Genome-Wide Identification, Characterization, and Expression Analysis of the Grapevine Superoxide Dismutase (SOD) Family

Superoxide dismutase (SOD) is an essential enzyme of the plant antioxidant system that responds to oxidative damage caused by adverse conditions. However, little is known about the SOD gene family in Vitis vinifera (Vv). In the present study, ten SOD genes, including 6 copper/zinc SODs, 2 iron SODs, and 2 manganese SODs, were identified in the grapevine genome where they were unevenly distributed on 12 chromosomes. Ten VvSOD genes were divided into three main groups based on phylogenetic analysis, subcellular localization, and the distribution of conserved protein motifs. Additionally, many cis-elements related to different stresses were found in the promoters of the 10 VvSOD genes. Syntenic analysis revealed that VvMSD1 and VvMSD2 were derived from segmental duplication, and VvCSD4 and VvCSD5 belong to a pair of tandemly duplicated genes. Gene expression analysis based on microarray data showed that the 10 VvSOD genes were expressed in all the tested tissues. Interestingly, the segmentally duplicated gene pair (VvMSD1 and VvMSD2) exhibited differential expression patterns in various organs. In contrast, the tandemly duplicated gene pair (VvCSD4 and VvCSD5) displayed similar expression patterns in the tested organs. Our results provide a basis for further functional research on the SOD gene family in grapevine.


Introduction
Grapevine (Vitis vinifera) is one of the most cultivated and economically valuable fruit crops in the world. However, abiotic stresses, particularly drought and salt stress, threaten the growth of grapevine globally, thereby affecting fruit yield and quality [1][2][3]. The most common result of such stress is the generation of toxic reactive oxygen species (ROS). Membrane damage, protein oxidation, DNA lesions, and even irreparable metabolic dysfunction and cell death can occur as a result of excessive ROS (such as superoxide anions, hydroxyl radicals, hydrogen peroxide, and singlet oxygen) [4,5].
To cope with ROS toxicity, plants have developed efficient and sophisticated antioxidative response systems, including the synthesis of low relative molecular mass antioxidant molecules (e.g., l-ascorbic acid and glutathione) and various enzymes. Among these enzymatic components, the superoxide dismutases (SODs) constitute the first line of defense against ROS by catalyzing the dismutation of the superoxide O 2 to O 2 and H 2 O 2 [6,7]. In plants, SODs have been detected in the roots, leaves, fruits, and seeds where they play important roles in protecting cells from oxidative damage [8].
Multiple SOD isozymes exist in plants and are classified into three types based on their metal cofactor, protein folds, and subcellular distribution: copper/zinc-(Cu/Zn-) SOD, manganese-(Mn-) SOD, and iron-(Fe-) SOD, and these SODs are located in different compartments of the cell [7,9,10]. Cu/Zn-SODs are mainly distributed in the cytosol, chloroplasts, peroxisomes, and/or the extracellular space, while Fe-SODs mainly occur in the chloroplasts, as well as in peroxisomes and mitochondria, and Mn-SODs are mainly localized not only in the mitochondria but also in different types of peroxisomes [11]. In addition, a new type of SOD, nickel (Ni)SOD, was first discovered, cloned, and characterized in Streptomyces [12]. However, no evidence for NiSOD has been found in plants [13].
In the present study, we identified the SOD gene family in grapevine on the genome-wide scale. Genomic organization, gene structure, motif composition, subcellular localization, syntenic analysis, and phylogenetic relationships were analyzed using bioinformatics. Then, the gene ontology and putative promoters of the grapevine SODs were also investigated, and the cis-elements involved in stress responses were analyzed. Putative transcription factors (TFs) which may regulate the VvSOD genes were predicted. Furthermore, we studied the expression patterns of the grapevine SOD gene family under abiotic stress (salt, drought, cold, and heat) using microarray data and a real-time quantitative-(qRT-) PCR detection system. Finally, coexpression networks of VvSOD genes and TF genes under the four abiotic stresses were generated based on the qRT-PCR data. This was the first comprehensive study of the SOD gene family in grapevine and may provide valuable information for understanding the classification, evolution, and putative functions of the grapevine SOD family on the genome-wide scale.

Materials and Methods
We downloaded the most recent version (V 2.1) of a 12X assembly of the grapevine (V. vinifera) genome from CRIBI (http://genomes.cribi.unipd.it/grapevine/) for identifying SOD genes. To identify members of the SOD gene family in grapevine, the full-length protein sequences from A. thaliana (TAIR locus number: AT1G08830.1, AT2G28190, AT5G18100, AT4G25100, AT5G51100, AT5G23310, AT3G56350, and AT3G10920) were used as BLASTp queries against all the grapevine protein sequences with the threshold expectation value set to 1.0. All the hits were further submitted to Pfam analysis (http://pfam.xfam.org/) to verify the presence of the SOD domain. Physicochemical characteristics of the SOD proteins were calculated using the ProtParam tool (http://web.expasy.org/protparam/), including the number of amino acids, molecular weight, theoretical isoelectric point (pI), aliphatic index, and grand average of hydropathicity (GRAVY) [25].
ProtComp9.0 server (http://linux1.softberry.com/) was used to predict the subcellular localizations of the SOD proteins [22]. Conserved protein motifs were predicted by MEME Suite (http://meme-suite.org/tools/meme) with the default settings, except that the number of motifs was set to 8, and the minimum and maximum motif widths were set to 20 and 150 amino acids [26]. The web-based Gene Structure Display Server (GSDS; http://gsds.cbi.pku.edu.cn/index .php) program was utilized to illustrate exon and intron organization for the grapevine SOD genes.

Phylogenetic Analysis.
To investigate the phylogenetic relationships of the SODs among grapevine and A. thaliana, multiple SOD protein sequences were aligned, and an unrooted phylogenetic tree was constructed in MEGA 6.06 [27]. The phylogenetic tree was constructed using the neighbor-joining (NJ) method and Jones-Taylor-Thornton (JTT) model. In the phylogenetic tree, the degree of support for a grouping pattern was evaluated using bootstraps (1000 replicates).

Chromosomal
Locations and Syntenic Analysis. The chromosomal locations of the grapevine SOD genes were verified from the CRIBI database (http://genomes.cribi. unipd.it/grapevine/), and chromosomal images were drawn using MapInspect software [28].
Duplication patterns of the VvSOD genes were assigned based on their locations. Genes located in a 200 kb region of a chromosome or a scaffold were defined as a gene cluster derived from tandem duplications [29]. Segmentally duplicated genes were identified as genes located on duplicated chromosomal blocks and were determined using MCScanX software (http://chibba.pgml.uga.edu/mcscan2/), which detects gene duplication events using an E-value threshold of 10 −5 [30].
To detect the selection mode of the segmentally duplicated gene pairs, the ratio of nonsynonymous substitutions to synonymous substitutions was evaluated. Firstly, the CDSs (nucleotide coding sequences) of the VvSOD genes were aligned using Muscle in MEGA6.06. Then, nonsynonymous substitutions (Ka), synonymous substitutions (Ks), and the ratio between them (Ka/Ks) were calculated using MEGA 6.06.

GO Analysis and
Promoter Sequence Analysis. The gene ontology (GO) term IDs of each VvSOD gene member in grapevine were obtained from the CRIBI database (http://genomes.cribi.unipd.it/DATA/V2/annotation/bl2go. annot.txt). The annotations of the GO term IDs were collected from the Gene Ontology Consortium (http://www. geneontology.org). Ten grapevine SOD proteins were assessed based on their molecular functions, biological processes, and cellular localizations.
The 2000 bp upstream sequences of the coding region of each VvSOD gene were downloaded from the CRIBI database. Then, the cis-elements in the promoters of each VvSOD gene were predicted using the PlantCARE server (http:// bioinformatics.psb.ugent.be/webtools/plantcare/html/) [31].

Prediction of Potential Regulatory Interactions between
TFs and VvSODs. The TFs which possess overrepresented targets in the VvSOD gene sets were detected from the Plant Transcriptional Regulatory Map (PlantRegMap) with a regulation prediction tool (http://plantregmap.cbi.pku. edu.cn/regulation_prediction.php) [32]. 2000 bp upstream sequences of the coding region of each VvSOD gene were used as the input.
Expression analyses of SOD genes and TF genes in response to abiotic stresses were based on microarray data (including accession number GSE31594 and GSE31675) downloaded from the NCBI GEO datasets. Four abiotic stresses, including cold, salt, heat, and polyethylene glycol (PEG), were analyzed. The fold changes compared with the corresponding control in each experiment were used to generate heatmaps using the R package (https://www. r-project.org/) pheatmap.

Coexpression Network and Interaction of VvSODs and TF
Genes. For the four stress treatments, the Pearson correlation coefficient (PCC) value (Table S4) was calculated between each pair of VvSODs and TF genes using gene expression values from microarray data in SPSS v 20.0 (IBM Corp., Armonk, NY, US) [28]. Coexpressed gene pairs were filtered with a PCC cutoff of 0.95. The coexpression network map was generated by Cytoscape 3.3 (http://www. cytoscape.org).

Plant Material and Experimental
Treatments. For the cold, heat, and salt treatments, six-week-old "Pinot Noir" grapevine (V. vinifera, the sequenced genotype PN40024) plantlets were used in this study. The plantlets were maintained at a photoperiod of 16 h light/8 h dark and a temperature of 23°C. Low temperature and heat stress treatments were applied by placing the plants at 4°C (drug storage box, HYC-360, Haier) and 42°C (intelligent artificial climate box, RXZ-380C, Jiangnan Instrument Factory, Ningbo) with a photoperiod of 16 h light/8 h dark, respectively [28]. Leaf samples were harvested from both treatments at 0, 6, 12, and 24 h after initiation of the treatments. Plants subjected to 200 mmol/L NaCl were sampled at 0, 6, 12, and 24 h after treatment. Three independent biological replications were performed for each treatment. Tissue samples were immediately frozen in liquid nitrogen and stored at −70°C until analysis.
For drought stress treatment, in vitro-grown grapevine plants (PN40024) were maintained on half-strength Murashige and Skoog medium containing 0.3 mg/L indole 3-butyric acid (IBA) and placed at 25°C in a culture room under a photoperiod of 16/8 h and a light intensity of 100 μmol·m -2 ·s -1 . Five-week-old tissue-cultured grapevine plants were transplanted into pots and acclimatized in the growth chamber with 16 h light at 24°C/8 h dark at 22°C and 70%-80% relative humidity [34]. Plants grown in pots were fully watered for the first four weeks, after which water was withheld to impose the drought treatment. The grapevine leaf samples were collected at 0 (control), 2, 4, and 8 days after the treatment, respectively.
2.8. Total RNA Extraction and cDNA Synthesis. Total RNA was extracted from the leaves using the Plant Total RNA Isolation Kit Plus. The total RNA was eluted in 30 μL of RNase-free water and stored at −70°C. The purity and density of the extracted RNA was assessed on a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, USA) using RNase-free water as a blank.
Single-stranded cDNA was synthetized using a Takara PrimeScript RT reagent kit with a gDNA eraser (Takara, Dalian, China) and oligo dT primers as described by the manufacturer's protocol. Approximately 2 μg of total RNA in a single 20 μL reaction was converted to single-stranded cDNA using standard thermal cycling conditions.
2.9. qRT-PCR Analysis of VvSOD Genes. qRT-PCR analyses were conducted on a Roche LightCycler 480 PCR system using Lightcycler 480 SYBR Green I Master. The oligonucleotide primers (Table S3) for amplifying specific VvSODs were designed using the PrimerQuest Tool (http://sg.idtdna.com/ Primerquest/Home/Index). The 10 μL reaction volume contained 5 μL SYBR Green I Master, 1 μL of diluted cDNA, 1 μL of each primer (10 μM), and the addition of ddH 2 O to bring the total volume to 10 μL. The thermocycle protocol was as follows: denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 10 s, 58°C for 20 s, and 72°C for 10 s. At the end, a melting curve was generated from 65 to 97°C. Three replicates were used for each sample. Expression levels were calculated using the 2 -ΔΔT method and the housekeeping gene (actin-101-like, VIT_012s0178g00200) was used as a reference gene for normalizing the expression of the VvSOD genes [35]. Fold differences were visualized by normalizing all of the data based on setting the expression level at 0 h as a value of 1 (values above 1 and below 1 were considered as up-and downregulated, respectively) [36].

Genome-Wide Identification of the SOD Gene Family in
Grapevine. Based on a BLAST search using the known A. thaliana SOD protein sequences and the identification by the Pfam database, a total of 10 SOD genes were identified in the grapevine genome. Based on the domain analysis, the 10 grapevine SOD genes were classified into 3 groups (Cu/ZnSODs, FeSODs, and MnSODs), including 6 VvSOD proteins with a copper-zinc domain (Pfam: 00080) and 4 SOD proteins with an iron/manganese superoxide dismutase alpha-hairpin domain (Pfam: 00081) or an iron/manganese superoxide dismutase, C-terminal domain (Pfam: 02777). Based on the domain analysis and chromosome location, the 10 grapevine genes were named as follows: VvCSD1, VvCSD2, VvCSD3, VvCSD4, VvCSD5, VvCSD6, VvMSD1, VvMSD2, VvFSD1, and VvFSD2 (Table 1).
The physicochemical analysis indicated that the length of the amino acid sequences, MWs, pIs, aliphatic indexes, and GRAVY values varied among these VvSOD proteins. Considerable variations in the number of amino acids among these VvSOD proteins were observed, ranging from 79 to 329 aa. The predicted molecular weight of the 10 VvSOD proteins varied from 8594.7 to 37779.23 Da. The results revealed that four Fe-MnSODs were basic and five Cu/Zn-SODs were acidic, except for the slightly basic VvCSD3. The GRAVY numeric values of the three VvSOD proteins Chloroplast (VvCSD1, VvCSD2, and VvCSD5) were positive, indicating that they are hydrophobic proteins and the rest of the VvSODs are hydrophilic proteins ( Table 1). The ProtComp9.0 program was used to predict the subcellular localizations of the VvSOD proteins. Among them, six Cu/ZnSODs were cytoplasmic, two MnSODs were in the mitochondria, and two FeSODs were located on the chloroplast (Table 1). These results were in accordance with a previous study [8,37].
The MEME server was used for conserved motif analysis, and a total of eight conserved motifs were identified ( Figure 1). Among them, motif 3 was observed in all six Cu/ZnSODs, while motifs 5, 6, and 7 were found in four Fe-MnSODs ( Figure 2). The exon numbers varied among the 10 VvSODs, ranging from 4 to 9 ( Figure 2). It is worth noting that the tandemly duplicated gene pair (VvCSD4 and VvCSD5) and segmentally duplicated gene pair (VvMSD1 and VvMSD2) both exhibited the same exon numbers.

Phylogenetic
Analysis of SOD Genes. Twenty-eight SOD protein sequences from grapevine and A. thaliana (Table S1) were used to construct an unrooted phylogenetic tree ( Figure 3). The SODs were divided into three major groups, namely, Cu/ZnSODs, MnSODs, and FeSODs. The FeSODs and MnSODs were clustered within a large clade with a high bootstrap value (99%), implying that the FeSODs and MnSODs originated from a common ancestor. Additionally, species-specific duplications occurred between VvCSD4 and VvCSD5 and VvMSD1 and VvMSD2.

Chromosomal Locations and Syntenic Analysis. The 10
VvSOD genes were located on seven chromosomes, including VvCSD1 on chromosome 2; VvCSD2 and VvMSD1 on chromosome 6; VvCSD4, VvCSD5, and VvCSD6 on chromosome 14; and VvCSD3, VvFSD1, VvMSD2, and VvFSD2 on chromosomes 8, 10, 13, and 16, respectively (Figure 4). One segmental duplication event was detected between gene VvMSD1 and VvMSD2 on chromosomes 6 and 13. Moreover, VvCSD4 and VvCSD5 were identified as a pair of tandemly duplicated genes. The Ka/Ks ratios were calculated to better understand the duplication and functional divergence of the duplicated VvSOD genes during their evolutionary course. As shown in Table S2, the Ka/Ks ratio of the segmentally duplicated gene pair (VvMSD1 and VvMSD2) was 0.29, indicating that they underwent purifying selection. The Ka/Ks ratio of the tandemly duplicated gene pair (VvCSD4 and VvCSD5) was 1.29, which indicated that they were driven by positive selection during their evolutionary process.
3.4. GO Analysis of the VvSOD Proteins. Biological processes, molecular functions, and cellular components of genes are characteristics of genes or gene products that elucidate the diverse molecular functions of proteins [38]. GO analysis was performed to further characterize the predicted functions of the 10 VvSOD proteins ( Figure 5). The "molecular function" data showed that all VvSOD genes were involved in the "superoxide dismutase activity" (GO:0004784). According to the "biological process" results, all of the VvSOD genes participated in the "oxidation-reduction process" (GO:0055114). Moreover, the grapevine SOD genes may be involved in the biological processes responding to biotic stimulus and abiotic stimulus, such as strong light, ozone, salt, and UV stresses. In addition, they may also participate in some biological metabolic processes, such as superoxide metabolic processes, tRNA metabolic processes, and rRNA processing.

Analysis of cis-Elements in Putative VvSOD Gene
Promoters. To further determine the regulatory roles of VvSODs under various stresses, the cis-elements in the VvSOD gene promoter sequences were researched. The cis-elements were divided into four major subgroups: stress-responsive, hormone-responsive, light-responsive, and MYB binding site ( Figure 6). Among 10 VvSODs, nine of them possessed several MYB binding sites. Furthermore, we found that abundant light-responsive cis-elements existed in the VvSODs, particularly in the VvMSD2 gene with a number of 19.

Expression Analysis of the VvSOD Genes Based on
Microarray Data. To gain more insight into the role that VvSODs play in plant growth and development, we analyzed the expression patterns of the VvSOD genes in different grapevine tissues and organs based on the microarray data from 54 grapevine samples [33]. Interestingly, VvMSD2 displayed consistently high expression in all 54 tissues, whereas VvMSD1 exhibited lower expression levels compared with the others (Figure 7). VvCSD4 and VvCSD5 exhibited similar expression patterns in all the tested tissues. This similarity was also evident in gene VvFSD1 and VvFSD2. Additionally, two genes (VvMSD1 and VvMSD2) demonstrated distinct tissue-specific expression patterns. Furthermore, four Affymetrix microarray datasets were specifically analyzed for determining the expression profiles of VvSODs and TF genes in response to abiotic stresses, which included PEG, salt, heat, and cold. Six out of 10 probes corresponding to VvSOD genes were found (VvCSD1, VvCSD2, VvCSD3, VvCSD4, VvMSD2, and VvFSD1). Eight out of 21 probes corresponding to TF genes were found (VIT_18s0122g00380, VIT_17s0000g01260, VIT_ 08s0007g06270, VIT_04s0008g01470, VIT_05s0020g03880, VIT_19s0014g01680, VIT_15s0048g02870, and VIT_ 10s0003g00140). In the cold, salt, and PEG stress, the expression levels of VvCSD1 and VvCSD2 were higher than those of the other four genes (Figure 8(a)). VvFSD1 was expressed at an extremely low level in these three treatments. However, VvCSD3 and VvFSD1 were highly induced under the heat stress treatment. In each abiotic stress, there existed a single TF gene which responded strongly. However, the other was expressed at the similar level (Figure 8(b)). For example, listed genes, VIT_10s0003g00140 in cold stress, VIT_15s0048g02870 in salt and PEG stress, and VIT_ 08s0007g06270 in heat stress, were all altered strongly by 8 times compared with the control. VIT_10s0003g00140, VIT_15s0048g02870 and VIT_08s0007g06270 belong to ERF, HD-ZIP, and SBP TF family, respectively (Table S6).

Expression Analysis of VvSOD Genes Based on qRT-PCR.
Previous studies have demonstrated the important roles that SOD genes play in the plant response to abiotic stresses. The The Jones-Taylor-Thornton model with P-distance was chosen to conduct the phylogenetic tree. The bootstrap value was set as 1000. The SOD genes were classified into two major groups and five subfamilies (groups a, b, c, d, and e which have red, blue, yellow, purple, and green branches, respectively).
analysis of microarray data also indicated that the VvSOD genes were induced by different stresses. To gain more insight into the relative expressional diversity of eight members of the VvSOD gene family to cold, heat, salt, and drought, we detected the expression levels using qRT-PCR ( Figure 9). Our results demonstrated that all eight VvSOD genes were induced noticeably by the four stresses, exhibiting different expression levels. All eight members were upregulated in the cold treatment, except for VvCSD3 and VvFSD2 (Figure 9). Among the six upregulated genes, VvCSD1, VvCSD2, and VvFSD1 were altered strongly by 3-7.7 times compared with the control. Upon heat stress, expression of VvCSD1 increased by 12-fold, 11-fold, and 9-fold at 6 h, 12 h, and 24 h, respectively. VvCSD2, VvCSD3, VvFSD2, and VvMSD2 were also upregulated at 6 h with a lower expression level compared with that in VvCSD1. However, the other three genes (VvCSD4, VvCSD6, and VvFSD1) were downregulated during the treatment periods (Figure 9). In the salt treatment, the eight genes showed the lowest expression levels compared with the other three stresses. The expression levels of most of the genes exhibited slight variations, except VvCSD3 and VvCSD6, which were upregulated more than 2-fold. Upon drought stress, the expression level of VvCSD1, VvCSD2, VvCSD4, and VvCSD6 peaked at 12 h, but six genes (VvCSD1, VvCSD2, VvCSD3, VvFSD1, VvFSD2, and VvMSD2) were downregulated at 24 h. Only two genes (VvCSD3 and VvMSD2) were downregulated at 6 h and maintained similar levels at the following time points.       Figure 8: Expression profiles of the response of the grapevine VvSODs (a) and TF genes (b) to several abiotic stresses. Microarray data (accession number GSE31594 and GSE31675) downloaded from the NCBI GEO datasets. Four abiotic stresses, including salt, polyethylene glycol (PEG), heat, and cold, were analyzed. The fold changes compared with the corresponding control in each experiment were used to perform heatmaps that were generated by RStudio software. three gene pairs showed a negative correlation (Figure 10(a)). Only one gene pair was positively correlated with one another under salt stress (VvFSD1 and VIT_08s0007g06270) (Figure 10(b)). Interestingly, correlated gene pair (VvCSD3 and VIT_17s0000g01260, noted by yellow node) was also predicted to have regulation interaction (Table S6). Additionally, the result inferred that there was a negative correlation between VIT_ 04s0008g01470 with two VvSOD genes (VvCSD3 and VvMSD2) under PEG treatment (Figure 10(c)).
In the present study, we identified 10 SOD genes in the grapevine genome that could be classified into three SOD types (six Cu/ZnSODs, two FeSODs, and two MnSODs) based on the domain and phylogenetic tree analyses. The 10 SOD genes possessed different intron numbers. A previous study reported that Cu/Zn-SODs containing different intron numbers and positions showed no exon-intron structural similarities in related species [42]. It is worth noting that one tandem pair of VvSOD genes (VvMSD1 and VvMSD2) exhibited similar intron/exon organization patterns.
The promoters of VvSODs contain large amounts of light-responsive cis-elements that may contribute to light stress resistance. Gupta et al. detected increased resistance to oxidative stress in transgenic plants with an overexpressed Cu/Zn-SOD gene [43]. Transgenic A. thaliana plants overexpressing a miR398-resistant form of CSD2 accumulated more CSD2 mRNA than plants overexpressing a regular CSD2 and as a result were much more tolerant of intensive light [44]. Similar observations have also been reported in Nicotiana tabacum and N. plumbaginifolia [45,46]. All VvSOD genes except for VvFSD2 possessed the MYB binding site. It had been reported that MYB transcription factors played regulatory roles in developmental processes and defense responses in plants [47]. The existence of abundant MYB-related cis-elements indicated that it may involve in the regulation of the expression of VvSOD genes.
By predicting TFs which may regulate VvSODs, several TFs were obtained (Table S6). These TFs belonged to a different TF protein family, including SBP, MYB, WRKY, ERF, CPP, LBD, TCP, HD-ZIP, AP2, and Dof family. Many of them had been found involve in a variety of biological functions like flower development, biotic and abiotic stresses, cell proliferation, hormonal response, and carbohydrate metabolism [48][49][50][51][52][53][54]. It has been reported that TFs regulate the target genes by both activate way and repress way [55,56]. The coexpression results of VvSODs and TF genes also showed evidence for this conclusion. TF regulation prediction showed that VIT_17s0000g01260 may interact with VvCSD3. Coexpression result identified this prediction with a negative relationship. However, the roles of TFs in VvSOD regulatory networks, as well as their interaction mechanism, remain to be fully elucidated through experimental verification and bioinformatic analysis.
Gene duplication provides new genetic material and novel genes and occurs via tandem, segmental, or genome-wide duplication [57][58][59][60]. In our study, only one segmental duplication event (VvMSD1 and VvMSD2) and one tandem duplication event (VvCSD4 and VvCSD5) were identified. The expression profiles of the 10 VvSOD genes in the different tissues indicated that most of the VvSOD genes were expressed, and a few genes displayed tissue-specific expression patterns. VvCSD4 and VvCSD5, identified as tandemly duplicated gene pairs, exhibited similar gene expression patterns. Moreover, VvMSD1 and VvMSD2, the segmentally duplicated gene pair, showed different expression patterns in all the tested tissues. Li et al. reported that the attribution of duplication mode to the expression divergence implies a different evolutionary course of duplicated genes [61]. Adams et al. found that organ-specific and biased expression or silencing of duplicated gene pairs existed in Gossypium hirsutum [62]. A similar phenomenon was also reported in wheat [63].
The qRT-PCR results revealed that every VvSOD gene responded to at least one abiotic stress treatment performed in our study ( Figure 9). Notably, only VvCSD1 was activated strongly under heat stress. Moreover, VvCSD1 also expressed highly under cold stress. This suggest that VvCSD1 may play a vital role in the low/high temperature defense. VvCSD6 showed the highest expression level compared with other genes under drought stress. Promoter analysis revealed that VvCSD6 harbored 3 MYB binding sites which involved in drought inducibility, which could explain the significant expression of VvCSD6 under drought stress. Under salt stress, only the genes VvCSD3 and VvCSD6 altered obviously. Taken together, the expression pattern under various stress conditions suggested different VvSOD genes could perform its own function under different stresses. This research constitutes the first comprehensive study of the SOD gene family in grapevine and may provide valuable information for elucidating the classification, evolution, and putative functions of the grapevine SOD family on the whole-genome scale.

Conclusion
In the present study, we identified 10 SOD gene grapevine, including three types (6 Cu/Zn SODs, 2 FeSODs, and 2 MnSODs), which were unevenly distributed on 12 chromosomes. Syntenic analysis revealed that VvMSD1 and VvMSD2 were derived from segmental duplication, and VvCSD4 and VvCSD5 belong to a pair of tandemly duplicated genes. Promoter analysis indicated that VvSODs contain large amounts of light-responsive cis-elements which may contribute to light stress resistance. Furthermore, the expression profiles of the 10 VvSOD genes in the different tissues and abiotic stresses indicate that a few genes displayed tissue-specific expression patterns and VvSODs play roles in different aspects of abiotic stress. Finally, coexpression networks of VvSODs and TF genes under the four abiotic stresses were generated based on the microarray data. VIT_ 17s0000g01260 predicted to regulate VvCSD3, coexpression analysis verified the corelationship of these two genes. This suggested that VIT_17s0000g01260 may participate in the regulation of VvCSD3. However, the regulation mechanism needs further investigation.
This was the first comprehensive study of the SOD gene family in grapevine and may provide valuable information for understanding the classification, evolution, and putative functions of the grapevine SOD family on the genome-wide scale.

Data Availability
Previously reported microarray data were used to support this study and are available at the GEO dataset (https:// www.ncbi.nlm.nih.gov/geo/). The accession numbers are in the following: GSE36128, GSE31594, and GSE31675. The qRT-PCR data used to support the findings of this study are included within the supplementary information file (Table S5).

Conflicts of Interest
The authors declare that there is no conflict of interest regarding the publication of this paper.