Physiological response and proteomics analysis of Reaumuria soongorica under salt stress

Soil salinity can severely restrict plant growth. Yet Reaumuria soongorica can tolerate salinity well. However, large-scale proteomic studies of this plant’s response to salinity have yet to reported. Here, R. soongorica seedlings (4 months old) were used in an experiment where NaCl solutions simulated levels of soil salinity stress. The fresh weight, root/shoot ratio, leaf relative conductivity, proline content, and total leaf area of R. soongorica under CK (0 mM NaCl), low (200 mM NaCl), and high (500 mM NaCl) salt stress were determined. The results showed that the proline content of leaves was positively correlated with salt concentration. With greater salinity, the plant fresh weight, root/shoot ratio, and total leaf area increased initially but then decreased, and vice-versa for the relative electrical conductivity of leaves. Using iTRAQ proteomic sequencing, 47 177 136 differentially expressed proteins (DEPs) were identified in low-salt versus CK, high-salt versus control, and high-salt versus low-salt comparisons, respectively. A total of 72 DEPs were further screened from the comparison groupings, of which 34 DEPs increased and 38 DEPs decreased in abundance. These DEPs are mainly involved in translation, ribosomal structure, and biogenesis. Finally, 21 key DEPs (SCORE value ≥ 60 points) were identified as potential targets for salt tolerance of R. soongolica. By comparing the protein structure of treated versus CK leaves under salt stress, we revealed the key candidate genes underpinning R. soongolica’s salt tolerance ability. This works provides fresh insight into its physiological adaptation strategy and molecular regulatory network, and a molecular basis for enhancing its breeding, under salt stress conditions.

Soil salinization is one of the main environmental factors limiting the sustainable development of agriculture and forestry worldwide 1 . Salinization impairs the productive potential of soil, destroys the habitat of plants, reduces the diversity of communities, and disrupts the ecological chain, which leads to the degradation or loss of ecosystem functions. According to recent statistics, the total area of saline-alkali land globally has reached 954 million hectares, with an annual expansion rate of 10% 2,3 . Especially at risk are arid and semi-arid areas, where climate conditions characterized by little precipitation and strong evaporation further promote the accumulation of salt in the soil surface. Therefore, it is imperative that researchers cultivate new varieties of salt-tolerant plants and use more salt-tolerant plants in regional cultivation.
To cope with the ion toxicity and osmotic stress caused by salt stress, plants have evolved a suite of adaptive mechanisms to minimize such damage to cells and tissues 4 . These mainly consist of morphological adaptations 5 , regulation of osmotic substances 6 , defensive functioning of the antioxidant enzyme system 7 , changes in the photorespiration pathway 8 , and ion zone isolation in cells 9 . In addition, some salt-secreting plants can form salt glands and use them or vesicles to secrete excess salt out of the body to avoid a large accumulation of salt ions 10 . In recent years, with rapid advances in proteomics technology, the proteomics approach has provided a powerful shortcut to predict the response of plants to salt stress conditions. Using proteomics technology to reveal differences in protein expression of plants under salt stress is now a research hotspot in the post-genomic era. Previous studies have analyzed the protein composition and change of cell and subcellular structures in different tissues and organs of salt-stressed plants, such as in Lobular 11 , okra 12 , rice 13 , alfalfa 14 , and licorice 15 , thereby uncovering many salt-reactive proteins conferring salt stress-resistance traits. These include aquaporins, ribosomal protein, heat shock proteins protein kinases, ornithine decarboxylase, ascorbate peroxidase as well Table 1, when compared with control A (i.e., 0 mM NaCl), both the fresh weight and root/shoot ratio of R. soongorica in group B (i.e., 200 mM NaCl) were significantly higher. However, both fresh weight and root/shoot ratio gradually decreased in group C (i.e., 500 mM NaCl). When the NaCl concentration reached that of group C (i.e., 500 mM NaCl), the growth of R. soongorica was significantly inhibited. The fresh weight of above-ground and root tissues was respectively 43.82% and 50.99% that of the control, and these differences were significant (P < 0.05). Under the NaCl treatment with the B concentration, the water content of the above-ground and root tissues, as well as the total leaf area of leaves, exceeded that of the control. However, when the NaCl concentration was C, the water content of above-ground and root tissues were significantly lower than those of the control.
Effects of NaCl concentrations on relative conductivity and proline content of R. soongorica leaves. Under salt stress, Fig. 1-I shows the responsive changes in the relative conductivity and proline content of R. soongorica leaves. The relative conductivity decreased at first and then increased with an increasing NaCl stress concentration, and the differences were statistically significant. Meanwhile, leaf proline content also increased significantly ( Fig. 1-II). These results indicated that R. soongorica could adapt to a salt stress environment by adjusting its leaf-level proline content. Under low salt stress, the cell membrane system of R. soongorica leaves was not damaged by stress, and its cell membrane had strong stability and could adequately adapt to a certain salt environment.

Number of differentially expressed proteins (DEPs).
Compared with group A, 47 DEPs were obtained from group B, of which 36 proteins were up-regulated and 11 proteins were down-regulated. Compared with group A, 177 DEPs were obtained from group C, with 126 of them up-regulated and the other 51 proteins down-regulated. Compared with group B, 136 DEPs were obtained from group C: 67 and 69 that were up-and down-regulated, respectively ( Fig. 2-I). The identified proteins with significantly different expressions were statistically analyzed, and a certain number of proteins were found common among the three groups, as depicted in Fig. 2-II. Evidently, different proteins appeared in the three groups, and there were 12, 83, and 65 specific proteins in the A vs. B, A vs. C, and B vs. C comparison groups, respectively. In both A vs. B and A vs. C groups, 31 differential protein sites were found. Among these, 27 differential protein sites were up-regulated and 4 were down-regulated. In the A vs. B and B vs. C groups, there were 8 differential protein sites, 3 up-regulated and 1 down-regulated, while the other four DEPs showed opposite expression patterns in the two comparison groups. There were 67 differential protein sites in both A vs. C and B vs. C groups: 37 were up-regulated and 30 down-regulated. These results indicated significant differences in protein expression occurred between low salt (B) and high salt (C) conditions in R. soongorica seedlings. Notably, 83 proteins were only expressed in A vs. C Hierarchical clustering analysis for DEPs under salt stress. As seen in Fig. 3, each column in the graphic represents a sample, and each row represents a protein; color represents the relative expression level of a given protein in the group of samples. On the left is the tree of protein clustering: the closer the branches of two proteins are, the closer their expression levels are, namely, the closer the trends in their variation. By analyzing the up-regulation and down-regulation of different proteins in different sample groups, we can tell that the similarity between the three repeated samples in each group is very high, which would support screening the DEPs accordingly. Figure 4 shows that under the salt stress response, DEPs are involved in different biological processes. These included RNA processing and modification, chromatin structure and dynamics, energy   www.nature.com/scientificreports/ production and conversion, amino acid transport, nucleotide transport, and metabolism, among others. There were 29 differential proteins in the A vs. B comparison group that could be annotated and functionally classified by the COG database. These differential proteins were mainly involved in translation, ribosomal structure and biogenesis, function unknown, post-translational modification, protein turnover, and chaperones biological process, of which 21 were up-regulated and 8 were down-regulated. In the A vs. C comparison group, 112 differ-  www.nature.com/scientificreports/ ential proteins were annotated and functionally classified by COG database, these chiefly involved in translation, ribosomal structure and biogenesis, function unknown, general function prediction only, biological process of energy production and conversion, of which 78 and 34 respectively were up-and down-regulated. There were 86 differential proteins found in the B vs. C comparison group that could be annotated and functionally classified by the COG database. They were mostly involved in translation, ribosomal structure and biogenesis, general function prediction only, post-translational modification, protein turnover, and chaperones biological process, with 35 up-regulated and 51 down-regulated.

Functional classification of DEPs according to the Clusters of Orthologous Groups (COGs) under salt stress.
Gene ontology (GO) enrichment analysis for DEPs under salt stress. After their GO annotation, differential proteins were classified according to the functional categories of molecular function (MF), cell component (CC), and biological process (BP). Major biological functions performed by the DEPS could be determined by a GO significance analysis. In the A vs. B control group analysis, 276 GO items were obtained (P < 0.05), consisting of 194 BP items, 31 CC items, and 51 MF items, with 14 differential proteins annotated by GO. These DEPs were mainly enriched in translation, response to external stimulus, intracellular structure, ribosome and structural constituent of ribosome, and metabolic process, etc. In the A vs. C control group analysis, 495 GO items were obtained (P < 0.05), namely 274 BP items, 88 CC items, and 133 MF items, with 82 differential proteins annotated by GO. These DEPs were mainly enriched in translation, biosynthetic process, ribosome, membrane, structural constituent of ribosome, oxidoreductase activity, and in other ways. In the B vs. C control group analysis, 390 GO items were obtained (P < 0.05), comprising 213 BP items, 78 CC items, and 99 MF items, with 50 differential proteins annotated by GO. These DEPs were mainly enriched in translation, photosynthesis, cytoplasm, small ribosomal subunit, structural constituent of ribosome, and RNA binding, etc. (Fig. 5).

KEGG pathway enrichment analysis of DEPs.
In the face of salt stress, protein functioning depends on the synergistic action of multiple proteins, resulting in significant changes in terms of their abundance. Pathway analysis can provide a more comprehensive and systematic understanding of the biological process each protein is relevant to, and thus point to and reveal the metabolic network of salt stress. In order to further understand the biological functions of the uncovered DEPs, their KEGG enrichment analysis was performed. These results showed that ( Fig. 6) in the A vs. B comparison group, differential proteins were significantly enriched (P < 0.05) to six metabolic pathways (sesquiterpenoid and triterpenoid biosynthesis, glucosinolate biosynthesis, plantpathogen interaction, ribosome, etc.). The differential proteins of A vs. C comparison group were significantly enriched to 16 metabolic pathways (linoleic acid metabolism, C5-dibasic acid metabolism, porphyrin and chlorophyll metabolism, etc.). Finally, the differential proteins in the B vs. C comparison group were significantly enriched to 13 metabolic pathways (glucosinolate biosynthesis, nitrogen metabolism, SNARE interactions in vesicular transport, etc.).
Screening of key DEPs in leaves of R. soongorica seedlings under salt stress. Based on the COG database, 29, 112, and 86 DEPs in A vs. B, A vs. C, and B vs. C comparison groups could be annotated and functionally classified. Combined with the differential protein interaction regulatory network, protein points with a node connection number > 1 were selected and repeated proteins in each comparison group were integrated to further screen out the 72 DEPs. The abundance of 34 DEPs increased and 38 DEPs decreased under the salt stress treatments. The most varied proteins were involved in translation, ribosomal structure and biogenesis, amounting to 20 of them, of which 17 belonged to the ribosomal protein family (RP). Six ribosomal proteins (RPL2-A, RPL12A, RPS23B, RPS6B, RPL30A, and RPS16C) were up-regulated, while the expression levels of another 11 (RPL5, RPS13, RPS10, RPS2D, RPS11C, RPS20B, RPL21A, RPL27, RPS9, RPL29, and RPL7AA) were down-regulated under the NaCl stress. These results showed that R. soongorica seedlings could tolerate stress by synthesizing and degrading proteins in response to salt stress conditions. Furthermore, some proteins (GUN4, MRL1) with pronounced expression differences but not any reported functions were also found. These will be investigated in planned future work. The score of each protein was calculated by Mascot search software 22 . If the score was more than 60, the protein was considered reliable. Finally, four categories of concern were determined Scientific Reports | (2022) 12:2539 | https://doi.org/10.1038/s41598-022-06502-2 www.nature.com/scientificreports/  www.nature.com/scientificreports/ and their related DEPs artificially grouped: those proteins related to plant energy and metabolism, those proteins associated with photosynthesis, those proteins related to plant defense and stress resistance, and those participating in protein synthesis, processing, and degradation ( Table 2).

Discussion
Effect of NaCl concentration on growth indicators and leaf physiological indexes of R. soongorica seedlings. Salinity is undoubtedly a worsening worldwide problem, being a major abiotic stress affecting the growth, development, and productivity of plants 23 . Nevertheless, there are differences in the mechanisms of tolerance to salt stress among different species of plants. Osmotic stress is a direct response of plants to salt stress, and plants can mitigate their incurred damage by regulating intracellular regulatory substances 24,25 .
In addition, the generation and transport of biomass is also an important factor in assessing the degree of salt stress in plants 26 . Studies have shown that salt stress reduces plant biomass synthesis, and that plant leaves are capable of responding to salt stress via the rapid accumulation of proline [27][28][29] . In our experiment, under the stress condition of low salt B (200 mM NaCl), the fresh weight and root/shoot ratio of R. soongorica seedlings increased, and low salt had a significant promotional effect on their growth. Yet when the NaCl concentration reached C (500 mM NaCl), the seedlings' fresh weight and root-shoot ratio decreased, and their growth was inhibited significantly. These results, which are consistent with those of Nasim et al. 30 for butterfly pea could be explained by the high salt tolerance of R. soongorica in the imposed salinity range; specifically, by it effectively relying on Na + and Cl − accumulation to regulate cell permeability and maintain expansion, and via effective K + homeostasis to maintain stomatal functioning in leaves. The inhibition of seedling growth under high salt conditions is likely due to the limited ability to isolate Na + and Cl − in vacuoles 31 . The proline content in R. soongorica leaves increased significantly with a greater salt concentration, a result consistent with findings of previous studies, possibly because the proline synthesis gene was activated or the expression of proline degradation gene was inhibited under salt stress 28 . Proteins associated with photosynthesis. Photosynthesis is the most important process in plant metabolism 38 .

DEPs in leaves of
Rubisco is a key enzyme in the dark reaction of photosynthesis and plays a central role in carbon fixation 39 . Under salt stress, rubisco protein's expression in Prunus mume leaves was down-regulated 40 , but its expression in Haloxylon salicornicum leaves was up-regulated 41 . We found that rubisco (RCA) expression was down-regulated www.nature.com/scientificreports/ in the A vs. C and B vs. C comparison groups, suggesting that down-regulated rubisco expression may lead to reduced light energy utilization in R. soongorica leaves while under salt stress. In plants, PSAN is the subunit that mediates LHCII energy transfer to the photosystem I (PSI) core and it figures prominently in fostering efficient electron transport from plastocyanin to P700. In eukaryotic photosynthetic organisms, the PSI subunit PSAF is involved in the docking of the electron-donor proteins plastocyanin and cytochrome c6 42 ; however, salt-alkali stress can significantly reduce the binding stability between the subunits of PSI 43 . In our study, the expression levels of both PSAN and PSAF in R. soongorica were decreased under salt stress, perhaps because salt stress inhibits the electron transfer mechanism in its leaves. Photosystem II (PSII) is prone to photoinduced damage; hence, it is continuously repaired to maintain its function. The Psb27 protein interacts with the CP43 subunit of PSII and participates in this repair of PSII, and in Arabidopsis two PSB27-like proteins (PSB27-H1 and PSB27-H2) were found involved in PSII repair 44 . Our study found that the PSB27-1 protein was down-regulated by 0.61 and 0.65 times in the A vs. C and B vs. C comparison groups, respectively, which could be explained by damage to both the donor side and recipient side of PSII, and severe photoinhibition of both PSII and PSI. These www.nature.com/scientificreports/ results suggest that these four photosynthesis-related proteins may play important roles in the leaf response to salt stress.
Proteins related to plant defense and stress tolerance. Pyrroline-5-carboxylic acid reductase (P5CR) is a terminal enzyme that functions critically in proline biosynthesis 45 . To improve their salt tolerance, the leaves of Sorghum bicolor 46 and sweet potato 47 can accumulate large amounts of proline via overexpression of the P5CR protein under salt stress. In our study, the P5CR protein was up-regulated by 1.51 and 3.11 times under low and high salt stress, respectively, which further explains why the proline content increased under salt stress as shown in Fig. 1-II. It has been suggested that proline could induce the expression of responsive genes in response to physiological stress caused by too much salinity. Previous reports have revealed that salt stress induced the expression of FC2 in Arabidopsis leaves 48 . Likewise, in R. soongorica, high salt stress caused the FC2 to increase by 1.68 times. This shows that under high salt conditions, FC2 protein expression can effectively control the transport and distribution of metal in the cell, enabling each leaf cell to reach a steady state equilibrium, which demonstrates a certain tolerance to salt stress. SNARE proteins drive vesicle transport and transport membrane functioning, taking cargo to target sites within and on the cell surface, thereby contributing to cell homeostasis, morphogenesis, and pathogen defense 49 . SYP proteins are a family of QC-SNARE proteins unique to plants. In Arabidopsis, SYP121 interacts with KAT1 and KC1 (K channels) to regulate K + currents in the plasma membrane 50 . The up-regulated expression of SlSYP51.2 protein is known to enhance tomato plant's tolerance of salt stress 51 . In our study, both SYP71 and SYP131 proteins were up-regulated under high salinity (A vs. C), suggesting these proteins may similarly regulate the activity of metal ion channels and thus improve R. soongorica's tolerance of salt stress. www.nature.com/scientificreports/ Protein synthesis, processing, and degradation. Ribosome synthesis can trigger the nucleolar stress response and activate p53, thus maintaining the stability of the intracellular environment 52 . The ribosome consists of two parts, the case ribosomal protein (RP) and ribosomal RNA. The RP not only maintains the configuration of RNA, but also participates in the synthesis, transport, and localization of proteins 53 . Salinity reduced the ribosomal protein expression abundance in creeping bentgrass 54 . Li et al. 55 found that for upland cotton under salt stress, the abundance of two ribosomal proteins decreased whereas that of two ribosomal proteins increased. Our data showed the abundance of three ribosomal proteins (RPL2-A, RPL12A, and RPS6B) increasing significantly under low salt stress, which suggests the overall protein synthesis level of R. soongorica seedlings increased under conditions of low salt stress and promoted this plant's growth. Under high salt stress, the abundance of two ribosomal proteins (RPL2-A and RPL12A) increased significantly, and the abundance of four ribosomal proteins (RPS10, RPS2D, RPS9 and RPL7AA) decreased significantly. Yet at the same time the expression of a peptide chain releasing factor (AT2G47020) was significantly up-regulated, indicating that the overall protein synthesis level of R. soongorica seedlings is decreased under conditions of high salt stress. The differential regulation of different ribosomal proteins in the translation mechanism suggests that R. soongorica seedlings cope with high salt stress by balancing ribosomal proteins' synthesis and degradation. In April 2020, uniform full-sized seeds were selected and disinfected for 8 min with 1% NaClO and rinsed six times with ultra-pure water. Cleaned seeds were then sown in a plug tray (8.5 cm height × 4.5 cm diameter, with drainage holes at the bottom) filled with vegetative soil, quartz sand, and vermiculite (3:1:1), and sprinkler-irrigated with underground water. They were cultivated in a greenhouse at a temperature of 25 ± 1 °C under 50% humidity with ventilation and natural light. In July 2020, local farming soil was selected for use in the pot experiments; this soil type is that of irrigated desert soil. Before transplanting, an intelligent soil nutrient analyzer (TPY-6A, Top Yunong, Zhejiang, China) was used to determine the available phosphorus, ammonium nitrogen, salinity, and pH of the tested soil, which were 26.6 mg/kg, 10.0 mg, 0.2%, and 8.3, respectively. Next, uniformly germinated seedlings were transferred into plastic pots containing 2.5 kg of soil (pot dimensions: 23 cm wide at the top, 13 cm wide at the bottom, and 14 cm in height). Intelligent watering control systems were used to maintain the soil water content close to field capacity (i.e., 60%). The salt treatments were applied after 1 month of slow seedling. Seedlings were thinned to four plants per pot, with five pots in each group to which 0 (control, A), 200 (low salt, B), or 500 (high salt, C) mM NaCl was supplied, for a total of three treatment groups (15 pots, 60 seedlings in all). To reduce measurement error, each group was tested three times (Table 3). According to the experimental design, the corresponding NaCl solution was prepared with deionized water, and using a syringe it was evenly poured around the root system of R. soongorica plants. To avoid osmotic shock in seedlings caused by a salt shock reaction, the target concentration was reached over a 24-h-period via gradual salt applications. The NaCl treatment for 24-48 h was set as the NaCl treatment at day 0, and the relevant indexes were measured after 3 days.

Morphological and physiological indexes determination. Determination of plants' wet weight, dry
weight, and ratio of roots to shoots. The 10 plants from each treatment group were washed with distilled water and dried with absorbent paper. We then counted their main stem, branches, and leaves above the rhizosphere, as the above-ground plant parts. The fresh weight (FW) of roots and above-ground parts of the plants was obtained and their mean values per plant calculated. Then they were dried-at 105 °C in an oven for 30 min, and www.nature.com/scientificreports/ dried again to constant weight at 80 °C-and reweighed. We calculated their root/shoot ratio (R/T) = dry weight of underground part/dry weight of above-ground parts.
Measurement of a leaf 's total surface area. The total surface area of single plant leaf was measured with a leaf area analyzer (LI-3000C, Legol Tech, Beijing, China). The area of two clean plastic sheets was first measured to eliminate error. The R. soongorica leaves were removed with scissors and tweezers, and individually placed between the two plastic sheets and passed under the scanning head, to obtain their total surface area.
Determination of leaf relative conductivity and proline content. The relative conductivity of each sample was measured according to Hu et al. 's 56 measurement method, with some modifications applied. We weighed 0.3 g of R. soongorica fresh leaves, rinsed off any surface stains, dried them with absorbent paper, placed them in a 50 ml conical bottle, added 20 ml of deionized water, and placed them at room temperature for 3 h, next we measured the conductivity of the solution with a conductivity meter (DDS-W, Bant instrument, Shanghai, China), recorded that, then put them into a thermostatic water bath at 100 °C, and let it boil for 15 min, cool down, and finally determined the conductivity of the solution as R2. The relative conductivity (%) in leaves was calculated as conductivity/the final conductivity × 100%. The proline content of the R. soongorica leaves was determined as described by Sajid et al. 57 .
Analysis and identification methods of differential proteomics in leaves of R. soongorica under salt stress. The differential proteomics analysis and protein identification conducted for leaves of R. soongorica followed the described methods of Kumaravel et al. 58 , Holáet et al. 59 , and Kumar et al. 60 , with some modifications introduced. The preparation of proteome libraries and their deep sequencing were both performed by the Naomi Metabolic Technology Corporation (Suzhou, China).
Protein extraction and quantification. For this, a given leaf sample was taken from ultra-low temperature refrigerator (-80 °C), ground it into fine powder in liquid nitrogen at low temperature, and put it into an EP tube. Then a 100-μL powder subsample was added into a new EP tube, to which 500 μL methanol was added, mixed well, the allowed to rest on ice for 10 min, after which it was centrifuged (14 000 × g, 4 °C) for 5 min and the ensuing supernatant removed (repeated four times). Finally, the methanol residue in the precipitate was cleaned with 1 × PBS, centrifuged (14 000 × g, 4 °C) for 5 min, and the precipitate then collected. To each sample, 500 μL of an 8 M-urea lysis buffer was added, and this ultrasonicated on ice for 10 min (power: 15%, ultrasonic 3 s, stop 3 s). The protein was quantified by SDS-PAGE electrophoresis.
Trypsin digestion. According to the quantitative results of the electrophoretic diagram, 200 μL of protein lysate was taken from each sample and placed in a centrifugal tube for an enzyme digestion. We added 2 μL of 1 M DTT (DL-Dithiothreitol, Promega, Beijing, China) to each tube, mixed it, heated it at 56 °C for 15 min, and the briefly spun it by centrifuging, after which it was cooled to room temperature. Each sample was divided into four tubes: 50 μL per tube, to which 150 μL of 50 mM ABC (Vectastain ABC Kit, Jinpan, Shanghai, China) was added to attain a 2 M-urea concentration. Next, we added 1.5 μg trypsin (Putai, Hangzhou, China) to each tube, mixed it with a pipette gun, cut by enzymatic at 37 °C for 4 h, then added 1.5 μg trypsin, cut by enzymatic at 37 °C overnight. After this enzyme digestion process, 20 μL 10% FA (Fisher Scientific, A117-50, Fisher, America) was added and centrifuged at 14 000 × g, from which supernatant was removed for desalting. Finally, one sample was taken from each group, and each analyzed by liquid chromatography-tandem mass spectrometry for 1 h.
Peptide fragment labeling and fractionation. The labeled peptide was dissolved in a 0.1 M TEAB (triethylaminecarbonate) buffer (pH = 8.5). We took out a tube containing 0.1 mg of the TMT (TMT10plex™, Thermo Scien-tific™, America) reagent, added 12 μL of acetonitrile, vortexed it for 10 s, then placed it at room temperature for 5 min, and the repeated the vortex (3 times) to ensure the TMT reagent was fully mixed in. Then 10 μL of the peptide (10 μg) was removed and added to the corresponding TMT reagent, according to labeling information table (Table 4). Different samples were labeled with a different labeling reagent, and these thoroughly mixed by vortexing, and allowed to sit at room temperature for 1 h. After that, the labeling reaction was stopped. A 1-μg sample was taken from this mixture and mixed with 150 μL of 1% FA for desalting. The mass spectrometry method was used for detection, for which the labeling efficiency must be higher than 95% to reach the standard. The remaining mixture samples were stored at -80 °C to be separated. The eluent was mixed into two fractions by the strong cation exchange (SCX) method, and then the two fractions were added to different C18 reversedphase columns. The peptides were eluted by an 80-μL CAN (cerium ammonium nitrate, Fisher Scientific, Pittsburgh, America) solution. Finally, the eluted peptides were mixed into six fractions and stored at -80 °C for the mass spectrometry detection. www.nature.com/scientificreports/ Mass spectrometry analysis. Orbitrap Fusion Lumos (Thermo Fisher Scientific, USA) mass spectrometer was used for the data collection. The ion source was nano current electrospray ion source (NSI) with a spray voltage of 2200 V and an ion transport capillary temperature of 320 °C. The mass spectrometry data were collected in the positive ion mode by data-dependent acquisition mode (DDA). Orbitrap was used for full scans at level 1. Secondary mass spectrometry acquisition was performed by fragmentation of parent ions with 38% high-energy collision dissociation (HCD), and the ensuing fragment ions were detected in Orbitrap.
Qualitative and quantitative analysis of proteins. After the completion of MS scanning, the total ion flow chromatogram of MS signal was obtained. After the mass spectrum data were inputted into Proteome Discoverer software (PD) (v2.2, Thermo Fisher Scientific), the software first screened the mass spectrum. This mass spectrum data was searched via the Sequest operation program embedded in the PD software. The same software conducted a qualitative analysis according to the Sequest search results and the spectrum (after the first screening step). By extracting the signal value of TMT reporting ions, the protein quantification value was recalculated, this represented here by its median, where the protein quantification value was the sum of the obtained peptide quantification values.
Statistical and bioinformatics analysis. Microsoft Excel 2016 was used for data processing, and Origin 8.0 software was used for plotting. SPSS 22.0 analysis software was used for analysis of variance (ANOVA). The functional enrichment database did not provide the enrichment analysis of R. soongorica leaves. Accordingly, the identified protein sequences were compared with the background libraries of GO and KEGG via BLAST searches. For their relative quantification, protein expression abundance was set to determine the statistical significance of the difference and accurately identify the DEPs induced by salt stress. When the protein difference multiple was > 1.5 and its P-value < 0.05, it was considered an up-regulated protein; when the difference multiple was < 0.66 and the P-value < 0.05, the protein was designated as down-regulated. The functions of all identified proteins were determined by searching GO (Gene Ontology) analysis in UniProt database (http:// www. unipr ot. org), and proteins were classified according to their main functions. We used the string-DB (http:// string-db. org/) protein interaction database (selecting Arabidopsis thaliana) to analyze the interaction of the compared and differentially expressed proteins.
Ethics approval and consent to participate. The

Conclusion
As a salt-secreting plant, R. soongorica underwent a series of changes in its growth and differentially expressed proteins in its leaves while under controlled salt stress conditions. In terms of its growth indexes, low salt (200 mM·L -1 ) significantly promoted the vegetative growth (total leaf area, total fresh weight, root shoot ratio) of R. soongorica seedlings, while increasing the proline content of their leaves. Proteomic analysis revealed that energy-and metabolism-related proteins (P5CR, CSY4) and ribosomal proteins (RPL2-A, RPL12A, RPS6B) were up-regulated under low salt stress. However, the growth of R. soongorica seedlings was significantly inhibited under high salinity (500 mM·L -1 ). The reason for this may be that high salt decreases the abundance of proteins associated with photosynthesis (RCA, PSAF, PSAN, PSB27-1), ribosomal proteins (RPS10, RPS2D, RPS9, RPL7AA) yet it increases the abundance of a peptide chain-releasing factor (AT2G47020). Meanwhile, R. soongorica may respond to a high salt stress environment by up-regulating the expression of proteins related to energy and metabolism (VHA-A, VHA-E1, AHA4, CSY4, PCK1), defense and anti-stress related proteins (P5CR, FC2, SYP71, SYP131) and ribosomal proteins (RPL2-A, RPL12A) in its leaves. These proteins play an important role as a potential target protein conferring the salt tolerance ability of R. soongorica. This study lays a foundation for better understanding the molecular regulation mechanism underlying the salt stress response of R. soongorica.