The role of three heat shock protein genes in the immune response to Aeromonas hydrophila challenge in marbled eel, Anguilla marmorata

Heat shock proteins (HSPs) are highly conserved molecular chaperones that play critical roles in both innate and adaptive immunity. However, little information about HSPs from marbled eel Anguilla marmorata is known. In this study, the full-length Amhsp90 (2527 bp), Amhsp70 (2443 bp) and Amhsc70 (2247 bp) were first cloned from A. marmorata, using rapid amplification of cDNA ends, containing open reading frames of 2181, 1932 and 1950 bp in length, and encoding proteins with 726, 643 and 649 amino acids, respectively. The deduced amino acid sequences of three Amhsps shared a high homology similarity with other migratory fish. Real-time fluorescent quantitative polymerase chain reaction was used to evaluate tissue-specific distribution and mRNA expression levels of three Amhsps subjected to infection with Aeromonas hydrophila. The mRNA expression of three Amhsps in eight tested tissues, namely liver, heart, muscle, gill, spleen, kidney, brain and intestine, of juvenile A. marmorata was evaluated to reveal the major expression distribution in liver, intestine, muscle and heart. After pathogen challenge treatments, mRNA transcriptions of three Amhsps revealed a significant regulation at various time points in the same tissue. All these findings suggest that Amhsps may be involved in the immune response in A. marmorata.

(1.0 × 10 8 CFU ml −1 for each individual), whereas the control fish were injected with 0.1 ml of 0.85% NaCl. After treatment, the fish were returned to the culture tanks, and the tissue samples (liver, muscle, intestine and heart) from every three fish per group were randomly collected at 1, 3, 6, 12, 24, 48 and 72 h post-injection.

RNA extraction and cDNA synthesis
Total RNA was isolated from the samples of all untreated and treated fish at each sampling time point to explore tissue-specific distribution and the effect of pathogen challenge on mRNA expression levels of Amhsp90, Amhsp70 and Amhsc70. Liver, kidney, spleen, gill, muscle, heart, brain and intestine samples were collected from A. marmorata for RNA extraction. Total RNA was extracted using rapid extraction kit (BioTeke, Beijing, China). The quality of RNA integrity and cDNA production by reverse transcription was checked with 1.0% agarose gel electrophoresis. Reverse transcription templates were synthesized using HiScript™ QRT SuperMix (Vazyme, NJ) according to the manufacturer's protocols. To perform the rapid amplification of cDNA ends (RACE), we used the universal primer A mix primer and gene-specific primers, and the gene cloning was conducted using Clontech Advantage 2 PCR kit from Takara (Dalian, China). The amplified products were cloned into pMD18-T vectors and sequenced by Beijing Genomics Institute (Beijing, China). The detailed procedures were performed according to the manufacturer's instructions. All primers are listed in table 1.

Sequence analysis and phylogenetic analysis
The sequences were obtained from polymerase chain reaction amplification, the ORF and RACE were assembled using DNA star software to assemble the full-length cDNA, and the full-length cDNA sequence was subjected to homology analysis. Similarity searching of amino acid sequences was conducted with BLAST in NCBI (http://www.ncbi.nlm.nih.gov/BLAST/). The isoelectric points of deduced proteins were predicted using ExPASy (http://www.au.expasy.org/). Translation of cDNAs and multiple sequence alignments was conducted with DNAMAN software (Lynnon Biosoft, Quebec, Canada), and characteristic motifs and domains were predicted using the simple modular architecture research tool (SMART; http://smart.embl-heidelberg.de/) and INTERPROSCAN (http://www.ebi.ac.uk/ interpro/). To examine the evolutionary relationships among the hsp90 and hsp70 family members in other species, a phylogenetic tree of different vertebrate hsp genes based on amino acid sequences was constructed by the neighbour-joining method and bootstrapped for 1000 replicates using MEGA v. 5 program (http://www.megasoftware.net/megamac.php).

Tissue distribution and mRNA expression of heat shock proteins
The real-time fluorescent quantitative polymerase chain reaction (RT-qPCR) method with β-actin as an internal control was used to explore the mRNA expression levels of Amhsp90, Amhsp70 and Amhsc70 in various tissues, including liver, heart, muscle, gill, spleen, kidney, brain and intestine, of untreated A. marmorata. RT-qPCR was performed following the manufacturer's protocol of the kit of SYBR Green Master (Roche, Basel, Switzerland). The primers for RT-qPCR (e.g. Amhsp90-RT-F/Amhsp90-RT-R) are listed in table 1. The experiments were carried out in triplicate with a total volume of 20 µl in ABI stepone™ plus (Applied Biosystems, USA), containing 10 µl of SYBR Green Master, 4 µl of cDNA (dilution to 5 ng µl −1 ) and 3 µl of each forward or reverse primer (2 µmol l −1 ). RT-qPCR was programmed at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, and 55°C for 1 min, and a final extension at 72°C for 60 s. To confirm the specificity of the amplification, the dissociation curve was analysed for amplified products to ensure an obvious amplification peak. The expression level of Amhsps was calculated by 2 − C T method and subjected to statistical analysis [19]. Similarly, RT-qPCR was also used to explore mRNA expression of Amhsps in liver, muscle, intestine and heart after being challenged with A. hydrophila.

Statistical analysis
Statistical analysis was performed using SPSS 19.0. The relevant values in this study were analysed through one-way analysis of variance followed by Tukey's test. Statistical significance was considered at p < 0.05, and highly significant difference was considered at p < 0.01. All data were expressed as mean standard errors (s.e.) in terms of relative mRNA expression.

Multiple sequence alignment and phylogenetic analysis
Amhsp90 amino acid sequence showed high identity with hsp90-alpha in Salmon salar (90%), Danio rerio (86%); Amhsp70 and Amhsc70 showed high identity with hsp70 and hsc70 in S. salar (87%, 96%) and D. rerio (90%, 96%). The alignment analysis between Amhsp70 and Amhsc70 amino acid sequences showed the identity of 83.41%. Conserved sequence and characteristic motifs were identified in the deduced amino acid sequences of Amhsp90, Amhsp70 and Amhsc70. The alignment results showed that the amino acid sequences of hsp90 family and hsp70 family have significant differences. To examine the relationships among hsp90, hsp70 and hsc70, the phylogenetic tree was established by MEGA v. 5.0 based on the neighbour-joining method through amino acid sequences. Different hsp90, hsp70 and hsc70 family members were selected from other vertebrate species, respectively. The phylogenetic tree showed that these proteins were divided into two major groups. One group comprised hsp70 family, and the other group contained hsp90 family. All constitutive form hsc70 and inducible form hsp70 were clustered in the branch of hsp70 family (figure 4).

Tissue-specific distribution of Amhsp genes
Tissue-specific distribution analysis using RT-qPCR method showed that three hsp mRNAs were ubiquitously expressed in all detected tissues of liver, heart, muscle, gill, spleen, kidney, brain and intestine. The mRNA transcripts of three Amhsps were expressed at a relatively high level in liver, intestine, muscle and heart, at the lowest level in spleen and at a moderate level in other examined tissues ( figure 5).   Amhsp70 (b) and Amhsc70 (c) genes from A. marmorata in brain, gill, spleen, intestine, liver, kidney, muscle and heart was explored using RT-qPCR methods. Each sample was run in triplicate. Deviation bars are the standard errors. The A. marmorata β-actin gene was used as an internal control to calibrate the cDNA template for all samples.

Expression of Amhsps in response to Aeromonas hydrophila injection
The temporal expression profile of Amhsp90 was observed after bacterial challenge. In liver, the expression level of Amhsp90 revealed a rapid upregulation within 1 h after the challenge with A. hydrophila and a peak level at 6 h, and then exhibited a decreasing trend from 48 to 72 h (figure 6a). In intestine, Amhsp90 reached the highest level at 12 h (figure 7a). In muscle and heart, the expression of Amhsp90 was rapidly upregulated from 3 to 24 h and reached the highest level at 24 and 6 h, and then dropped rapidly (figures 8a and 9a).   Upregulated expression of Amhsp70 mRNA from 1 h post-challenge in liver and muscle was observed (figures 6b and 8b). After treatment, the inducible Amhsp70 showed a significantly high expression at 6 h in intestine (figure 7b) and heart (figure 9b). From 6 to 72 h, Amhsp70 showed a highly significant difference in intestine, but it showed a highly significant difference in heart only between 6 and 24 h. In heart, the expression level of Amhsp70 was rapidly upregulated from 6 to 24 h. At 24 h, the expression reached the peak level, and then levelled off at 48 h.
The expression of Amhsc70 mRNA presented a fluctuating trend in muscle (figure 8c) and liver (figure 6c). In intestine, Amhsc70 mRNA expression level basically remained unchanged when compared with that in the control group, and reached the maximum level at 12 h (figure 7c). In heart, Amhsc70 showed a significant upregulation at a middle phase, and then rapidly decreased in a short time (figure 9c).

Discussion
HSPs are stress response proteins as a 'dangerous signal' to protect the immune system and the immune cells involved in the protection of cytoplasm components, including all kinds of biological factors, such as bacterial infection [24]. It can be used as the immune system to identify the important antigens for two reasons: first, the mRNA expression of hsps in most organisms revealed an obvious increase in the process of immune response when pathogens are engulfed by macrophages, in order to protect the organism and to maintain life force. Second, HSP is highly conserved, and the immune system can easily identify these highly conservative molecules. When bacteria invade an organism, the organism may release certain cell toxins, and promote intracellular cytokine synthesis and secretion caused by the variation of protein or polypeptide chain fragments. These abnormal proteins can be induced by hsp genes in cells with high expression efficiency [25].  In this study, we obtained full-length cDNAs of Amhsp90, Amhsp70 and Amhsc70 of A. marmorata for the first time. They are similar to most of the known HSPs in teleost fish. Moreover, multiple sequence alignment results also indicated that hsps were highly conserved, suggesting that Amhsps may share a similar function with other known hsps (figure 10). In the phylogenetic tree, three Amhsps were clustered together with teleost. Amhsp70 and Amhsc70 were clustered into a major branch and all belonged to hsp70 family ( figure 4). The topological structure displayed in the phylogenetic tree is in good agreement with traditional taxonomy. The molecular information of Amhsps will be more useful for further exploring the expression of hsp genes, such as thermal stress, and this gene sequence information expands the gene database, and provides a theoretical basis for further studies on other hsp genes in other teleost fishes.
The Amhsps exhibited a similar gene expression pattern in different tissues. A broad tissue distribution of hsp genes was observed in A. marmorata, which revealed high expression in liver, intestine, heart and muscle. Specifically, Amhsp70 and Amhsc70 were expressed at high levels in liver, and Amhsp90 mRNA expression was the highest level in heart. In comparison with M. amblycephala, S. senegalensis, S. ocellatus and O. niloticus, there were some different expression patterns with tissue-specific distribution [26][27][28][29]. Therefore, the distribution of hsp genes is varied in different tissues owing to different species. There is a tissue-or species-specific profile in response to the challenge with A. hydrophila.
According to the tissue-specific distribution of three hsp genes, four tissues (liver, muscle, intestine and heart) were selected as candidates. In liver, Amhsp90, Amhsp70 and Amhsc70 were rapidly upregulated within 1 h of A. hydrophila challenge and peaked at 6, 24 and 6 h (p < 0.01; figure 6). The experimental results showed that three Amhsps were sensitive to bacterial stimulation in liver, especially at the early stage following injection. The initial upregulation may be due to bacterial toxicity, and Amhsps gene could be activated quickly and transferred to the cytosol [30], which was in accordance with the expression patterns of ScHSP70s following A. hydrophila infection and M. amblycephala [12,31]. In the whole experiment process, the upregulation of hsps may be a protective mechanism, because hsps could bind to the damaged or misfolded proteins to restore their original structures [24,32].
In muscle, three Amhsps showed a similar dynamic trend. Amhsp70 and Amhsc70 mRNA levels reached a maximum at 12 h and then gradually decreased after bacterial challenge. However, Amhsp90 significantly increased until the challenge for 24 h and then sharply reduced. In C. macrocephalus, hsc70-1 was nearly constant and hsc70-2 revealed a continuous increase. Once the intramuscular injection is applied, bacteria can immediately affect muscle lesions, and the high levels of these three Amhsps in muscle may reflect cell protection of hsps [5]. In liver and muscle tissues, three Amhsp genes showed different expression patterns, which may be due to liver tissue as the most important metabolic organ and defence organ with rapid mRNA expression of Amhsps with more intensity in the early phase during 72 h of infection. However, muscle tissue is located under the skin. When the fish were injected with A. hydrophila, it was easy to cause damage of skin and muscle. Therefore, the mRNA expression of Amhsp genes increased with an increase in swimming time, the contact surface of the wound may be enlarged, and the time point of high expression of Amhsps in muscle is relatively delayed when compared with liver.
There are very few reports about the expression patterns of intestinal mRNA owing to bacterial infection in fish. A surprising discovery of this study is that the mRNA expression levels of Amhsp90 and Amhsc70 reveal the peak level at 12 h (p < 0.01). The expression level of Amhsp90 exhibits a rapid increase to the peak level at 12 h (p < 0.01) post-infection and a sequential decrease. In comparison with Amhsp90, Amhsc70 mRNA expression reveals a sharp rise at 12 (p < 0.01) and 24 h (p < 0.05), and there is no significant change at other time points. At the same time, Amhsp70 mRNA expression from 6 (p < 0.01) to 24 h (p < 0.01) shows an increasing trend. In contrast, in channel catfish, hsp90 does not show a significant fold change in intestine at 3 h, 24 h and 3 days after Edwardsiella ictaluri infection [33]. This study suggested that three Amhsps played an important role in immune stress. Likewise, a clear time-dependent mRNA expression pattern of Amhsp90 in heart was also observed when infected with A. hydrophila. Amhsp70 and Amhsc70 mRNA expression exhibited an increase at the middle phase, reached the peak at 24 (p < 0.01) and 12 h (p < 0.01), and began to decrease. According to an early report, hsp60 has a significant increase at 4 h after A. hydrophila challenge in heart of grass carp.
In teleost fish, reports regarding heart and intestinal immune mechanisms associated with hsps are rather few. Our results suggest that Amhsps plays an important role in the intestine and heart of flower eel. Intestine digestive tube contains diffused lymphoid tissue, lymphocytes, macrophages and plasma cells, which are involved in immune defence. When intestinal mucosa was infected by bacteria after the stimulation of lymphoid tissue within the mucosal immune response, then endocrine immune globulin was produced in the digestive tube to prevent bacterial invasion in the digestive tract, thus regulating the compositions of the intestinal mucosal immune system [34]. Therefore, the infection of bacteria caused interesting expression patterns. At the same time, HSP is an important stress protein to protect myocardial cells against myocardial injury, can participate in the repair and restoration of ion channel redox balance, reduce the release of oxygen free radicals, and can also be used as an antioxidant for free radicals as a molecular chaperone to protect cells from damage [35]. Thus, in heart, the expression of Amhsps is very concentrated and intense.
In general, this is the first time of cloning cDNAs of Amhsp90, Amhsp70 and Amhsc70. Bioinformatic analysis has confirmed that three Amhsps belong to hsp90 family and hsp70 family, respectively, and are ubiquitously expressed in eight tested tissues. We have also demonstrated that three Amhsps present more rapid and sensitive expression in liver and muscle after A. hydrophila challenge, whereas a relatively delayed sensitivity was observed in intestine and heart. These Amhsp genes may be involved in the regulation of A. hydrophila response in A. marmorata.
Animal ethics. The following information is supplied relating to ethical approvals (i.e. approving body and any reference numbers): the experiments were conducted on A. marmorata that is regarded as species under second-class protection in China. All experiments were performed according to the Guideline for the Care and Use of Laboratory Animals in China. This study was also approved by the Ethics Committee of Experimental Animals at Nanjing Normal University (research permit number: SYXK2015-0028). The location is not privately owned or protected in any way. Data accessibility. Data file on tissue distribution and gene expression from this study can be found in the electronic supplementary material.
DNA deposition. The following information regarding the deposition of DNA sequences is supplied: GenBank accession nos. KT274760-KT274762.