Evolutionary analysis of the OSCA gene family in sunflower (Helianthus annuus L) and expression analysis under NaCl stress

Plant material

The sunflower salt-tolerant inbred line 19S05 (bred by Bayannur Institute of Agricultural and Animal Sciences) was selected for this study. This material had a plant height of 200 cm, number of leaves was approximately 30, leaves were green, without branches, and the inclination of the disc was 4. The diameter of the disc was 20 cm, yellow flowers were ligulated, and the weight of 100 seeds was 16 g. The grains were narrow oval, 2.00 cm long and 0.95 cm wide, with brown white edges, no streaks, and strong salt tolerance. Selected and full 19S05 seeds were washed with water, surface sterilized with 3% H₂O₂ for 10 min, and then placed in plastic pots filled with sterilized soil. After culturing four true leaves, a 200 mM NaCl stress treatment was performed. Root samples were taken at 0 h of treatment and then at 1 h, 3 h, 6 h, 12 h, and 24 h after treatment. Samples were then snap-frozen in liquid nitrogen (three biological replicates for each treatment) and stored in a −80 °C refrigerator.

Identification and cioinformatics analysis of OSCA genes

Helianthus annuus, Lactuca sativa var. Angustata, Gossypium hirsutum, Glycine max (Linn.) Merr, Arabidopsis thaliana, Oryza sativa, Vitis vinifera, and Zea mays genomic and proteomic data were downloaded from the Ensembl Plants database (http://plants.ensembl.org/index.html) (Bolser et al., 2016). Arctium lappa, Cichorium endivia, and Cichorium intybus genomic and proteomic data were downloaded from the NCBI database (https://www.ncbi.nlm.nih.gov/). Chrysanthemum morifolium genomic and proteomic data were downloaded from the Chrysanthemum genome database (http://www.amwayabrc.com/zh-cn/). Carthamus tinctorius L. genomic and proteomic data were downloaded from the NGINX database (http://118.24.202.236:11010/filedown/). Additionally, we used the OSCA gene domain Hidden Markov model (PF02714) and HMMER software (http://hmmer.org/) to identify protein sequences validated in the NCBI-CDD (https://www.ncbi.nlm.nih.gov/cdd/) database (Mistry et al., 2021). ExPASy software (http://cn.expasy.org/tools) was used on the physicochemical data to calculate the protein, and EuLoc software performed subcellular localization prediction (Wilkins et al., 1999; Chang et al., 2013).

Phylogenetic and collinear analysis of the OSCA gene family

The OSCA protein sequences from Helianthus annuus, Arctium lappa , Chrysanthemum morifolium, Cichorium endivia, Cichorium intybus, Lactuca sativa var. Angustata, Carthamus tinctorius, Gossypium hirsutum, Glycine max (Linn.) Merr, Arabidopsis thaliana, Oryza sativa, Vitis vinifera, and Zea mays were used for multiple sequence alignment. A phylogenetic tree was then constructed using the neighbor-joining method (the bootstrap method value was set to 1,000 and the remaining parameters were set to default values) (Hall, 2013). The nwk file was obtained using MEGA11 software, and a phylogenetic tree of the OSCA gene was made. The related collinearity map was constructed using the circle gene view function in TBtools (Chen et al., 2020; Wang et al., 2012).

Chromosomal location, gene structure, and motif analysis

The chromosomal location information for the members of the sunflower OSCA gene family was extracted using TBtools software, and the chromosomal location map was drawn using Mapchart software (Chen et al., 2020; Voorrips, 2002). A phylogenetic tree of Helianthus annuus, Arctium lappa L., Chrysanthemum morifolium, Cichorium endivia, Cichorium intybus, Lactuca sativa var. Angustata, and Carthamus tinctorius L. OSCA sequences was constructed using MEGA11 software (Wang et al., 2012). OSCA motif types were analyzed using Multiple Expectation maximizations for Motif Elicitation (MEME: http://meme-suite.org/) (Bailey et al., 2009). Based on the mRNA sequence of the OSCA gene, the intron-exon structure was analyzed. Graphs were made using TBtools software (Chen et al., 2020).

Analysis of cis-acting elements of the sunflower OSCA gene

The upstream 2,000 bp sequence of the sunflower OSCA gene was extracted, and the possible cis-acting elements were predicted using the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). After screening, the TBtools software was used to map (Chen et al., 2020).

RNA-seq analysis

Transcriptome sequencing of different tissues of sunflower under different hormone and salt stress conditions was downloaded from NCBI (SRP092742 and PRJNA866668). Fastp software was used for filtering, removing adapters, and quality control of the raw data. Hisat2 software aligned the filtered data to the sunflower reference gene, and StringTie software was used for expression quantification (Kim et al., 2019; Pertea et al., 2015). Gene expression was assessed using the FPKM method, and the expression data were normalized to log10 (FPKM+1). Then, the expression heatmap was drawn with TBtools software (Chen et al., 2018).

qRT −PCR

Primers were designed in specific regions of the sunflower OSCA gene sequence using Primer 6.0 software. The HaActin gene was chosen as the reference gene for qRT-PCR analysis (Table S1; Li & Brownley, 2010). Root tissue cDNA was used as the template to detect the expression of candidate genes by qRT −PCR, and each sample was repeated three times. The total reaction system was 20 µl, qRT −PCR program: 94 °C for 30 s; 94 °C for 5 s, 60 °C for 15 s, 72 °C for 10 s for 38 cycles; and 4 °C for the end. Each group of data was replicated three times, and the 2^−ΔΔCt method was used for relative quantitative analysis (Tanino et al., 2017).

WGCNA

The DEseq2 software package was used to identify differential genes under salt stress for WGCNA analysis. We selected β = 6 to process the original relation matrix in order to get a non-proportional adjacency matrix. The adjacency matrix was further transformed into a topological overlap matrix (TOM). The minimum number of genes in a module was 30, the threshold for merging similar modules was 0.25 (cutHeight =0.25), and the network type was “signed” (type =“signed” or networkType =“signed”). Cytoscape−3.8.2 software was used for visualization (Shannon et al., 2003). We performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses using the cluster Profiler software package in R language.

Identification of the OSCA gene family

We systematically studied the changes in the copy number of the OSCA gene family during the sunflower’s evolution. First, using HMMsearch the OSCA gene was comprehensively searched from seven Compositae species (Helianthus annuus, Arctium lappa, Chrysanthemum morifolium, Cichorium endivia, Cichorium intybus, Lactuca sativa, and Carthamus tinctorius). The search results were validated against the NCBI-CDD database. Helianthus annuus, Arctium lappa, Chrysanthemum morifolium, Cichorium endivia, Cichorium intybus, Lactuca sativa, and Carthamus tinctorius contained 15, 17, 20, 19, 19, 21, and 14 OSCA sequences, respectively. We named them HaOSCA1.1∼HaOSCA3.1 according to the 15 sunflower sequences and the phylogenetic tree of 12 other plants. The protein encoded by sunflower OSCA family genes contained 648-838 amino acid residues, the length of the open reading frame (ORF) was 1,947–2,517 bp, the theoretical isoelectric point was 7.61−9.94, and the relative molecular weight was 73.42–96.62 kDa. The protein subcellular localization showed that HaOSCA was localized to the endoplasmic reticulum, vacuole, chloroplast, golgi apparatus. and mitochondrion (Table 1).

Fifteen HaOSCA genes were distributed on 10 sunflower chromosomes (Chr03, Chr04, Chr05, Chr06, Chr07, Chr09, Chr12, Chr15, Chr16, and Chr17) (Fig. 1). Among these, Chr07 and Chr16 staining contained the most (three) OSCA genes. Except for Chr10, which contained two OSCA genes, the other chromosomes contained only one OSCA gene.

Evolutionary analysis of the OSCA gene family

To better understand the evolutionary relationship of the sunflower OSCA family genes, we constructed a phylogenetic tree of the full-length OSCA protein sequences of Helianthus annuus, Gossypium hirsutum, Glycine max (Linn.) Merr, Arabidopsis thaliana, Oryza sativa, Vitis vinifera, and Zea mays (Fig. 2A). According to the results, the evolutionary tree was divided into four groups. Compared with Gossypium hirsutum, Glycine max (Linn.) Merr, Arabidopsis thaliana, Oryza sativa, Vitis vinifera, and Zea mays, there was no sunflower OSCA protein in Group 4 (Fig. 2), which may have been caused by the loss of genes in the sunflower during the evolution process. The number of OSCA proteins in Group 1 and Group 2 subgroups was consistent both in the sunflower and in Gossypium hirsutum, Glycine max (Linn.) Merr, Arabidopsis thaliana, Oryza sativa, Vitis vinifera, and Zea mays (Fig. 2B). The number of OSCA genes in sunflower, Arabidopsis thaliana, and Glycine max (Linn.) Merr was basically the same, and the sunflower genes of the same branch were more closely related to those in Arabidopsis thaliana and Glycine max (Linn.) Merr. Except for soybean and cotton, the other crops in Group 3 all contained an OSCA sequence, which also indicated that the Group 3 subgroup of the OSCA gene family was more conserved during the plant’s evolution. Although Group 3 and Group 4 had relatively few members, they were retained in the evolution of plants, suggesting that they may play important roles in biological processes. However, the loss of members of sunflower Group 4 may have also been caused by functional redundancy of genes, which requires further study.

To further explore the evolutionary relationship of the sunflower OSCA gene family, we selected the sunflower OSCA gene as the core and constructed a colinear relationship between sunflower and Gossypium hirsutum, Glycine max (Linn.) Merr, Arabidopsis thaliana, Oryza sativa, Vitis vinifera, and Zea mays OSCA genes (Fig. 3). We found that the sunflower OSCA had 18 collinear gene pairs with Arabidopsis thaliana, 36 with Gossypium hirsutum, 12 with Glycine max (Linn.) Merr, eight with Oryza sativa, and 10 with Vitis vinifera. There were 11 collinear gene pairs with Zea mays (Fig. 3). The greatest collinearity between sunflower OSCA and cotton may have been related to cotton having a larger genome and the largest number of OSCA genes. In addition to cotton, the sunflower OSCA had greater collinearity with Arabidopsis thaliana and Glycine max (Linn.) Merr, which was consistent with the results of the phylogenetic tree analysis.

To further study the evolutionary relationship of the sunflower OSCA gene family, we constructed a phylogenetic tree based on the OSCA protein sequences of six Compositae species (Arctium lappa L., Chrysanthemum morifolium, Cichorium endivia, Cichorium intybus, Lactuca sativa var. Angustata, and Carthamus tinctorius L.) and Helianthus annuus (Fig. 4A). Except for Helianthus annuus and Carthamus tinctorius L., which lacked the Group 4 subgroup, all Asteraceae contained four subgroups (Fig. 4A). The Cichorium endivia and Lactuca sativa var. Angustata Group 4 subgroups contained three and four OSCA genes, respectively, while Arctium lappa L., Chrysanthemum morifolium, Cichorium endivia, and Carthamus tinctorius L all contained one, which was consistent with the results from Arabidopsis thaliana, Glycine max (Linn.) Merr, Oryza sativa, and Zea mays (Figs. 2B and 4B). Except for Chrysanthemum morifolium Group 3, which contained four sequences, there was no difference in the subgroups of Group 1, Group 2, and Group 3 based on Compositae.

We also constructed the collinear relationships for the OSCA of seven species of Compositae and found that there were three pairs of HaOSCA (HaOSCA1.1 and HaOSCA2.1, HaOSCA1.2 and HaOSCA2.2, HaOSCA1.3 and HaOSCA1.5) with tandem repeats in sunflower genes (Fig. 5A), indicating that these three gene pairs may have been derived from tandem repeats. At the same time, five pairs of HaOSCA (HaOSCA1.5 and HaOSCA1.6, HaOSCA2.2 and HaOSCA2.1, HaOSCA1.2 and HaOSCA1.1, HaOSCA1.8 and HaOSCA1.7, HaOSCA1.3 and HaOSCA1.1) showed very high identities (over 84.77%), implying that they may have arisen from segments or whole genes (Fig. 5A). Additionally, the ratio of non-synonymous to synonymous (Ka/Ks) between duplicate gene pairs was calculated, and the Ka/Ks values of all gene pairs were less than 1, indicating that these gene pairs may have undergone purifying selection (Table S2). Helianthus annuus had 32 collinear gene pairs with Arctium lappa L., 32 with Chrysanthemum morifolium, 29 with Cichorium endivia, 31 with Cichorium intybus, and 39 with Lactuca sativa var. Angustata, and 18 collinear gene pairs with Carthamus tinctorius L.. In addition to having lower collinearity with Carthamus tinctorius, Helianthus annuus OSCA had more collinearity with other Compositae species, which indicates that the OSCA gene family was conserved in the evolution of Compositae. Moreover, Helianthus annuus and Lactuca sativa var. Angustata had the most collinear gene pairs and were also close to Lactuca sativa var. Angustata OSCA in the evolutionary tree, which indicates that the OSCA of Helianthus annuus and Lactuca sativa var. Angustata may have the same biological function.

Gene structure and motif analysis of the OSCA gene of Compositae

To further understand the composition of the sunflower OSCA gene, we compared the OSCA gene structure of seven Compositae species. All OSCA genes, except Group 4, contained introns (Fig. 6). Members of the same subfamily had similar conserved motifs, and Group 4 had the simplest gene structure and least number of motifs (Fig. 6). The Pfam website (http://pfam.xfam.org/family/pf02714#tabview=tab0) showed that the RSN1_7TM domain generally contained seven transmembrane domains. In this article, motif 2, motif 3, motif 4, motif 6, motif 7, motif 8, and motif 9 showed that they may contain multiples of the seven transmembrane domains. Only a few OSCA genes (CtOSCA1.5, CiOSCA2.2, and LsOSCA4.1) contained a motif, although they were relatively long. CmOSCA2.2 was the longest and contained 10 exons, but it also contained eight motifs. The LsOSCA4.4 gene had the shortest length and contained one exon but only one motif (Motif 3). This indicates that the distribution of OSCA gene motifs in Compositae was highly correlated with the length of the genes.

Analysis of cis-acting elements in the promoter of the sunflower OSCA gene

Transcription factors (TFs) are proteins that can bind to the promoter regions of genes in a sequence-specific manner in order to regulate transcription, thereby regulating plant functions, including responses to environmental factors, growth, and development (Warren, 2002). Analysis of the 2,000 bp upstream sequence of the sunflower OSCA gene start codon showed that there were multiple stress and hormone response elements in the upstream sequence of the HaOSCA gene, with different types and numbers (Fig. 7). Three main categories of cis elements were found in the promoter sequences of HaOSCA genes. The first category was involved in phytohormones, such as abscisic acid (ABA), jasmonic acid (JA), auxin, gibberellins (GA), and salicylic acid (SA). The second category was associated with stresses, such as low-temperature responsiveness, defense, and stress responsiveness. The last category was mainly MYB binding sites. Importantly, all 14 HaOSCA genes, except HaOSCA1.8, contained the JA-responsive element (TGACG-motif and CGTCA-motif), and the ABA responsive element (ABRE) was found in almost all gene promoters (Table S3). Interestingly, cis-acting elements were not found in the promoter region of HaOSCA1.8. These results showed that HaOSCA may affect hormone signal responsiveness and stress adaptation. No cytokinin-responsive elements were identified in the HaOSCA promoter regions.

Expression analysis of the sunflower OSCA gene

Gene expression patterns are usually closely related to gene functions. To gain insight into the expression patterns of sunflower OSCA genes, we first analyzed the expression patterns of 15 sunflower OSCA genes in nine tissues (Fig. 8A). The results showed that there were nine sunflower OSCAs (HaOSCA1.1, HaOSCA1.2, HaOSCA1.4, HaOSCA1.6, HaOSCA1.7, HaOSCA1.8, HaOSCA2.5, HaOSCA2.6, and HaOSCA3.1) in pollen and stamen. The expression was higher than in other tissues, and the remaining six had higher expression levels in root, stem, leaves, bract, corolla, ovary, and ligament, which indicated that sunflower OSCA had a wide range of tissue expression types.

Plant hormones play an important role in the processes of growth, development, and stress. We analyzed the expression pattern of the sunflower OSCA gene induced by six hormones (ABA, BR, JA, GA, IAA, and SA) (Fig. 8B). Six genes (HaOSCA1.3, HaOSCA1.5, HaOSCA1.8, HaOSCA2.2, HaOSCA2.3, and HaOSCA2.4) were more strongly induced by ABA than other hormones, six genes (HaOSCA1.2, HaOSCA1.4, HaOSCA1.6, HaOSCA1.7, HaOSCA2.5, and HaOSCA2.6) were more strongly affected by IAA than other hormones, and HaOSCA1.1 was most strongly affected by JA.

Next, we analyzed the expression pattern of the sunflower OSCA gene under salt stress. All sunflower OSCA gene expression levels changed after NaCl treatment (Fig. 8C). Seven OSCA genes (HaOSCA1.1, HaOSCA1.2, HaOSCA1.5, HaOSCA1.7, HaOSCA2.2, HaOSCA2.6, and HaOSCA3.1) were upregulated, three (HaOSCA1.4, HaOSCA1.6, and HaOSCA2.5) were significantly downregulated, and the expression levels of the remaining five (HaOSCA1.3, HaOSCA1.8, HaOSCA2.1, HaOSCA2.3, and HaOSCA2.4) also changed, but not significantly (Fig. 8C). These results indicated that the 10 OSCA genes (HaOSCA1.1, HaOSCA1.2, HaOSCA1.4, HaOSCA1.5, HaOSCA1.6, HaOSCA1.7, HaOSCA2.2, HaOSCA2.5, HaOSCA2.6, and HaOSCA3.1) may play a role in the sunflower’s response to salt stress.

qRT−PCR of the sunflower OSCA gene under salt stress

With the increase of soil salinization, salt stress has become the most important abiotic stress faced by the sunflowers. Based on the previous expression analysis, we found that 10 genes (HaOSCA1.1, HaOSCA1.2, HaOSCA1.4, HaOSCA1.5, HaOSCA1.6, HaOSCA1.7, HaOSCA2.2, HaOSCA2.5, HaOSCA2.6, and HaOSCA3. 1) may be involved in the response of sunflower to salt stress. qRT −PCR was used to study the expression patterns of these 10 genes under NaCl stress. Compared with levels before stress (0 h), the expression of seven genes (HaOSCA1.1, HaOSCA1.2, HaOSCA1.5, HaOSCA1.6, HaOSCA2.2, HaOSCA2.6, and HaOSCA3.1) was significantly changed at different points in time during treatment (Fig. 9). HaOSCA1.6 was significantly downregulated and the remaining six genes were significantly upregulated. The expression levels of HaOSCA2.6 and HaOSCA3.1 reached more than a 10-fold change at 6 h of NaCl stress. These results suggest that HaOSCA2.6 and HaOSCA3.1 may play an important role in the response of sunflower to salt stress.

Protein interaction of sunflower OSCA gene

Genes typically perform their biological functions and signal transduction pathways through interacting networks, so studying the underlying interaction networks associated with gene families can provide a better understanding of their functions. To elucidate the role of OSCA genes in salt stress in the sunflower, we used WGCNA to analyze OSCA genes (HaOSCA1.1, HaOSCA1.2, HaOSCA1.5, HaOSCA1.6, HaOSCA2.2, HaOSCA2.6, and HaOSCA3.1) and the salt coexpression network of all differentially expressed genes under stress. Genes were clustered using dynamic shearing and divided into modules. By calculating the eigenvectors of each module and merging similar modules, a total of seven gene coexpression modules were obtained (Fig. 10A). In this study, OSCA genes in each module were selected as core genes, and gene interaction network diagrams were drawn using core genes and their interacting genes (Fig. 10B and Table S4). To further uncover the function of the network, we performed an enrichment analysis of all interacting genes (Figs. 10C and 10D). The GO enrichment results showed that the biological processes involved were mainly cellular component organization or biogenesis, biological regulation, multicellular organismal process, response to stimulus and cellular process, cellular components are extracellular regions, cells and extracellular regions, molecular functions are transporter activities, antioxidant activities, transcription regulator activities, and molecular carrier activity (Fig. 10C). The KEGG pathway was mainly alpha-linolenic acid metabolism, glycerolipid metabolism, peroxisome, and tryptophan metabolism (Fig. 10D). This network further demonstrated the complex functions and potential roles of the OSCA gene family in sunflower salt-stressed species. The results of this study lay a foundation for further research on the functions and molecular mechanisms of these genes in the sunflower under salt stress and provide a reference for further study of the sunflower OSCA gene family.