An inhibitor of apoptosis protein (EsIAP1) from Chinese mitten crab Eriocheir sinensis regulates apoptosis through inhibiting the activity of EsCaspase-3/7-1

Inhibitor of apoptosis proteins (IAPs) maintain the balance between cell proliferation and cell death by inhibiting caspase activities and mediating immune responses. In the present study, a homolog of IAP (designated as EsIAP1) was identified from Chinese mitten crab Eriocheir sinensis. EsIAP1 consisted of 451 amino acids containing two baculoviral IAP repeat (BIR) domains with the conserved Cx2 Cx6 Wx3 Dx5 Hx6 C motifs. EsIAP1 mRNA was expressed in various tissues and its expression level in hemocytes increased significantly (p < 0.01) at 12–48 h after lipopolysaccharide stimulation. In the hemocytes, EsIAP1 protein was mainly distributed in the cytoplasm. The hydrolytic activity of recombinant EsCaspase-3/7-1 against the substrate Ac-DEVD-pNA decreased after incubation with rEsIAP1. Moreover, rEsIAP1 could directly combine with rEsCaspase-3/7-1 in vitro. After EsIAP1 was interfered by dsRNA, the mRNA expression and the hydrolytic activity of EsCaspase-3/7-1 increased significantly, which was 2.26-fold (p < 0.05) and 1.71-fold (p < 0.05) compared to that in the dsGFP group, respectively. These results collectively demonstrated that EsIAP1 might play an important role in apoptosis pathway by regulating the activity of EsCaspase-3/7-1 in E. sinensis.


Results
The sequence characteristics and phylogeny of EsIAP1. A novel sequence of EsIAP1 (GenBank accession numbers MF351747) was identified from E. sinensis genome database. The open reading frame of EsIAP1 was of 1,356 bp, encoding a predicted polypeptide of 451 amino acids with calculated molecular weight of approximately 50 kDa. SMART analysis demonstrated that EsIAP1 contained two BIR domains (BIR1 and BIR2). The conserved cysteine and histidine residues and the spacing between them in the reported BIR2 domains (Cx 2 Cx 6 Wx 3 Dx 5 Hx 6 C) were also identified in the BIR2 domain of EsIAP1 (Fig. 1a). The deduced amino acid sequences of BIR1 and BIR2 domains of EsIAP1 shared high sequence similarities with the corresponding domains of other IAPs, such as those from L. vannamei IAP1 (40.5% and 50.7%), Mus musculus XIAP (43. (36.1% and 57.5%), and Penaeus monodon IAP (40.5% and 50.7%) (Fig. 1b). To evaluate the evolutional relationship of EsIAP1, a phylogenetic tree was constructed based on the amino acid sequences of 13 IAP members by the neighbor-joining method. EsIAP1 was firstly clustered with other arthropod IAPs in the phylogenetic tree, and then grouped with invertebrate IAPs, and finally clustered into the vertebrate XIAPs and c-IAPs (Fig. 1c).
Tissue distribution of EsIAP1 mRNA and subcellular localization of EsIAP1 in hemocytes. The mRNA transcripts of EsIAP1 could be detected in all the examined tissues, including hemocytes, hepatopancreas, heart, gill, brain and muscle with the highest expression level in hepatopancreas, which was 5.24-fold (p < 0.01) of that in muscle. The expression level of EsIAP1 mRNA in gill, hemocytes, heart and brain was 3.51-fold (p < 0.01), 3.41-fold (p < 0.01), 2.16-fold (p < 0.05) and 1.61-fold (p > 0.05) of that in muscle, respectively (Fig. 2a).
The purified rEsIAP1 was employed to prepare polyclonal antibody (Fig. 2b, Supplementary Fig. S1). A clear band with 55 kDa was revealed by western blotting assay, indicating the high recognition specificity of the polyclonal antibody against EsIAP1 (Fig. 2b, Supplementary Fig. S1). Pre-immune serum was used as negative control and no bands were detected (Fig. 2b, Supplementary Fig. S1). Western blotting assay of the native tissue sample with EsIAP1 antibody revealed that there was a distinct band of 50 kDa (Fig. 2c). Immunofluorescence assay was performed to detect the localization of EsIAP1 in hemocytes. The nucleus stained by DAPI was observed in blue, and the positive signal of EsIAP1 was in green. The positive fluorescence signals were mainly observed in the cytoplasm of hemocytes according to the merged pictures (Fig. 2d).  www.nature.com/scientificreports www.nature.com/scientificreports/ compared with that in PBS group, p < 0.01), reached the highest level of 13.86-fold (p < 0.01) at 24 h, and finally down-regulated to 2.44-fold (p < 0.01) at 48 h (Fig. 3a). After A. hydrophila stimulation, the relative expression level of EsIAP1 mRNA kept at quite low level and there was no significant difference from 0 to 24 h compared with that in PBS group. However, it increased significantly (2.71-fold of control group, p < 0.01) at 48 h post A. hydrophila stimulation (Fig. 3b).
The mRNA expression of EsCaspase-3/7-1 and the activity of caspases in hemocytes after the gene silencing of EsIAP1. To further explore the function of EsIAP1 in vivo, the dsRNA-induced RNA interference (RNAi) was used to inhibit the expression of EsIAP1. The mRNA expression of EsIAP1 in hemocytes decreased to 0.47-fold (p < 0.05) (Fig. 4a), while the mRNA expression of EsCaspase-3/7-1 increased to 2.26-fold (p < 0.05) (Fig. 4b) at 24 h post the injection with sequence-specific dsRNA targeting EsIAP1 compared to that in dsGFP group. After EsIAP1 was silencd, the activity of caspase towards Ac-DEVD-pNA in hemocytes increased to 1.71-fold (p < 0.05) compared to that in the dsGFP group. While the activity of caspase-1 towards Ac-YAVD-pNA and caspase-6 towards Ac-VEID-pNA in hemocytes increased to 1.21-fold (p > 0.05) and 1.25-fold (p > 0.05) of that in the dsGFP group, respectively (Fig. 4c).

Discussion
Apoptosis represents a fundamental biological process that relies on the activation of caspases 34 . IAPs are a family of negative regulators of both caspases and cell death 35 . In the present study, a novel IAP was identified from E. sinensis (designated EsIAP1). There were two BIR domains identified in EsIAP1, which was the typical domain of IAP family 6 . The deduced amino acid sequences of BIR1 and BIR2 domains in EsIAP1 shared high similarities (36.1%~53.5% and 45.9%~58.9%, respectively) with the corresponding domains of other IAP proteins (Fig. 1a,b). Moreover, the conserved spacing of cysteine and histidine residues (Cx 2 Cx 6 Wx 3 Dx 5 Hx 6 C) in the other reported BIR2 domains were also found in EsIAP1, which was suggested to contribute to a novel zinc-binding fold 6 . These results suggested that EsIAP1 was a typical IAP family member. In the phylogenetic tree, EsIAP1 was firstly grouped with the crustacean IAPs to form a separated clade, then grouped with other arthropod IAP proteins, and finally clustered with vertebrate IAP proteins (Fig. 1c). These evidences collectively indicated that EsIAP1 belonged to the IAP family in crustaceans.
As regulators of the apoptotic machinery, IAPs play important roles in many physiological processes, including homeostasis maintenance, development of tissues, and immune responses 22,36,37 . In the present study, the mRNA transcripts of EsIAP1 could be detected in all the examined tissues inculding hemocytes, hepatopancreas, gill, muscle, brain, and heart (Fig. 2a). Similarly, the transcripts of CgIAP2 were also detected in various tissues in oyster C. gigas 25 . The constitutive expression profile of EsIAP1 indicated that it might involve in www.nature.com/scientificreports www.nature.com/scientificreports/ many physiological processes of crabs. It has been reported that IAPs could regulate the activity of caspases, further modulate cell cycle proliferation and receptor-mediated signal transduction 9 . The higher level of EsIAP1 mRNA was observed in immune-associated tissues, including hemocytes, hepatopancreas and gill, which might be attributed to the cellular metabolism and innate immunity 38,39 . Crustacean hemocytes play important roles in the host immune response, including recognition, phagocytosis and cell communication 33,40,41 . Moreover, the high level of EsCaspase-3/7-1 and EsCaspase-3-like were observed in hemocytes 28,29 . The hemocytes were thus chosen as target to analyze the expression of EsIAP1. In the present study, the location of EsIAP1 in hemocytes was observed by immunofluorescence assay, and the positive signal was found to be mainly distributed in the cytoplasm, which was similar as the previous reports in other species [42][43][44][45] , possibly for the sake of binding to cytoplasmic caspase to regulate hemocyte apoptosis. LPS, a vital component of the outer wall of gram-negative bacteria, could trigger caspase-mediated apoptosis pathway 42,46 . The apoptosis pathway is regulated by initiator caspases (such as aspase-8 and caspase-10), which can be triggered by death receptor (like Fas, TNFR1 and TRAIL-R1/R2) to initiate the activity of effector caspases [43][44][45] . In the present study, the expression level of EsIAP1 mRNA was significantly up-regulated after LPS and A. hydrophila stimulations (Fig. 3a,b). It has been reported that apoptosis pathway could be activated after LPS and A. hydrophila stimulations in crustacean 47,48 . In C. gigas, CgIAP2 was proposed to play a role in apoptosis inhibition in the immune defense against bacterial challenge 25 . Some crustacean IAPs such as PmIAP and LvIAP1 were suggested to be central to the regulation of hemocyte apoptosis 24,49 . Theses results suggested that EsIAP1 might exert important roles in immune defenses by regulating the apoptosis pathway in E. sinensis.
IAPs could regulate apoptosis through controlling caspase activities and caspase-activating platform formations, which also appeared to be important determinants of the responses of cells to endogenous or exogenous cellular injuries 13 . It was reported that c-IAP1 and c-IAP2 could directly bind to the activated caspase-3 and -7 to inhibit their activities in vertebrates 10 . In the present study, EsCaspase-3/7-1, an effector caspase, identified previously from E. sinensis 28 , were employed to investigate the binding activity of EsIAP1 with caspase. After rEsCaspase-3/7-1 was incubated with rEsIAP1 or Z-VAD-FMK, the hydrolyzing activity of rEsCaspase-3/7-1 was significantly decreased (Fig. 5e). This result was in coincidence with the observation that IAPs could inhibit the activation of effctor caspase 13,20 . The direct combination of rEsIAP1 with rEsCaspase-3/7-1 confirmed by pull down assay might explain the decrease of rEsCaspase-3/7-1 hydrolyzing activity after incubation with rEsIAP1 in vitro. These results suggested that rEsIAP1 could inhibit the hydrolytic activity of rEsCaspase-3/7-1 by interacting with rEsCaspase-3/7-1. Furthermore, the expression of EsCaspase-3/7-1 mRNA in hemocytes of crabs were significantly increased after the interference of EsIAP1, indicating the inhibitory regualtion of EsIAP1 on EsCaspase-3/7-1. The hydrolytic activity of caspase-3 was increased in hemocytes rather than caspase-1 and -6 after EsIAP1 was silenced. These results showed that EsIAP1 could regulate EsCaspase-3/7-1 and further inhibit hemocyte apoptosis in vivo. Similarly, the number of circulating hemocytes was increased in LvIAP1-silenced shrimp because of the extensive apoptosis 24 . Some mammalian IAPs, such as c-IAP1 and c-IAP2, were also found to be involved in signaling cascades, and play important roles in TNF-induced apoptosis 50 . Therefore, it was speculated that EsIAP1 could inhibit apoptosis by regulating EsCaspase-3/7-1 in E. sinensis.
Caspases are activated to gain the full catalytic activity after being proteolytically cleaved to initiate apoptosis 51 . EsCaspase-3, -7 and -8 are characterized to play crucial roles in Cadmium-induced apoptosis 27 , and EsCaspase-3/7-1 and EsCaspase-3-like protein are found to be involved in innate immune response and induce apoptosis under pathogen stimulation 28,29 . IAPs inhibit such proteolytically activated caspases, and further regulate apoptosis 52 . The loss or inhibition of cIAP1, cIAP2 and XIAP causes the majority of cells to be sensitized to death receptor to induce cell death 53 . In Drosophila, DIAP1 normally inhibits both initiator and effector caspases 54,55 . In summary, this study suggested that LPS and bacterial challenge could activate the apoptosis pathway in E. sinensis. EsIAP1 could inhibit apoptosis by directly combining with EsCaspase-3/7-1 and inhibit www.nature.com/scientificreports www.nature.com/scientificreports/ its hydrolytic activity (Fig. 5f). These results provided novel idea to understand the modulatory role of IAP in apoptosis pathway in crustaceans.

Materials and Methods
Crabs, collection of tissues and immune stimulations. The crabs with an average weight of 20 g were collected from a commercial farm in Lianyungang, China, and cultured in aerated freshwater at 20 ± 2 °C for one week before processing 28 . Six crabs were sacrificed for determining the expression of EsIAP1 mRNA in different tissues. The tissues including muscle, heart, brain, hepatopancreas and gill were collected from crabs to detect the mRNA expression of EsIAP1 according to the previous study 56 . The hemolymph drawing from the last pair of walking legs from each crab by using a syringe was mixed with anticoagulant solution (510 mM NaCl, 100 mM glucose, 200 mM citric acid, 30 mM sodium citrate, 10 mM EDTA·2Na, pH 7.3) at a ratio of 1:1, and the hemocytes were harvested by centrifugation 57 . Tissues from two crabs were pooled together as one sample and there were three duplicates for each assay according to the previous methods 58 . The crabs were treated by the injections www.nature.com/scientificreports www.nature.com/scientificreports/ of 100 μL Aeromonas hydrophila (10 7 CFU mL −1 ) and 100 μL lipopolysaccharide (500 μg mL −1 ) according to the previous reports 59 , respectively. Ninety crabs were employed and randomly divided into three groups. According to previous study, a volume of 100 μL alive A. hydrophila (1 × 10 7 CFU mL −1 ) or lipopolysaccharide (LPS from Escherichia coli 0111:B4, L2630, Sigma Aldrich, USA; 100 µg mL −1 ) resuspending in PBS (40 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 2 mM KH 2 PO 4 , pH 7.4) was injected into the arthrodial membrane of the last pair of walking legs in the stimulation groups, respectively 28,59,60 . The crabs received an injection of 100 µL PBS were employed as control group. Six crabs were randomly sampled from each group at 0, 6, 12, 24 and 48 h after treatments.
RNA extraction and cDNA synthesis. TRIzol reagent (Invitrogen) was uesd for the extraction of total RNA from E. sinensis tissue samples, and the first-strand cDNA was synthesised by using the PrimeScript ™ real-time PCR kit (Takara, Japan) according to the manufacture's instruction.
Sequence analysis of EsIAP1. The sequence of IAP genes was analyzed by BLASTP (http://blast.ncbi. nlm.nih.gov/Blast.cgi) in the genome database (PRJNA305216) of E. sinensis 61 . The primers (EsIAP1-F and -R) were designed to clone the open reading frame (ORF) of EsIAP1. The multiple sequence alignments were created by Clustal W. The conserved domain was identified through the SMART (http://smart.embl-heidelberg.de/). MEGA6.0 package was used to construct phylogenetic tree.
Purification of recombinant protein and preparation of polyclonal antibody. The full-length ORF sequences of EsIAP1 and EsCaspase-3/7-1 were amplified with specific primers (rEsIAP1-His-F and -R, rEsCaspase-3/7-1-F and -R) ( Table 1). The PCR products were inserted into pET-22b vector (Novagen) with a His-tag. rEsIAP1-GST-F and -R (Table 1) were used to amplify EsIAP1, and the PCR products were inserted into the pGEX4T-1 vector (GE Healthcare) with a GST-tag. All those recombinant plasmids were transformed into E. coli BL21 (DE3) competent cells. These prokaryotic proteins were purified by a Ni 2+ chelating sepharose column or GST-resin, following the manufacturers' instructions. Their concentrations were measured by BCA kit (Beyotime). The preparation of antiserum was performed as previously described 62 . Western blotting and immunohistochemistry analysis of EsIAP1. The western blotting assay was performed according to the previous report 28 . Recombinant protein was separated by SDS-PAGE, and transferred onto nitrocellulose membrane. After blocking for 1 h with 5% non-fat milk in TBST, the membrane was incubated successively with 1/1000 diluted poly-antibody of anti-EsIAP1 as first antibody and 1/10,000 diluted goat-anti-mouse IgG (Sangon) as secondary antibody. After washing in TBST, the membrane dipped in ECL substrate system (Thermo Scientific) for 2 min, then imaged by Amersham Imager 600 (General Electric Company).
The hemocytes were resuspended with DMEM (Sangon) and then added into poly-L-lysine pre-coated dishes. After fixed with 4% paraformaldehyde (PFA, diluted in PBS), the dishes were blocked with 3% fetal bovine serum albumin (diluted in PBS) at 37 °C for 30 min, followed by washing with PBST (PBS with 0.1% tween-20). The dishes were then successively incubated with 1/1000 diluted anti-EsIAP1 antibody at 37 °C for 1 h and 1/1,000 diluted Alexa Fluor 488-labeled goat-anti-mouse antibody at 37 °C for 1 h. After final washing with PBST, DAPI (1 µg/mL in PBS) was used to stain the nucleus, and the dishes were observed by fluorescence microscope (ZEISS).

RNA interference.
The RNA interference assay of EsIAP1 was performed according to the previous report 32 .
T7 promoter linked primers (GFP-RNAi-F and -R, EsIAP1-RNAi-F and -R) were used to amplify the cDNA sequence of dsGFP (657 bp) and dsEsIAP1 (1,356 bp), respectively. The dsRNAs of EsIAP1 and GFP were diluted www.nature.com/scientificreports www.nature.com/scientificreports/ with PBS to the final concentration of 0.5 μg μL −1 . The crabs were treated by the injections with 100 μL PBS, dsGFP and dsEsIAP1, respectively. The untreated crabs were used as blank group. Six crabs from each group were randomly sampled at 0 and 24 h post injections. The hemocytes were divided into two parts, and one aliquot of hemocyte sample was used to estimate the silencing efficiency, while the other was used for the measurment of caspase activity.

Quantitative real-time PCR (qRT-PCR) analysis of mRNA expression. qRT-PCR was conducted
according the previous reports 63 . Two primers, EsIAP1-qRT-F and -R, were used in qRT-PCR to detect the expression of EsIAP1. The fragment amplified by primers of Es-β-actin (Es-β-Actin-F and -R) were employed as reference. The gene expression analysis was performed using the 2 −∆∆Ct method 63 , and all data were given in terms of relative mRNA expression of mean ± S.E. (N = 3).
Pull down assay. The pull down assay was carried out according to the previous report 63 . The proteins of rEsIAP1 (GST) and rEscaspase-3/7-1 (His) (30 μg) were mixed with 20 μL of glutathione resin (for GST-tagged proteins) or charged nickel-nitrilotriacetic acid beads (for His-tagged proteins), respectively. The mixture (resin and binding proteins) was incubated at room temperature for 2 h with slight rotation, and then washed for three times by centrifuging at 500 × g for 3 min to remove the unbound proteins. The tested protein (rEsCaspase-3/7-1-His and rEsIAP1-GST), without GST tag or His tag, was added into the mixture containing the nickel-nitrilotriacetic acid beads or glutathione resin, and gently rotated at room temperature for 2 h. After washing three times, the mixture was analyzed by SDS-PAGE.
Hydrolyzing activity analysis of caspases in EsIAP1-interfered crabs. The hydrolyzing activity of caspases in hemocytes was examined according to the method described by previous study 29 . The hydrolytic activity of the crab hemocyte protein was detected at 0 and 24 h after the injection of EsIAP1-dsRNA. The protein concentration of the supernatant was measured using the Bradford Protein Assay Kit (Beyotime) and adjusted to 1 mg mL −1 with lysate buffer. The hydrolytic activity of caspases was examined with the substrate Ac-YAVD-pNA, Ac-DEVD-pNA and Ac-VEID-pNA using the caspase-1, -3 and -6 activity assay kit (Beyotime, Shanghai, China) under the manufacturer's manual. The absorbance values of the reaction mixture was monitored at 405 nm using Spectra Max 190 (Molecular Devices, Sunnyvale, CA, USA). The different absorbance values represented the cleavage and release of pNA. The blank group was used as the reference. Statistical analysis. The data (represented as mean ± S.E., N = 3) were calculated by using the 2 −∆∆Ct method 65 , and analyzed with t-test. Significant differences across controls were indicated with an asterisk at p < 0.05, and two asterisks at p < 0.01.