Eight NHX protein sequences of A. thaliana were obtained from the TAIR database (https://www.arabidopsis.org/) (Philippe et al., 2012). On this basis, the local BLAST program was used to search for NHX proteins in the database of three cultivars of Cucurbita genome (C. moschata, C. maxima, C. pepo) (http://cucurbitgenomics.org/). The NHX candidate proteins must meet the requirements: the protein sequence identity of AtNHX protein with other protein more than 70% and e<10⁻¹⁰. Na⁺_H⁺_Exchanger domain (PF00999) of NHX candidate proteins from Cucurbita were confirmed by PFAM (http://pfam.xfam.org/) (Finn et al., 2016) and SMART (http://smart.emblheidelberg.de/) (Letunic et al., 2012). Theoretical isoelectric point (pI) and molecular weight (MW) of NHX candidates were predicted by ExPASy software (https://web.expasy.org/compute\upi/) (Bjellqvist et al., 1993). The TMHMM server v. 2. 0 software (http://www.cbs.dtu.dk/services/TMHMM/) (Moller et al., 2001)was used to predict the membrane domain of all proteins.

To elucidate the phylogenetic relationships of NHX protein families in Cucurbita moschata, C. maxima, C. pepo, P. trichocarpa and A. thaliana, NHX proteins from P. trichocarpa were firstly extracted from previous literature (Tian et al., 2017). Moreover, NHXs protein sequences in C. moschata (Cmo), C. maxima (Cma), C. pepo (Cp), P. trichocarpa (Pt) and A. thaliana (At) were blast by ClustalW program (Larkin et al., 2007). Finally, a circular phylogenetic tree was constructed using MEGA 7.0 software (Kumar et al., 2016) and the Maximum Likelihood method (Guindon and Gascuel, 2003) was applied with bootstrap value set to 1000 replicates.

To further analyze the conserved domains of NHX family members in Cucurbita, the MEME suite online program (MEME 5.3.3, http://meme-suite.org/tools/meme) (Bailey et al., 2006) was used to analyze and draw the motifs. The operating parameter were set as: the base width was 6-50 aa and the maximum number of motifs was 15.

To clarify the structural characteristics of NHX genes in Cucurbita, the CDS sequences of 26 NHX genes were compared with the corresponding genomic DNA sequences from the corresponding genomic database. Finally, the Gene Structure Display Server (GSDS, http://gsds.cbi.pku.edu.cn/) (Hu et al., 2014)was used to map exon-intron structure of NHX genes.

To clarify the distribution of NHX gene on chromosomes in Cucurbita, we first obtained the starting positions of CmoNHX, CmaNHX, and CpNHX genes on chromosomes from the genome databases of three Cucurbita cultivars, and finally the analysis were performed using TBtools (https://github.com/CJ-Chen/TBtools) (Chen et al., 2020).

To further analyze the collinearity of NHX genes among C. moschata (Cmo), C. maxima (Cma) and C. pepo (Cp), and their collinearity with A. thaliana NHX genes, MCScanX software and Circos-0.69 software (Krzywinski et al., 2009) were used to analyze and visualize the collinearity of NHX genes.

To identify the cis-acting elements of NHX genes in Cucurbita, the promoter sequences (1500 bp sequence upstream of start codon) of all CmoNHXs, CmaNHXs and CpNHXs were extracted from the genome of Cucurbita (http://cucurbitgenomics.org/). The cis-acting elements of these NHX genes were then predicted using the PlantCARE website (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) (Lescot et al., 2002). We analyzed the cis-acting elements related to growth and development, hormone and abiotic stress, with emphasis on the cis-acting elements related to salt stress.

To determine the response of CmoNHXs and CmaNHXs to salt stress, we excavated the transcriptome data (BioProject: PRJNA464060) published in 2018 (Niu et al., 2018) and analyzed the transcription profile of NHXs in the leaf mesophyll and leaf vein of the C. moschata cultivar, “Rifu” and C. maxima cultivar, “Rimu” under salt stress. Sequencing samples were obtained from 100 mM NaCl treated for 24 hours. 0 mM NaCl was used as a control. The expression value was calculated in terms of reads per kilobase of exon model per million mapped reads (RPKM).

We cloned the CmoNHX1 gene’s nucleotide sequence (Table S4) and inserted it into PRI101-GFP to create the recombinant vector PRI101-GFP-CmoNHX1 in order to study the subcellular localization of the CmoNHX1 protein. The recombinant vector, control vector PRI101-GFP and plasma membrane marker: pm-rbCD3-1008 were transformed into Agrobacterium tumefaciens (GV3101) by freeze-thaw method (Jyothishwaran et al., 2007), respectively. According to the method of Wu et al. (2010), CmoNHX1 was transiently expressed in tobacco. After 40 hours of dark culture, laser confocal microscopy was used to detect cell fluorescence.

To construct CmoNHX1 ectopic expression vector, we constructed CmoNHX1 fragment into pTCK303 using recombinant vector PRI101-GFP-CmoNHX1 as template (Table S4), and the recombinant vector was termed as pTCK303-CmoNHX1. Transformation of constructed recombinant vector into GV3101 by freeze-thaw method (Jyothishwaran et al., 2007). In addition, transgenic A. thaliana plants were obtained by floral dip method (Xu et al., 2010). T2 generation transgenic plants were finally obtained through multi-generation selfing and GUS staining.

We used A. thaliana plants that were heterologously expressed in CmoNHX1 from the T2 generation as well as wild type plants to better understand how CmoNHX1 responded to salt stress. In order to irrigate the plants at the seedling stage, a 100 mM NaCl solution was made, and 50 mL of it was poured twice, once every three days, into each hole. Water was used as the control. Every time a symptom appeared, pictures were taken. Following the prior procedure, RNA extraction, reverse transcription, and qRT-PCR were carried out at this time. (Yuan et al., 2019).

A total of eight AtNHX protein sequences from Arabidopsis thaliana as query sequences were used to identify NHX proteins in three cultivars of the Cucurbita genus, including C. moschata, C. maxima, and C. pepo. Using a series of screening procedures, a total of 26 NHX proteins were discovered, which were named CmoNHX1 through CmoNHX9, CmaNHX1 through CmaNHX9, and CpNHX1 through CpNHX8. The naming convention was based on the name of the corresponding AtNHX genes and their position on the chromosome, from the first to last chromosomes, and from top to bottom. More information about these NHX proteins can be found in Table S1 of the study.

Sequence analysis of 26 NHX genes showed that the length of open reading frame in 9 CmoNHXs ranged from 903 bp (CmoNHX6) to 3264 bp (CmoNHX9), and the corresponding number of amino acids obtained by translation ranged from 300 aa to 1087 aa (Table 1). The pI and MW of CmoNHXs were from 5.41 to 9.13 and from 33.328 98 KDa to 122.789 68 KDa, respectively. Nine CmoNHXs contained 5 to 12 transmembrane domains (Table 1). Table 1 also displays the length of the open reading frame, the length of the amino acid, the size of the pI, and the MW in CmaNHXs and CpNHXs. These findings accurately capture the diversity and conservation of NHXs’ biological and structural features.

The study you mentioned constructed a phylogenetic tree of NHX proteins from Cucurbita moschata, C. maxima, C. pepo, Arabidopsis thaliana, and Populus trichocarpa, to better understand the evolutionary relationships within the plant NHX protein family (Figure 1). Based on the amino acid sequence identity of 92%, the NHX proteins were classified into three subfamilies, namely Endo, Vac, and PM. All the NHX proteins were classified into these subfamilies according to the classification of AtNHX proteins. The phylogenetic analysis revealed that the Endo subfamily contained three CmoNHX proteins, three CmaNHX proteins, two CpNHX proteins, two AtNHX proteins, and one PtNHX protein. The PM subfamily included one CmoNHX protein, one CmaNHX protein, and two AtNHX proteins. The Vac subfamily contained the majority of NHX proteins, including five CmoNHX proteins, five CmaNHX proteins, six CpNHX proteins, four AtNHX proteins, and five PtNHX proteins. The study found that the NHX proteins of Cucurbita, Arabidopsis, and Populus were mostly distributed in the Vac subfamily, followed by the PM subfamily, and finally the Endo subfamily. The PtNHX proteins were mostly found in the Vac subfamily, followed by the Endo subfamily, and lastly the PM subfamily. This information provides insights into the evolutionary history and diversification of NHX proteins in plants. The homology of NHXs among the three domesticated species of Cucurbita is significantly higher than that of A. thaliana and P. trichocarpa, when seen from the perspective of evolutionary lineages (Figure 1).

Twenty-six Cucurbita NHX proteins are further grouped into three subfamilies (Endo, PM, and Vac) based on the evolutionary tree (Figure 2A). All 26 NHX proteins were found to possess the conserved Na⁺_H⁺_Exchanger domain (PF00999) after conducting a conserved domain analysis (Figure 2B). Furthermore, we looked at the motifs of 26 NHX proteins and found a total of 15 motifs (Figure 2C; Figure S2). Motif 11 was present in all NHX proteins. Furthermore, motif 7 and 12 were exclusive to the Vac subfamily, and motif 13 was present only in the PM and Endo subfamilies (Figure 2C). Motifs 5 and 2 are almost always present with motif 11. Overall, PM proteins look very similar to the Vac proteins with some domains missing.

A gene’s biological function is significantly influenced by the intron-exon distribution pattern. All NHX genes in the Vac subfamily had 12-16 exons, according to an examination of the intron-exon structure. Exon counts for all NHX genes in the Endo subfamily varied greatly (10-24 exons). There were many exons in the NHX genes of the PM subfamily (20-23). Similar exon numbers and intron lengths can be seen in genes belonging to the same branch, such as CmaNHX4 and CpNHX4, CmaNHX1 and CmoNHX1 (Figure S1). However, certain homologous genes clearly display distinct intron-exon structural variations. As an illustration, CmoNHX2 had 20 exons, but CmaNHX2 had 23 exons and lengthier introns (Figure S1).

The distribution of 26 NHX genes on the chromosomes revealed that 9 CmoNHXs are found on chromosomes Cmo01, Cmo04, Cmo08, Cmo10, Cmo11, Cmo13, Cmo17, and Cmo18 (Figure 3A). The distribution pattern of CmaNHXs and CmoNHX genes is comparable (Figure 3B). On the corresponding chromosomes Cmo13 and Cma13, there are two NHX genes, respectively (Figures 3A, B). The CpNHX genes are primarily found on the chromosomes Cp02, Cp04, Cp10, Cp11, Cp12, Cp17, and Cp20 (Figure 3C), where two genes are located on chromosome Cp17. Differences between the distribution on Cp vs Cmo/Cma reflect larger differences in chromosome synteny.

MCScanX software was used to examine the collinearity and further investigate the evolutionary relationship between the NHX gene families in C. moschata, C. maxima, C. pepo, and A. thaliana. The findings revealed that, between CmoNHXs and CmaNHXs, CpNHXs, AtNHXs, 13, 11, and 5 syntenic gene pairs were identified, respectively (Figure 4; Table S2). In addition, between CmaNHXs and CpNHXs, AtNHXs, a total of 10, 4 syntenic gene pairs were discovered, respectively (Figure 4; Table S2). Two syntenic gene pairs between CpNHXs and AtNHXs were found (Figure 4, Table S2). We discovered that the homology of C. moschata, C. maxima, and C. pepo was substantially higher than that of A. thaliana based on the collinearity analysis of NHX genes in four species.

This study examined the CmoNHX, CmaNHX, and CpNHX promoter sequences and discovered 755 cis-acting elements in the promoter regions of 26 NHX genes. They included CmoNHXs, CmaNHXs, and CpNHXs, which each had 252 cis-acting elements, 249 cis-acting elements, and 254 cis-acting elements. The majority of these cis-acting elements were connected to hormone response, abiotic stress, and growth and development factors (Figure 5; Table S3). We focused on examining the cis-acting elements associated with salt stress because of the significance of the NHX gene under salt stress. Although ABRE, TGA-element, TGACG/CGTCA-motif are related to hormone response, G-box and GT1-motif are related to light response, MBS is related to drought stress response (Figure 5; Table S3), but related studies have found that these cis-acting elements are all involved in the response to salt stress (Yamniuk and Vogel, 2004; Saeediazar et al., 2014). In the present study, we discovered that the CmoNHX1, CmaNHX1, and CpNHX1 promoters contain the most ABRE, G-box, and TGACG/CGTCA elements (Figure 5; Table S3). On the basis of the aforementioned analysis, we hypothesized that the genes CmoNHX1, CmaNHX1, CpNHX1, CmoNHX5, CmaNHX5, and CpNHX5 may be crucial in salt stress.

The different expression patterns of NHX genes in C. moschata and C. maxima suggest that they might have different mechanisms to improve salt tolerance under salt stress. After NaCl treatment, the transcription levels of CmoNHX2, CmoNHX5, CmoNHX6, and CmoNHX7 genes increased greatly in the mesophyll and vein of C. moschata, while the transcription level of CmoNHX1 genes fell dramatically (Figure 6). For instance, after NaCl treatment, the transcription levels of CmoNHX1 in the mesophyll and veins were considerably decreased by 55.44% and 69.04%, respectively, in comparison to the control treatment (Figure 6). In mesophyll, there was no apparent change between the NaCl treatment and the control treatment in the transcript levels of any CmaNHX genes. The transcription levels of CmaNHX6, CmaNHX7, and CmaNHX8 under salt stress were significantly higher than those under control treatment, and the transcription level of CmaNHX1 under salt stress was 51.26% of that under control treatment, despite the fact that CmaNHX2, CmaNHX3, CmaNHX4, CmaNHX5, and CmaNHX9 did not change significantly under salt stress in the veins (Figure 7). These findings imply that NHXs might improve salt tolerance in two cultivars via various expression patterns. Overall, the transcriptional changes of NHX genes in response to salt stress provide insights into the molecular mechanisms of salt tolerance in Cucurbita cultivars.

Eight positive seedlings in the T1 generation were grown from CmoNHX1-heterologously expressed A. thaliana plants in order to investigate the function of CmoNHX1. On the leaves of T2 plants, we carried out hygromycin resistance gene identification (Figure 8A) and GUS staining validation (Figure 8B), after which we arbitrarily selected the OE-1 and OE-3 lines for additional investigation.

The transgenic A. thaliana plants and wild-type were both given NaCl treatments in order to study how the CmoNHX1-heterologously expressed A. thaliana plants responded to salt stress. Under water treatment, the phenotype revealed no discernible difference between transgenic plants and wild-type plants. However, when exposed to salt stress, the leaves of A. thaliana plants heterologously expressing CmoNHX1 clearly yellowed and wilted in contrast to wild-type plants (Figure 8C). No significant difference between the transgenic lines OE1 and OE3 (Figure 8E). CmoNHX1 expression under salt stress was considerably lower than it was after water treatment, according to qRT-PCR analyses (Figure 8D). These findings imply that CmoNHX1 decreases the salt tolerance of plants under salt stress.

The discovery of the CmoNHX1 gene’s temporary expression in tobacco demonstrated that GFP, a fusion protein created with CmoNHX1, emits light on the cytoplasmic membrane and overlaps with the red fluorescent protein of the membrane marker: pm-rbCD3-1008 (Figure 9). This suggests that CmoNHX1 is located on the cytoplasmic membrane.