The potato (diploid cultivar Solanum phureja, DM1-3 516 R44) plants used in this study were obtained from Institute of Vegetable and Flowers, Chinese Academy of Agricultural Sciences (CAAS). The potato was grown in a growth chamber at 26 °C/18 °C (day/night) with a 16:8 light: dark cycle and 60-70% relative humidity according to (Ali et al., 2014). The roots of 7-8-leaves-old plantlets were watered with 200 mM NaCl solution (Ma et al., 2021). And the leaves were sprayed with 100 μM ABA, 50 μM MeJA, 350 μM GA and 50 μM SA, respectively. When spraying, moisten the positive and negative sides of all leaves with condensed water droplets without dropping. After the spraying, the plants were immediately wrapped in black plastic bags and treated only once (Yu et al., 2021). Then, the 1, 2, 3, 4 and 5 d (0 d as control) treated plant leaves were respectively quickly frozen in liquid nitrogen at -80 °C for later use (Li et al., 2021). And each treatment was repeated three times.

All protein sequences were obtained from potato genome data (SolTub_3.0) ¹ . First, the HMM profile for the SOS1s domain (PF00999) was downloaded from the Pfam server ² . Then, the HMMER program ³ was used to identify the SOS1 proteins in the potato genome (Liang et al., 2017). Finally, the SOS1 (Na⁺/H⁺ exchanger, NHX) domain of all putative SOS1 proteins were determined through CDD ⁴ and SMART databases ⁵ . A total of 37 putative SOS1 genes were identified.

Biophysical characteristics of SOS1 proteins were analyzed through ExPASy webserver ⁶ (Wang T. et al., 2021) and NetPhos 3.1 ⁷ (Naureen et al., 2023). The online prediction tool UniProt ⁸ (Ilzhöfer et al., 2022) was applied to predict the tertiary structures of potato SOS1s. Subcellular location of protein was predicted using the Cell-PLoc 2.0 prediction tool ⁹ . The physical positions of the StSOS1s along each chromosome were identified from the potato genome database and the distribution of StSOS1s was plotted (Xiang et al., 2016).

The 2000 bp upstream region of the ATG start codon was submitted to PlantCARE ¹⁰ (Koul et al., 2019) to identify the cis-acting elements and calculate the number of each element. These promoter sequences were represented as word clouds with the help of the WordArt tool ¹¹ (Sharma et al., 2021).

The conserved motifs in StSOS1s were identified to use the MEME website ¹² (Multiple Em for Motif Elicitation) (Zhang et al., 2021) with the maximum number of motifs was set to 10. Figures of phylogenetic tree along with gene conserved motifs and CDS/UTR structure of StSOS1s were drawn with TBtools (v1.098) (Chen et al., 2020) software. Gene Structure Display Server (GSDS) ¹³ (Sun et al., 2022) and MEME webserver were employed for gene structure analysis.

RNA-Seq data (fragments per kilobase of exon per million mapped, FPKM) (NCBI accession number ERP000527) in potato DM genotype (Wang J. et al., 2021) was used to analyze the expression level of StSOS1 genes. PlantRegMap ¹⁴ (Li H. et al., 2020) was used to functionally re-annotate the proteome of up or down-regulated genes and to plot gene ontology (GO) annotations. Protein-protein interaction (PPI) enrichment was computed by STRING ¹⁵ (Fayez et al., 2022) tool, in which Cytoscape software was used for reconstructing the PPI network, modules and to detect the relationship between overall targeted genes.

The SOS1 protein sequences of Arabidopsis, tomato, pepper and tobacco were downloaded from the EnsemblPlants (Contreras-Moreira et al., 2022). Homologous sequences were fed into the MEGA7 software and the Clustalw program was used to perform multi-sequence alignment. The results of the output multi-sequence alignment were used to construct an evolutionary tree using the proximity method (He et al., 2022). The collinearity of the sequences of potato with other four species was extracted using TBtools (Zhang C. et al., 2022).

The leaves samples were ground into powder in liquid nitrogen, total RNA was extracted using TransZol Up Plus RNA kit (Trans, Beijing, China), following the manufacturer’s instruction. Then the extracted RNA was employed as a template with TransScript ^® One-Step gDNA Removal and cDNA Synthesis SuperMix for qPCR (Trans, Beijing, China) for the first strand cDNA synthesis. All primer sequences used in this study were designed by Primer Blast website ¹⁶ of NCBI ( Table S1 ). The RT-qPCR was performed on a QuantStudio-3 system (Thermo Fisher Scientific, Shanghai, China). The reaction system was 20 µL (cDNA 1 µL, SOS1-F 0.4µL, SOS1-R 0.4µL, SuperMix 10 µL, DyeII 0.4µL, Water 7.8 µL). The reaction system was 94 °C 30 s, (94 °C 5 s, 60 °C 30 s) ×40 Cycles. Three replications were performed and the expression values were calculated by using the 2^−ΔΔCT method (Mo et al., 2022).

For the localization and expression of StSOS1-13 in potato, the CDS without the stop codon was cloned into pCAMBIA1300. Firstly, the complete coding region of StSOS1-13 (1 734bp) was amplified from the cDNA by PCR using a pair of primers with a homologous arm and inserted into the pCAMBIA1300 vector linearized by the restriction enzyme NcoI. Then, the obtained pStSOS1-13-GFP fusion plasmid was converted into Escherichia coli DH5α for verified by bacterial liquid PCR and company sequencing (Sangon, Shanghai, China), further inserted into individual Agrobacterium tumefaciens strain GV3101 cells and a single colony was selected for PCR positive identification. Finally, the expression vectors were injected into tobacco leaves for the transient expression experiments (Luo et al., 2022). GFP expression was analyzed using scanning confocal laser microscopy.

To identify the SOS1s family members in potato, the similar protein sequences were searched in the HMMER program with the query sequence SOS1s motif (PF00999). The SMART tool was then used to confirm whether the candidates contained the Na⁺/H⁺ exchanger (NHX) domain. In total, 37 SOS1 genes were retrieved from the potato genome and renamed StSOS1-1 to StSOS1-37 based on their relative linear order on each chromosome, following the widely used nomenclature ( Figure 1 ). Meantime, we found four pairs of tandem duplicated genes existed in 37 StSOS1 genes. The analysis showed that there was one pair of tandem duplicated genes (StSOS1-2 and StSOS1-3) on Chr1, one pair (StSOS1-7 and StSOS1-8) on Chr2, one pair (StSOS1-26 and StSOS1-27) on Chr6, and one pair (StSOS1-30 and StSOS1-31) on Chr9.

We further determined the biophysical properties of the potato SOS1 genes including the locus ID, protein length (aa), predicted protein molecular weight (MW), isoelectric points (pI), and NHX domain. The statistical results showed that the protein length ranged from 209 (StSOS1-15) to 1153 (StSOS1-1) amino acids, the average amino acids length and molecular weights ranged from 22.51 KDa (StSOS1-15) to 127.86 KDa (StSOS1-1). PI varying from 4.96 (StSOS1-20) to 10.12 (StSOS1-3). The subcellular localization of these StSOS1s predicted through Cell-PLoc 2.0 tool revealed that most of the StSOS1 proteins were localized in the plasma membrane ( Table 1 ). The results of the NetPhos 3.1 server revealed that StSOS1 proteins were phosphorylated, and phosphorylated residues were Serine (Ser), threonine (Thr) and tyrosine (Tyr) ( Table S2 ), among which serine prediction sites ranged from 11 (StSOS1-24) to 72 (StSOS1-1). The threonine prediction sites ranged from 5 (StSOS1-15) to 32 (StSOS1-1), and the tyrosine prediction sites ranged from 0 (StSOS1-15 and StSOS1-20) to 9 (StSOS1-21). Three-dimensional protein models were constructed by sequence similarity search using UniProt PDB database and the homology modeling was predicted by DS Visualizer ( Figure S1 ). The structures of StSOS1-7, StSOS1-8, StSOS1-9, StSOS1-16, StSOS1-17, StSOS1-21, StSOS1-22, StSOS1-25, StSOS1-28, StSOS1-29, StSOS1-31, StSOS1-32, StSOS1-36, and StSOS1-37 are similar and suggest shared functionality, as do StSOS1-18 and StSOS1-19. These provide an initial basis for understanding the molecular function of the StSOS1 proteins.

To clarify which hormonal, environmental stress, or developmental-related signal elements are involved in these StSOS1s, we performed a promoter analysis using the PlantCARE server. A large number of basic components were discovered in the upstream sequence (2000 bp) regions, including WRE3, GATA-motif, CAT-box and G-Box, but also P-box, TCA-element, AuxRR-core, TGACG-motif, ABRE and ERE hormonal response-related elements; as-1, LTR, ARE, GC-motif, MBS environmental stress-related components and A-box development-related elements ( Figures 2A, B ). Hormonal response elements were detected in the promoters of 37 potato StSOS1 genes, including 15 SA, 19 MeJA, 26 ABA and 30 auxin response. The cis-elements involved in the GA response are present in all promoters of StSOS1s. The promoters of 10, 16, and 20 StSOS1 genes contained MYB binding sites involved in low-temperature response, defense and stress response cis-elements and drought-inducibility, respectively ( Figure 2C ). These results suggest that the StSOS1 genes may play a critical role not only in phytohormones, but also in biological and abiotic responses in the potato.

In order to better understand the relationship between the structure and function of these StSOS1 proteins, gene structure and conserved motifs were analyzed to construct individual phylogenies. Depending on the different branches of the evolutionary tree, it has been found that the motif architectures remain consistent within the same evolutionary branch, and thus they may have a similar function ( Figures 3A, B ). The results showed that the number of intron in StSOS1 genes ranged from two (StSOS1-22, StSOS1-23, StSOS1-34) to 20 (StSOS1-13, StSOS1-26, StSOS1-27). Furthermore, closely related genes share a similar structural architectures with different introns lengths ( Figure 3C ). The shortest StSOS1 protein was just 209 aa in length (StSOS1-15), while the longest was StSOS1-1, with a length of 1153 aa ( Table 1 ). The functional sites in the conserved motifs were analyzed using the Eukaryotic Linear Motif resource server (ELM) and the results showed that there was a great functional divergency among these sites and most of the functional sites are related to phosphorylation, kinase phosphorylation, binding and sorting signal responsible for the interaction ( Table S3 ).

To investigate the biological function of StSOS1s in different tissues, expression profiles of all identified StSOS1 genes were analyzed in six different tissues, including roots, tubers, stolons, leaves, whole mature flowers, and mature whole fruit ( Figure 4A ). Of all the 21 StSOS1 genes, StSOS1-2 exhibits the highest levels of expression in almost all the tissues except the tubers. Some members of StSOS1 exhibit highly tissue-specific expression, such as the expression of StSOS1-10, StSOS1-16, StSOS1-17, StSOS1-19, StSOS1-22, StSOS1-23, StSOS1-32, and StSOS1-35 throughout the mature flower, suggesting that the StSOS1 genes exhibit differential tissue-specific expression patterns. Then we analyzed spatio-temporal expression patterns in stolon, tuber pith, tuber peel, tuber cortex, young tuber, mature tuber and tuber sprout using RNA-seq data ( Figure 4B ). It showed that two genes (StSOS1-14 and StSOS1-32) had a very low abundance in these tissues or organs. StSOS1-16, StSOS1-19, and StSOS1-31 were predominantly expressed in stolon; StSOS1-6 and StSOS1-13 were predominantly expressed in tuber sprout. StSOS1-1 was highly expressed in tuber pith, tuber peel, tuber cortex, young tuber and tuber sprouts. To have a better understand the function of StSOS1s under biotic stress, the expression pattern was observed responding to Phytophthora infestans, β-aminobutyric acid (BABA) and benzothiadiazole (BTH) treatment ( Figure 4C ). StSOS1-2 was the only member to exhibit down-regulation under all three biotic stress conditions. Some genes show up-regulation, in particular one type of stress treatment; StSOS1-6, StSOS1-13, StSOS1-28 and StSOS1-29 showed up-regulation only in response to BABA treatment. For the abiotic stresses and phytohormones responsiveness of StSOS1s, we analyzed their transcript profiling in response to three abiotic stress and four phytohormone conditions mannitol, water-stress, heat, IAA, GA₃, BAP and ABA ( Figure 4D ). StSOS1-6 was found to be highly up-regulated in the three stress conditions of mannitol, water-stress and ABA. StSOS1-10, StSOS1-17, StSOS1-22, StSOS1-23, StSOS1-32, and StSOS1-35 showed low or no expression in the eight tissues.

To identify functions of up and down-regulated genes, GO analysis was performed and genes belonging to different categories of Biological Processes (BP), Molecular Functions (MF) and Cellular Compartments (CC) were identified ( Figure 5 ). The BP categorized results showed that the up-regulated genes were significantly enriched in transport and cellular process. For MF, these up-regulated states enriched in transport activity. Moreover, up-regulated genes in the CC category are significantly enriched in both membrane and membrane-like components. In addition, the most significantly enriched GO terms for down-regulated genes were detection of hydrogen transport (BP), and transporter activity and antiporter activity (MF). It is important to note that the membrane integral, the membrane intrinsic and membrane fraction are all present in both up- and down-regulated genes in CC. The difference is that the up-regulated genes have a late endosome while the down-regulated genes do not. In summary, most GO terms are involved in membrane transport and composition, suggesting that they are likely to play an important role in maintaining proper ion homeostasis in the cytoplasm.

The verification of PPI is a defining aspect of molecular biology. PPI analysis was conducted to analyze the interactions among the SOS1s ( Figure S2 ). The biological pathways and cellular compartments (retrieved from the GO) associated with these proteins were similar. Here, the interaction network between 96 SOS1-related genes was also mapped using the STRING database and Cytoscape software for function analyse, seven clusters were identified, including the pathways of biological regulation, membrane, ion transmembrane transporter activity, calcium ion binding, potassium ion transmembrane transport, response to salt stress and cellular process ( Figure S3 ). Only 25 StSOS1s interact with other genes, and the most PPI was observed between proteins involved in potassium ion transmembrane transport, response to salt stress and cellular processes. These studies inform the biochemical mechanism of StSOS1 and provide a new reference for the interplay between ion homeostasis and transmembrane transport during plant salt tolerance.

For the evolutionary relationship of SOS1s among Arabidopsis, tomato, pepper, potato and tobacco, we extracted and compared the protein sequences of SOS1s in these species, and constructed the phylogenetic tree of neighbor junction (NJ) ( Figure 6A ). Potato SOS1s are named based on their position relative to orthologs from four other species on the tree. 134 SOS1 candidates of five species were grouped into four distinct classes (I-IV) based on sequence conservation. Among them, the subgroup I had 13 members (11.19%), subgroup II 27 (20.14%) and subgroup III 37 (27.61%), respectively. The subgroup IV contained 57 genes and had the most members (42.54%). The phylogenetic relationships indicate that the SOS1 proteins in the potato are more strongly homologous to pepper and tomato than to Arabidopsis and tobacco. Gene duplication has always played a key role in the expansion of genes and the occurrence of novel functions of genes. To explore the evolution of SOS1 genes, we studied the replication patterns of the five species and performed genetic correlation analysis ( Figure 6B ). The results showed that there were 17, 8, 5, and 1 SOS1 members participating in the potato-tomato, potato-pepper, potato-Arabidopsis and potato-tobacco synteny relations, respectively. Among the above collinear gene pairs, StSOS1-11 with SlSOS1-23, CaSOS1-10, AtSOS1-58, respectively; StSOS1-28 with SlSOS1-26, CaSOS1-1, AtSOS1-47, respectively; and StSOS1-37 with SlSOS1-26, CaSOS1-37, AtSOS1-55, respectively, had simultaneously collinear relations.

The SOS pathway plays an important role in maintaining proper ion homeostasis in the cytoplasm and in regulating plant tolerance to salinity. However, there is limited information on SOS1’s response to potato salt stress. In order to investigate the potato response to salt stress, the StSOS1 genes were analyzed using the transcriptomic data of potato exposed to NaCl treatment. Only 21 StSOS1 genes showed differential gene expression pattern and were identified and visualized in a heat map ( Figure 7A ). Furthermore, six StSOS1 genes in potato leaves of different grow stages under salt stress were randomly selected and quantitative analyzed by RT-qPCR ( Figures 7B–G ). These results suggested that these six genes were significantly differentially up-regulated under salt stress, which may positively regulate salt tolerance in the potato, this is not consistent with the heat map, which may be related with different levels of expression under different levels of salt stress treatment. StSOS1-2, StSOS1-6 and StSOS1-28 occurred two up-regulated expressions phenomenon under salt stress, this could be related to the response period of the SOS1 signaling pathway. Notably, the expression of StSOS1-13 was 14-fold higher at 3 d after salt treatment compared to expression levels before salt stress, and then reached 32-fold higher at 4 d, suggesting that StSOS1-13 may be an important candidate gene involved in the salt stress response.

To further understand potential function changes in StSOS1-13 gene in response to abiotic stress, RT-qPCR was used to analyze the expression patterns of the selected StSOS1-13 gene in phytohormone treatment ( Figure 8 ). It was observed that the StSOS1-13 was up-regulated on exposure to ABA, GA, and SA treatment, and the magnitude of up-regulation was higher in ABA treatment as compared to GA, SA treatment. Conversely, for the MeJA treatment, expression in the leaves decreased after 0-2 d and then increased continuously, with the highest levels of expression in the leaves at 5 d. Overall, these results indicated that StSOS1-13 may play a critical regulatory role in response to abiotic stress.

Detecting the subcellular localization of StSOS1-13 is essential to elucidate their function. The subcellular localization of StSOS1-13 predicted by the Cell-PLoc 2.0 tool revealed that the StSOS1-13 protein was localized in the plasma membrane. To further verify the location of StSOS1-13 protein, the full-length coding sequence of StSOS1-13 deleted stop codon was fused with green fluorescence protein (GFP) and the transient expression was performed under the control of 35S promoter in tobacco. The results showed that the StSOS1-13 protein is localized in the plasma membrane ( Figure 9 ), this is consistent with the result of bioinformatics analysis.