Evolutionary Expansion and Expression Analysis of the Oligopeptide Transporter Gene Family Between Cucurbita Moschata and Cucurbita Maxima

Background: Pumpkin is an important non-saline economic vegetable, salt stress often restricts the growth of pumpkin roots and the transportation and balance of mineral ions in the body. Oligopeptide transporter (OPT) plays an important role in transporting small peptides, secondary amino acids, Glutathione and minerals. However, information about the family of OPT in pumpkin is still limited. Results: In this study, 45 OPT transporters were identied from two cultivated species of Cucurbita moschata and Cucurbita maxima. Phylogenetic analysis showed that these OPT families were divided into two evolutionary branches: OPT clade and yellow stripe-like (YSL) clade. All of these genes contain the typical structure of OPT superfamily, OPT clade contains 12 conserved motifs, while the YSL clade contains 7 conserved motifs. There are tandem gene replication events on chromosomes 13, 16 and 18 of Cucurbita moschata and Cucurbita maxima, and 17 and 18 pairs of genes were collinear with Arabidopsis thaliana, respectively. Promoter element analysis showed that there were many cis-acting elements in the upstream promoters of these genes in Cucurbita moschata and Cucurbita maxima, which responded to 10 kinds of stress, especially hormones (MeJA and ABA) and hypoxia. The expression patterns based on transcriptome data sources showed that some OPT genes were organ-specic and tissue-specic, which might be involved in plant functional development. Transcriptome and qRT-PCR verication tests showed that CmoYSL7, CmaOPT15 and CmaYSL7 of Cucurbita moschata and Cucurbita maxima might play a key role in regulating the balance of metal ions between leaf mesophyll and leaf veins under salt stress. Conclusions: Overall, the data obtained from our study contribute to a better understanding of the complexity of the OPT genes family in pumpkins. These results will provide new insights into the mechanism of salt tolerance and the structure and function of OPT genes in pumpkin. OPT genes among multi-species, we constructed a phylogenetic tree (Fig. 1) using CmoOPTs and CmaOPTs amino acid sequence sequences and OPT amino acid sequences in Arabidopsis and rice. The results showed that the OPT genes of the four plants were divided into OPT clade (type A) and YSL clade(type B), and the OPT clade was subdivided into A1-A4, YSL clade was subdivided into B1-B3. In the OPT clade, A3 contained at least 4 OPT genes of pumpkin, A4 contained 12 OPT genes of pumpkin, and A1 and A2 both contained 6 OPT genes of pumpkin. In the YSL clade, B2 contained at least 2 YSL genes of pumpkin, B3 contained at most 8 YSL genes of pumpkin, and B1 contained 6 YSL genes of pumpkin. while that of 15 pairs of gene duplication events in the Cucurbita maxima chromosome was 0.32–18.17 MYA. It that although these genes were conserved in sequence, there were differences in evolutionary time.

until recently [27]. Up to now, the effect of salt stress on the expression pro le of OPT family members in seedlings of different pumpkin varieties has not been clear.
In this study, 45 OPTs family genes were identi ed from two pumpkin varieties (Cucurbita moschata and Cucurbita maxima) grown in different regions. Then these OPTs family members were analyzed by gene structure, phylogenetic tree, chromosome location, gene replication, Ka/Ks, collinearity between pumpkin group and other families, and expression patterns in tissues and organs. To clarify the changes in OPTs family members' expression pro le under salt stress, the differential expression patterns of OPTs between Cucurbita moschata and Cucurbita maxima were analyzed based on published transcriptional ancestral data and RT-qPCR test. Combined with this information, new insight was provided for the analysis of the mechanism of salt tolerance and the structure and function of the OPT gene in pumpkin.

Results
Genome-wide identi cation of OPT genes in pumpkin To identify the members of the OPT gene family, we searched the Cucurbita moschata and Cucurbita maxima genome databases with 17 AtOPTs (AtOPTs and AtYSLs), and further identi ed 21 and 24 OPT genes in two pumpkin genomes by using OPT domain (PF03169) and HMMER software. These genes are named as CmoOPT1-13, CmoYLS1-8, CmaOPT1-15 and CmaYLS1-8 according to the location and distribution of these genes on chromosomes (Fig. 1, Table  S1). In Cucurbita moschata, CmoOPT1-13 and CmoYLS1-8 encoded amino acid lengths from 659 to 1360, pI from 8.38 to 9.33, GRAVY from 0.075 to 0.515, and MW from 72899.67 Da to 127699.53 Da, predicted transmembrane structures from 12 to 26. The prediction of protein subcellular localization showed that except for CmoOPT2, CmoOPT3 and CmoYSL1 in Cucurbita moschata, all proteins were predicted to be located in the cell membrane.

Evolutionary analysis of OPTs in pumpkin
To study the relationship of OPT genes among multi-species, we constructed a phylogenetic tree ( Fig. 1) using CmoOPTs and CmaOPTs amino acid sequence sequences and OPT amino acid sequences in Arabidopsis and rice. The results showed that the OPT genes of the four plants were divided into OPT clade (type A) and YSL clade(type B), and the OPT clade was subdivided into A1-A4, YSL clade was subdivided into B1-B3. In the OPT clade, A3 contained at least 4 OPT genes of pumpkin, A4 contained 12 OPT genes of pumpkin, and A1 and A2 both contained 6 OPT genes of pumpkin. In the YSL clade, B2 contained at least 2 YSL genes of pumpkin, B3 contained at most 8 YSL genes of pumpkin, and B1 contained 6 YSL genes of pumpkin.

Motifs analysis of OPTs protein in pumpkin
Through the conservative domain analysis of 45 OPT genes in Cucurbita moschata and Cucurbita maxima, as shown in Fig. 2A, these genes can be divided into two clades: OPTs and YLSs, the OPTs clade is divided into A1-A4 subfamily, and the YLSs clade is divided into B1-B3 subfamily. All of these genes in Fig. 2B contained the typical structure of OPT superfamily, and the corresponding structure contained different numbers of a motif. Except for CmaOPT9, CmaOPT12 and CmoOPT7, other OPT genes in OPTs clade have 12 conserved motif patterns (such as Motif 5-9-3-7-8-10-6-4-14-2-1-11), while YSL genes in YLSs clade have 7 conserved motif patterns (such as Motif 9-3-3-13-4-15-12). There were 9 conserved motifs in CmaYSL4 and CmoYSL4 in B2 subfamily, while 8 conserved motifs in B1 and B3 subfamilies. From the intron and exon point of view, the number of exons in the OPTs clade ( Fig. 2C) ranged from 4 to 27, while that in the YSLs clade ranged from 6 to 13. B1 and B3 subfamilies contained 8 and 7 introns, respectively, while the B2 subfamily also contained 8 introns, but their distribution was different from that of the B1 subfamily.
Multiple sequence alignment of OPT genes n pumpkin By multiple sequence alignment of the amino acid sequences of the Cucurbita moschata (21 OPT genes) and Cucurbita maxima (24 OPT genes) by OPT (29) and YLS (16) clades, it was found that 29 OPT amino acid sequences shared 12 transmembrane domains (Fig. S2), and that they also shared the structure of the NPG and KIPPR motifs, containing 26 and 27 amino acid residues, respectively. The YLS amino acid sequences in the YLS clade contained 13 transmembrane domains (Fig. S3).

Chromosome location of OPT genes and collinearity analysis of in pumpkin genome
According to the distribution of OPT genes in Cucurbita moschata and Cucurbita maxima in chromosomes, it could be seen that these genes were mainly distributed on 9 chromosomes, of which chromosomes 1, 2, 3, 4, 9, 11 and 18 of the two genomes. There was the same number of genes among 9 chromosomes (Fig. 3). Cucurbita moschata contained 6, 4 and 4 genes on chromosomes 13, 16 and 18, respectively, while Cucurbita maxima contained 3, 5, and 4 genes on chromosomes 13, 16 and 18, respectively. There were gene tandem replication events on these chromosomes. To explore the collinearity of OPT genes in the two genomes, this analysis was performed using the MCScanX method ( Fig. 3A and 3B, Table S2). We found that there were 4 pairs of fragment duplications in Cucurbita moschata chromosomes, however, there were 15 pairs of fragment duplication phenomenon in Cucurbita maxima chromosomes. These fragment duplications mainly occur between chromosomes 13, 16 and 18 of the two genomes. By analyzing the Ka, Ks, Ka/Ks and the evolutionary age of the two genomes, it was found that the Ka/Ks of the gene pairs existing in two genomes were all less than 1, which tended to be puri ed selection, indicating that OPT genes in the evolutionary process were relatively conservative, and had not undergone major changes (Table S2). From the perspective of evolution time, the evolutionary time of 4 pairs of gene duplication events in the Cucurbita moschata chromosome was 4.56-18.02 MYA, while that of 15 pairs of gene duplication events in the Cucurbita maxima chromosome was 0.32-18.17 MYA. It showed that although these genes were conserved in sequence, there were differences in evolutionary time.
Collinearity of OPT genes between pumpkin and other families The collinear analysis was performed with the corresponding gene blocks of Cucurbita moschata (21 OPT genes), Cucurbita maxima (24 OPT genes) and Arabidopsis thaliana (16 OPT genes) (Fig. 4, Table S3). It was found that there were 24 pairs of gene collinear between Cucurbita moschata and Cucurbita maxima, mainly on chromosomes 1, 2, 4, 9, 11, 13, 16 and 18 of two genomes. There were 17 pairs of genes collinear between Cucurbita moschata and Arabidopsis thaliana, which were mainly concentrated on chromosomes 1,2,4,9,11,13,16,18 in Cucurbita moschata and 1, 3, 4, 5 chromosomes in Arabidopsis thaliana. There were 18 pairs of genes collinear between Cucurbita maxima and Arabidopsis thaliana, which was consistent with the location of chromosomes distributed in Cucurbita maxima and Arabidopsis thaliana.

Promoter analysis of OPT genes in pumpkin
To further clarify the function of the OPT family members of Cucurbita moschata and Cucurbita maxima, we selected the 2000 bp promoter region upstream of the CDS sequence to visualize the cis-acting elements by using TBtools software (Fig. 5). There were many cis-acting elements in the upstream promoters of these genes. They all responded to 10 kinds of stress (hormone response, anaerobic response, defense and resistance response, drought induction, light response, low-temperature response, etc.). In Cucurbita moschata, the cis-acting elements responsive to light regulation were distributed in all genes, while CmoOPT12 responsive to hormone (MeJA) contained as many as 16 cis-acting elements, followed by CmoOPT3 containing up to 14 cis-acting elements, and CmoYSL5 responsive to ABA contained 10 cis-acting elements (Fig. 5A). In Cucurbita maxima, there were at least 5 cis-acting elements in response to light regulation, while CmaYSL5 in response to hormone-like (ABA) contained as many as 10 cis-acting elements; CmaOPT2, CmaOPT11 and CmaYSL6 in response to IAA all contained 6 cis-acting elements; CmaOPT14 and CmaOPT10 responsive to hypoxia contained 14 and 8 cis-acting elements, respectively (Fig. 5B).

Expression pro les of OPT genes in different tissues in pumpkin
To analyze the expression of OPTs family genes in different tissues using common transcriptome data (Fig. 6, Table S4), we found that the expression abundance of CmaOPT1, CmoOPT1, CmoOPT4, CmoOPT11 and CmaYSL1 was lower in different tissues. In addition, the expression abundance of all genes in fruit was the lowest compared with other tissues. The expression levels of CmoYSL1, CmoYSL2, CmoYSL3, CmoYSL6, CmoOPT2, CmoOPT6 and CmoOPT8 in Cucurbita moschata root were higher than those in other tissues. The expression levels of CmaOPT3, CmaOPT5-10, CmaOPT12, CmaOPT15, CmaSYL2, CmaSYL4-5 and CmaSYL8 in Cucurbita maxima leaves was higher than those in other tissues.

Expression pro les of OPTs in pumpkin under salt treatment and qRT-PCR veri cation
To explore the response of OPTs in pumpkin vein and mesophyll to salt stress, previous RNA-seq data were analyzed (Table S5). There was a signi cant difference in the expression levels of OPTs family members in mesophyll and vein between Cucurbita moschata and Cucurbita maxima seedlings under salt stress (Fig. 7). Compared with the control treatment, the expression levels of CmoOPT2, CmoOPT5, CmoOPT6, CmoOPT7, CmoOPT8, CmoOPT12, CmoYSL1, CmoYSL2, CmoYSL3, CmoYSL4, CmoYSL5 and CmoYSL8 in the mesophyll of Cucurbita moschata were up-regulated under the NaCl treatment, while the expression levels of CmoOPT3, CmoOPT9, CmoOPT10, CmoOPT11, CmoOPT13, CmoYSL6 and CmoYSL7 were down-regulated (Fig. 7A). Compared with the control treatment, the expression levels of CmoOPT5, CmoOPT8, CmoOPT12, CmoYSL4 and CmoYSL7 in the vein of Cucurbita moschata was up-regulated under the NaCl treatment, while the expression levels of CmoOPT1, CmoOPT3, CmoOPT10 and CmoYSL6 were down-regulated (Fig. 7B). Compared with the control treatment, the expression levels of CmaOPT2, CmaOPT3, CmaOPT9, CmaOPT12, CmaOPT13, CmaYSL1 and CmaYSL7 in the mesophyll of Cucurbita maxima was up-regulated under the NaCl treatment, while the expression levels of CmaOPT5, CmaOPT6, CmaOPT8, CmaOPT11, CmaOPT14, CmaOPT15 and CmaYSL8 in the mesophyll of Cucurbita maxima was down-regulated ( Fig. 7C). Compared with the control treatment, the expression levels of CmaOPT4, CmaOPT6, CmaOPT12, CmaOPT15 and CmaYSL8 in the vein of Cucurbita maxima was up-regulated under the NaCl treatment, while the expression levels of CmaOPT2, CmaOPT9, CmaOPT12, CmaYSL1, CmaYSL6 and CmaYSL7 was down-regulated ( Fig. 7D).
To further verify the response of OPT family members to salt stress, we applied salt stress to the seedling stage of Cucurbita moschata "Baimi". It was found that after 12 hours of salt stress, the relative expression of CmoOPT3 and CmoYSL7 in the mesophyll of "Baimi" under salt stress was signi cantly lower than that of the control treatment, while the relative expression of CmoOPT5-8, CmoOPT12-13, CmoYSL4 and CmoYSL8 was signi cantly higher than that of the control treatment (Fig. 7E). The relative expression of CmoOPT3, CmoYSL5 and CmoYSL7 in the vein of "Baimi" under salt treatment was signi cantly higher than that of the control treatment, and the relative expression of CmoOPT8, CmoOPT10 and CmoOPT12 was signi cantly higher than that of the control treatment (Fig. 7F).

Discussion
Evolutionary relationship of OPTs in Cucurbita moschata and Cucurbita maxima Pumpkin (Cucurbita moschata Duch.). is cultivated by C. moschata (Cucurbita moschata Duch.), C. Pepo (Cucurbita Pepo L.), C. maxima (Cucurbita maxima Duch.) and several wild species [29]. C. moschata and C. maxima are both economic vegetables and widely cultivated, and the genomes of C.moschata and C. maxima are only recently reported [27]. In this study, a total of 45 OPT family members in two cultivars were found by searching the published pumpkin genome data. The comprehensive analysis of the phylogenetic tree, exon/intron gene structure and conserved motif of these genes in two cultivated species enables us to make some predictions and generalizations on the possible origin and interrelationship of OPTs. Phylogenetic analysis of two cultivated species, rice and Arabidopsis showed that the members of the OPT family were divided into two branches (OPT clade and YSL clade), which is basically consistent with the previous classi cation of Arabidopsis thaliana [13], rice [6] and several other species [10], indicating that the members of the OPT family were relatively conservative in the process of evolution. The structural differences of exons/introns can be used to determine the evolutionary history of gene families to some extent. From the point of view of introns and exons (Fig. 2C), the range of exon number of OPT clades was wider than that of the YSL clade, indicating that the variation of OPT clade members was larger in the process of evolution, while YSLs clade members were relatively conservative. There were results showed that the introns and exons of OPTs members in the OPT clade of pumpkin experienced strong gene replication in the process of evolution. This is also con rmed by the collinearity (Fig. 3 and Fig. 4) of OPT genes among C.moschata, C. maxima and Arabidopsis thaliana. In addition, the gene replication of C.moschata and C. maxima mainly occurred on chromosomes 13, 16 and 18. According to the evolutionary time, the chromosome gene repetition event of C.moschata occurred at 4.56-18.02 Mya, while that of C. maxima occurred at 0.32-18.17MYA. Some studies have shown that the earliest replication event of Cruciferae is expected to occur in 34MYA [30]. This showed that the gene replication of the two cultivated species occurred at almost the same time. At the same time, the Ka/Ks values of gene pairs in OPT and YSL clade members were less than 1 (Table S2), which further indicates that these genes were puri ed in the process of evolution.

Tissue expression of OPT genes in Cucurbita moschata and Cucurbita maxima
The expression pro les of family members in crops can provide useful clues for gene function [25]. In this study, the expression levels of 45 OPT genes were analyzed using publicly available RNA sequence data from the genomes of Cucurbita moschata and Cucurbita maxima [27]. It was found that the expression abundance of all genes in the fruits of the two cultivars was the lowest compared with other tissues. The expression abundance of 7 genes in Cucurbita moschata was higher than that in other tissues, and the expression abundance of 13 genes in Cucurbita maxima was higher than that in other tissues (Fig. 7). The expression levels of CmoYSL3, 5, 7 and CmaYSL3, 5, 7 were the highest in the roots, and these genes belonged to the B3 branch, in which the OsYSL12 and OsYSL13 gene was closely related to its evolution. It has been reported that the highest expression of OsYSL12 and OsYSL13 is in the roots of 14-day-old seedlings in rice [31]. These showed that these OPT genes were tissue-speci c, and had multiple functions in different tissues. On the other hand, some genes might play a role in speci c tissues. Some studies have shown that AtOPT3 [32] and AtOPT6 [16] in Arabidopsis thaliana expressed in seed tissues during development, which might be involved in the process of embryo development. OsOPT1, OsOPT3, OsOPT4 and OsOPT7 were expressed in rice embryos [6]. In this study, except for A3 clade, other genes were closely related to these genes in rice, suggesting that these OPTs might be related to embryo development. In addition, little is known about the role of OPT genes on stress response in pumpkin. Through promoter element analysis and tissue expression analysis, we obtained some clues about the role of pumpkin OPT genes on stress response [33]. Some studies indicated that GT1-motif and TGACG-motif were identi ed as salt stress response elements [34], ABRE, G-box, MBS and TGA-element had a regulatory role under salt stress. Therefore, we speculated that OPTs in Cucurbita moschata and Cucurbita maxima might play a key role in salt stress.

Expression pro le of OPT genes in Cucurbita moschata and Cucurbita maxima under salt stress
In plants, Na + is either distributed in specialized cells [35], or isolated in vacuoles in the mesophyll, so the cytoplasmic Na + level of cytoplasm in functional cells is kept at a low level [36,37]. The ion changes in plants under salt stress are closely related to K + , Mg 2+ , Ca 2+ and microelements. Functional diversi cation is the result of the evolutionary expansion of gene families through gene duplication, usually accompanied by changes in the expression pro les of gene family members. The OPT family is divided into two subfamilies (OPT subfamily and YSL subfamily) [11], in which the YSL subfamily is considered to be able to transport iron compounds, and the OPT subfamily is considered to be able to transport oligopeptides with 45 amino acid residues [38]. However, some members of the OPT subfamily have also been identi ed as iron transporters, such as AtOPT3 [13], OsOPT1, OsOPT4 and OsOPT7 [6]. According to the published RNA-seq data, the transcriptional data of CmoOPTs and CmaOPTs in leaf mesophyll and leaf vein under salt stress was analyzed. It was found that the expression levels of CmoYSL7, CmaOPT15 and CmaYSL7 in the mesophyll and veins of the two cultivated species were opposite under salt treatment. It was speculated that these genes played a key role in regulating the balance of metal ions between mesophyll and veins under salt stress (Fig. 7). The qRT-PCR technology was used to further verify the expression of these genes in the veins and mesophyll of Cucurbita moschata (Fig. 8). This result is almost similar to the transcriptional result. In this study, the response of pumpkin to salt stress was discussed from the expression pro le of OPTs family genes, but the speci c function and mechanism of these OPT genes in pumpkin need to be further studied.

Conclusions
In summary, the genomic analysis of OPT genes family in Cucurbita moschata and Cucurbita maxima was carried out, and the phylogeny, gene structure and homologous repetition history of OPTs family were discussed. We identi ed a total of 45 OPTs in the pumpkin genome and provided genetic information such as chromosome location, exon-intron structure, conserved domain and repetitive genes. In particular, we examined the expression pro les of these OPT genes in different tissues. At the same time, we detected the response of 45 OPT genes to salt stress and veri ed by qRT-PCR, and found that several key genes played a role in regulating the balance of metal ions between mesophyll and veins. These data may provide valuable information for the functional study of the gene family in the future.

Genome-wide identi cation of OPT genes in pumpkin
We downloaded the whole genome data from the Cucurbita moschata and Cucurbita maxima genome database (http://cucurbitgenomics.org/), and identi ed the possible members of the OPTs gene family in these two genomes by BLASTP tool. Seventeen OPT proteins [10] from Arabidopsis thaliana were used as query sequences to search the two genome databases. We choose the protein sequences (E value is lower than 1E − 10 value) as the candidate proteins. The obtained sequences were further con rmed through the CDD (conservative domain database) in NCBI and the PFAM website (http://pfam.xfam.org/) download OPT domain (PF03169). The integrity of the OPT domain in candidate genes was veri ed by ExPASy (http://prosite.expasy.org/)). Each OPTs gene was given a unique name according to its exact location on the chromosome (each chromosome from top to bottom). Prediction of the transmembrane helix was performed with TMHMM Server v. 2.0 (http://www.cbs.dtu.dk/services/TMHMM/) [39]. The molecular weight, isoelectric point and grand average hydropathy (GRAVY) values of each OPT gene were estimated using the ExPASy Proteomics Server (https://web.expasy.org/tools/protparam.html/) [40]. The subcellular location of OPTs gene was predicted using the Plant-mPLoc (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi) [41].
Multiple sequence alignment and phylogenetic tree analysis of OPT genes in pumpkin The OPT protein sequences were downloaded from the genomes of Arabidopsis thaliana, Oryza sativa, Cucurbita moschata and Cucurbita maxima, respectively. The protein sequences of these species were compared by ClustalW2 (http://www.genome.jp/tools-bin/CLUSTORW) software, and the phylogenetic analysis of the OPT proteins was constructed using the neighbor-joining method in MEGA 7.0 (1000 Bootstrap repeats) [42].

Structural analysis and chromosome mapping of OPT genes in pumpkin
The exon/intron organizations of the OPT genes were analyzed using the Gene structure display server 2.0 (GSDS2.0; http://gsds.cbi.pku.edu.cn/). The CDS of each gene was compared with its corresponding genomic DNA sequence, [43]. The structural motif of OPTs protein sequence was annotated using online software Multiple EM for Motif Elicitation (MEME) (http://meme-suite.org/tools/meme) [44], and the maximum value of motif site was set to 15. The results of the exon/intron and motifs were visualized by the TBtools software (https://github.com/CJ-Chen/TBtools) [45].

Gene duplication and evaluation of Ka, Ks and Ka/Ks in pumpkin
To identify gene duplication events, all CDS sequences of wheat OPT genes were subjected to BLAST searches against each other (identity > 80%, e value < 1e − 10 ) by using the local Blast program. Gene alignment coverage was then acquired by pair-wise alignment using the previously calculated method: Gene alignment coverage = (alignment length-mismatch length) / the length of larger genes. When the gene alignment coverage is more than 0.75, it is considered to be a duplicate gene pair. In addition, in the 100 kb region, two genes separated by ve or fewer genes are considered as tandemly duplicated genes [46]. The values of Ka/ Ks between Ka and Ks and between paired genes were calculated by DnaSP software (http://www.ub.edu/dnasp/) [47]. For the timing of duplication events, the formula: T = Ks/2λ × 10-6 Mya was used to calculate divergence time (T) in millions of years (Mya), where λ = 6.5 × 10 − 9 represented the rate of replacement of each locus per herb plant year [48]. All the OPTs gene location information was obtained from the genome database of Cucurbita moschata and Cucurbita maxima. The chromosome distribution of these genes and tandem repeats in Cucurbita moschata and Cucurbita maxima were visualized using the Advanced Circos tool (https://github.com/CJ-Chen/TBtools) [45].

Analysis of cis-acting elements of OPT genes promoter in Pumpkin
To analyze the promoter elements of OPT genes in Cucurbita moschata and Cucurbita maxima, we downloaded the sequence of the transcriptional initiation site (ATG) pre-2000 bp of these genes from the pumpkin database. The cis-acting elements of these genes were predicted by online tool PLANTCARE (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/) [49] and visualized by TBtools (https://github.com/CJ-Chen/TBtools) [45].
Tissue expression of OPT genes in pumpkin and its response to salt stress To analyze the expression patterns of CmoOPTs and CmaOPTs in different organs (root, stem, leaf, fruit), the transcriptome data (BioProject: PRJNA385310) of different organs of Cucurbita moschata and Cucurbita maxima were downloaded [27]. To determine the response of Cucurbita moschata and Cucurbita maxima OPTs gene family to salt stress, we analyzed two different Cucurbita cultivars, "Rifu" (Cucurbita moschata) and "Rimu" (Cucurbita maxima). The transcriptome data in 2018 (BioProject: PRJNA464060) was excavated and analyzed the transcription pro les of OPTs in the leaf veins and leaf mesophylls under salt stress.
All the published transcriptome data were represented by RPKM (Reads per kilobase of exon model per million mapped reads), which has been converted to log 2 (RPKM) when plotting heat map. The heat map was visualized using the TBtools Heatmap tools (https://github.com/CJ-Chen/TBtools) [45].

Experimental materials and stress treatment
To further analyze the response of OPTs in Cucurbita moschata to salt stress, the experiments were performed with "Baimi9" as the research material. They were provided by the pumpkin team of School of Horticulture and Landscape Architecture, Henan Institute of Science and Technology. The seeds were sown in a tray containing a matrix-meteorite (3:1) mixture and grown in a plant growth chamber. The arti cial growth conditions were set as light intensity of 350 µmol/m2/sec, 65% relative humidity and 25 ℃,16 h light / 16 ℃, 8 h dark. The two-month-old seedlings were cultured in Hoagland solution, pH 6.5. After 5 days of adaptation, some of the seedlings were cultured with 75 mM NaCl. Leaf veins and leaf mesophylls were collected at 12 h after the NaCl treatment.
The control plants (CK) refers to the seedlings cultured normally without NaCl treatment. Each treatment has three independent biological replications. The Control and salt-treated samples were frozen in liquid nitrogen and stored at -70 ℃ for further analysis.

Quantitative real-time PCR (qRT-PCR) analysis
Total RNA was removed with RNase-free DNase I (Takara, Tokyo, Japan) to avoid genomic DNA contamination. First-strand cDNAs were synthesized using the PrimeScript RT Reagent Kit with gDNA Eraser (Takara, Dalian, China) according to the manufacturer's protocol. The qRT-PCR assays were performed with the Primer Script RT Reagent Kit (Takara, Dalian, China). The speci c primer of OPTs gene in pumpkin was designed on Primer-BLAST program (https://www.ncbi.nlm.nih.gov/tools/primer-blast/index.cgi) ( Table S6). The Cucurbita moschata β-Actin was used as the internal reference gene. Data were analyzed with Opticon monitor software (Bio-Rad). All primers for qRT-PCR were designed using Primer 6.0 software and primer sequences are listed in Table   S3. The PCR conditions were as follows: 95 °C for 10 s and 40 cycles of 95 °C for 5 s and 60 °C for 30 s. The qRT-PCRs were performed using an ABI Step One Table 1 Properties and locations of the predicted OPT proteins in Cucurbita moschata and Cucurbita maxima.  Figure 1 Phylogenetic tree of OPT family proteins among Arabidopsis thaliana, Oryza sativa, Cucurbita moschata and Cucurbita maxima. The OPT family proteins were classi ed into two OPT clade (type A) and YSL clade(type B), and the OPT clade was subdivided into A1-A4, YSL clade was subdivided into B1-B3. At, Arabidopsis thaliana; Os, Oryza sativa; Cmo, Cucurbita moschata; Cma, Cucurbita maxima.   Cis-acting elements distribution and statistical analysis of OPT genes promoters. The numbers in circles depicts the quantity of the predicted cis-acting elements in the promoter region.

Figure 7
Expression pro les and qRT-PCR veri cation of OPTs in pumpkin vein and mesophyll under salt treatment. A-D, Expression elvels of OPT genes was recalculated with transcriptome data (BioProject: PRJNA464060). All data were standardized by Log2RPKM(NaCl)/RPKM(Control). NaCl: NaCl treatment, Control: normal conditions. E-F, The data represented the relative expression of OPT genes at 12 h after NaCl treatment. Control referred to untreated plants (control plants) under normal conditions. The data was calculated via the 2−ΔΔCt method, and the reference gene (β-Actin) was used to correct the expression level of target genes. Errors bars represent the standard errors of three biological replicates, and asterisks indicated a signi cant difference in expression level between NaCl and Control treatments (*P< 0.05, **P< 0.01)

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.