further analyze their transcriptional expressions under different treatments by using quantitative real-time PCR. In particular, the consistency between the protein and mRNA levels of the following proteins was compared: transcription factor basic helix-loop-helix 145 (bHLH145, spot 13), superoxide dismutase 4 (SOD, spot 18), heat-shock protein (HSP, spot 19), oxygen-evolving enhancer protein 1 (OEE1, spot 14), abscisic acid stress ripening protein (Asr, spot 31) and glutamine synthetase 1;1 (GS1;1, spot 33). The detailed list of the corresponding primer pairs and the detailed PCR procedures used for the real-time PCR are summarized in Supplemental Table S4. As shown in Figure 6, four selected genes showed a similar pattern with their corresponding protein, except for bHLH 145 and GS1;1.

Comparisons of Expression Patterns between Protein and Transcript of Selected DEPs
To further verify the results of the proteome, we randomly selected six DEPs to further analyze their transcriptional expressions under different treatments by using quantitative real-time PCR. In particular, the consistency between the protein and mRNA levels of the following proteins was compared: transcription factor basic helix-loop-helix 145 (bHLH145, spot 13), superoxide dismutase 4 (SOD, spot 18), heat-shock protein (HSP, spot 19), oxygen-evolving enhancer protein 1 (OEE1, spot 14), abscisic acid stress ripening protein (Asr, spot 31) and glutamine synthetase 1;1 (GS1;1, spot 33). The detailed list of the corresponding primer pairs and the detailed PCR procedures used for the realtime PCR are summarized in Supplemental Table S4. As shown in Figure 6, four selected genes showed a similar pattern with their corresponding protein, except for bHLH 145 and GS1;1.

H2S but Not Other NaHS Derivatives Contributes to Chlorophyll Content Recovery Decreased by NaCl Treatment
NaHS, which in solution dissociates to Na + and HS -(of which the latter associates with H + to

H 2 S but Not Other NaHS Derivatives Contributes to Chlorophyll Content Recovery Decreased by NaCl Treatment
NaHS, which in solution dissociates to Na + and HS − (of which the latter associates with H + to produce H 2 S), is widely used to examine the biological effects of H 2 S [19]. In contrast to the results from NaHS, treatment with other compounds failed to induce endogenous H 2 S accumulation and to alleviate the reduction in chlorophyll content caused by NaCl exposure (Supplementary Figure S2). Accordingly, this study provides evidence that NaHS contributes to chlorophyll content recovery by HS − , rather than Na + . NaHS-associated responses are H 2 S-specific in K. obovata.

Exogenous H 2 S Alleviates Growth and Photosynthesis Inhibition Induced by NaCl in K. obovata Seedlings
Parameters such as relatively low fresh weight, relatively low dry weight, and reduced photosynthesis are potential indicators of salinity stress in plants [11]. In this study, significant variations in the studied physiological traits of K. obovata under different concentrations of NaHS with 400 mM of NaCl were recorded. When the concentration of H 2 S reached 500 µM, the leaf dry weight and photosynthesis decreased ( Figure 1). These findings are in agreement with those of Chen et al. [20], who reported a similar decline in the biomass and photosynthesis of Spinacia oleracea after treatment with a high concentration of NaHS. The change in Ci was analyzed as an indicator of the dominant factor of stomatal or non-stomatal limitations to photosynthesis. Ci showed a reverse trend to Gs, which suggests that non-stomatal closure is the dominant limitation to photosynthesis [21]. In the present research, the decrease in Ci showed a reverse trend to Gs under various concentrations of NaHS, which indicated that non-stomatal limitations dominated.
The previous study indicated that H 2 S could indirectly enhance the photosynthesis reaching its maximal values, as well as the maximal photochemical efficiency of photosystem II (F v/ F m ) [20]. In the present research, the NaCl and H 2 S treatment also increased F v/ F m (Supplementary Figure S3). Additionally, the proteomics analyses revealed six proteins involved in the light reactions of photosynthesis, including electron transfer, light-harvesting, and the light-induced oxidation of water. The level of ferredoxin-NADP reductase (FNR, spot 2), which catalyzes the removal of the electron from photosystem I to NADP + , and otherwise nicotinamide adenine dinucleotide phosphate (NADPH) is oxidized to produce reducing power via ferredoxin (Fd) [22], which is recovered under the NaCl and H 2 S treatment. The O 2 evolving complex 33kD family protein (OEC spot 12) has been termed as a chloroplast manganese-stabilizing protein due to its functions in assembly, stabilization and protection of the manganese cluster required for oxygen evolution in higher plants [23]. Though the accumulation of OEC was down-regulated in two groups, the accumulation of OEC was up-regulated relative to the NaCl treatment, which implied that H 2 S maintained the photosynthetic electron transport activity during K. obovata exposure to high salt stress. Additionally, NaCl-induced the down-regulation of photosynthetic electron transfer C (PETC, spot 10), which is an important subunit of the Cytb 6 f complex that participates in linear electron transport [24], was alleviated by the addition of H 2 S. Nevertheless, plastocyanin (PC, spot 1) was up-regulated under NaCl, which is consistent with the study on K. candel under high salinity did by Wang et al. [25]. On another hand, the abundances of PC remained unchanged under the combined NaCl and NaHS treatment. The results of proteomics analyses are in agreement with physiological measurements that showed that the treatment with NaHS alleviated the NaCl-caused inhibition of photosynthesis.
Photosynthetic carbon assimilation is driven by ribulose-1,5-bisphosphate (RuBP) carboxylase (Rubisco). Ribulose-1,5-bisphosphate carboxylase large subunit (RBCL, spot 8) was down-regulated by the NaCl treatment and unchanged under the NaCl and NaHS treatment. It is noteworthy that ribulose activase small isoform precursor (RBCS, spot 36) showed a reverse trend that was unchanged under the NaCl treatment and down-regulated by the NaCl and NaHS treatment. The capacity of photosynthesis is associated with the level of the RBCL/RBCS ratio and the higher photosynthetic rate exhibited the higher RBCL/RBCS ratio in rice leaves [26]. The up-regulation of the RBCL/RBCS ratio by the NaHS addition under salt treatment in our study indicated that the alleviation of Rubisco inhibition inevitably leads to the improvement of photosynthesis.
Phosphoglycolate phosphatase (PGP, spot 25) is involved in photorespiration, which is required to give carbon from 2-phosphoglycolate back into metabolism [27]. The allocation of photosynthetic electron transport to photorespiration has been found to be enhanced with the decrease of the chlorophyll content [27]. In this study, the NaCl treatment strongly up-regulated PGP, which indicated that the allocation of photosynthetic electron transport to the Calvin cycle was reduced [28]. The accumulation of PGP was not significantly different from the control treatment by the NaHS addition.
The photosynthetic electron transport chain converts light into chemical energy, supplying ATP and NADPH to drive the carbon dioxide reduction and fixation processes [29]. ATP synthase F1 complex (CF1) epsilon subunit (atpC, spot 7) is involved in the energy supply needed for the carbon dioxide reduction and fixation processes. Though the accumulation of atpC was down-regulated in our study, the accumulation of atpC was modified relative to the NaCl treatment, thus implying the NaHS addition might have had a positive effect to supply the energy.

H 2 S Rescues the Primary Metabolism Altered by NaCl
In this study, triosephosphate isomerase (TPI, spots 16), and phosphoglycerate kinase (PGK, spot 35) were found to be down-regulated or unchanged under high salinity, respectively. PGK plays a crucial role in catalyzing the ATP-dependent phosphorylation of phosphoglycerate in the Calvin cycle [30]. The two glycolytic enzymes were all positivity modulated by the NaHS addition, suggesting much more energy was produced by glycolysis.
All levels of plant function are affected by nitrogen metabolism [31]. In higher plants, glutamine synthetase (GS2 and GS1:1, spots 32 and 33, respectively) is a crucial enzyme of primary ammonium assimilation and nitrogen metabolism [32]. Glutamine synthetase participates in the synthesis of GSH through the glutamate biosynthesis pathway; thus, the overexpression of glutamine synthetase leads to more GSH formation [33,34]. It is interesting to note that the accumulation of glutamine synthetase was positivity modulated by the NaHS addition in our study.
HSP (spot 19) and 20 kDa chaperonin family protein (Cpn 20, spot 26) were involved in protein synthesis, a process that was positively modulated by the NaHS addition. As molecular chaperones, HSP and chaperonin family protein participate in protein transport, protein folding, and protein assembly processes [35]. During abiotic stress, chaperonins help plants combat increasing amount of incorrectly folded proteins [35].
Previous publications have reported that the TCA cycle provides essential precursors for amino acid biosynthesis and energy metabolism. ATP synthase subunit d (ATP5PD, spot 9) and ATP synthase beta subunit (ATP5F1B, spot 29) are located in the mitochondria. The ATP synthase is an important enzyme that catalyzes energy production by synthesizing ATP from ADP [36], providing energy to the TCA cycle. In mammals, increased oxidative stress represents an important factor for isolated disorders of ATP synthase [37], and there is a connection between ATP synthase and oxidative stress. Moreover, it has recently been predicted that a protein network in which ATP synthase is involved can contribute to protect plants against photo-oxidative damage [38]. In our study, the NaCl and H 2 S treatment significantly up-regulated the accumulation of ATP5PD and ATP5F1B. In addition, nucleoside diphosphate kinase 1 (NDPK1, spot 30), a housekeeping enzyme that maintains levels of CTP, GTP, and UTP in cells, was identified to be down-regulated under high salinity and significantly up-regulated by the NaHS addition. A recent study showed that NDPK grants transgenic plants tolerance to multiple stresses, including salt and extreme temperatures [39].
Remarkably, H 2 S rescued the primary metabolism altered by NaCl in our study. In a previous study, Li et al. also reported that increment in energy metabolism by H 2 S might be responsible for abiotic stress alleviation [40].

H 2 S Relieves Oxidative Stress Induced by NaCl
The salt-induced disruption of normal plant metabolism results in the accumulation of harmful ROS [41]. Excess ROS not only cause growth inhibition but also programmed cell death (PCD). The accumulation of H 2 O 2 and consequent increase in malondialdehyde and REL are indicators of salt-induced oxidative damage to membranes [42]. In the present study, the NaHS pretreatment reduced the accumulation of REL under salt stress, which is in agreement with a previous study by Shi et al. [43]. In addition, Actin7 isoform 1 (Actin7, spot 23)-which is thought to aid in the stability of organelle membranes in plant cells and to induce PCD with the stabilization of actin depolymerization [44,45]-was significantly up-regulated under the NaCl and H 2 S treatment in our study. Meanwhile, plants trigger various enzymatic and non-enzymatic antioxidants to detoxify ROS and prevent cellular damage [46]. In this study, compared with the NaCl treatment, the content of GSH was found to be induced by application of H 2 S (Figure 2). Ascorbate (AsA) and GSH play important roles in the AsA-GSH cycle [47]. As an electron donor, APX reduces H 2 O 2 into water by using AsA, and monodehydroascorbate can react with GSH to produce AsA and GSSG catalyzed by DHAR [47]. In our study, the increase in GSH content and the activity of APX indicate that H 2 S promoted the AsA-GSH cycle that was responsible for the removal of H 2 O 2 under high salinity. In agreement, APX (spot 37) protein expression was found to be up-regulated under the NaCl and H 2 S treatment. Moreover, copper/zinc superoxide dismutase CSD2A-1 (CSD2, spot 17) was found to be up-regulated under the NaCl and H 2 S treatment, but SOD (spot 18) showed the opposite trend. The SODs were divided into four types according to the different metal cofactors in the catalytic site, and CSD2 is mainly found in higher plants [48]. We propose that CSD2 plays a critical role in a direct and rapid mechanism of ROS detoxification under high salinity. The SODs (spot 18) might be iron SOD, which was found in the cytoplasmic of plant or the bacterial cells and played no major role in K. obovata leaves [48]. In short, based on the above results, we can infer that H 2 S increased non-enzymatic antioxidants to detoxify ROS and to prevent cellular damage under high salinity.
Moreover, some abiotic stress-related proteins, such as cysteine proteinase (CP, spot 4), pyridoxine biosynthesis PDX1-like protein 3 (PDX1, spot11), and alcohol dehydrogenase (ADH, spots 20) were also found to respond to high salinity and the NaCl and H 2 S treatment. An increase in the CP activity occurs just before the first senescence symptoms became visible in most plants [49]. CP is involved in PCD, and the balance between the levels of CP and phytocystatins is kept in a certain state through the antagonistic activities of abscisic acid (ABA) and gibberellins [50,51]. In our study, the accumulation of CP was significantly down-regulated by the NaHS addition. PDX1 is regarded as pyridoxal phosphate-binding protein [52]. Pyridoxal phosphate-binding proteins protect plants from high ion concentrations by inducing cysteine biosynthesis or the synthesis of cystatin to inhibit PCD [53]. We observed that PDX1 was significantly up-regulated to inhibit PCD under the NaCl and H 2 S treatment. ADH catalyzes the regeneration of NAD + from the reduction of acetaldehyde to ethanol [54]. H 2 S has been shown to inhibit ADH activity in the roots of the freshwater marsh species Spartina alterniflora [16], but the accumulation of ADH was down-regulated under the NaCl treatment in our study and did not change significantly with the NaHS addition. We observed that H 2 S helped alleviate the reduction of ADH under the NaCl treatment.

H 2 S Regulates Hormone Biosynthesis and Transcription Factor
Abscisic acid stress ripening protein (Asr, spot 31) is a downstream protein that participates in ABA signaling pathways [55]. The bHLH transcription factors serve as a negative feedback regulatory loop in ABA signaling in Arabidopsis thaliana [56]. In the present study, Asr and bHLH145 were up-regulated under the salt treatment and did not change significantly under the NaCl and H 2 S treatment, which indicated the signal transduction of ABA was inhibited by the addition of NaHS. Furthermore, in the promoter regions of the genes encoding CP exist ABA-responsive elements, and the transcription of CP is repressed by ABA through DOF (DNA binding One Zinc Finger) [51]. Additionally, NaHS up-regulated the accumulation of basic region-leucine zipper (bZIP, spot 27), which has been found to positively alter the adaptation of plants to adverse circumstances [57,58].

Protein-Protein Interaction Networks Analysis
The DEPs in the PPI network exhibited a higher enrichment, indicating a higher degree of modularization; therefore, the DEPs were divided into two modules to investigate their interactions by using the Molecular Complex Detection (MCODE) ( Figure 5). In module A, chloroplastic drought-induced stress protein (CDSP), NAD(P)-binding Rossmann-fold superfamily protein isoform 1 (FLDH, spot 24), FNR, GS2, OEE, PC, PETC, and RBCS were enriched in response to stimulus. The key node in this module was GS2. In module B, six nodes (APX, Cpn20, TPI, PGK, 2-Cys peroxiredoxin (Prx), and SOD) were involved in response to abiotic stimulus, and another two nodes (RBCL, ATP5F1B) were related to the Calvin cycle. The two modules implied that response to abiotic stimulus influenced other biological processes.

Plant Growth and Treatment
Mature propagules of K. obovata were collected from the mangrove forest in the National Nature Reserve for Mangroves in Zhangjiang River Estuary (23 • 55 N, 117 • 26 E), Yunxiao County, Fujian Province, PR China, where salinity levels range from 8% to 20%. Healthy propagules of similar size (20 cm in length) were selected and pre-cultivated in a plastic pot (dimension of 12 cm in diameter and 11 cm in depth) with clean sands. The pots were placed in a growth chamber with a temperature of 25/30 • C (night/day), a relative humidity of 70%, and a photoperiod of 8 h dark/16 h light with around 1250 µmol m −2 s −1 . A 1/4 strength Hoagland's nutrient solution was used to cultivate the hypocotyls. The solutions were replaced every week.
The treatments of NaHS and NaCl were set up when the 3rd pair of leaves appeared for physiological assays. The healthy and uniform seedlings were randomly cultivated in three groups, and each group had three replicates.
In the first group, plants were supplied with NaHS and NaCl for 7 days. Different concentrations of NaHS at 0, 50, 100, 200, 350 and 500 µM (The average of H 2 S was 243.1 ± 234.9 µM in the pore water samples of mangrove forest sediment) [59] were prepared in the 1/4 strength Hoagland's nutrient solution in the presence or absence of 400 mM of NaCl. After 7 days of treatment, photosynthesis, chlorophyll content, and dry weight were measured to determine the optimal concentration of NaHS for the salinity-induced growth inhibition.
The second group was supplied with a 1/4 strength Hoagland's nutrient solution containing 400 mM of NaCl in the presence of NaHS (200 µM), Na 2 SO4 (200 µM), NaHSO 3 (200 µM), NaHSO 4 (200 µM) and CH 3 COONa (200 µM). After 7 days of treatments, the contents of chlorophyll and endogenous H 2 S were measured to distinguish the actual role of H 2 S/HS − from the other compounds.
According to the results from above two experiments, the third group was supplied with a 1/4 strength Hoagland's nutrient solution containing both 400 mM of NaCl (NaCl) and 400 mM of NaCl and 200 µM of NaHS (NaCl and H 2 S). The same volume of a 1/4 strength Hoagland's nutrient solution was used as the control (CK). After chlorophyll fluorescence measurements, the second pair of leaves from the apex of the growing shoots was harvested, rapidly frozen in liquid nitrogen, and stored at −80 • C for a proteomic study and further biochemical analysis.

Measurements of Leaf Photosynthetic Pigment, Photosynthetic Rate and Chlorophyll Fluorescence Quenching
The contents of chlorophyll were measured according to the method of Lichtenthaler with little modifications [60]. Eighty percent (v/v) acetone was used to extract chlorophyll, and the content was estimated according to the absorbances at 470, 646 and 663 nm.
The net photosynthetic rate (Pn) of fully expanded leaves, as well as transpiration (Tr), stomatal conductance (Gs), and internal CO 2 concentration (Ci), were determined by using a portable photosynthesis system (Li-6400, Li-Cor, Lincoln, NE, USA). Those measurements were carried out in the morning from 9:00 to 11:30. The recovery coefficient of Pn was calculated according to the following equation: Pn NaCl /Pn CK . Here, Pn NaCl and Pn CK stand for the leaf photosynthetic rate measured in the NaCl and CK treatments.
We applied a fluorometer for measuring chlorophyll fluorescence (Li-6400, Li-Cor, Lincoln, NE, USA). Fv/Fm was calculated by using (F m − F 0 )/F m , where minimum (dark) fluorescence (F 0 ) was obtained by applying to measure light pulses at low frequency (0.03 µmol m −2 s −1 for 1 s). The maximum fluorescence (F m ) was determined by applying a saturating light pulse (6000 µmol m −2 s −1 for 0.8 s) to a dark-adapted sample [61].

Determination of Dry Weight and the Content of Endogenous H 2 S
Endogenous H 2 S content was determined according to the procedure of Zhang et al. [62], and the method of Chen et al. was employed to determine the dry weight [63].

Measurements of Oxidative Stress and Antioxidant System Activity
The H 2 O 2 content was measured according to the method of Hung and Kao [64]. Lipid peroxidation was determined by measuring thiobarbituric acid reactive substances (TBARS), according to the method of Yan et al. [65].
For the relative electrolyte leakage (REL) measurement, fresh leaves (0.2 g) were cut, added to deionized water (20 mL), and degassed for 10 min at room temperature. Using an electrical conductivity meter (DDS-11A), the initial reading of conductivity (E1) was recorded. Then, the solution containing plant materials was incubated at 100 • C for 15 min. The final reading of conductivity (E2) of the solution was recorded after cooling. Additionally, the reading of conductivity for deionized water (E0) was measured. Finally, the REL was estimated based on the following equation: Superoxide dismutase (SOD) activity was determined according to the method described by Beauchamp and Fridovich [66]. Ascorbate peroxidase (APX) activity was estimated according to the method of Chen and Asada [67].
The total content of glutathione (GSH) was measured by a GSH kit (JBI, Nanjing, China). Leaves (0.3 g) were ground with 0.3 mL of H 3 PO 3 (25%) and 0.9 mL of an NaH 2 PO 4 -EDTA buffer (0.1 M, pH 8.0). The homogenates were separated at 10,000× g at 4 • C for 20 min. The supernatant was used for GSH content measurement [68]. The absorbance at 420 nm was measured according to the manufacturer's instructions.

Protein Extraction from K. Obovata Leaves and 2-DE PAGE and Image/Data Analysis
Proteins were extracted by a phenol extraction procedure and followed by methanolic ammonium acetate precipitation, according to the method described by Delaplace et al. [69]. We rehydrated 17 cm immobilized pH 4-7 gradient strips (BioRad, CA, USA) with 340 µL of a rehydration buffer (containing a 1 mg protein sample) in the tray overnight. An Ettan immobilized pH gradient (IPG)phor3 system (GE Healthcare Amersham Bioscience, Little Chalfont, UK) was used for isoelectric focusing (IEF), following the conditions described by Shen et al. [70]. In 12.5% acrylamide gels, we observed gel electrophoresis. Each experiment was repeated three times. Each treatment was performed for three biological replications for quantitative analysis.
We stained SDS-PAGE gels with Coomassie Brilliant Blue R-250, and these gels were scanned by Uniscan M3600 (China) at 600 dpi. Gels were analyzed by the PDQuest software (Version 8.0, Bio-Rad) [71]. Afterward, DEPs were obtained by pairwise comparison with a fold change ≥2.0 and a Student's t-test (p < 0.05).

In-Gel Digestion, Identification and Classification of Differentially Expressed Proteins
We excised the DEPs with a more than 2.0-fold change from 2-DE gels [70]. Gel slices were incubated with 25 mM of NH 4 HCO 3 and acetonitrile for four times to remove the color of coomassie brilliant blue-R250 (CBB-R250). Trypsin/Lys-C Mix (Promega, Madison, WI, USA) was used to digest the proteins. Trifluoroacetic acid (0.5-1%) was added to terminate digestion. The supernatant was collected at 12,000× g for 10 min. A MALDI-TOF-TOF mass spectrometer was used to analyze the supernatant for mass spectrometry identification (Applied Biosystems, Massachusetts, USA), according to the method of Hu et al. [71]. The search parameters were set as follows: database NCBInr (release date: 2018.12.01); taxonomy Viridiplantae (green plants); peptide mass ranged from 10 to 130 kDa; the coverage of protein sequence had to reach a minimum of 10%; proteins had scores higher than 60 (p < 0.05); results with a confidence interval % (C.I.%) value higher than 95% were considered to be an identification.
The functions of the identified proteins were determined in http://www.uniprot.org/uniprot, and then the proteins were divided into five groups according to their biological functions in the plant. The subcellular localization of identified proteins was obtained by the protein subcellular localization prediction tool (WoLF PSORT, http://wolfpsort.hgc.jp/), as well as from previously published papers when possible.
The STRING database was used to predict the protein-protein interactions (PPIs) [72], and Cytoscape was used to visualize significant protein-protein associations in the PPI network. A combined score of >0.4 was selected as the cut-off value. Then, module analysis was carried out by the Molecular Complex Detection (MCODE) plugin to highlight the biological significance of gene modules that respond to the NaCl and/or NaHS treatment in K. obovata leaves [73]. Subsequently, the BiNGO tool was used to visualize the level of enriched GO terms.

Quantitative Real-Time PCR Analysis
For the total RNA extraction, the frozen K. obovata leaves (0.1g) were ground in liquid nitrogen and extracted by using a Total RNA Kit (TaKaRa, Dalian, China). We observed the RNA quality and integrity with an ultraviolet spectrophotometer (Cary 50, Varian, USA) and agarose gel electrophoresis. The RNA was used to synthesized cDNAs with moloney murine leukemia virus (M-MLV) reverse transcriptase First-Strand cDNA synthesis kit (TaKaRa, Dalian, China), and the cDNA mixture was used as a template for subsequent PCRs. The primers used for real-time PCR are shown in Supplementary  Table S4. A 10 µL real-time PCR mixture contained 2 µL of primers, 2 µL of cDNA, and 6 µL of SYBR Green (Sangon, Shanghai, China). Five independent biological replicates were used to perform gene expression. The relative gene expression was calculated by the 2 −∆∆CT method, and actin was used as an an internal control [74]. The Bio-Rad iQ5 Multicolor Real-Time PCR Detection System (Bio-Rad, Hercules, CA) was used to run qRT-PCR.

Statistical Analysis
A two-tailed Student's t-test (p < 0.05) and a Duncan's multiple range test were used for statistical analysis. Pearson's correlation among the amendment's characteristics was evaluated with SPSS Statistics for Windows (Version 22.0, IBM Corp, Armonk, NY). The direct and indirect relationships among leaf drought weight, chlorophyll content, net photosynthetic rate, intercellular carbon dioxide, stomatal conductance and transpiration rate were evaluated through path analysis by using edgeR package. The heatmap of the DEPs was drawn by using the R software "heatmap package."

Conclusions
Physiological and proteomic evidence in this study support beneficial role of H 2 S on K. obovata exposed to high salinity: (1) H 2 S increased carbon fixations and electron transfer, maintaining photosynthesis under high salinity; (2) H 2 S stimulated glycolysis to generate precursor metabolites, energy, and protein synthesis; (3) H 2 S increased non-enzymatic antioxidants to detoxify ROS and prevent cellular damage under high salinity; and (4) H 2 S affected the ABA signaling pathway, which led to a sustained plant adaptation to high salinity. As shown in Figure 7, we propose a working model to illustrate the detailed mechanism by which H 2 S alleviates NaCl-induced inhibition. In conclusion, these findings have important implications for understanding the functional role of H 2 S in the salt tolerance of K. obovata. Figure 7. The proposed regulatory networks of H2S on the salt tolerance of K. obovata seedling leaves. The left and right dots stand for the NaCl and NaCl and H2S treatments, respectively. The red color of dot indicates the up-regulated change, the green color of dot indicates the down-regulated change, and the white color of dot indicates no change in comparison to the control. The left and right arrows stand for the NaCl treatment and the NaCl and H2S treatment, respectively. The red color arrow indicates physiological indexes of increased change, and the green color arrow indicates physiological indexes of decreased change. ADH: alcohol dehydrogenase; APX: cytosolic ascorbate peroxidase; Asr: abscisic acid stress ripening protein; atpC: ATP synthase CF1 epsilon subunit; ATP5F1B: ATP synthase beta subunit; ATP5PD: ATP synthase subunit d, mitochondrial; CP: cysteine proteinase; Cpn 20: 20 kDa chaperonin family protein; Cpn 60: TCP-1/cpn60 chaperonin family protein; CSD2: copper/zinc superoxide dismutase; CDSP: chloroplastic drought-induced stress protein; FLDH: NAD(P)-binding Rossmann-fold superfamily protein isoform 1; FNR: ferredoxin-NADP reductase; GS2: glutamine synthetase 2; GS1:1: glutamine synthetase 1;1; HSP: heat-shock protein; MAN: Glycosyl hydrolase superfamily protein; MDH: malate dehydrogenase; NDPK: nucleoside diphosphate kinase; OEE1: oxygen-evolving enhancer protein; OEC: O2 evolving complex 33kD family protein; PC: plastocyanin; TPI: triosephosphate isomerase; PCD: programmed cell death; PDX1: pyridoxin biosynthesis; PETC: photosynthetic electron transfer C; PGK: phosphoglycerate kinase; PGP: phosphoglycolate phosphatase; Prx: 2-Cys peroxiredoxin; RBCL: ribulose-1,5bisphosphate carboxylase large subunit; RBCS: RuBisCO activase small isoform precursor; and SOD: superoxide dismutase. The detailed information for each spot is shown in Table 1.
Supplementary Materials: Supplementary Materials can be found at www.mdpi.com/xxx/s1; Figure S1. Net photosynthetic rate (Pn) recovery coefficient of K. obovata seedling leaf under different concentrations of NaHS; Figure S2. H2S and/or HS -, but not other compounds derived from NaHS, contribute to (A) chlorophyll content and (B) endogenous H2S accumulation; Figure S3. Effects of NaHS on the chlorophyll fluorescence characteristics of K. obovata seedling leaves; Figure S4. Protein-protein interactions (PPIs) network analysis of 27 identified proteins from the NaCl-and NaHS-treated K. obovata leaves; Figure S5. Gene ontology analysis of the proteins identified from the leaves of K. obovate; Table S1. Co-relations among different physiological indices in K. obovata leaves; Table S2. The detail information of the peptide sequence, expected and observed m/z values, and sequence coverage for the different expressed proteins identified by MALDI-TOF/TOF MS in the leaves of K. obovata under the salinity and NaHS treatments; Table S3. The modules of the protein-protein interaction networks and nodes biological process (GO) analysis; Table S4. Sequences of forward and reverse primers used The left and right dots stand for the NaCl and NaCl and H 2 S treatments, respectively. The red color of dot indicates the up-regulated change, the green color of dot indicates the down-regulated change, and the white color of dot indicates no change in comparison to the control. The left and right arrows stand for the NaCl treatment and the NaCl and H 2 S treatment, respectively. The red color arrow indicates physiological indexes of increased change, and the green color arrow indicates physiological indexes of decreased change. ADH: alcohol dehydrogenase; APX: cytosolic ascorbate peroxidase; Asr: abscisic acid stress ripening protein; atpC: ATP synthase CF1 epsilon subunit; ATP5F1B: ATP synthase beta subunit; ATP5PD: ATP synthase subunit d, mitochondrial; CP: cysteine proteinase; Cpn 20: 20 kDa chaperonin family protein; Cpn 60: TCP-1/cpn60 chaperonin family protein; CSD2: copper/zinc superoxide dismutase; CDSP: chloroplastic drought-induced stress protein; FLDH: NAD(P)-binding Rossmann-fold superfamily protein isoform 1; FNR: ferredoxin-NADP reductase; GS2: glutamine synthetase 2; GS1:1: glutamine synthetase 1;1; HSP: heat-shock protein; MAN: Glycosyl hydrolase superfamily protein; MDH: malate dehydrogenase; NDPK: nucleoside diphosphate kinase; OEE1: oxygen-evolving enhancer protein; OEC: O2 evolving complex 33kD family protein; PC: plastocyanin; TPI: triosephosphate isomerase; PCD: programmed cell death; PDX1: pyridoxin biosynthesis; PETC: photosynthetic electron transfer C; PGK: phosphoglycerate kinase; PGP: phosphoglycolate phosphatase; Prx: 2-Cys peroxiredoxin; RBCL: ribulose-1,5-bisphosphate carboxylase large subunit; RBCS: RuBisCO activase small isoform precursor; and SOD: superoxide dismutase. The detailed information for each spot is shown in Table 1.
Supplementary Materials: Supplementary Materials can be found at http://www.mdpi.com/1422-0067/21/1/118/s1; Figure S1. Net photosynthetic rate (Pn) recovery coefficient of K. obovata seedling leaf under different concentrations of NaHS; Figure S2. H 2 S and/or HS − , but not other compounds derived from NaHS, contribute to (A) chlorophyll content and (B) endogenous H 2 S accumulation; Figure S3. Effects of NaHS on the chlorophyll fluorescence characteristics of K. obovata seedling leaves; Figure S4. Protein-protein interactions (PPIs) network analysis of 27 identified proteins from the NaCl-and NaHS-treated K. obovata leaves; Figure S5. Gene ontology analysis of the