Genome-Wide Identiﬁcation, Expression Proﬁle, and Alternative Splicing Analysis of CAMTA Family Genes in Cucumber ( Cucumis sativus L.)

: The calmodulin-binding transcription activator ( CAMTA ), as one of the most distinctive families of transcription factors, plays an important role in plant growth and development and in the stress response. However, it is currently unknown whether CAMTA exists in cucumbers and what its function is. In this study, we ﬁrst identiﬁed four CAMTA genes in the cucumber genome using a genome-wide search method. Subsequently, we analyzed their physical and chemical properties, gene structure, protein domains, and phylogenetic relationships. The results show that the structure of CsCAMTAs is similar to that of other plants, and a phylogenetic analysis divides them into three groups. The analysis of cis-acting elements shows that most CsCAMTAs contain a variety of hormones and stress-related elements. The RT-PCR analysis shows that CsCAMTAs have different expression levels in different tissues and can be induced by IAA, ABA, MeJA, NaCl, and PEG. Finally, we analyzed the expression pattern of CsCAMTAs’ alternative spliceosomes under salt and drought stress. The results show that the expression levels of the different spliceosomes are affected by the type of stress and the duration of stress. These data indicate that CsCAMTAs participate in growth and development and in the stress response in cucumbers, a ﬁnding which lays the foundation for future CsCAMTAs’ functional research. ber. The results show the CsCAMTA genes in various at different levels FPKM > 0). Among them, CsCAMTA1 , CsCAMTA3 , and CsCAMTA4 are constitutive expressions (FPKM > 1 in all samples). CsCAMTA1 has the highest expression in the roots of 4-week-old seedlings; CsCAMTA2 and CsCAMTA3 have the highest expression levels in female ﬂowers; CsCAMTA4 has the highest expression in 12-week-old cucumber roots. CsCAMTA1 and CsCAMTA3 have low expression levels in young leaves, while CsCAMTA3 and CsCAMTA4 have the lowest expression levels in 1-week-old fruits. This indicates that CsCAMTAs may mainly act in cucumber roots, stems, leaves, and female ﬂowers, but have weak effects on the growth and development of fruits.


Introduction
As one of the most important secondary messengers in plant signal transmission [1], Ca 2+ signals play an important role in plant growth, development, and response to external stimuli [2]. External environmental stimuli can cause spatial and temporal changes in cytosolic-free Ca 2+ concentration ((Ca 2+ ) cyt), thereby stimulating a series of downstream reactions. Subsequently, calmodulin (CaM), calcium-dependent protein kinase (CDPK) and calcineurin B-like protein (CBL) that are present in the cell bind to Ca 2+ and convert extracellular signals into intracellular signals [3]. In the downstream of the CaM signaling pathway, CaM can combine with a variety of transcription factors (TFs) to cause highly specific responses. The calmodulin-binding transcription activator (CAMTA), as one of them, exists widely in organisms and plays a very important role [4].

Identification of the CAMTA Family Members in Cucumber
First, the cucumber genome sequence and protein sequence information file were downloaded from the National Center for Biotechnology Information (NCBI) (https:// www.ncbi.nlm.nih.gov/; accessed on 12 February 2021) database. The hidden Markov model (HMM) of the calmodulin-binding transcription activator was downloaded from the plant genome database (https://phytozome.jgi.doe.gov/; accessed on 20 February 2021). Second, using the HMM model as a template, the hmmsearch function of HMMER3.0 was used to compare all protein sequences of the cucumber to obtain the target protein. To determine further whether the identified protein belongs to the CAMTA gene family, we used SMART (http://smart.embl-heidelberg.de/; accessed on 20 February 2021) to analyze the protein domain and deleted the ones that did not contain GC-1, ANK repeats, and the IQ domains protein. Finally, we obtained the members of the cucumber CAMTA gene family, and used Expasy (https://web.expasy.org/protparam/; accessed on 18 March 2021) to analyze the physical and chemical properties of these genes. The subcellular localization of CsCAMTAs' protein is predicted using the online software PlantmPLoc (http://www. csbio.sjtu.edu.cn/bioinf/plant-multi/; accessed on 18 March 2021) [22].

Gene Structure and Protein Conserved Domain Analysis
The exon-intron structure information of the CsCAMTA genes was extracted from the genome gff3 annotation file through Tbtools software. This information is visualized through the Gene Structure Display Server (GSDS v2.0; http://gsds.gao-lab.org/; accessed Agronomy 2021, 11, 1827 3 of 15 on 2 April 2021) [23]. We used SMART (http://smart.embl-heidelberg.de/; accessed on 15 April 2021) to analyze the protein domain and subsequently used Illustrator for BioSequence (IBS) software to draw a schematic diagram of the protein domain.

Phylogenetic Tree and Cis-Acting Elements Analysis
The protein sequences of Arabidopsis, tomatoes, rice, and soybeans were obtained from the Arabidopsis Information Resource (TAIR), the plant genome resources (Phytozome), and NCBI. The multiple sequence alignment of proteins was performed by ClustalW. We used the Neighbor-joining (NJ) method of MEGA-X, set the bootstrap replicates value to 1000, and constructed a phylogenetic tree of 39 CAMTA protein sequences [24]. The EvolView tool (http://www.evolgenius.info; accessed on 10 May 2021) was used to draw the phylogenetic tree [25]. The up-stream 2000 bp DNA sequences of CsCAMTA genes were obtained from the cucumber genome sequence, and cis-acting elements in the promoter region were analyzed in the PlantCARE database (http://bioinformatics.psb.ugent.be/ webtools/plantcare/html/; accessed on 19 May 2021) [26].

Transcript Analysis of CsCAMTA Genes in Different Plant Tissues
The expression data of the cucumber CAMTA gene in different periods and different organs were downloaded from the Short Read Archive (SRA) database of NCBI (accession number: SRP071224), and RNA-Seq data of the cucumber was analyzed in reference to the method of Wei et al. [27]. The expression data were converted with Log (RPKM + 1) to calculate gene expression levels. The heat map of the expression profile of CsCAMTAs was produced by using HemI [28].

Plant Materials and Treatments
Cucumber seeds ('Chinese long' inbred line 9930) were provided by Shenzhen Genomics Institute, Chinese Academy of Agricultural Sciences. We soaked the seeds in water at 55 • C for 15 min, then placed them on damp filter paper where they germinated overnight in the incubator at 25 • C. After the cotyledons were fully expanded, the seedlings were moved to the hydroponic box for cultivation. Yamazaki cucumber nutrient solution was used and replaced every three days. The environment of the growth room was controlled to have a photoperiod of 14/10 h (light/dark), an air temperature of 28/18 • C (day/night), and a light intensity of 200 µmol.m −2 s −1 . Stress treatments were carried out at the two-leaf seedling stage including PEG6000 (15%), NaCl (150 mM). We collected leaf samples for RT-PCR experiments after treating for 0, 3, 6, 12, 24, and 48 h. [18]. During the hormone treatment, we transferred the two-leaf stage seedlings to a nutrition solution containing IAA (10 µM), ABA (100 µmol/L), and methyl (Me)-JA (100 µmol/L), and then collected leaves at 0, 3, 6, 12, 24, 48 h for RT-PCR experiments [21]. Collected samples were immediately frozen in liquid nitrogen and stored at −80 • C for analysis.

RNA Extraction and Quantitative RT-PCR
Total RNA was isolated using the MiniBEST Plant RNA Extraction Kit (TaKaRa, Dalian, China). The FastQuant First Strand cDNA Synthesis Kit (Tiangen, Beijing, China) was used to synthesize cDNA according to the manufacturer's protocol. RT-PCR was performed using the SuperReal PreMix Plus kit (TIANGEN, Beijing, China) and Roche LightCycler instrument. There were three biological replicates per treatment. The primers used for RT-PCR were designed using prime5 software, as shown in Table S1. The cucumber ACTIN gene was used to normalize relative expression levels. The 2 −∆∆Ct method was used to analyze the data.

Identification and Characterization of Calmodulin-Binding Transcription Activator (CAMTA) Genes in Cucumber
Based on the completed genome sequences, using biological information technology to analyze and identify the cucumber genome, we found that there are four CAMTA gene family members in the cucumber. According to the degree of similarity with the aligned sequences, they were named CsCAMTA1-4 in order ( Table 1). The CsCAMTA genes distribute on Chr-4, -6, -7, and the amino acid length varies from 916 (CsCAMTA4) to 1102 (CsCAMTA1). Then, we analyzed their physicochemical properties and found that the pI of CsCAMTAs all concentrate between 5.59 (CsCAMTA1) and -7.59 (CsCAMTA2). CsCAMTA1 and CsCAMTA3 are slightly acidic, while CsCAMTA2 and CsCAMTA4 are weakly alkaline. Due to their grand average of hydropathicity being less than 0 and the instability index being greater than 75, they belong to hydrophilic labile proteins. In addition, subcellular localization prediction results showed that all CsCAMTAs exist in the nucleus.

Genomic Structure and Protein Domain Analysis of CsCAMTA Members
GSDS, a gene structure analysis software, was used to analyze the structure of CsCAMTA members, and we found that the number of introns in CsCAMTAs is between 11 and 12 ( Figure 1). Among them, the intron number of CsCAMTA1 and CsCAMTA3 is 12; the intron number of CsCAMTA2 and CsCAMTA4 is 11. The gene structures of different members are relatively similar. ACTIN gene was used to normalize relative expression levels. The 2 −ΔΔCt method was used to analyze the data.

Identification and Characterization of Calmodulin-Binding Transcription Activator (CAMTA) Genes in Cucumber
Based on the completed genome sequences, using biological information technology to analyze and identify the cucumber genome, we found that there are four CAMTA gene family members in the cucumber. According to the degree of similarity with the aligned sequences, they were named CsCAMTA1-4 in order ( Table 1). The CsCAMTA genes distribute on Chr-4, -6, -7, and the amino acid length varies from 916 (CsCAMTA4) to 1102 (CsCAMTA1). Then, we analyzed their physicochemical properties and found that the pI of CsCAMTAs all concentrate between 5.59 (CsCAMTA1) and -7.59 (CsCAMTA2). CsCAMTA1 and CsCAMTA3 are slightly acidic, while CsCAMTA2 and CsCAMTA4 are weakly alkaline. Due to their grand average of hydropathicity being less than 0 and the instability index being greater than 75, they belong to hydrophilic labile proteins. In addition, subcellular localization prediction results showed that all CsCAMTAs exist in the nucleus.

Genomic Structure and Protein Domain Analysis of CsCAMTA Members
GSDS, a gene structure analysis software, was used to analyze the structure of CsCAMTA members, and we found that the number of introns in CsCAMTAs is between 11 and 12 ( Figure 1). Among them, the intron number of CsCAMTA1 and CsCAMTA3 is 12; the intron number of CsCAMTA2 and CsCAMTA4 is 11. The gene structures of different members are relatively similar. For a better understanding of these genes, we analyzed the structure of the proteins encoded by these genes. The results show that the four CsCAMTA members all contain a CG-1 DNA binding domain (Pfam03859), ankyrin repeats (Pfam12796), IQ motifs (Pfam00612), and CAMBD ( Figure 2). According to the existence of TIG (Pfam01833), CsCAMTAs can be divided into two groups. CsCAMTA1, CsCAMTA3, and CsCAMTA4 belong to one group, while CsCAMTA2 belongs to another group, suggesting that there are differences in the types of functional domains of CsCAMTAs, implying that there may For a better understanding of these genes, we analyzed the structure of the proteins encoded by these genes. The results show that the four CsCAMTA members all contain a CG-1 DNA binding domain (Pfam03859), ankyrin repeats (Pfam12796), IQ motifs (Pfam00612), and CAMBD ( Figure 2). According to the existence of TIG (Pfam01833), CsCAMTAs can be divided into two groups. CsCAMTA1, CsCAMTA3, and CsCAMTA4 belong to one group, while CsCAMTA2 belongs to another group, suggesting that there are differences in the types of functional domains of CsCAMTAs, implying that there may be differences in their functions. The number of ANK domains and IQ motifs in CsCAMTAs varies from 1 to 3.

Phylogenetic Analysis of CAMTA Family Genes
To understand the evolutionary history or genetic relationship of CsCAMTA m bers better, we used the NJ algorithm to construct a phylogenetic tree among Arabido tomatoes, rice, soybeans, and cucumbers. As shown in Figure 3, 39 CAMTAs (6 AtC TAs, 7 SlSRs, 7 OsCAMTAs, 15 GmCAMTAs, 4 CsCAMTAs) are divided into three gro Among them, CsCAMTA1 and CsCAMTA2 belong to group A and have the highes mology with GmCAMTA5 and GmCAMTA7. CsCAMTA4 belongs to group B and h closer relationship with GmCAMTA8. CsCAMTA3 is in group C and has upper homo with AtCAMTA4. It can be seen from the entire evolutionary tree that CsCAMTAs the highest homology with GmCAMTAs, relatively low homology with SlSRs and AtC TAs, and the lowest homology with OsCAMTAs.

Phylogenetic Analysis of CAMTA Family Genes
To understand the evolutionary history or genetic relationship of CsCAMTA members better, we used the NJ algorithm to construct a phylogenetic tree among Arabidopsis, tomatoes, rice, soybeans, and cucumbers. As shown in Figure 3, 39 CAMTAs (6 AtCAMTAs, 7 SlSRs, 7 OsCAMTAs, 15 GmCAMTAs, 4 CsCAMTAs) are divided into three groups. Among them, CsCAMTA1 and CsCAMTA2 belong to group A and have the highest homology with GmCAMTA5 and GmCAMTA7. CsCAMTA4 belongs to group B and has a closer relationship with GmCAMTA8. CsCAMTA3 is in group C and has upper homology with AtCAMTA4. It can be seen from the entire evolutionary tree that CsCAMTAs have the highest homology with GmCAMTAs, relatively low homology with SlSRs and AtCAMTAs, and the lowest homology with OsCAMTAs. be differences in their functions. The number of ANK domains and IQ motifs in CsCAM-TAs varies from 1 to 3.

Phylogenetic Analysis of CAMTA Family Genes
To understand the evolutionary history or genetic relationship of CsCAMTA members better, we used the NJ algorithm to construct a phylogenetic tree among Arabidopsis, tomatoes, rice, soybeans, and cucumbers. As shown in Figure 3, 39 CAMTAs (6 AtCAM-TAs, 7 SlSRs, 7 OsCAMTAs, 15 GmCAMTAs, 4 CsCAMTAs) are divided into three groups. Among them, CsCAMTA1 and CsCAMTA2 belong to group A and have the highest homology with GmCAMTA5 and GmCAMTA7. CsCAMTA4 belongs to group B and has a closer relationship with GmCAMTA8. CsCAMTA3 is in group C and has upper homology with AtCAMTA4. It can be seen from the entire evolutionary tree that CsCAMTAs have the highest homology with GmCAMTAs, relatively low homology with SlSRs and AtCAM-TAs, and the lowest homology with OsCAMTAs.

Cis-Acting Regulatory Elements in the Promoters of the CsCAMTAs
To explore the possible response mechanism of CsCAMTAs to various external stimuli, we used the Plant CARE database to analyze the 2000 bp, cis-acting element concentrated distribution region of four CsCAMTA genes' promoter regions. The results show that the predicted cis-acting elements can be divided into three categories: light signal response, hormone signal response, and abiotic stress response (Table 2). In this region, CsCAMTA members contain 6-12 cis-acting elements (Figure 4). Among them, anaerobic responsive element (ARE) exists in all four CsCAMTAs. It suggests that CsCAMTAs may function when an anaerobic reaction occurs. The abscisic acid (ABA)-response element (ABRE), auxin response element (TGA-element), and MeJA-responsive element (CGTCA-motif) are all present in CsCAMTA1, indicating that it may be sensitive to hormones. CsCAMTA3 has a drought response element (MBS), low-temperature-response element (LTR), and ARE, indicating that it plays a crucial role in the response to abiotic stress.

Cis-Acting Regulatory Elements in the Promoters of the CsCAMTAs
To explore the possible response mechanism of CsCAMTAs to various external uli, we used the Plant CARE database to analyze the 2000 bp, cis-acting element con trated distribution region of four CsCAMTA genes' promoter regions. The results s that the predicted cis-acting elements can be divided into three categories: light s response, hormone signal response, and abiotic stress response (Table 2). In this re CsCAMTA members contain 6-12 cis-acting elements ( Figure 4). Among them, anae responsive element (ARE) exists in all four CsCAMTAs. It suggests that CsCAMTAs function when an anaerobic reaction occurs. The abscisic acid (ABA)-response ele (ABRE), auxin response element (TGA-element), and MeJA-responsive element (CGT motif) are all present in CsCAMTA1, indicating that it may be sensitive to horm CsCAMTA3 has a drought response element (MBS), low-temperature-response ele (LTR), and ARE, indicating that it plays a crucial role in the response to abiotic stres   ber. The results show that the four CsCAMTA genes express in various tissues at different levels ( Figure 5; FPKM > 0). Among them, CsCAMTA1, CsCAMTA3, and CsCAMTA4 are constitutive expressions (FPKM > 1 in all samples). CsCAMTA1 has the highest expression in the roots of 4-week-old seedlings; CsCAMTA2 and CsCAMTA3 have the highest expression levels in female flowers; CsCAMTA4 has the highest expression in 12-week-old cucumber roots. CsCAMTA1 and CsCAMTA3 have low expression levels in young leaves, while CsCAMTA3 and CsCAMTA4 have the lowest expression levels in 1-week-old fruits. This indicates that CsCAMTAs may mainly act in cucumber roots, stems, leaves, and female flowers, but have weak effects on the growth and development of fruits.
Agronomy 2021, 11, x FOR PEER REVIEW 7 cucumber. The results show that the four CsCAMTA genes express in various tissu different levels ( Figure 5; FPKM > 0). Among them, CsCAMTA1, CsCAMTA3 CsCAMTA4 are constitutive expressions (FPKM > 1 in all samples). CsCAMTA1 ha highest expression in the roots of 4-week-old seedlings; CsCAMTA2 and CsCAMTA3 the highest expression levels in female flowers; CsCAMTA4 has the highest expressi 12-week-old cucumber roots. CsCAMTA1 and CsCAMTA3 have low expression lev young leaves, while CsCAMTA3 and CsCAMTA4 have the lowest expression levels week-old fruits. This indicates that CsCAMTAs may mainly act in cucumber roots, s leaves, and female flowers, but have weak effects on the growth and developme fruits.

Expression Patterns of CsCAMTA Genes under Hormone and Abiotic Stress
To understand the expression pattern of CsCAMTA genes under different horm and abiotic stresses, we treated cucumber seedlings with three hormones (IAA, MeJA) and two stress factors (NaCl, PEG). The results are shown in Figures 6 and 7. Figure 6, we can see that the expression of CsCAMTA1 is significantly up-regulated 3 h of IAA and MeJA stimulation, while it takes 6 h to increase the expressio CsCAMTA1 during ABA treatment. The performance of CsCAMTA2 is significantl ferent from other CsCAMTAs, and its expression is significantly down-regulated aft of treatment with the three hormones. The expression patterns of CsCAMTA3 CsCAMTA4 are similar. At 6 h after IAA treatment, the expression levels of both ar regulated, and at 48 h after MeJA treatment, the expression levels reach their maxim However, the difference is that CsCAMTA4 is down-regulated after 3 h ABA treatm Under abiotic stress treatment, different CsCAMTA genes were up-regulate down-regulated to varying degrees after a certain period of stress (Figure 7). Unde salt stress, CsCAMTA1, CsCAMTA3, and CsCAMTA4 show a trend that there is no s icant change in the short-term (0, 3, 6, 12, and 24 h) stress, but they are all up-regulat the long-term (48 h) stress. After 24 h of salt stress, the expression of CsCAMTA1 is s icantly up-regulated to 3.0-fold relative to the control, while CsCAMTA3 and CsCAM are up-regulated by more than 4.0-fold and 6.0-fold, respectively. However, short salt stress inhibited the expression of CsCAMTA2, and there is no significant change Figure 5. Tissue-specific expression analysis of CsCAMTAs. RNA-seq data were obtained from NCBI (accession number: SRP071224). S1, roots of 4-week-old seedlings; S2, stem of 4-week-old seedlings; S3, cotyledon of 4-week-old seedlings; S4, true leaf of 4-week-old seedlings; S5, root; S6, stem; S7, young leaf; S8, old leaf; S9, female flower; S10, male flower; S11, flesh of 1-week-old fruit. Red and blue indicate high and low levels of expression level, respectively. The color bar represents the expression values.

Expression Patterns of CsCAMTA Genes under Hormone and Abiotic Stress
To understand the expression pattern of CsCAMTA genes under different hormones and abiotic stresses, we treated cucumber seedlings with three hormones (IAA, ABA, MeJA) and two stress factors (NaCl, PEG). The results are shown in Figures 6 and 7. From Figure 6, we can see that the expression of CsCAMTA1 is significantly up-regulated after 3 h of IAA and MeJA stimulation, while it takes 6 h to increase the expression of CsCAMTA1 during ABA treatment. The performance of CsCAMTA2 is significantly different from other CsCAMTAs, and its expression is significantly down-regulated after 3 h of treatment with the three hormones. The expression patterns of CsCAMTA3 and CsCAMTA4 are similar. At 6 h after IAA treatment, the expression levels of both are up-regulated, and at 48 h after MeJA treatment, the expression levels reach their maximum. However, the difference is that CsCAMTA4 is down-regulated after 3 h ABA treatment.
Under abiotic stress treatment, different CsCAMTA genes were up-regulated or downregulated to varying degrees after a certain period of stress (Figure 7). Under the salt stress, CsCAMTA1, CsCAMTA3, and CsCAMTA4 show a trend that there is no significant change in the short-term (0, 3, 6, 12, and 24 h) stress, but they are all up-regulated in the long-term (48 h) stress. After 24 h of salt stress, the expression of CsCAMTA1 is significantly up-regulated to 3.0-fold relative to the control, while CsCAMTA3 and CsCAMTA4 are up-regulated by more than 4.0-fold and 6.0-fold, respectively. However, short-term salt stress inhibited the expression of CsCAMTA2, and there is no significant change after long-term stress. Under drought stress, CsCAMTA1 and CsCAMTA3 are down-regulated after 3 h stress, but a slow upward trend appears after long-term stress. There is no significant change in CsCAMTA2. CsCAMTA4 is significantly up-regulated after 48 h under both stresses.

Alternative Splicing Analysis for CsCAMTA Genes
To understand the expression patterns of different transcripts produced by alternative splicing in response to stress responses of CsCAMTA genes, we designed specific primers for different transcripts of CsCAMTA genes based on the annotation information of NCBI.
Under stress treatment, the expression pattern of each splicing isoform of CsCAMTA genes in leaves is similar to that of normal transcripts, but some splicing isoforms show specific expression patterns at different periods. As shown in Figure 8, under salt stress, most genes are down-regulated under short-term treatment and are significantly induced after long-term treatment (48 h), such as CsCAMTA1.3, CsCAMTA2.3, CsCAMTA3.2, etc. Agronomy 2021, 11, x FOR PEER REVIEW after 3 h stress, but a slow upward trend appears after long-term stress. There is n nificant change in CsCAMTA2. CsCAMTA4 is significantly up-regulated after 48 h both stresses.

Alternative Splicing Analysis for CsCAMTA Genes
To understand the expression patterns of different transcripts produced by al tive splicing in response to stress responses of CsCAMTA genes, we designed speci mers for different transcripts of CsCAMTA genes based on the annotation informa NCBI.
Under stress treatment, the expression pattern of each splicing isoform of CsCA genes in leaves is similar to that of normal transcripts, but some splicing isoforms specific expression patterns at different periods. As shown in Figure 8, under salt

Alternative Splicing Analysis for CsCAMTA Genes
To understand the expression patterns of different transcripts produced by alternative splicing in response to stress responses of CsCAMTA genes, we designed specific primers for different transcripts of CsCAMTA genes based on the annotation information of NCBI.
Under stress treatment, the expression pattern of each splicing isoform of CsCAMTA genes in leaves is similar to that of normal transcripts, but some splicing isoforms show Agronomy 2021, 11, 1827 9 of 15 specific expression patterns at different periods. As shown in Figure 8, under salt stress, most genes are down-regulated under short-term treatment and are significantly induced after long-term treatment (48 h), such as CsCAMTA1.3, CsCAMTA2.3, CsCAMTA3.2, etc. In contrast, CsCAMTA1.2 is induced after 6 h of NaCl treatment but is down-regulated after 48 h. This may indicate that the sensitivity of different gene spliceosomes to stress is various, and the period of the function is different. Compared with salt stress, CsCAMTA2.1 and CsCAMTA4.1 are induced quickly after 3 h of PEG treatment (Figure 9), suggesting that they are more sensitive to drought stress. CsCAMTA3.1 plays a major role in salt stress, but under drought stress, CsCAMTA3.2 has a more significant effect. These results indicate that alternative splicing plays a crucial role in CsCAMTAs' response to stress. FOR PEER REVIEW 9 of 17 after 48 h. This may indicate that the sensitivity of different gene spliceosomes to stress is various, and the period of the function is different. Compared with salt stress, CsCAMTA2.1 and CsCAMTA4.1 are induced quickly after 3 h of PEG treatment (Figure 9), suggesting that they are more sensitive to drought stress. CsCAMTA3.1 plays a major role in salt stress, but under drought stress, CsCAMTA3.2 has a more significant effect. These results indicate that alternative splicing plays a crucial role in CsCAMTAs' response to stress.
Agronomy 2021, 11, x FOR PEER REVIEW  The asterisk (*) indicates that the expression level of the stress group is significantly different from that of the control group (* p < 0.05, ** p < 0.01, one-way ANOVA, Tukey test).

FOR PEER REVIEW 11 of 17
Agronomy 2021, 11, x FOR PEER REVIEW

Discussion
Calcium signaling is a pathway that transmits extracellular sig biological response through changing the intracellular Ca 2+ conce studies have shown that a variety of stimulating factors can cause c Ca 2+ concentration, thereby mediating important biological respo cluding high temperature, low temperature, salt, pathogenic bac species, and hormones. CaM is the most important multifunction Ca 2+ in cells, and it has a high affinity and specificity with Ca 2+ [3] CAMTA/SR, NAC, WRKY, MYB, MADS-box, and bZIP, have been with CAM to regulate plant growth, development, and the biotic sponse [1]. Members of the CAMTA gene family have been identifie and play an important role in plant hormones and the abiotic stre port, we screened the members of the CAMTA family in the cucum four in total ( Table 1). The size of the CAMTA gene family in cucum The asterisk (*) indicates that the expression level of the stress group is significantly different from that of the control group (* p < 0.05, ** p < 0.01, one-way ANOVA, Tukey test).

Discussion
Calcium signaling is a pathway that transmits extracellular signals to an intracellular biological response through changing the intracellular Ca 2+ concentration [29]. Previous studies have shown that a variety of stimulating factors can cause changes in intracellular Ca 2+ concentration, thereby mediating important biological response processes [30], including high temperature, low temperature, salt, pathogenic bacteria, reactive oxygen species, and hormones. CaM is the most important multifunctional receptor protein for Ca 2+ in cells, and it has a high affinity and specificity with Ca 2+ [3]. Many TFs, including CAMTA/SR, NAC, WRKY, MYB, MADS-box, and bZIP, have been identified to interact with CAM to regulate plant growth, development, and the biotic and abiotic stress response [1]. Members of the CAMTA gene family have been identified in many eukaryotes, and play an important role in plant hormones and the abiotic stress response. In this report, we screened the members of the CAMTA family in the cucumber genome and found four in total ( Table 1). The size of the CAMTA gene family in cucumbers is similar to that in bananas and Arabidopsis with five and six members, respectively [5,31]. However, it is far less than that in soybeans with 15 members [11], wheat with 15 members, and Brassica napus with 18 members [13,16]. This indicates that the evolutionary distribution of the CAMTA gene family in the entire plant kingdom is uneven. Brassica napus and soybeans are tetraploid, so they have undergone multiple rounds of genome-wide replication during the evolution process, resulting in a relatively large number of CAMTA members. However, bananas are triploid, and cucumbers and Arabidopsis are diploid; thus, they have relatively few members of the CAMTA gene family.
We analyzed the structure of four CsCAMTA genes and found that the number of introns in all CsCAMTAs ranges from 11 to 12. This number is almost three times the average number of introns in cucumber genes (4.39), which may lay the structural foundation for the occurrence of AS [32]. The number of introns of CsCAMTA1 and CsCAMTA3 is 12; the number of introns of CsCAMTA2 and CsCAMTA4 is 11 (Figure 1). This is similar to the results of previous identifications on other species, indicating that members of the CAMTA gene family are conservative in gene structure [14]. Just as the CAMTA identified in other species, the four CsCAMTAs contain all the conserved domains of the typical CAMTA protein, including the CG-1 domain, TIG domain, ankyrin (ANK) repeat domain, IQ domain, and CaM binding domain (CaMBD) (Figure 2). According to the existence of TIG, CsCAMTAs are divided into two groups, which are consistent with previous studies [13]. On the one hand, CsCAMTA without the TIG domain may affect the DNA non-specific interaction of transcription factors and protein dimerization. On the other hand, the absence of the TIG domain may be the structural basis for the expansion and evolution of its family members [33,34]. According to the results of previous studies, IQ motifs can be combined with CaM in a Ca 2+ -dependent or Ca 2+ -independent manner to transmit signal substances, while CAMBD can transmit signals through the combination with Ca 2+ /CaM complexes [4,35]. Regarding how CsCAMTAs interact with CaM, our research found that all CsCAMTAs contain IQ motifs and CAMBD, which indicates that CsCAMTAs can not only bind to CaM in a calcium-dependent manner but also can bind to CaM in a calcium-independent manner. This may make the signal transmission more stable and accurate.
To understand the relationship between cucumber CAMTA members and other species, a phylogenetic tree of CAMTA members of cucumbers, Arabidopsis, tomatoes, rice, and soybeans was constructed, which divides the 39 CAMTAs into three groups. The four CsCAMTAs members fell into all three groups (Figure 3), which suggests that the structure and function of CsCAMTAs are highly conserved during plant evolution [36]. In group A, the number of CAMTA members of cucumbers and other species is more than the number of members distributed in groups B and C, indicating that CAMTA is undergoing rapid adaptive evolution in group A [37]. In addition, in the same group, the closest members have similar gene structures and may have similar functions. Previous studies have confirmed that AtCAMTA1, AtCAMTA2, and AtCAMTA3 play an important role in plant defense against pathogenic bacteria and in response to low temperatures and salt stress [38]. Thus, we speculate that CsCAMTA1 and CsCAMTA2 may have similar functions to them. Later experiments also proved that CsCAMTA1 is indeed up-regulated under long-term salt stress treatment (Figure 7).
The cis-elements in the promoter region, as the binding sites of transcription factors, play a vital role in the regulation of gene expression, especially for the regulation of gene expression in response to biotic and abiotic stress [39]. We have identified many cis-acting elements in CsCAMTA genes, including the W-box, Circadian, TGA-box, ABRE, TGACG-motif, CGTCA-motif, AuxRR-core, MBS, LTR, and ARE ( Figure 5). An interesting phenomenon is that CsCAMTA3 does not have MeJA response elements (TGACG-motif, CGTCA-motif), but it has a clear response to MeJA treatment ( Figure 6). CsCAMTA2 contains a drought response element (MBS), but it does not respond significantly under drought stress (Figure 7). This phenomenon was also found in poplar and soybean CAMTA family members [11,21], indicating that the relationship between corresponding elements and the occurrence of responses is not one-to-one. The specific relationship may need to be further explored in detail.
The function of CAMTA in the process of plant growth and development has been extensively studied. In cotton, the positive correlation between the expression of Gh-CAMTA2A.2 and GhCAMTA7A and fiber strength proved their important role in the development of cotton fiber [40]. Yang et al. [8] found that SlSRs act as candidate signal junctions for connection development and ethylene-and calcium-mediated signals during tomato fruit development and maturation. In Arabidopsis, CAMTA1 and CAMTA5 regulate the expression of the organ development gene AVP1, thereby controlling auxin fluxes and then regulating the occurrence of organs [41,42]. In our study, CsCAMTAs have different expression intensities and spatially differentiated expressions in different tissues. Compared with CsCAMTA1, CsCAMTA3, and CsCAMTA4, CsCAMTA2 is expressed very low in almost all tissues. This phenomenon also occurs in rape. BnCAMTA3A2 and BnCAMTA3C2 have low expression levels in almost all tissues [16], which may be due to redundant functions of these genes or transcriptional silencing/post-transcriptional silencing effects [43]. The expression levels of CsCAMTA3 and CsCAMTA4 in female flowers and roots are significantly higher than their expression levels in fruits. This is similar to the high expression levels of wheat TaCAMTA1-D and TaCAMTA3-D in reproductive ears and seedling buds, respectively [13]. Such expression patterns exist in almost all CAMTA families that have been identified, indicating that CsCAMTAs are involved in the growth and development of cucumbers at various growth stages, but the period and intensity of action between members are different, and the specific functions need to be further verified.
The response of CAMTA gene family members to hormonal and abiotic stresses has been reported in many species. In Arabidopsis, CAMTA1 takes part in auxin signaling and responds to salt stress [44]. A previous study also found that the expression of maize ZmCAMTAs is regulated by stress-related hormone signaling molecules (IAA, SA, ABA, and JA), which suggests that ZmCAMTAs may respond to stress through hormone signaling pathways [10]. In our experiments, CsCAMTA1 and CsCAMTA3 are significantly increased after 48 h of salt stress treatment. Meanwhile, ABA and MeJA significantly increase the expression of CsCAMTA1 and CsCAMTA3, which indicates that CsCAMTA1 and CsCAMTA3 might increase the salt tolerance of cucumber seedlings through ABA and MeJA signal transduction pathways. AtCAMTA1 induces the expression of photosynthesis-related genes and changes membrane integrity by generating ABA under drought stress [45]. Interestingly, the expression of CsCAMTA4 is up-regulated under 48 h salt and drought stresses, but it has no obvious response to ABA treatment, indicating that CsCAMTA4 may improve the drought tolerance of cucumber seedlings through an ABA-independent pathway. The detailed signal process of CsCAMTA genes under various environmental stimuli requires further study.
As a post-transcriptional modification mechanism, alternative splicing increases the complexity of gene expression and the diversity of protein expression to a certain extent. In the analysis of gene families, the phenomenon that a gene generates multiple transcripts through alternative splicing often occurs, including in the pepper WRKY family, Arabidopsis PPR family, and human KIR family [46][47][48]. The CAMTA gene family is no exception. In previous studies, Wei et al. [21] found that under cold stimulation, the expression patterns of alternative spliceosomes of poplar PtCAMTAs differ between different tissues and different varieties, which indicates that alternative splicing may play a key regulatory role in plant development and the response to environmental stimuli. Previous studies have shown that CAMTA plays an important role in the regulation of salt and drought stresses [18,20]. Therefore, we analyzed the expression patterns of CsCAMTAs' alternative spliceosomes under salt and drought stresses. According to its expression results, we found that the expression level of CsCAMTAs' spliceosome at different time points under stress is various. Most of the spliceosome tends to be inhibited by short-term stress and induced by long-term stress, such as CsCAMTA1.4, CsCAMTA2.3, and CsCAMTA3.2, etc. This is contrary to the tendency that PtCAMTA1.2, 2.3, and 6.2 in poplar leaves are induced by short-term stress treatment [21]. This difference may be because the poplar and cucumber respond differently to stress: under short-term stress, plants may down-regulate CsCAMTAs to maintain the balance of plant growth and metabolism, while under longterm stress, CsCAMTAs are up-regulated to participate in a series of metabolic pathways to maintain plant growth and development. However, the response mode of PtCAMTAs may be the opposite. Some spliceosomes have different major genes under different stresses. CsCAMTA3.1 is the major gene under salt stress, while under PEG treatment, the effect of CsCAMTA3.2 is more significant (Figures 8 and 9). This phenomenon also occurs in Brassica rapa BrRS2Z5 [49], probably because different stresses may induce the expression of the different spliceosomes to respond to the corresponding stress more efficiently [50]. However, the specific model needs further experimental verification.

Conclusions
In general, in this study, we identified four cucumber CAMTA gene family members and analyzed their gene structure, conserved domains, and phylogenetic relationships. The results show that the CsCAMTA gene family is highly conserved in the evolutionary process. The analysis of the cis-acting shows that the cucumber CAMTA genes have a genetic basis for responding to multiple hormones and stress. Tissue-specific analysis indicates that CsCAMTAs are expressed in multiple tissues, but the expression levels are different, indicating that different CsCAMTAs may be involved in the growth and development of cucumbers in different periods. The expression patterns under hormonal and abiotic stresses indicate that the participation of CsCAMTAs in the plant response to stress may be through stress-related hormonal signal pathways, and alternative splicing may also be involved. Our research provides evidence for the involvement of the CAMTA gene family in cucumbers' growth and development and response to stress, and lays a theoretical foundation for further exploration of the functions of CsCAMTAs.