Genome-Wide Identification, Expression Diversication of Dehydrin Gene Family and Characterization of CaDHN3 in Pepper (Capsicum annuum L.)

Dehydrins (DHNs) play a crucial role in enhancing abiotic stress tolerance in plants. Although DHNs have been identified and characterized in many plants, there is little known about Capsicum annuum L., one of the economically important vegetable crops. In this study, seven CaDHNs in the pepper genome were identified, which could be divided into two classes: YnSKn- and SKn-type, based on their highly conserved domains. Quantitative real-time PCR (qRT-PCR) results showed that the seven DHN genes were expressed in all tissues and might be involved in the growth and development of pepper. The gene expression profiles analysis suggested that most of the CaDHN genes were induced by various stresses (low temperature, salt and mannitol) and signaling molecules (ABA, SA and MeJA). Furthermore, the CaDHN3 (YSK2)-silenced pepper plants showed obvious lower resistance to abiotic stresses (cold, salt and mannitol) than the control plants (TRV2:00). So the CaDHN3 might act as a positive role in resisting abiotic stresses. This study lays the foundation for further studies into the regulation of their expression under various conditions.


Introduction
Unfavorable environmental conditions such as cold, high salinity, or drought, limit the growth, development, and distribution of plants which are unable to move away and reduce agricultural productivity [1]. In response to these stresses, plants employ multiple mechanisms to increase their tolerance to various abiotic stresses, such as accumulation of compatible osmolytes (soluble sugars, glycine betaine,and proline) and synthesis of hydrophilic proteins such as dehydrins (DHNs) [2][3][4].
Dehydrins, group 2 late embryogenesis abundant (LEA) proteins, are expressed during late embryogenesis, as well as in vegetative tissues when subjected to drought, low temperature, conserved sequences, a K-segment and its variants. Candidate CaDHN genes were aligned with ClustalW2 online software (http://www.ebi.ac.uk/Tools/msa/clustalw2), and the genes with different sequences between the two cultivars were identified. The primer pairs (S1 Table), whose specificity were detected by NCBI Primer BLAST (http://www.ncbi.nlm.nih.gov/tools/primerblast/index.cgi?LINK_LOC=BlastHome), were designed to amplify these differing quences with Primer Premier 5.0 (Premier Biosoft International, PaloAlto, CA, USA), and CM334 and Zunla-1 sequences for the same gene were then aligned to confirm the correct sequences. The 7 identified DHN proteins were mapped to the chromosomes via MapDraw [31].

Characterization and comparison of deduced DHN proteins
The deduced pepper amino acid sequences were aligned by Clustal omega online software (http://www.ebi.ac.uk/Tools/msa/clustalo/), and adjusted manually when necessary. Protein MW (molecular weight), pI (isoelectric point), GRAVY (grand average of hydropathy) and composition of amino acids were predicted using Compute pI/MW tool (http://web.expasy. org/computer_pi/). The sequence algorithm NetPhosK (Expasy) was used to predict phosphorylation sites in pepper proteins, with its probability limit set to 60%.

Phylogenetic analysis, exon/intron struction determination and identification of conserved motifs
The software MEGA 6.0 [32] was used to construct the unrooted phylogenetic trees of the fulllength protein sequences, by Neighbor-Joining method [33] and Minimum Evolution method with the parameters (p-distance and completed deletion). Phylogenetic trees were estimated using 1000 bootstraps replicates. The exon/intron construction of the CaDHNs were determined based on alignment of cDNA and pepper genomic sequences using Gene Structure Display Server program (GSDS, http://gsds.cbi.pku.edu.cn/index.php) [34]. Extraction of motif from 7 CaDHN protein sequences were programed using the program of MEME (http:// meme-suite.org/tools/meme), with the following parameters: Normal mode, maximum number of motifs was 10 and distribution of motif sites was any number of repetitions.

Plant materials and treatments
The cold-tolerant pepper cultivar P70 was provided by the pepper research group in Northwest A&F University, China. Pepper seedling were cultivated in a growth chamber (25/20°C day/night temperature and 16/8h day/night photoperiod cycle).
When the 6-8 true leaves expansions period, and growth of about 50 days, pepper seedlings were used for various abiotic stresses and plant hormone treatments. The seedlings of P70 were subjected to 6°C for cold stress, incubated in 300mM NaCl for salt treatment and 300mM mannitol for osmotic stress, and sprayed 0.57mM ABA or 5mM SA, by using the method followed by Guo et al [35]. The spraying of 50μM MeJA were performed as described previously [36]. The leaves were collected at 0, 1, 3, 6, 12 and 24 h after all treatments. To evaluate the expression levels of CaDHNs in all tested tissues under normal conditions, six different tissue (roots, stems, leaves, flowers, fruits and seeds) were collected as described previously [25]. All experiments were performed in three biological replicates for each treatment.

RNA isolation and qRT-PCR analysis
Total RNA was extracted from frozen samples according to the instruction of Total RNA kit (Bio Teke, Beijing, China). Reverse transcription was performed using the Primescript™ first strand cDNA Synthesis Kit (TaKaRa, Dalian, China), following the manufacturer's protocol. The gene encoding the ubiquitin-conjugating protein UBI-3 (GeneBank accession no. AY 486137.1) from pepper was used as the reference gene [37,38]. qRT-PCR was performed in triplicate Using SYBR Premix Ex Taq II (TaKaRa, Dalian, China) on an IQ5.0 Bio-Rad iCycler thermocycler (Bio-Rad, Hercules, CA, USA). Each reaction system was performed as described by Guo et al [35]. The following qPCR reaction systems were used: 95°C for 1min, followed by 45 cycles of 95°C for 15s, 57°C for 20s, and 72°C for 30s. The fluorescence data was collected during the 57°C step. Primers, designed by Primer Premier 5.0, are listed in S2 Table. Relative expression levels of pepper DHN genes were determined using the comparative threshold method (2 -ΔΔCt ).

Virus-induced gene silencing (VIGS) assay of CaDHN3 in pepper
The pTRV2: CaDHN3 construct was engineered to include a 443 bp fragment of CaDHN3 cloned from P70 leaves cDNA template using a gene-specific primer pair (forward primer 5 0 -AATATGGCACATAACGGTACTAG-3 0 and reverse primer 5 0 -CGGGATCCCTCCAAAGTGAT GATGATAAGGT-3 0 ). The underlined nucleotides contained a BamH I restriction site. The resulting PCR product which was sequenced in Quintara Biosciences Company (Wuhan, China), was inserted into pTRV2 vector to form pTRV2:CaDHN3. Agrobacterium tumefaciens strain GV3101 containing pTRV1 was respectively mixed with pTRV2:00, TRV2-CaPDS or TRV2-CaDHN3 at 1:1 ratio (OD600 = 0.5-0.8 for each construct). The mixtures were injected into pepper leaves, and plants were grown as described by Wang et al [36]. All silencing assays were performed using three replicates, and 50 plants were used for each repetition.
Total chlorophyll content was measured using spectrophotometric method after extracting into 80% (v/v) acetone [35]. Electrolyte leakage was measured according to the method described by Dionisio-Sese and Tobita [39].

Statistical analysis
All obtained data were subjected to analysis of variance (ANOVA) using SPSS software. The analyzed data were expressed as means ± standard error (SE) of two biological replicates in all measured parameters except for Mn-SOD and POD which were performed using three replicates. The mean separation was analyzed using the Duncan's multiple range test, taking p < 0.05 as a significant difference.

Genome-wide identification of DHN genes in pepper
The conserved amino acid sequence of dehydrin (Pfam: PF00257) was used to search in the pepper genome database (PGP) (http://pepper.genome.snu.ac.kr/), using the HMMER (v3.0) software. Meanwhile the sequences of DHN members from Arabidopsis [16] and tomato [21] were used to blast against PGP. As a result, a total of 7 candidate DHN genes were identified in pepper (Table 1). In addition, providing a simplified nomenclature for each identified gene, we adopted the acronyms of CaDHN1 to CaDHN7, based on their order of appearance from chromosome 1-12. Because a dehydrin gene of pepper had been identified and named to CaDHN1, so here, we named the other six genes from CaDHN2 to CaDHN7. The 7 DHNs were then mapped to 3 of 12 pepper chromosomes, with four present on chromosome 2; one on chromosome 4 and two on chromosome 8 (S1 Fig). but remaining regions displayed relatively low identity among the seven genes. Furthermore, NLSs, with an RRKK motif, were identified in five DHN proteins (CaDHN2, CaDHN3, CaDHN4, CaDHN5, and CaDHN6) (Fig 1). Based on the number and order of Y-, S-and Kmotifs, the seven DHNs were classified as YSK2-(CaDHN2, CaDHN3 and CaDHN5), Y3SK2-(CaDHN4 and CaDHN6) and SK3-type (CaDHN1 and CaDHN7) proteins (Fig 1; Table 1).
Generally, DHNs contain an abundance of Gly and polar amino acid, but lack Cys and Trp [9,14]. Analyzing the amino acid compositions of seven DHN proteins, we found that they shared the common feature (S3 Table). The CDS (full-length coding sequence) sizes of 7 CaDHN proteins varied from 399 bp (CaDHN3) to 675 bp (CaDHN6), with respective deduced proteins of 132-224 amino acids in length. The predicted molecular weights were between 13.82 kDa (CaDHN3) and 23.04 kDa (CaDHN6). All members of the CaDHN family were found to be highly hydrophilic, with GRAVY values ranging from -0.885 to -1.602 and pIs (isoelectric point) from 5.41 to 7.30 (Table 1). Among the 7 CaDHN genes, 5 members (CaDHN2, CaDHN3, CaDHN4, CaDHN5 and CaDHN6) that belonged to YnSKn-type, possessed a higher pI than the SKn-type DHN (CaDHN1 and CaDHN7). In addition, we predicted many phosphorylation sites of each DHN protein, with SKn-type (CaDHN1 and CaDHN7) containing a higher number of casein kinase 2 (CK2) phosphorylation sites than putative protein kinase C (PKC) phosphorylation sites, and YnSKn-type containing a higher number of PKC sites than CK2 sites except CaDHN3 and CaDHN5. Beyond that, we also identified a conserved dehydrin motif LXRXXS phosphorylated by an Snf1-related kinase (SnRK2-10) [40].

Phylogenetic analysis, exon/intron struction determination and identification of conserved motifs
To better understand CaDHN proteins we performed the NJ methods using MEGA 6.0 to obtain the phylogenetic tree based on full-length amino acid sequences of pepper DHN proteins (Fig 2a). The seven CaDHN proteins were classified into three categories: class-I, CaDHN2, CaDHN3 and CaDHN5; class-II, CaDHN4 and CaDHN6; and class-III, CaDHN1 and CaDHN7.
The structures in exon/intron distribution CaDHN genes were analyzed based on the alignments of their coding region sequences with respective genomic full-length sequences, and shared a highly conserved exon/intron structures, with one intron and the 0 intron phase (Fig  2b). The length of introns ranged from 106bp (CaDHN4) to 672bp (CaDHN2).
The conserved K-segment, S-segment and Y-segment are also identified as important motifs, which were extracted by MEME based on the CaDHN proteins (Fig 2c). As a result, four significant motifs were obtained (S4 Table). Among these motifs, motif 1 and motif 3 were  respectively identified as K-and Y-segments which had high similarities with previous K-and Y-motifs of DHNs. In addition, the motif 2 was identified as S-segment based on a track of Ser residue. The motif 4 was only found in Sk3-type DHNs (CaDHN1 and CaDHN7), which might play a role in the function of the two DHNs (Fig 2c). The distributions of motifs were consistent with the classification of all CaDHN proteins. In order to comprehensively analyze the evolutionary relationship between CaDHN genes and DHN genes from Arabidopsis, rice, barley and tomato, we constructed an unrooted phylogenetic tree of dehydrin proteins. Based on our phylogenetic results, the DHNs could be divided into five groups: YnSKn-, SKn-, Kn-, YK-and KS-type proteins (Fig 3). The classification results of the CaDHN proteins were consistent with the above multiple alignments of CaDHN amino acid sequences (Fig 1), phylogenetic groups (Fig 2a) and the presence of conserved segments (Fig 2c). Interestingly, pepper only had two types and lacked YK-, KS-and Kntype DHNs. However, more groups are present in Arabidopsis, barley and tomato.

Expression analysis of CaDHN genes in pepper tissues
In order to elucidate the additional information on the functions of different members of DHN family in pepper, the expression of all CaDHNs under normal growing conditions was analyzed in six different tissues (roots, stems, leaves, flowers, fruits, and seeds) using qRT-PCR. Expression patterns of the seven pepper DHN genes were significantly different. In general, CaDHN 1, 2, 7, 4 are constitutively expressed in all the tissues, although at different levels and with the lowest expression detectable in the leaves. CaDHN 3, 4, 5 are quiet silent and, perhaps, can be activated at different development or under different conditions (such as stress or hormone treatment). The expression of CaDHN2 and CaDHN7 was high in all the tissues, except leaves which were not detected. Concerning the leaves only CaDHN1 and CaDHN4 exhibit appreciable expression levels. The expression of CaDHN1 was at a high level in all tested tissues. However, CaDHN6 was not detected in any tissue. In addition, we found that the expression of CaDHN1 and CaDHN7 (SK3-type), were higher in fruits than other tissues (Fig 4).

Response of CaDHN genes expression to various abiotic stresses
To determine whether CaDHNs exhibited stress-responsiveness, we analyzed the expression levels of all genes in the leaves of pepper plants subjected to various stress conditions, using qRT-PCR. In response to chilling (6°C), the accumulations of mRNA were induced, with levels peaking at 12h (CaDHN7) or 24h (CaDHN1, CaDHN2, CaDHN3 and CaDHN4). Most genes were obviously up-regulated, except CaDHN5 and CaDHN6 which were having no obvious change petterns, while CaDHN4 was especially increased 260-fold at 24h of cold stress (Fig 5a).
Under salt stress treatment, expression of CaDHN5 and CaDHN7 increased 134-fold and 420-fold respectively. In addition, the expression of CaDHN1, CaDHN2, and CaDHN3 increased by more than 20-fold. Compared with the above genes, the increase multiples levels of CaDHN6 was smaller. Whereas, CaDHN4 was down-regulated (Fig 5b).
Under osmotic stress treatment, CaDHN5 and CaDHN7 were rapidly up-regulated to 169and 68-fold, respectively. Other genes, except CaDHN4 that was down-regulated, were up-regulated more or less (Fig 5c).

Expression of CaDHN genes in response to various signaling molecules
Abscisic acid (ABA) and salicylic acid (SA)/jasmonic acid (JA) generally mediate the responses of plants to abiotic and biotic stress, respectively. So we investigated DHN expression in pepper leaves treated with ABA, SA or MeJA, respectively, aiming to explore whether the response of pepper DHNs under stress conditions was related to these molecules. Results showed that by exogenous ABA treatment, the majority of DHN genes were significantly up-regulated, especially CaDHN3 and CaDHN5. The exceptions were CaDHN1 and CaDHN4-the former exhibiting not appreciably altered and the latter being down-regulated (Fig 5d). For SA application, all the genes were up-regulated, especially previous research on CaDHN1 that was strongly induced, with the exception of CaDHN2 and CaDHN5, which showed relatively stable in their expression (Fig 5e). Finally, after MeJA application, most genes were rapidly up-regulated, especially CaDHN1 and CaDHN3 increased by more than 10-fold. Whereas the exception was CaDHN6, which showed slightly down-regulation (Fig 5f).

CaDHN3-silenced pepper plants reduces tolerance to abiotic stresses
CaDHN3, a YSK2-type dehydrin and 132bp amino acids in length, was up-regulated under all of the tested abiotic and hormone treatments. In addition, previous studies showed that YSK2type dehydrins played an important in enhancing stress resistance [10,11]. Because of the above results, we were interested in the characterization of CaDHN3. We studied the function of CaDHN3, using the virus induced gene silencing (VIGS) method. When bleaching was evident on the positive control plants (inoculated with TRV2-PDS) (S2 Fig), we detected the silencing efficiency with young leaves of CaDHN3-silenced plants (inoculated with TRV2-CaDHN3) and negative control (inoculated with TRV2). Compared to negative control, the CaDHN3 silencing rate dramatically reached nearly 85% (Fig 6b) under non-stress conditions. At the same time, the other six genes of pepper family were not silenced (S3 Fig). Whereas the CaDHN7 was up-regulated to 17-fold. The results showed that there might be a functional interaction between CaDHN3 and CaDHN7. Above all, VIGS was successful and effective for CaDHN3 gene silencing in pepper.
Under 6°C cold stress for 24h, obvious seriously wilting appeared in CaDHN3-silenced plants, while control leaves exhibited slight withering (Fig 6a). In order to confirm the influence of CaDHN3 silencing in the cold stress defense response, electrical conductivity was measured in control and silenced pepper plants. After 24h of 6°C cold treatments, the electrical conductivity content was significant higher than that of the negative control plants (Fig 6c). In addition, to elucidate the possible mechanism of CaDHN3 in tolerance to cold stress, the expression levels of Mn-SOD and POD were examined in control (pTRV2:00) and CaDHN3silenced (pTRV2:CaDHN3) plants (Fig 6). The results showed that Mn-SOD and POD could be  cold-induced expression in control plants and CaDHN3:silenced plants (Fig 6d and 6e). However, at 24h of cold stress, the expressions of Mn-SOD and POD genes in CaDHN3-silenced plants were significantly lower than in control plants.
To determine whether the silencing of CaDHN3 led to reduced tolerance to salt and osmotic stresses, leaf discs (0.5 cm in diameter) from control (pTRV2:00) and CaDHN3-silenced (pTRV2:CaDHN3) plants were exposed to different concentrations of NaCl solution (0, 200 mM, 300mM and 400 mM) and mannitol solution (0, 300 mM, 400 mM and 500 mM) respectively with continuous lighting for 3d (Fig 7a and 7b). After 3 days, CaDHN3-silenced leaf discs were more yellow and even white symptoms in high concentrations of NaCl solution (300mM and 400 mM) than those of control plants. The same conditions were observed in mannitol-treated plants. The chlorophyll contents of the CaDHN3-silenced were also obvious reduced as compared to controls, especially under the stress of high concentrations of salt and mannitol (Fig 7c and 7d).

Discussion
Dehydrins are believed to play a fundamental role in the response of plants to various abiotic stresses [11,41,42]. They are multigene families and the functions of the dehydrin genes have been characterized in many plants, such as Arabidopsis [16], rice [20], barley [17][18][19], tomato [21] and so on. Nevertheless, little is known about the pepper DHN family. In this study, a total of 7 DHN family members had been identified in pepper based on the pepper genome (Table 1) [17,28].
Although pepper DHN protein were all highly hydrophilic, they had obvious differences in pI, the number of conserved motifs, CK2 and PKC kinase sites. The pepper YnSKn-type DHNs were higher pI than the SKn-type DHNs. That is, the YnSKn-type DHNs were more easy to combine into negatively charged membrane than SKn-type DHNs [7]. Phosphorylation of the dehydrins may play a role in functional regulation of stressed plant cells and modulate membrane binding of DHNs [43]. SKn-type (CaDHN1 and CaDHN7) contained a higher number of casein kinase 2 (CK2) phosphorylation sites than putative protein kinase C (PKC) phosphorylation sites, and the YnSKn-type (CaDHN2, CaDHN4 and CaDHN6) contained a higher number of PKC sites than CK2 sites except CaDHN3 and CaDHN5. These findings were in agreement with the suggestion of grapevine DHNs family, which YnSKn-type DHNs are mainly phosphorylated by PKC and SKn-type DHNs are mainly phosphorylated by CK2 [4,22,23].
A phylogenetic tree of the DHN family proteins from pepper, Arabidopsis, barley, rice and tomato were constructed. Overall, the members from pepper DHN family except CaDHN4 showed the closer phylogenetic relationship with those of tomato, agreeing with a common ancestor (Fig 3) [28]. The number of pepper dehydrin family was similar to other gene families, such as Arabidopsis and tomato. But it only contained five YnSKn-type DHNs and two SKntype DHNs, lacked YK-, KS-and Kn-type DHNs which might have been lost in pepper species (Yang et al. 2012). Moreover, the number of the YnSK2-type DHNs (5 in pepper, 3 in Arabidopsis, 8 in barley, 5 in rice and 2 in tomato) accounted for over 50 percent of total DHN members in the phylogenetic tree, making it the largest part. It was consistent with the result that the YnSK2-type DHN proteins were one of the most abundant neutral or alkaline proteins in the nature [11]. The exon/intron structural analysis showed that all of the pepper DHN genes  contained only one intron and the 0 intron phase (Fig 2b). Guo thought it might be due to a conservative evolution pattern and the length of the insert introns might affect the functional divergences [44].
The expression profiles of CaDHN genes were distinct but partially overlapping. Rorat (2006) indicated that DHNs were present in nearly all vegetative tissues during normal growth conditions [11]. Similarly, pepper DHN genes expressed in all tissues (roots: 5/7, stems: 5/7, leaves: 2/7, flowers: 4/7, fruits: 4/7 and seeds: 4/7), which suggested that CaDHN genes might be involved in the growth and development of pepper. Furthermore, in our study, there was a same conclusion with suggestion that different types of DHNs could localize in the same tissues under the normal growth conditions, for example, CaDHN2 (YSK2) and CaDHN7 (SK3) were found to have the same tissue expression patterns. However same types of DHNs could localize in the different tissues under the normal growth conditions. These results could be used to further understand the function of CaDHN genes in growth and development of pepper.
DHN genes exist as multi-gene families in various plants, and have been well-characterized. Previous studies suggested that SKn and Kn proteins might be largely up-regulated by cold stress, while YnSKn-type DHNs might mainly participate in plant acclimation to salt, ABA and desiccation stresses [11,42]. For example, in barley, almost all of the DHN genes encoding for YnSK2 dehydrins were up-regulated by both dehydration and ABA, but not by low temperature. In contrast, DHN5 (K9) and DHN8 (Sk3) were up-regulated by cold treatment. Arabidopsis LTI30 (K6) and wheat WCS120 (K6) were mainly induced by low temperature stress [45,46]. However, the fact are not invariable. There are large differences among species. Such as, in Ginseng, the majority of DHN genes containing various types (SKn-, YnSKn-and KStype) could be up-regulated by cold, drought and ABA treatment [47]. The BjDHN1 and BnDHN1 that belong to Y3SK2-type DHNs, were showed to be up-regulated under cold stress [48]. MnDHN2 and MnDHN4 (Y2SK3-and YSK3-type) could be powerful induced by low temperature stress [23]. In our study, seven members of the pepper DHN family exhibited very distinct patterns of expression (Fig 5). CaDHN1, CaDHN2, CaDHN3 and CaDHN7 were induced by cold, NaCl and mannitol stresses. While CaDHN7 was also highly up-regulated by these treatment, which suggested a possible role in response to abiotic stress in pepper. On the contrary, CaDHN6 showed no any obvious up-regulation under these stresses. It was worth noting that CaDHN4 (Y3SK2) was strongly expressed in low temperature conditions and down-regulated by NaCl and Mannitol stresses, which were similar to MdDHN2 and MdDHN4 [23]. We believed CaDHN4 might has an important role in cold condition. Many studies have indicated that plant hormone (ABA, SA and MeJA) play a key role in adaptive environmental stresses and plant developmental processes [49,50]. Dehydrins proteins have an important role in defenses against stress, in either an ABA-dependent or ABA-independent pathway. Most of the PgDHN genes and MnDHN genes could be induced by ABA [23,47]. The Physcomitrella patens DHN gene was induced by ABA [12]. In our experiments, exception for CaDHN1 and CaDHN4, the other pepper DHN genes were obvious up-regulated by ABA treatment, which indicated they might be involved in ABA-dependent pathways. These pepper DHN genes had different expression pattern in ABA stress. The similar result, Allagulova thought it might be due to the differences of the number of ABA response element in promoter regions [51]. SA and MeJA had been reported to involve in plant stress tolerance, and were important endogenous signaling molecule [36,52,53]. Our results showed that out of seven pepper DHN genes, five genes were induced by SA while in case of MeJA six genes were induced.
To determine the function of the CaDHN3 in abiotic stress responses, we performed VIGS, which is widely used for a rapid detection of pepper genes involved in response to abiotic stresses [54,55]. Previously, we analyzed the function of CaDHN1, using the VIGS method [25]. In addition, Chae et al (2015) also analyzed the function of CaLEA1 gene using this method [56]. In our study, CaDHN3-silenced pepper plants exhibited obvious lower resistance to abiotic stresses than the control plants.
In addition, Mn-SOD and POD were involved in inhibiting the production of reactive oxygen species (ROS) and protecting the cell membrane during the low temperature [57]. Mn-SOD gene strongly response to low temperature and oxidative stress reaction [58,59]. Chilling stress enhanced the activities of POD enzymes [59,60]. In our study, we found that the CaDHN3 gene knockout suppressed the expression of Mn-SOD and POD (Fig 6), which might be lead to more ROS accumulation in the CaDHN3:silenced pepper than in the TRV2:00 pepper plants under stress conditions. These results suggested that CaDHN3 played an important role in improving pepper stress tolerance and might act as a positive regulator of stress-responsive gene expression.

Conclusions
In conclusion, we identified 7 CaDHNs in the pepper genome. They were divided into two classes (YnSKn-and SKn-type). The transcripts of seven CaDHN genes were distinct but partially overlapping expression profiles, and the DHNs were expressed in all tissues, suggesting that CaDHN genes might be involved in the growth and development of pepper. The gene expression profiles analysis suggested that most of the CaDHN genes were induced by various stresses (low temperature, salt and mannitol) and signaling molecules (ABA, SA and MeJA). Further more, the CaDHN3-silenced pepper plants showed obvious lower resistance to abiotic stresses (cold, salt and mannitol) than the control plants (TRV2:00). So CaDHN3 belonging to Y3SK2-type dehydrin, might act as a positive role in resisting abiotic stress. This comprehensive analysis would be an important information for further studies to elucidate the function roles of CaDHNs in pepper.