Genome-wide identification of OSCA gene family and their potential function in the regulation of dehydration and salt stress in Gossypium hirsutum

Abstract

Background

Cotton (Gossypium hirsutum) provides the largest natural fiber for the textile manufacturing industries, but its production is on the decline due to the effects of salinity. Soil salt-alkalization leads to damage in cotton growth and a decrease in yields. Hyperosmolality-gated calcium-permeable channels (OSCA) have been found to be involved in the detection of extracellular changes which trigger an increase in cytosolic free calcium concentration. Hyperosmolality-induced calcium ion increases have been widely speculated to be playing a role in osmosensing in plants. However, the molecular nature of the corresponding calcium ion channels remains unclearly. In this research work, we describe the OSCA genes and their putative function in osmosensing in plants by carrying out genome-wide identification, characterization and functional analysis of the significantly up-regulated OSCA gene, GhOSCA1.1 through reverse genetics.

Result

A total of 35, 21 and 22 OSCA genes were identified in G. hirsutum, G. arboreum, and G. raimondii genomes, respectively, and were classified into four different clades according to their gene structure and phylogenetic relationship. Gene and protein structure analysis indicated that 35 GhOSCA genes contained a conserved RSN1_7TM (PF02714) domain. Moreover, the cis-regulatory element analysis indicated that the OSCA genes were involved in response to abiotic stress. Furthermore, the knockdown of one of the highly up-regulated genes, Gh_OSCA1.1 showed that the virus-induced gene silenced (VIGS) plants were highly sensitive to dehydration and salinity stresses compared with the none VIGS plants as evident with higher concentration levels of oxidant enzymes compared with the antioxidant enzymes on the leaves of the stressed plants.

Conclusion

This study provides the first systematic analysis of the OSCA gene family and will be important for understanding the putative functions of the proteins encoded by the OSCA genes in cotton. These results provide a new insight of defense responses in general and lay the foundation for further investigation of the molecular role played by the OSCA genes, thereby providing suitable approaches to improve crop performance under salinity and drought stress conditions.

Background

Salt and dehydration stresses are the major forms of abiotic stress factors which limit the growth and development of the plant (Liu et al. 2010). A number of researchers have tried to explore the mechanism of salt and dehydration stress responses, although it is complicated (Nakashima and Yamaguchi-Shinozaki 2013; Qiu et al. 2011; Ullah and Sun 2018). Therefore, some potential signal pathways were proved in salt and dehydration stress response (Munns 2005; Zhu 2016). Moreover, a number of stress-responsive genes have been found to play a significant role in enhancing plants adaptation to various forms of abiotic stress factors such as drought and salinity stress (Magwanga et al. 2018). Furthermore, several investigations have been done in order to understand the plant’s response or regulatory mechanism under salt and/or drought stress conditions (Deng et al. 2018; Sanchez-Barrena et al. 2004; Taji et al. 2004; Wu et al. 1996; Zhu et al. 2018; Zhu 2016). Salt-Overly-Sensitive (SOS) pathway was the first abiotic stress response signal pathway to be discovered in plants (Zhu 2000). Moreover, studies on the SOS pathways have shown that calcium ions are integral in the SOS salt-dehydrative responsive pathways in plants (Da and Ploy 2012; Siaud et al. 2010). In this pathway, the cytosolic calcium signal was sensed by the EF-hand calcium-binding protein (SOS3) under salt stress. Then, SOS3 interacts with and activates SOS2, a serine/threonine protein kinase (Ishitani et al. 2000). Previous studies showed that plants have development ABA-independent and ABA-dependent signal pathway to perceive and response to dehydration stress (Nakashima and Yamaguchi-Shinozaki 2013; Podia et al. 2018). Dehydration-responsive elements (DRE) play an important role in the ABA-independent pathway (Gupta et al. 2014; Pardo et al. 1998). The ABA-responsive element (ABRE) is involved in the ABA-dependent signal cascade pathway (Yoshida et al. 2014). However, the osmotic stress response is an important and common mechanism to regulated salt and dehydration stress, the mechanism underlying the early response to osmotic stress in plants remains undiscovered (Shavrukov 2012).

Hyperosmolality-induced change in Ca²⁺ level was widely speculated to be involved in osmotic stress regulation in plants (Zhu 2002). The intracellular free calcium concentration is increased under dehydration and salt stress in plants (Knight et al. 1997; McAinsh and Pittman 2009). The hyperosmolality-induced free calcium concentration increase (OICI) is the very first process to mitigate the effects of osmotic stress (Knight et al. 1997). Furthermore, the osmotic stimuli-gated Ca²⁺ permeable channels, osmosensors, and the regulated free calcium concentration have been observed in bacteria under osmotic stress (Árnadóttir and Chalfie 2010). Moreover, the AtOSCA, encoding a membrane protein, was involved in osmotic stress response as a hyperosmolality gated calcium-permeable channel in Arabidopsis thaliana. Fifteen and 11 OSCA family genes were identified in Arabidopsis and Oryza sativa (Kiyosue et al. 1994; Li et al. 2015), respectively. In Arabidopsis, early response to dehydration (ERD) genes were cloned and thought to be involved with dehydration-induced osmotic stress. ERD4 encodes a protein which contains a conserved DUF221 domain (Rai et al. 2012). The conserved DUF221 domain, including seven transmembrane regions, was renamed RSN1_7TM domain (PF02714) (Ganie et al. 2017). The previous study has shown that OSCA genes encode a protein, which contains a highly conserved RSN1_7TM domain (Camargo et al. 2007; Ganie et al. 2017; Rai et al. 2012; Shinozaki and Yamaguchi-Shinozaki 2000). Therefore, identifying the OSCA gene family will provide a potential resource to improve the deep understanding of regulation to dehydration and salt stress.

In this study, a total of 35, 21, 22 OSCA family members were identified in Gossypium hirsutum, G. arboreum and G. raimondii, respectively. Physical and chemical characteristics of the protein encoded by the GhOSCA genes were analysed. Phylogenetic relationships, chromosome location, gene and protein structure analysis were performed among these OSCAs. Furthermore, OSCA gene family member expansions were deeply analyzed for better understanding by performing the gene duplication events analysis. Expression levels in various organs/tissues and under dehydration and salt stress were analysis in our study. Gene silencing of GhOSCA1.1 proved the potential function of the novel OSCA gene and its involvement in enhancing dehydration and salt-induced osmotic stress response in cotton. These results provide a new insight into defense responses in general and lay the foundation for future crop improvement.

Materials and methods

Plant material, dehydration and salt stress treatment

G. hirsutum var. marie-galante 85 (MAR85) was selected for functional analysis of the GhOSCAs under dehydration and salt stress. The G. hirsutum accessions of MAR85 are known to be distributed in Guadeloupe and Guatemala, and were introduced from USDA-ARS Southern Agricultural Research Center in College Station, Texas, USA and perennially preserved in the National Wild Cotton Nursery (Sanya, Hainan), and managed by Institute of Cotton Research, Chinese Academy of Agricultural Sciences (ICR, CAAS). The seeds of MAR85 were first germinated at 28 °C in a 16 h light/8 h dark cycle and then transplanted in a normal hydroponic solution with a Hoagland solution for a period of 3 weeks. After 3 weeks and with a fully expanded third leaf, the seedlings were exposed to salinity and drought stress, by adding 300 mmol·L^-1 of sodium chloride (NaCl) solution and 17% PEG6000, salinity and drought stress, respectively. The tissues examined were the roots and leaves, in which the samples were collected at 0 h, 3 h, 12 h, and 48 h after salt-alkali stress treatment. The samples were immediately frozen under –80 °C awaiting RNA extraction for RT-qPCR (quantitative real-time polymerase chain reaction) validation.

Identification of OSCAs in G. hirsutum, G. arboreum, and G. raimondii

Genes and proteins annotated in G. hirsutum, G. arboreum, and G. raimondii were downloaded from the COTTONGEN database (https://www.cottongen.org/). For the two cotton genomes, G. hirsutum (AD) and G. arboreum (A), their annotations were obtained from the Cotton Research Institute, Nanjing Agricultural Unversity website (http://mascotton.njau.edu.cn/) while the sequences for G. raimondii was obtained from phytozome (https://phytozome.jgi.doe.gov/pz/portal.html). The OSCA genes family members of Arabidopsis and rice, which were used for identified candidate OSCA genes of G. hirsutum, G. arboreum and G. raimondii, were retrieved from UNIPROT (https://www.uniprot.org/). AtOSCAs and OsOSCAs were aligned with the protein sequences of the G. hirsutum, G. arboreum and G. raimondii with the default parameter by local BLASTP software. The conservative RSN1_7TM domain (PF02714) of the OSCA family was used to further reconfirm the candidate OSCAs of G. hirsutum, G. arboreum and G. raimondii by PFAM database (https://pfam.xfam.org/) and online CD-search tool of NCBI (https:// www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi) (Marchler-Bauer et al. 2016). The biophysical characters of the encoded proteins were computed using the ExPASy ProtParam tool (http://us.expasy.org/tools/protparam.html). Prediction of the subcellular localization of the proteins encoded by the OSCA gene family using WoLFPSORT (https://wolfpsort.hgc.jp/).

Mapping, phylogenetic tree construction, and gene structure analysis of the OSCA gene family

Mapping of GhOSCA genes was performed using Mapchart software (Voorrips 2002). The exon/intron structures of individual OSCA genes were determined by Gene Structure Display Server (GSDS 2.0) (Hu et al. 2014). Full-length sequences of GhOSCA proteins were first aligned with the ClustalX program (http://www.clustal.org/clustal2/) (Larkin et al. 2007), and the phylogenetic trees was constructed by using two methods, the neighbor-joining (NJ) method with 1 000 bootstrap replicas, and the Maximum likehood to validate the phylogentic tree (Fan et al. 2018; Kumar et al. 2016) and the Poisson model by using MEGA 7.0 software (http://www.megasoftware.net). Meanwhile, the orthologous gene pairs of GhOSCA in A, D genomes, At and Dt subgenomes were searched via InParanoid software (http://inparanoid.sbc.su.se/cgi-bin/index.cgi). Additionally, the d_S and d_N substitution rates were calculated with the PAL2NAL web server (http://www.bork.embl.de/pal2nal#RunP2N), which uses the CODEMAL program of PAML.

RNA extraction and quantitative and real-time PCR

Results of RNA-seq were validated via quantitative real-time PCR (RT-qPCR) experiments and real-time PCR analyses were performed as the user manual of the TransScript II All-in-One First-Strand cDNA Synthesis SuperMix for PCR (TransGen Biotech) and the SYBR Premix Ex Taq II kit (Roche) described. The housekeeping gene was Ghactin7 (Forward sequence: 5’ATCCTCCGTCTTGACCTTG3’; Reverse sequence: 5’TGTC CGTCAGGCAACTCAT3’). The gene-specific primers designed using Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primer-blast/) tool and primers are listed in Table 1. The experiments of quantitative real-time PCR were performed using three biological replicates for each tissue sample and at least three technical replicates of each biological replicate. The value of genes folds change was calculated using the 2^-ΔΔC_T method.

Table 1 Primers used in RT-qPCR analysis of salt and drought response genes

Vector construction and procedure for VIGS in cotton availability of supporting data

The TRV2 (Tobacco rattle virus) vectors construct TRV2:00, TRV2:CLA1 and TRV2:GhOSCA1.1 which were prepared and introduced into Agrobacterium tumefaciens strain LBA4404. In order to monitor the silencing efficiency, the TRV2:CLA1 vector was constructed as a visual marker. Primers were used to generate TRV2 vector forward sequence “GTGAGTAAGGTTACCGAATTCCAGCGTAATTGCAGGCAGTG” and reverse sequence “CGTGAGCTCGGTACCGGATCCGAACAGGTGTCACGGTA GCA”. The Agrobacterium culture was Agroinfiltrated into two expanded cotyledons of 10-day-old soil-grown seedling of Marie-galante 85 (MAR85). The cotton seedlings were planted in a 26 °C and 16 h light/8 h dark cycle. At least 24 seedlings were inoculated for each construct. At 14 days after Agrobacterium inoculation when VIGS was established, the silenced seedlings were posted to salt and drought. At 20 days after salt-alkali stress treatment, the leaf samples were collected for expressed level, malondialdehyde (MDA), proline (PRO) and superoxide dismutase (SOD) assay.

Determination of water loss rate, malondialdehyde, superoxide dismutase, and proline assays

After VIGS infusion at the three-leaf stage of the cotton seedlings growth stage, nine cotton leaves of similar size were taken from TRV2:00, TRV2:CLA1 and TRV2:GhOSCA1.1, respectively. Leaves were cultured in an artificial climate incubator at 28 °C. Three repeats were set up. Each hour interval, the leaves were weighed, and the water loss rate of the isolated leaves was counted [Leaf sater loss rate (%) = (Leaf fresh weight–Leaf dry weight) *100%/Leaf fresh weight]. To detect the content of MDA and PRO and activity of SOD, leaves of MAR85 were collected after 48 h post to salt-alkali stress. The corresponding assay kits (Beijing Solarbio Science & Technology Co., Ltd.) were used for determining the content of MDA and PRO and the activity of SOD.

Results

Identification of OSCA genes family in the cotton genome

To explore members of the OSCA gene family in G. hirsutum, G. arboreum and G. raimondii, 16 AtOSCAs and 11 OsOSCAs protein sequences were used as a query to screen protein databases of G. hirsutum, G. arboreum, and G. raimondii genome. A total of 35, 21 and 22 candidate OSCAs of G. hirsutum, G. arboreum and G. raimondii were obtained, respectively. In previous studies, 15, 11, 10 and 21 OSCA genes were identified in Arabidopsis, rice, maize, and soybean, respectively (Gu et al. 2018). A large number of OSCA gene family members (Shan et al. 2005) in G. hirsutum may be related to the whole genome replication of cotton. But queerly, compared with the number of OSCA genes of diploid A and D genome donor species, G. arboreum (Magwanga et al. 2018) and G. raimondii (Magwanga et al. 2019b), the allotetraploid species G. hirsutum (Shan et al. 2005) showed fewer OSCA members. This result suggested that there was possible gene loss and/or as a result of chromosome rearrangement during the history of chromosome doubling and plant evolution. The results were in agreement with previous findings in other plant gene members such as the LEA genes, in which 157, 89 and 85 proteins encoded by the LEA genes were identified in G. hirsutum, G. raimondii and G. arboreum, respectively (Magwanga et al. 2018).

Furthermore, the OSCA genes of three different Gossypium species have various characteristics (Table 2). The length of the OSCA gene sequences ranged from 900 bp to 26 539 bp. The gene with the highest length of 26 539 had the highest level of intron interruption compared with all other members of the OSCA genes in G. hirsutum. The length of OSCA coding sequences ranged from 300 bp to 3 678 bp in three different cotton species. Interestingly, the length and number of OSCA introns are quite different in three Gossypium species. Above all, the various lengths of gene sequences among the OSCA gene family in cotton were the difference of intron structure. From Table 2, it can be found that the theoretical isoelectric point and molecular weight of OSCA protein have little difference, indicating that the physical and chemical properties of OSCA family genes have little difference. The isoelectric point (pI) of the majority of the GhOSCA proteins was alkaline except for GhOSCA4.1. The GRAVY values of the proteins were calculated as the sum of the hydropathy value of each residue, divided by the total number of the residues present in the sequences. Positive and negative GRAVY scores reflect hydrophobicity and hydrophilicity, respectively. Of all the three Gossypium species, the GRAVY scores of most GhOSCA proteins were positive, except GhOSCA1.14 and GhOSCA1.6 was negative, which indicated that most GhOSCA proteins were hydrophobic proteins. In addition, GhOSCAs contains multiple transmembrane domains. WoLF PSORT analysis found that most of OSCA family proteins were located in the plasma membrane, among which GhOSCA2.4, GhOSCA3.3, GhOSCA1.14, GhOSCA1.8, GhOSCA2.5, GhOSCA2.12, GhOSCA1.6, GhOSCA1.15, GhOSCA1.13, GhOSCA1.9, and GhOSCA1.7 may be located in chloroplasts and mitochondria.

Table 2 Physicochemical Properties of OSCA Gene

Phylogenetic tree relationship and gene structure analysis of the OSCA gene family in cotton

To explore the phylogenetic relationship of the cotton OSCA genes family, a phylogenetic tree was constructed using sequence protein of the OSCA gene in three different cotton species and Arabidopsis and rice. Totally, 62 OSCA genes were divided into two subfamilies (Subfamily I and Subfamily II). Subfamily I contained three groups, and Subfamily II contained one group. Each group consists of at least one of cotyledonous plants Arabidopsis and monocotyledonous plant rice, which indicates that the differentiation time of the OSCA gene family is earlier than that of mono-and cotyledons (Fig. 1). The third and fourth groups of OSCA members were small, but they were retained throughout the evolution of species, suggesting a significant role in a biological process. From Fig. 2, it can be seen that the numbers of G. arboreum and G. raimondii of the OSCA family genes were similar, and the corresponding relationship is almost one-to-one, whereas in the G. hirsutum the OSCA family gene has a high number of amplification, which is in accordance with the species evolution relationship.

Fig. 1

Phylogenetic analysis of OSCA protein in upland cotton, Arabidopsis and rice

Fig. 2

Phylogenetic analysis of OSCA proteins in G. hirsutum, G. arboreum and G. raimondii. Ga: Gossypium arboreum; Gr: Gossypium hirsutum

Through the genetic structure analysis some gene family evolution informations were obtained, and the difference between exon and intron distribution among the members of the OSCA family is compared (Fig. 3). The results showed that G. hirsutum, G. arboreum, and G. raimondii OSCA genes were divided into four groups according to the genetic structure, which was highly correlated with the classification based on the evolutionary tree. In the exon-intron composition mode, the same group is relatively similar and the difference is greater. This conserved genetic structure between genes in the same group is consistent with their close evolutionary relationship.

Fig. 3

The gene structure of all members of OSCA gene family in Gossypium hirsutum

Protein conserved domain and motility analysis of the OSCA gene family in G. hirsutum

Members of GhOSCA family highly conservative three-function domain structure, namely the late exocytosis and Cytosolic domain of 10 TM putative phosphate and Calcium-dependent channel. All members of the GhOSCA contained three conserved motifs except GhOSCA1.7, GhOSCA2.3, GhOSCA2.8, GhOSCA2.9, GhOSCA2.12, GhOSCA3.2, GhOSCA3.3 and GhOSCA3.4, which had one conserved domain. We used the MEME software to analyzed conserved motifs in the OSCA gene family (Fig. 4). Through the analysis of the conservative motif of the OSCA gene family, most members of the same group have a similar motif, suggesting that there are functional similarities in the same group. By multiple sequence alignment of amino acids, it was found that GhOSCA family protein had a high degree of sequence conservatism, especially calcium-dependent domain channel structure (Fig. 6). The protein sequences in the same group were highly conserved, but there were significant differences between groups, especially the Group IV of subfamily II and the three group sequences of the subfamily.

Fig. 4

The protein structure analysis of all members of OSCA gene family in G. hirsutum

Chromosome location and duplication analysis of the GhOSCA genes

To examine the genomic distribution of OSCA genes in G. hirsutum chromosomes, we investigated the chromosomal location of GhOSCA (Fig. 5). The result indicated that 31 GhOSCA genes were mapped onto 19 chromosomes, while four genes which could not obviously map to any chromosome were named GhOSCA1.7, GhOSCA2.1, GhOSCA3.2, GhOSCA3.3, respectively. We found the chromosomal location relatively uneven. Some chromosomes and chromosome regions have a higher density of GhOSCA genes while others do not. Fourteen GhOSCA genes were located on At-subgenome chromosomes, respectively, on A_h01, A_h05, A_h07, chrA_h08, A_h10, A_h11, chrA_h12, chrA_h13. GhOSCA1.7, GhOSCA2.1, GhOSCA3.2 and GhOSCA3.3 were mapped to the scaffold, A_h06, D_h05, A_h06, respectively. The remaining GhOSCA genes were located in the Dt-subgenome chromosomes. Interestingly, many genes were located in clusters, especially at the top of chromosomes A_h05, A_h11, D_h11. For example, Chromosomes A_h05 had the largest number of GhOSCA genes, with four members of GhOSCAs. This unbalanced distribution of GhOSCA genes on chromosomes suggested that genetic variation existed in the evolutionary process.

Fig. 5

The distribution of OSCA genes on chromosomes

Tandem and segmental duplication events are the main causes of gene-family expansion in G. hirsutum. Two or more genes located on the same chromosome, one following the other, confirms a tandem duplication event, while gene duplication on different chromosomes or within the same chromosome but not one following the other is designated a segmental duplication event. In order to understand potential gene duplication within the G. hirsutum genome, we analyzed the occurrence of tandem duplication and segmental duplication during the evolution of this gene family. According to whole genome analysis of gene duplication, we observed that 16 pairs of GhOSCA genes originating from segmental duplication, which deeply contributed to the expansion of the GhOSCA genes (Table 3). To calculate the evolutionary time of the GhOSCA gene family, synonymous (d_S) and non-synonymous (d_N) values were calculated using PAL2NAL. A d_S/d_N value of 1 suggested neutral selection; a d_S/d_N value of > 1 suggested positive selection; a d_S/d_N value of < 1 suggested purifying selection. We found that all GhOSCA genes had d_S/d_N values of less than 1, indicated that GhOSCA genes have evolved under the effect of purifying selection (Table 3).

Table 3 The dN/ds values for duplicate GhOSCA genes

Cis-regulatory element analysis in the promoter regions of GhOSCA genes

An extensive analysis of 1 500 bp upstream promoter region of GhOSCA genes, we found that cis-regulatory element included ABA-responsive elements (ABREs), low-temperature responsive elements (LTRs), defense and stress-responsive elements (TC-rich repeats), salicylic acid responsive elements (TCA-elements), heat stress-responsive elements (HSEs), MeJA-responsive elements (TGACG-motifs and CGTCA-motifs), MYB-binding sites (MBS) (Table 4). However, ABREs, TCA-elements, and TGACG-motifs belong to plant hormone-responsive elements. ABREs, TCA-elements, and TGACG-motifs are involved in ABA, SA and MeJA responsiveness, respectively. TCA-elements are the most abundant cis-regulatory hormone responsive element in the promoters of GhOSCA genes, as 27 gene members contained TCA-elements. Both CGTCA-motifs and TGACG-motifs were involved in the SA reaction. In total, 17 members contained ABRE-elements. The other important type of cis-regulatory elements in the upstream regions of GhOSCA genes are the environmental stress-related elements. In total, four types of elements were found that respond to four respective kinds of external environmental stresses. These were low-temperature-responsive (LTR), stress-responsive TC-rich repeats, heat-stress-responsive (HSEs) and drought responsive (MBSs). In total, 30 members contained TC-rich; 32 members contained HSEs; 26 members contained MBSs; and 17 members contained LTR-element. Among them, HSEs are the most enriched cis-regulatory element in all promoter sequences. We surmised that external environmental stress could induce the expression of GhOSCA genes through its response cis–regulatory element and further improve the resistance of plants to environmental stress.

Table 4 The cis-regulatory element analysis of GhOSCA promoters

Expression profiling of the GhOSCA genes under drought and salinity stress conditions

Gene expression pattern is usually related to the gene’s function. Previous studies have indicated that the OSCA gene plays an essential role in plant growth and development. To understand the expression profiles of these 35 GhOSCA genes in G. hirsutum, we used transcriptome data to assess the expression pattern under salt and drought stress. In the environment of drought and salt stress, different genes showed different expression patterns in the roots and leaves (Fig. 6). The analysis revealed that 16 GhOSCA genes (GhOSCA1.1/1.2/1.3/1.4/1.5/1.6/1.16/2.4/2.5/2.9/2.10/2.11/3.1/ 3.2/3.3/3.4) responded to salt and drought stresses, whereas the expression of other genes was not significantly altered under different stresses. Of which 7 GhOSCA genes (GhOSCA1.1/1.2/2.5/3.3/3.4/4.1/4.2) were notably up-regulated under salt and drought treatment based on the transcriptome data, and were selected for further analysis by RT-qPCR (Fig. 7).

Fig. 6

Expression analysis of GhOSCA genes in G. hirsutum under salt and drought stresses. The RNA-Seq expression profiles of G. hirsutum were used to identify the relative expression levels of GhOSCA genes. Levels of gene expression are depicted in different colors on the scale. Red color represents high expression and green color represents low expression

Fig. 7

Expression analysis of 10 selected GhOSCA genes using quantitative real-time RT-PCR (RT-qPCR). (a) RT-qPCR analysis of the selected GHOSCA genes under drought stress conditions, imposed by adding 17% of PEG-6000. (b) RT-qPCR analysis of the selected GHOSCA genes under salt stress conditions, imposed by adding 300 mM of NaCl solution. The relative expression level of 10 selected GhOSCA genes was normalized to the reference gene histone 2 in different tissues. The transcripts in non-stressed was set as 1 for each gene in different tissues. The bars show the standard deviation of three technical repeats. Different letters indicate significant differences in the expression levels of the genes in tissues at different time, 0 h, 24h and 48h of drought stress exposure, while for salt stress conditions, samples were taken at 0h, 3h, 12h and 48h of post salt stress exposure (ANOVA; P < 0.05). 0 h: normal conditions

Under salt stress, some of the GhOSCA genes were found to exhibit a moderately high expression level in root and leaf tissues. In contrast, GhOSCA1.1 and GhOSCA1.2 transcript levels were higher in roots. Moreover, GhOSCA2.2 and GhOSCA2.1 exhibited significantly higher levels of expression in roots, whereas in leaves it showed very low expression. However, two genes, GhOSCA3.1 and GhOSCA3.2 displayed an up-regulation tissues of all plant materials analysed. Moreover, GhOSCA1.3 and GhOSCA1.4 were significantly up-regulated in roots, while GhOSCA4.1 and GhOSCA4.2 were not significantly expressed under salt stress.

The number of genes induced by drought treatment was higher than in salt treatment, and they showed different expression levels. Here, we found that most GhOSCA genes were up-regulated in all organs except GhOSCA1.3, GhOSCA 1.4, GhOSCA 1.8, GhOSCA 1.9, GhOSCA 1.14, GhOSCA 1.16, and GhOSCA 1.17 which were down-regulated in most tissues. Moreover, GhOSCA3.3 and GhOSCA3.4 were highly up-regulated in leaves, but exhibited differential expression pattern on root tissues. However, GhOSCA1.16 and GhOSCA1.8 were significantly up-regulated in leaves, but GhOSCA3.1 and GhOSCA3.2 showed insignificantly expression under drought stress.

Increased salt and dehydration stress sensitivity in the GhOSCA1.1 virus-induced gene silenced plants

To further investigate the functions of GhOSCA1.1, specific primers were designed for reverse genetics by adopting virus induced gene silencing (VIGS) method. Agrobacterium strain of LBA4404 was transformed with three vectors, TRV2:CLA1, TRV: 00 and TRV2:GhOSCA1.1, respectively. A relatively tolerant upland cotton, MAR85 was used, the vector containing the knocked gene, and the positively controlled vector (TRV:00) were infused to the seedlings cotyledons, and were allowed to grow under normal conditions till the emergence of the third true leaf under hydroponic condition. The plants infused with an albino mutant designated CLA1–1 (for “cloroplastos alterados”, or “altered chloroplasts”) showed albino-like traits on their leaves. The CLA1–1 plants behave like wild-type in their capacity to etiolate and produce anthocyanins indicating that the light signal transduction pathway seems to be unaffected (Estévez et al. 2002). Albino leaves were observed in TRV2:CLA1 inoculated seedlings after 7 days of inoculation (Fig. 8a). The appearance of the albino-like trait showed that the vector used was effective, and the results were in agreement to previous findings in which PDS has been used to monitor the effectiveness of the vector in the knockdown of cytochrome P450 genes in upland cotton (Magwanga et al. 2019b). The VIGS plants, the positively controlled and the wild types were exposed to drought and salt stress, and the VIGS plants ability to tolerate the effects of drought and salt stress were highly compromised. There was significantly higher rate of water loss on the leaves of GhOSCA1.1 gene-silenced plants compared with the wild types and the positively controlled plants, the TRV2:00 infused plants (Fig. 8b). This result indicated that GhOSCA1.1 gene might be related to drought resistance. The expression level of GhOSCA1.1 was checked by RT-qPCR. Compared with TRV2:00 seedlings, the expression level of GhOSCA1.1 was up-regulated in 10 (Ganie et al. 2017) gene-silencing seedlings after 20 days of inoculation (Fig. 8c). The difference was not observed between infected seedlings. This result suggested that lower expression levels of GhOSCA1.1 could not alter the growth and development of cotton. Then, WT, TRV2:00 and TRV2:GhOSCA1.1 seedlings were exposed to salt stress (300 mmol·L^-1 NaCl) and dehydration stress. The leaves of TRV2:GhOSCA1.1 seedlings were withered and wilting, compared with WT and TRV2:00 seedlings after 2 days of salt stress treatment (Fig. 8d). A similar morphological character was observed after dehydration stress (Fig. 8e). Additionally, compared with WT and TRV2:00 seedlings after 2 days of salt and drought stress treatment, the dehydration rate, proline, and the SOD content were significantly lower in the VIGS plants. On the contrary, the MDA was higher in TRV2:GhOSCA1.1 seedlings (Fig. 8f). The higher concentration levels of the MDA in the leaf tissues of VIGS plants showed that the plants suffered more of oxidative stress compared with the wild types and the positively controlled plant under drought and salt stress conditions. The results obtained were in agreement with the previous findings in which the Gh_A05G2067 (GT-2) knocked out plants registered higher concentration levels of MDA, hydrogen peroxide and significant reduction on the concentration level of catalase (CAT), peroxidase (POD) (Magwanga et al. 2019a). Therefore, these results suggested that GhOSCA1.1 gene may improve salt and drought tolerance of cotton.

Fig. 8

VIGS validates the function of GhOSCA1.1 gene. a: The phenotypes of TRV2:CLA1, CK, TRV2:00 and TRV2:GhOSCA1.1 seedlings, b: The water loss rate of CK, TRV2:00 and TRV2:GhOSCA1.1 seedlings. c: the phenotypes of CK, TRV2:00 and TRV2:GhOSCA1.1 seedlings were observed 48 h after 17%PEG treatment. d: The silencing efficiency of GhOSCA1.1 gene in seedlings. e: The phenotypes of CK, TRV2:00 and TRV2:GhOSCA1.1 seedlings were observed 48 h after salt stress treatment. f: The activity of SOD in TRV2:00 and TRV2:GhOSCA1.1 seedlings after salt and drought stress treatment. g: The contents of MDA in TRV2:00 and TRV2:GhOSCA1.1 seedlings after salt and drought stress treatment. h: The contents of PRO in TRV2:00 and TRV2:GhOSCA1.1 seedlings after salt and drought stress treatment

Discussion

Effects of abiotic stress on cotton growth and yield quality, and their response mechanism

Xinjiang has become the largest cotton planting area in China, but the soil salinity and water shortage are serious stresses, which greatly limit the production and improvement of cotton fiber quality and yield (Zhang et al. 2014). Therefore, surveying the endogenous salt-resistant genes in the whole genome of Gossypium is a practical and imperative way to provide a resource for further enhancing the salt and drought stress resistance. In the long evolutionary process, plants have evolved some shared biological processes in response to abiotic and biotic stress (Ahmed et al. 2013; Bihmidine et al. 2014; Podia et al. 2018; Qiu et al. 2011; Reguera et al. 2014; Shavrukov 2012). For instance, salt and drought stresses both induce osmotic stress in the plant (Shavrukov 2012). Similarly, homeostasis of cellular osmotic is responsible for ensuring that cotton grows and develops normally under salt and drought stress (Shi et al. 2014; Zhang et al. 2014). In previous studies, AtOSCA was found to be involved in osmotic stress response as a hyperosmolality gated calcium-permeable channel in Arabidopsis thaliana (Yuan et al. 2014). Moreover, the AtOSCA protein contains a conservative trans-membrane domain, which was also found among the G. hirsutum OSCA protein. These discoveries provide a new insight to investigate the OSCA gene family of G. hirsutum under salt and drought stress. Furthermore, carrying out the expression analysis of the GhOSCAs genes under salt and dehydration stresses will facilitate selection of the potential target genes.

Phylogenetic analysis of the proteins encoded by the OSCA genes in cotton and other plants

Upland cotton provides the largest natural fiber for the textile industry in the world. G. hirsutum, allotetraploid upland cotton, contains A-subgenome and D-subgenome. Gossypium, dicotyledonous plants, diverged from its relatives approximately 10–15 million years ago (MYA). Researchers thought that G. arboreum and G. raimondii are the donor species of A-subgenome and D-subgenome, respectively. The allopolyploid kinds of cotton emerged about 1–2 MYA due to an intergenomic hybridization event between A and D genomes (Flagel et al. 2012; Senchina et al. 2003; Shan et al. 2005). Therefore, studying the phylogenetic relationship of OSCAs in G. arboreum, G. raimondii and G. hirsutum will enhance the understanding of OSCA gene family diversification during the history of evolution and domestication. OSCA genes of dicotyledonous plant cotton, Arabidopsis, and monocotyledonous plant rice were divided into four clusters, which were named Group I-IV based on the phylogenetic tree (Fig. 1). This result is consistent with previous studies (Li et al. 2015; Yuan et al. 2014). Interestingly, every group included OSCAs of cotton, Arabidopsis, and rice, and OSCAs of dicotyledonous cotton and Arabidopsis were clustered closer than OSCAs of the monocotyledonous plant rice, which indicated that OSCA family Group I-IV split long before the separation of cotton, Arabidopsis and rice. What is more, G. hirsutum D-subgenome and G. raimondii have the closest relationship, and G. hirsutum A-subgenome and G. arboreum have the closest relationship, which further supported G. arboreum and G. raimondii is the donor species of A-subgenome and D-subgenome, respectively. The exception to this is that GrOSCA2.1, GrOSCA2.6, GrOSCA2.7, GaOSCA2.3, GaOSCA2.6, GaOSCA2.9, GaOSCA2.8, and GaOSCA2.9 do not have a close relationship with any OSCA family gene of G. hirsutum. This result suggested gene-losing events haven occurred during the forming of allotetraploid upland cotton.

Gene structure, cis-regulatory element and gene expression analysis

Protein structure and gene structure are closely related to gene function. Previous studies have shown that OSCA genes in most higher plants contain three conserved domains, namely late exocytosis (Pfam13967), cytosolic domain of 10 TM putative phosphate transporter (Pfam14703, DUF4463) and calcium-dependent channel (Pfam02714, DUF221)(Yuan et al. 2014). In this study, GhOSCA1.7, GhOSCA2.1, GhOSCA2.3, GhOSCA2.12, GhOSCA2.8, GhOSCA2.9, GhOSCA3.1, GhOSCA3.2, GhOSCA4.1 and GhOSCA4.2 which contain RSN1_7TM superfamily domain, without the RSN1_7TM domain. In addition, due to the long intron length of GhOSCA1.6, gene length (26.5 Kb) is much larger than other genes of the OSCA gene family in G. hirsutum and GhOSCA1.6 contain a long Cnd2 super family domain. Those results suggested a more complex function of GhOSCA1.6. On the contrary, OSCA1.1-OSCA1.5 protein structures were similar to that of AtOSCA, which suggested that these five OSCA genes were putatively involved in osmotic stress response as a hyperosmolality gated calcium-permeable channel. Furthermore, we found the same groups of the GhOSCAs to have similar gene structure, suggested the most conserved duplication events occurred during the OSCA gene family expansion in the same group.

Gene expression patterns can provide important clues to gene function, which is thought to be related to the differentiation of promoter regions (Xue et al. 2008). Cis-regulatory regulatory elements contained in gene’ promoter regions play key role in conferring the developmental and environmental regulation of gene expression. In this research, members of the OSCA gene family contain a variety of environmental stress response elements, which can improve stress tolerance. There are more elements related to drought and ABA reaction, and fewer elements related to salt reaction. Based on the transcriptome results, we can find that GhOSCA1.1, GhOSCA1.9, GhOSCA1.14, GhOSCA1.1, GhOSCA2.12 were up-regulated significantly, but analysis of cis-regulatory elements found that they did not contain saline-alkali stress response element. This result indicates that when plants are under saline-alkali stress, they induce the expression of other stress responsive elements, or hormone responsive elements, so to regulate the gene expression thereby improving their tolerance to saline-alkali stress.

Knockdown of novel OSCA gene reveals their putative role in enhancing drought and salt stress in cotton

Dehydration and salt stress limited the cotton yield, although cotton is a typical plant with abiotic stress tolerance (Van Iersel and Oosterhuis 1996; Watanabe et al. 2000). Osmotic stress is an important phase to dehydration and salt stress response (Yuan et al. 2014). In the previous study, Osmoregulation occured during turgor-driven cell expansion of developing cotton fibers (Smart et al. 1998). Previously, Ca²⁺ and calmodulin-dependent signal pathway regulate salt and dehydration tolerance response in the plant (Pardo et al. 1998; Saijo et al. 2000). Previous studies have shown that AtOSCA genes were expression in leaves, flowers, and roots in Arabidopsis (Yuan et al. 2014). In this study, expression levels of GhOSCA genes in three different accessions of G. hirsutum races were investigated under salt and dehydration stress by RNA-seq. We found that GhOSCA genes expression pattern in the tissues analysis exhibited significant variation, and all the genes exhibited tissue specificity, which indicated that each member of the GhOSCA gene family played a specific role in different tissues/organs to regulate osmotic stress. Furthermore, we reconfirmed the transcriptional expression level by RT-qPCR. Interestingly, GhOSCA1.1, an orthologous gene pair to AtOSCA, was significantly up-regulated under salt and dehydration stress conditions, which demonstrated that GhOSCA1.1 was a potential gene with significant role in enhancing salinity and dehydration tolerance in cotton.

TRV2 vector of GhOSCA1.1 was constructed to investigate the salt and dehydration stress regulation by VIGS. The GhOSCA1.1-gene-silenced plant showed obvious wilting. Statistical analysis showed that the rate of water loss gradually increased VIGS-plants compared with their wild types. In particular, TRV2:GhOSCA1.1 seedlings showed a significantly higher rate of water loss and MDA concentration after drought stress exposure, but lower SOD and POD activity than controlled and the TRV:00 infused seedlings, which indicated that the sensitivity of TRV2:GhOSCA1.1 seedlings to drought and/or salt stresses was increased after post dehydration and salt stress treatment.

Conclusions

A total of 78 OSCA genes were identified in the three cotton species, in which 35, 21 and 22 proteins encoded by the OSCA genes were obtained in G. hirsutum, G. raimondii and G. arboreum, respectively. The genes phylogenetically grouped into four groups, which were in agreement with the previous findings. The physiochemical properties of the proteins encoded by the OSCA genes showed that the majority of the protein encoded by the OSCA genes in cotton ranged from − 0.245 to 0.706, which implied their GRAVY values were less than 1, and thus were hydrophobic in nature. Moreover, segmental duplication was found to be the major evolutionary mechanism underlying the duplication of the various OSCA genes in cotton. RT-qPCR analysis of the G. hirsutum OSCA genes under drought and salinity stress conditions, showed that Gh_A05G1480 (GhOSCA1.1) is evident by higher concentration levels of MDA and significant reduction in SOD and proline under drought and salt stress conditions, but when the gene was knocked down, the VIGS-plants showed increased sensitivity to drought and salt stress conditions. This study provides the first systematic analysis of OSCAs in cotton and provides a new insight of defense responses in general and lays the foundation for future crop improvement.