CgDM9CP-5-Integrin-MAPK Pathway Regulates the Production of CgIL-17s and Cgdefensins in the Pacific Oyster, Crassostrea gigas

Immune recognition is the first step of host innate immunity to sense invaders and trigger the following immune protection (1). Since pattern recognition theory was proposed in the 1990s (2), a lot of conserved recognizing protein domains have been identified in plants, invertebrates, and vertebrates to recognize the conserved pathogen-associated molecular patterns (PAMPs), such as the leucine-rich repeat domain (3), the lectin domain (4), and the Ig domain (5). These functional domains constitute most of the classical pattern-recognition receptors (PRRs), including the TLR, the NOD-like receptor, and the C-type lectin receptor in innate immunity, and even the variable lymphocyte receptor and BCR/TCR in adaptive immunity (6). Once these recognizing domains are occupied by corresponding ligands, the receptors immediately activate the downstream immune signal, for example, the MAPK and NF-κB signal, to induce inflammatory or defensive reactions (7). Although the three main framework domains have greatly accounted for the immune recognition mechanisms in living organisms, to our knowledge, there are exceptions, and novel recognizing protein domains are occasionally discovered, for example, the DM9 domain (8).

The DM9 domain was first identified in Drosophila melanogaster (9), and it was subsequently presumed to exist in most organisms except for plants and mammals (8). Several DM9 domain containing proteins (DM9CPs) have been reported, such as five in Platyhelminthes (10–12), four in Mollusca (8, 13–15), six in Arthropoda (9, 16–18), and five in fish (19, 20). These DM9CPs were found to act as PRRs to bind various PAMPs and microbial pathogens. The DM9CP in mosquito (Pfs47Rec) was composed of four tandem DM9 domains and it was able to bind Pfs47 of Plasmodium falciparum with high affinity (17). FgDM9-1 in Fasciola gigantica was able to agglutinate Gram-positive and Gram-negative bacteria with appropriate cell surface glycosylation patterns (10). CgCGL1 from the Pacific oyster (Crassostrea gigas) (renamed CgDM9CP-1) displayed strict specificity for mannose monomer and high-mannose-type N-glycans (8, 21). Whereas CgDM9CP-2 and CgDM9CP-3 could act as a PRR with binding capability to mannose, LPS, peptidoglycan (PGN), as well as various bacteria (13, 14), CgDM9CP-4 was indicated as a marker for the prohemocyte of oysters (15). Some of these DM9CP proteins were also reported to mediate various downstream immune response processes, such as phagocytosis (8, 22), agglutination activity (13), encapsulation (14), and hemagglutination (10). However, the detailed downstream signals of DM9CPs are still not clear.

C. gigas is marine bivalve belonging to the phylum Mollusca, which has developed a sophisticated innate immune system against pathogens (23). The comprehensive genomic annotation revealed a great number of genes constituting the oyster immune system, among which the members of PRR families exhibited significant expansions. The expanded PRRs are supposed to orchestrate a broad-spectrum and differentiated immune recognition and exhibit synergistic effects on the initiation of the downstream immune response (24). There are several DM9CPs with varied mRNA expression patterns identified in the genome of C. gigas, possibly indicating their diverse function in the immune response (13–15, 24). Among them, CgDM9CP-5 was expressed at the highest level in gill compared with other tissues (25). As filter feeders, oysters perpetually encounter environmental microbial pathogens in gill with continuous risk of infection. It is of great significance to investigate the role of CgDM9CP-5 in gill in defending against the invading microbial pathogens such as Vibrios. Integrins are important cell adhesion molecules found in C. gigas (26), among which a β integrin was identified as a membrane protein to mediate the phagocytosis toward Vibrio splendidus (27). These findings encouraged us to suspect that CgDM9CP-5 might function as an important recognition receptor in gill to trigger the downstream immune response via the membrane protein CgIntegrin. The objectives of the current study were to 1) investigate the function of CgDM9CP-5 to recognize and bind PAMPs and microorganisms, 2) identify the possible downstream signaling pathway mediated by CgDM9CP-5, and 3) explore the main immune reactions regulated by CgDM9CP-5 signal in gill of oysters.

Adult oysters (C. gigas) with a shell length of 9–13 cm were collected from a local farm in Dalian, Liaoning Province in China. They were cultured in the aerated seawater at 15–20°C for 10 d before processing. All of the experiments were performed following the animal ethics guidelines approved by the Ethics Committee of Dalian Ocean University.

Bacteria Staphylococcus aureus were obtained from the Microbial Culture Collection Center (Beijing, China). Escherichia coli Transetta (DE3) was purchased from TransGen Biotech (Beijing, China). V. splendidus was previously identified and preserved in our laboratory (28). Additionally, the fungi Yarrowia lipolytica was provided by Dr. Z. Chi from Ocean University of China.

Tissues including hepatopancreas, mantle, gonad, adductor muscle, haemolymph, labial palp, and gill were dissected and sampled from nine adult oysters. Each tissue from three individuals was pooled together as one sample. The hemolymph was collected and centrifuged at 800 × g, 4°C for 10 min to collect haemocytes. Other tissues were transferred into 1 ml of TRIzol reagent (Thermo Fisher Scientific, Waltham, MA) and RIPA buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, and 0.1% SDS) for RNA and protein extraction, respectively.

Mucus from gill and mantle was separately collected according to a previous report (29). Briefly, mucus was carefully collected using small sterile pieces of cotton balls, and the cotton balls were immersed in 5–10 ml of sterile seawater (SSW) on a rotating shaker at 4°C for 1 h. The resulting fluids were centrifuged (3000 × g, 30 min, 4°C), filter sterilized (0.22 μm syringe filters), and maintained at 4°C for the following experiments. A 25 μl aliquot of each fluid was used to determine protein concentrations with a Bradford protein assay kit (Beyotime Biotechnology, Shanghai, China) according to the manufacturer’s recommendations. Fluids were then diluted with SSW to a final protein concentration of 1 mg/ml.

Two hundred forty oysters were randomly divided into three groups with 80 individuals in each group. The oysters in the two stimulation groups received individually an injection of 100 μl of live V. splendidus suspension in SSW (1 × 10⁹ CFU/ml) and 100 μl of d-mannose (0.5 mg/ml in SSW, Sigma-Aldrich, M2069), respectively. The oysters that received an injection of 100 μl of SSW were employed as the control group. The oysters were returned to water tanks after the injections, and nine individuals from each group were randomly sampled for the gill collection at 0, 6, 12, 24, 48, and 72 h postinjection. The gills from three individuals were pooled together as one sample. All of these samples were stored at −80°C with TRIzol reagent (Thermo Fisher Scientific, Waltham, MA) for subsequent RNA extraction.

Total RNA was immediately extracted using TRIzol reagent according to the previous report (30). The quantity and quality of the extracted RNA were evaluated by NanoDrop 2000 and electrophoresis on 1% agarose gel, respectively. The first strand of cDNA was synthesized by using total RNA as the template and oligo(dT)-adaptors as primers according to the protocol of the manufacturer. The synthesis reaction was performed at 42°C for 30 min, and terminated by heating at 85°C for 5 s. The cDNA mix was diluted at 1:40 and stored at −80°C.

The primers CgDM9CP-5-F and CgDM9CP-5-R (Table I) were designed according to the sequence information (National Center for Biotechnology Information [NCBI] accession no. LOC117690720, acquired from NCBI at http://www.ncbi.nlm.nih.gov/) for PCR amplification. The PCR products were cloned into pMD18-T simple vector (Takara Bio, Shiga, Japan) and confirmed by DNA sequencing. The recombinant plasmid pMD18-T–CgIntegrin plasmid was obtained from previous research according to the sequence information (NCBI accession no. AB066348) (27) and confirmed by sequencing. The cDNA sequence and deduced amino acid sequence were analyzed using the BLAST algorithm (http://www.ncbi.nlm.nih.gov/blast) and the Expert Protein Analysis System (http://www.expasy.org/). The protein domain was predicted by an online platform (http://smart.embl.de/). Clustal X 1.81 software was used to create the multiple sequence alignment. An unrooted phylogenetic tree was constructed based on the sequence alignment by using the neighbor-joining algorithm and the Mega X program (31). The liability of the branching was tested by bootstrap resembling analysis (1000 pseudoreplicates).

The recombinant proteins of CgDM9CP-5 (rCgDM9CP-5) and CgIntegrin (rCgIntegrin) were produced and purified as previously described with some modification (32). The coding region of CgDM9CP-5 was amplified by the primers CgDM9CP-5-ExF and CgDM9CP-5-ExR (Table I). The PCR products were digested with NdeI and XhoI, gel-purified, and ligated into pET-30a plasmid to construct the expression plasmid pET-30a–CgDM9CP-5. The fragment of the CgIntegrin gene was amplified with the primers CgIntegrin-ExF and CgIntegrin-ExR (Table I). The PCR products were digested with BamHI and NdeI, gel-purified, and ligated into pMAL-c5x plasmid to construct the expression plasmid pMAL-c5x–CgIntegrin. The recombinant plasmids pET-30a–CgDM9CP-5 and pMAL-c5x–CgIntegrin were transformed into E. coli BL21 (DE3)-pLysS (TransGen Biotech, Beijing, China), and the positive strains with recombinant plasmids were incubated in Luria–Bertani medium (containing 50 mg/ml kanamycin and 100 mg/ml ampicillin, respectively) at 37°C with shaking at 220 rpm. When the culture medium reached OD₆₀₀ of 0.4–0.6, the cells were incubated for an additional 4 h with the induction of isopropyl β-d-thiogalactoside (IPTG) at the final concentration of 0.25 mmol/l. The rCgDM9CP-5 was purified by a Ni²⁺-chelating Sepharose column (Sangon Biotech, Shanghai, China) and desalted by extensive dialysis according to the previous description. The rCgIntegrin with the maltose-binding protein (MBP) tag was purified via an amylose-Sepharose column (New England Biolabs, Ipswich, MA).

The PAMP binding assay was performed according to the previous description with some modification (8). Briefly, PAMPs including LPS from E. coli 0111:B4 (Sigma-Aldrich, L2630), PGN from S. aureus (Sigma-Aldrich, 77140), d-(+)-mannose from wood (Sigma-Aldrich, M2069), and polyinosinic-polycytidylic acid (poly(I:C)) (Sigma-Aldrich, P9582) were dissolved in carbonate-bicarbonate buffer (0.0159 g/l Na₂CO₃, 0.0294 g/l NaHCO₃ [pH 9.6]) at a final concentration of 200 mg/l. One hundred microliters of PAMP solution was added to the wells to coat 96-well microliter plates (Costar) at 4°C for 12 h. The wells were washed with TBST for 5 min three times and then blocked with 200 μl of 3% BSA (w/v) in TBS at 37°C for 1 h. The plate was washed three times, and 200 μl of rCgDM9CP-5 at different concentrations (20, 10, 5, 2.5, 1.25, 0.625, 0.3125, and 0 nM) was added to the wells. The tag protein (Trx-His) (20, 10, 5, 2.5, 1.25, 0.625, 0.3125, and 0 nM) expressed by the pET-32a empty vector was added to the wells as the negative group at the same time. After incubation at 37°C for 1 h, the plate was washed three times with TBST and the bound protein was detected immunochemically. The plate was first incubated with 100 μl of the anti-6× His tag mouse mAb (1:1000 dilution in 3% BSA, Sangon Biotech, Shanghai, China) at 37°C for 1 h, followed by washing four times with TBST, and then incubated with 100 μl of HRP-labeled goat anti-mouse IgG (Beyotime Biotechnology, Shanghai, China, 1:2000 dilution in 3% BSA) as a second Ab. After washing four times with TBST for 15 min, 100 μl of tetramethylbenzidine was added to each well and incubated at room temperature in the dark for 15 min. The reaction was stopped by adding 50 μl of 2 mol/l HCl per well. The plates were finally read by using a Tecan Infinite M1000 PRO absorbance microplate reader (Tecan, Männedorf, Switzerland) at 450 nm. The readings of the rCgDM9CP-5 group and rTrx-His group were recorded and the K_d was calculated using GraphPad Prism 8 with a nonlinear regression curve fit and a one-site binding model analysis as K_d = A_max × [L]/A – [L], where A is the absorbance at 450 nm and [L] is the concentration of the rCgDM9CP-5. Each experiment was repeated in triplicate, and the results were given as the mean of three individual measurements ± SEM (n = 3).

The microbial binding assay was modified from the previously reported method (33). Gram-positive bacteria (S. aureus), Gram-negative bacteria (E. coli and V. splendidus), and fungi (Pichia pastoris) were employed to examine the microbial-binding ability of rCgDM9CP-5. The cultured microorganisms were washed three times and dissolved in carbonate-bicarbonate buffer at a final concentration of 10⁸ CFU/ml. An aliquot of 200 μl of microbes was immobilized on 96-well plates by incubating at 4°C overnight. The next ELISA reaction to detect the microbial binding activity was conducted following the procedure of the PAMP binding assay.

To prepare polyclonal Ab for CgDM9CP-5 and CgIntegrin, rCgDM9CP-5 and rCgIntegrin, were used to immune to 6-wk-old mice, respectively, as previously described (34), and the specificity of Ab was examined by Western blot. Sequence alignment clearly demonstrated that the amino acid sequence of oyster phospho-P38 sites (MTGYVATRWY) was identical to that of human phospho-P38 sites, and the amino acid sequence of oyster phospho-JNK sites (TPY) was also conserved with human JNK phosphorylation sites. The Abs of human phospho-P38 (Cell Signaling Technology, Danvers, MA), human phospho-JNK (Cell Signaling Technology, Danvers, MA), human-P38 (Beyotime Biotechnology, Shanghai, China), human-JNK (Beyotime Biotechnology, Shanghai, China) and β-tubulin (ABclonal, Wuhan, China) were used in the Western blot. The phosphorylation level of CgJNK and CgP38 was determined by Western blot.

The total proteins from gill were separated by SDS-PAGE and transferred to a 0.22 μm pore nitrocellulose membrane. After blocking with 5% BSA (in TBST, 50 mmol/l Tris-HCl, 150 mmol/l NaCl, and 1% Tween 20) at 4°C for 12 h, the membrane was incubated with mouse anti-CgDM9CP-5 Ab (diluted 1:500 in blocking buffer) at 37°C for 3 h, and then washed three times with TBST. HRP-labeled goat anti-mouse IgG (Beyotime Biotechnology, Shanghai, China, diluted 1:1000 in blocking buffer) was further incubated with the membrane as the second Ab at 37°C for 2 h. After washing three times with TBST, the membrane was incubated in a Western Lightning ECL substrate system (Thermo Fisher Scientific, Waltham, MA) and exposed to an automatic chemiluminescence gel imaging system (Amersham Imager 600, GE Healthcare Life Sciences, Marlborough, MA).

Quantitative real-time PCR (qRT-PCR) with the primers of CgDM9CP-5 RT-F/-R and CgIntegrin RT-F/-R (Table I) was performed to detect the relative mRNA expression. The CgEF (GenBank accession no. NM_001305313) fragment amplified with primers CgEF-RT-F and -R (Table I) was employed as an internal reference. The real-time PCR assay was carried out in an ABI QuantStudio sequence detection system (Applied Biosystems, Waltham, MA) using SYBR Premix Ex Taq (Takara Bio, Shiga, Japan) according to the manufacturers’ instructions (35).

Dissociation curve analysis of amplification products was performed at the end of each PCR to confirm that only one PCR product was amplified and detected. The relative expression level was analyzed by the 2^−ΔΔCt method (36).

An immunofluorescence assay was used to measure the expression level of CgDM9CP-5 protein after V. splendidus stimulation according to the previous description (37). Briefly, gill tissues were fixed first and dehydrated in 75, 80, 95, and 100% successive ethanol. The tissue samples were embedded in paraffin, and sections were prepared by a manual rotary microtome (RM2235, Leica Biosystems, Nussloch, Germany). After the paraffin was removed from the sections, the Ags were refolded in sodium citrate/hydrochloric acid buffer. The sections were blocked with 3% BSA at room temperature for 1 h and incubated successively with primary Ab of mouse anti-rCgDM9CP-5 and the secondary Ab of Alexa Fluor 488–labeled goat anti-mouse IgG H+L (Abcam, Cambridge, U.K.) at 37°C for 1 h. The nucleus was stained with DAPI dihydrochloride (Beyotime Biotechnology, Shanghai, China). As a negative control, the primary Ab was replaced with mouse preimmune serum. After extensive washing, the sections were covered with cover slides and observed under a fluorescence microscope (Axio Imager A2, Zeiss, Jena, Germany).

The subcellular localization of CgDM9CP-5 in gill cells was determined by an immunocytochemistry assay as previously described with some modification (38, 39). The entire gill tissues were harvested from oysters. Tissues were minced and incubated with Pronase (20 μg/ml) in Hanks’ buffer containing no Ca²⁺ or Mg²⁺ with gentle shaking overnight at 4°C. Supernatant was filtered and debris was eliminated by several washes and centrifugations at low speed (100 × g) in Hanks’ buffer at 4°C for 10 min. Cells were collected and resuspended in L15 medium (with additional saline 20.2 g/l NaCl, 0.54 g/l KCl, 0.6 g/l CaCl₂, 1.0 g/l MgSO₄, 3.9 g/l MgCl₂), and then the suspension was deposited on the clean slides (a drop on each) in the wet chamber. After settled in monolayers on slides, the haemocytes were fixed with 4% paraformaldehyde for 1 h and blocked by incubation in 3% BSA (fetal BSA diluted in TBST) at 37°C for 1 h. The supernatant was removed and the dishes were incubated with 500 μl of anti-rCgDM9CP-5 (diluted 1:500 in blocking buffer) as the primary Ab at 37°C for 1 h. After washing three times with TBST, the dishes were incubated with Alexa Fluor 488–labeled goat anti-mouse IgG H+L (diluted 1:1000 in blocking buffer) as the second Ab at 37°C for 1 h. After washing another three times with TBST, DAPI (Solarbio Life Sciences, Beijing, China, diluted 1:1000 in TBS) and DiI (Beyotime Biotechnology, Shanghai, China, diluted 1:1000 in TBS) were added into the dishes to stain the nucleus and membrane, respectively. As a negative control, the primary Ab was replaced with mouse preimmune serum. After the last three times of washing, the dishes were mounted in buffered glycerin for observation by a fluorescence microscope (Axio Imager A2, Zeiss).

A pull-down assay between rCgDM9CP-5 (His tag) and rCgIntegrin (MBP tag) was performed as previously described (40). rCgDM9CP-5 (His tag) and rCgIntegrin (MBP tag) were added to 2 ml of charged Ni-NTA beads (for His-tagged proteins) and incubated at room temperature with slight rotation for 2 h. The mixture (resin and binding proteins) was washed three times with TBS by centrifugation at 500 × g for 5 min to remove the unbound proteins, and then the proteins from the test tissues were added into the mixture and incubated with slight rotation at room temperature for another 2 h. After the resin was washed three times, the bound proteins were separated by 12% SDS-PAGE. rTrx-His (His) and rMBP pull-down assays to rCgIntegrin (MBP tag) and rCgDM9CP-5 (His tag) were conducted as controls, respectively.

The bio-layer interferometry (BLI) assay was performed on an Octet K2 system (ForteBio, Fremont, CA) according to a previous report (41) with PBS with Tween 20 as the running buffer. Ligand-free rCgDM9CP-5 (1000 nM) was biotinylated using a biotinylation kit (Genemore, Suzhou, China) for 60 min and then immobilized onto a streptavidin biosensor (ForteBio, Fremont, CA) that had been equilibrated in the running buffer for 10 min. The streptavidin biosensor was transferred into PBS with Tween 20 buffer containing rCgIntegrin (MBP tag) at the indicated concentration (170.5, 341, 682, 1364 nM) (for association) or the analyte-free buffer (for dissociation). The rMBP with the same concentration gradient as rCgIntegrin was used as a control. The data were analyzed and the binding constant was determined using software provided by ForteBio. The association and dissociation data were used to calculate equilibrium K_D values based on a 1:1 binding model.

A coimmunoprecipitation (Co-IP) assay was performed to confirm the interaction between CgDM9CP-5 and CgIntegrin proteins as previously described (40). The gills from oysters were lysed in RIPA buffer and centrifuged at 12,000 × g for 10 min at 4°C. The supernatant (500 μl, 1 mg/ml) was incubated with 20 μl of protein A+G beads with gentle shaking at 4°C for 30 min to remove nonspecific binding. After centrifugation at 3000 × g for 5 min, the supernatant was incubated with 50 μl of anti-rCgDM9CP-5 or anti-rCgIntegrin at 4°C overnight. Protein A+G beads (30 μl) were then added to the mixture and incubation was continued at 4°C for 1 h. After centrifugation at 3000 × g for 5 min, the beads were collected and washed with TBS three times. Subsequently, the pellets were suspended in 80 μl of electrophoresis sample buffer and denatured at 99°C for 10 min, followed by SDS-PAGE and Western blotting using anti-CgDM9CP-5 or anti-CgIntegrin Ab. Normal mouse IgG was used as a control.

The DNA fragments of CgDM9CP-5, CgIntegrin, and enhanced GFP (EGFP) were amplified with the primer pairs of CgDM9CP-5-Fi and CgDM9CP-5-Ri, CgIntegrin-Fi and CgIntegrin-Ri, and EGFP-Fi and EGFP-Ri, respectively (Table I). The amplified fragments were used as templates to synthesize dsRNA using T7 polymerase (Takara Bio, Shiga, Japan) according to the manufacturer’s instructions (34). Thirty-six oysters were randomly divided into four groups with nine individuals in each group. The oysters received individually an injection of dsCgDM9CP-5, dsCgIntegrin, and dsEGFP (100 μl, 1 μg/ml) and were employed as the dsCgDM9CP-5, dsCgIntegrin, and dsEGFP groups, respectively. The oysters without any treatment were employed as the blank group. To enhance the RNA interference (RNAi) efficiency, the second injection was performed at 12 h after the first injection, followed by the third injection at 12 h after the second injection. The oysters in the RNAi groups received an injection of 100 μl of V. splendidus at 12 h after the third injection with dsRNA. The gills were collected at 12 h after the V. splendidus injection, and those from three individuals were pooled together as one sample. Total RNA was extracted, and the RNAi efficacy in the mRNA level was evaluated by qRT-PCR with the primer pairs CgDM9CP-5-RT-F/CgDM9CP-5-RT-R and CgIntegrin-RT-F/CgIntegrin-RT-R (Table I), respectively. Total protein was extracted and the RNAi efficacy in protein level was evaluated by Weston blot with mouse anti-CgDM9CP-5 Ab and mouse anti-CgIntegrin Ab (27).

The expression levels of CgIL-17s (CgIL-17-3, CgIL-17-4, CgIL-17-5, and CgIL-17-6) (42–44) and Cgdefensins (Cg-Defh1, Cg-Defh2, and CgMolluscidin) (45, 46) in the gill of dsCgDM9CP-5 and dsCgIntegrin groups were detected by qRT-PCR with specific primers (Table I) at 12 h after V. splendidus stimulation, with the dsEGFP group as the control. There were three replicates for each experiment.

The inhibitors of P38 (BIRB796, Beyotime Biotechnology, Shanghai, China) and JNK (SP600125, Beyotime Biotechnology, Shanghai, China) were used to inhibit the activations of CgP38 and CgJNK, respectively. The oysters in the P38 inhibitor and JNK inhibitor groups received individually an injection with 50 μl of BIRB796 (1 μg/μl diluted in SSW containing 1% DMSO) and SP600125 (1 μg/μl diluted in SSW containing 1% DMSO), respectively. The same volume of 1% DMSO was employed as a control. At 1 h after the inhibitor injection, the oysters were stimulated with injections of 100 μl of V. splendidus, and SSW was used as a control. The total RNA was extracted from oyster gills for qRT-PCR assays to examine the mRNA transcripts of CgIL-17s (CgIL-17-3, CgIL-17-4, CgIL-17-5, and CgIL-17-6) and Cgdefensins (Cg-Defh1, Cg-Defh2, and CgMolluscidin).

Anti-rCgIntegrin serum (30 μl, prepared as described in Preparation of polyclonal Ab) was injected into the oysters with the same volume of preserum as a control. One hour later, the CgIntegrin-blockage oysters received an injection with 100 μl of V. splendidus (10⁶ CFU/ml). The total RNA was extracted from gills, and the mRNA transcripts of CgIL-17s (CgIL-17-3, CgIL-17-4, CgIL-17-5, and CgIL-17-6) and Cgdefensins (Cg-Defh1, Cg-Defh2, and CgMolluscidin) were examined by qRT-PCR. The gill proteins were extracted from the treated oysters, and the phosphorylation level of CgJNK and CgP38 was determined by Western blot.

All data are shown as mean ± SD. The two-sample Student t test was performed for the comparisons between groups. Multiple group comparisons were executed by one-way ANOVA and followed by a Tukey multiple group comparison test using SPSS 26.0 software. Differences were considered significant at p < 0.05 and extremely significant at p < 0.01.

A pair of gene-specific primers, CgDM9CP-5-F and CgDM9CP-5-R (Table I), were designed to amplify a 438 bp fragment of CgDM9CP-5 from the cDNA library of C. gigas. The open reading frame of CgDM9CP-5 encoded a polypeptide of 145 aa with a predicted molecular mass of 16.14 kDa and theoretical isoelectric point of 8.21. SMART analysis revealed that there were two putative DM9 domains (from aa 3 to 69, and from aa 77 to 143) in the deduced amino acid sequence of CgDM9CP-5 (Fig. 1A). CgDM9CP-5 was predicated to exist as a unique homodimer in which each protomer was composed of two DM9 domains related by a pseudo 2-fold axis (Fig. 1B).

The deduced amino acid sequence of CgDM9CP-5 shared high similarities with other reported DM9CPs, such as 47.8% identity with Oefarnesoic acid O-methyl transferase (FAMeT) from Ostrea edulis (AFJ91732.1), 31.9% identity with ZnNatterin-3 form Zootermopsis nevadensis (KDR19911.1), 29.9% identity with AgPRS1 from Anopheles gambiae (XP_001688783.1), 27.4% identity with DmCG13321 from D. melanogaster (NP_610831.1), and 27.0% identity with FgDM9-1 from F. gigantic (AMP43490.1) (Fig. 1C, Table II). An unrooted phylogenetic tree was constructed with the deduced amino acid sequences of CgDM9CP-5 and other reported DM9CPs using neighbor-joining method with a 1000 bootstrap test. There were two distinct branches of vertebrate and invertebrate DM9CPs in the phylogenetic tree. CgDM9CP-5 was first clustered with CgDM9CP-1–4 from C. gigas to form a sister group with DM9CPs from invertebrates such as A. gambiae, D. ananassae, D. melanogaster, and F. gigantica, whereas the DM9CPs from fish Oreochromis niloticus, Takifugu rubripes, Thalassophryne nattereri, and Xiphophorus maculatus were grouped together (Fig. 1D).

The recombinant plasmid (pET-30a–CgDM9CP-5) was transformed into E. coli BL21 (DE3)-pLysS. After IPTG induction, the whole-cell lysate of E. coli BL21 (DE3)-pLysS with pET-30a–CgDM9CP-5 was analyzed by SDS-PAGE. A distinct band with a molecular mass of 17 kDa was observed, which was consistent with the predicted molecular mass of CgDM9CP-5 together with an EHHHHHH tag (∼1.08 kDa) (Fig. 2A).

The purified rCgDM9CP-5 protein was used to prepare polyclonal Ab for the following functional verification assay. The distinct single band with the similar molecular mass as CgDM9CP-5 revealed by Western blot indicated the high specificity and efficiency of the polyclonal Abs (Fig. 2B). In the negative control group, no visible band was observed with mouse preimmune serum (data not shown).

The 639 bp cDNA fragment encoding the amino acid sequence from A⁵⁵ to E²⁶⁶ from the extracellular part of CgIntegrin was amplified and inserted into pMALc5x. The recombinant plasmid (pMALc5x-CgIntegrin) was transformed into E. coli BL21 (DE3)-pLysS. After IPTG induction, the whole-cell lysate of E. coli BL21 (DE3)-pLysS with pMALc5x-CgIntegrin was analyzed by SDS-PAGE. A distinct band with a molecular mass of 67 kDa was observed, which was consistent with the predicted molecular mass of CgIntegrin together with an MBP tag (∼45 kDa) (Fig. 2C).

The binding activities of rCgDM9CP-5 toward different PAMPs, including d-mannose, LPS, PGN, and poly(I: C), were determined by ELISA. rCgDM9CP-5 was able to directly bind d-mannose, LPS, PGN, and poly(I:C) in a concentration-dependent manner with a saturable process from 0 to 20 nM (Fig. 2D). The apparent K_d of rCgDM9CP-5 toward d-mannose, LPS, PGN, and poly(I:C) was calculated from the saturation curve, which was 1.757, 678.3, 900.5, and 4.023 nM, respectively.

The in vitro microbial binding assays showed that rCgDM9CP-5 displayed activity to bind P. pastoris, E. coli, V. splendidus, and S. aureus in a concentration-dependent manner with a saturable process from 0 to 20 nM (Fig. 2E). The apparent K_d of rCgDM9CP-5 toward P. pastoris, E. coli, V. splendidus, and S. aureus was calculated from the saturation curve, which was 0.355, 1.146, 2.194, and 2.790 nM, respectively.

A SYBR Green real-time PCR assay was employed to examine the RNA expression of CgDM9CP-5. The mRNA transcripts could be detected in all of the tested tissues, including the hepatopancreas, mantle, gonad, adductor muscle, haemolymph, labial palp, and gill (Fig. 3A). The highest expression level was detected in gill, which was 4154.57-fold higher (p < 0.01) than that in hepatopancreas. The mRNA expression of CgDM9CP-5 in gill was further examined at 0, 6, 12, 24, 48, and 72 h after d-mannose or V. splendidus stimulation. It was significantly upregulated at 12 h (7.77-fold of that in the SSW group, p < 0.01), reached the peak level (14.01-fold, p < 0.01) at 24 h, and then gradually decreased at 72 h (1.76-fold, p < 0.01) after mannose stimulation (Fig. 3B). The relative expression level of CgDM9CP-5 mRNA was upregulated at 6 h after V. splendidus stimulation (13.58-fold, p < 0.01), and then decreased to 1.03-fold of that in the control group at 72 h (Fig. 3C).

The expression level of CgIntegrin mRNA in oyster gill was examined at 0, 6, 12, 24, 48, and 72 h after the stimulation with V. splendidus. The relative mRNA expression level of CgIntegrin was upregulated at 6 h (3.88-fold, p < 0.01), reached the peak level (14.62-fold, p < 0.01) at 24 h, and then decreased to 1.42-fold of that in the control group at 72 h after V. splendidus stimulation (Fig. 3D).

The distribution of CgDM9CP-5 protein in various tissues was examined using a Western blot assay. CgDM9CP-5 and β-tubulin proteins could be detected clearly in gonad, gill, mantle, adductor muscle, labial palp, and haemocytes (Fig. 4A). The CgDM9CP-5 band in gill was relatively thicker compared with that in other tissues, which was 2.07-fold (p < 0.01) of that in gonad calculated by ImageJ grayscale analysis. CgIntegrin was also detected in the gill tissue by a Western blot assay (Fig. 4B). Meanwhile, CgDM9CP-5 was detected in gill mucus (Fig. 4C) and mantle mucus (Fig. 4D). The relative abundance of CgDM9CP-5 in gill tissue was 1.44-fold (p < 0.01) of that in gill mucus, and the relative abundance in gill mucus was 3.72-fold (p < 0.01) of that in mantle mucus.

The positive green signal of CgDM9CP-5 was observed in gill tissue by an immunofluorescence assay. After V. splendidus stimulation, the relative fluorescence intensity of CgDM9CP-5 at 12 h was 3.74-fold of that at 0 h by ImageJ statistical analysis (Fig. 4E). As a control, no signal was observed in the preimmune serum group.

An immunocytochemistry assay was used to detect the subcellular localization of CgDM9CP-5 in gill cells of C. gigas. The membranes of gill cells were stained by DiI in red, and the nucleus was stained by DAPI in blue fluorescence. The positive signals of CgDM9CP-5 labeled by Alexa Fluor 488 were indicated with green fluorescence, which were mainly observed on the membrane of cells (Fig. 4F).

The interaction between rCgDM9CP-5 and rCgIntegrin was analyzed by a His pull-down assay. Two distinct bands were observed in the elute liquid after the pull-down assay (Fig. 5A). In the control group with rTrx-His as the bait protein to pull down rCgIntegrin, only the rTrx-His band was observed in the elute liquid (Fig. 5B). When rCgDM9CP-5 was used as the bait protein to pull down rMBP, only the rCgDM9CP-5 band was observed in the elute liquid (Fig. 5C).

Additionally, BLI was used to determine the direct interaction between the purified rCgDM9CP-5 and rCgIntegrin. Two-fold serial dilutions of rCgIntegrin (170.5, 341, 682, 1364 nM) were allowed to associate with the biotin-labeled rCgDM9CP-5. The rCgIntegrin was bound to biotin-labeled rCgDM9CP-5 in a dose-dependent manner (Fig. 5D). A global fit of the multiconcentration data yielded a K_D value of 54.7 nM, which showed the binding affinity of the rCgIntegrin–rCgDM9CP-5 interaction (Fig. 5D). As a control, there was no binding signal when rMBP reacted with the biotin-labeled rCgDM9CP-5 (Fig. 5E).

Coimmunoprecipitation (co-IP) was further performed to confirm the interaction between CgDM9CP-5 and CgIntegrin. The band of CgDM9CP-5 in gill proteins was coimmunoprecipitated by CgIntegrin. The band of CgIntegrin in gill proteins was also coimmunoprecipitated by CgDM9CP-5 (Fig. 5F). As a control, no bands of CgDM9CP-5 or CgIntegrin were detected in gill proteins coimmunoprecipitated with IgG (Fig. 5F).

The mRNA transcripts of CgDM9CP-5 and CgIntegrin in the gill of oysters after injection with dsCgDM9CP-5 or dsCgIntegrin were examined by qRT-PCR. The expression levels of CgDM9CP-5 and CgIntegrin mRNA were downregulated significantly to 0.14- and 0.20-fold of that in the dsEGFP group (p < 0.01) (Fig. 6A, 6C), respectively. Meanwhile, the expressions of CgDM9CP-5 and CgIntegrin proteins in gill were also downregulated significantly, which were 0.38- and 0.41-fold of that dsEGFP group (p < 0.01) (Fig. 6B, 6D).

After CgDM9CP-5 was knocked down, the mRNA expressions of CgIntegrin in gill decreased significantly at 12 h after V. splendidus stimulation, which was 0.12-fold (p < 0.01) (Fig. 6E) of that in the dsEGFP group. The protein expression level of CgIntegrin in the gill tissue of oysters after the injection of dsCgDM9CP-5 was also detected by a Western blot assay. The bands of CgIntegrin in the dsCgDM9CP-5 group became thinner after V. splendidus stimulations, and the count values of these bands were 0.46-fold (p < 0.01) of that in the dsEGFP group (Fig. 6F).

The phosphorylation level of P38 and JNK in the gill of oysters after V. splendidus stimulation was detected by Western blot. The intensity of Cgp-P38 and Cgp-JNK bands in the gills increased after V. splendidus stimulation (Supplemental Fig. 1A).

The phosphorylation of P38 and JNK in the gill of oysters after RNAi with dsRNA injection and V. splendidus stimulation was further detected. In dsCgDM9CP-5-injected or dsCgIntegrin-injected oysters, the band indicating Cgp-P38 in gill became weaker, which was 0.73-fold (p < 0.01) or 0.62-fold (p < 0.01) of that in the EGFP-RNAi group by ImageJ grayscale analysis (Fig. 7A, 7B, Supplemental Fig. 2), respectively. The band of CgP38 did not change significantly in each group. Meanwhile, the band of Cgp-JNK became weaker in the gill of dsCgDM9CP-5-injected or dsCgIntegrin-injected oysters, which was 0.49-fold (p < 0.01) and 0.68-fold (p < 0.01) of that in dsEGFP group by ImageJ grayscale analysis (Fig. 7A, 7B), respectively. The band of CgJNK did not change significantly in each group.

The mRNA expression of CgIL-17s and Cgdefensins in the gill of oysters from the blank group increased after V. splendidus stimulation (Supplemental Fig. 1B). After CgDM9CP-5 or CgIntegrin was knocked down, the mRNA expressions of CgIL-17s and Cgdefensins in gill changed significantly at 12 h after V. splendidus stimulation. The expression levels of CgIL-17-3, CgIL-17-4, CgIL-17-5, CgIL-17-6, Cg-Defh1, Cg-Defh2, and CgMolluscidin were 0.17-fold (p < 0.05), 0.14-fold (p < 0.01), 0.11-fold (p < 0.01), 0.05-fold (p < 0.01), 0.15-fold (p < 0.01), 0.01-fold (p < 0.01), and 0.09-fold (p < 0.01) of that in dsEGFP-injected oysters, respectively (Fig. 8A, 8B).

In dsCgIntegrin-injected oysters, the mRNA expressions of CgIL-17-3 (0.47-fold, p < 0.05), CgIL-17-4 (0.37-fold, p < 0.01), CgIL-17-5 (0.07-fold, p < 0.01), CgIL-17-6 (0.08-fold, p < 0.01), Cg-Defh1 (0.33-fold, p < 0.01), Cg-Defh2 (0.06-fold, p < 0.01), and CgMolluscidin (0.12-fold, p < 0.01) all decreased significantly at 12 h after V. splendidus stimulation compared with that in dsEGFP-injected oysters, respectively (Fig. 8C, 8D).

After CgIntegrin was blocked by the anti-CgIntegrin serum, the bands of phospho-CgP38 and phospho-CgJNK in gill proteins became weaker after V. splendidus stimulations, and the count values of these bands were 0.48-fold (p < 0.01) and 0.64-fold (p < 0.01) of those in the preserum-injected group, respectively (Fig. 7C).

In anti-CgIntegrin serum-injected oysters, the mRNA transcripts of CgIL-17-3, CgIL-17-4, CgIL-17-5, CgIL-17-6, Cg-Defh1, Cg-Defh2, and CgMolluscidin were downregulated significantly after V. splendidus stimulation, which were 0.35-, 0.60-, 0.42-, 0.247-, 0.02-, 0.17-, and 0.27-fold (p < 0.01) of those in the preserum-injected group, respectively (Fig. 8E, 8F).

The mRNA expressions of CgIL-17s (CgIL-17-3, CgIL-17-4, CgIL-17-5, and CgIL-17-6) and Cgdefensins (Cg-Defh1, Cg-Defh2, and CgMolluscidin) after the injection of the P38 inhibitor (BIRB796) were examined to evaluate the function of CgP38 in mediating CgIL-17s and Cgdefensins production. In P38 inhibitor (BIRB796)–injected oysters, the mRNA transcripts of CgIL-17-3, CgIL-17-4, CgIL-17-5, and CgIL-17-6 were downregulated significantly after V. splendidus stimulation, which were 0.04-, 0.13-, 0.41-, and 0.04-fold (p < 0.01) of that in the SSW-injected group, respectively (Fig. 9A). Similarly, the mRNA transcripts of Cg-Defh1, Cg-Defh2, and CgMolluscidin decreased significantly in P38 inhibitor (BIRB796)–injected oysters after V. splendidus stimulation, which were 0.02-fold (p < 0.01), 0.02-fold (p < 0.01), and 0.20-fold (p < 0.01) of that in the SSW-injected group, respectively (Fig. 9B)

SP600125 was used to inhibit JNK activity, and the mRNA expressions of CgIL-17s (CgIL-17-3, CgIL-17-4, CgIL-17-5, and CgIL-17-6) and Cgdefensins (Cg-Defh1, Cg-Defh2, and CgMolluscidin) were examined to evaluate the function of CgJNK in regulating CgIL-17s and Cgdefensins production. In JNK inhibitor (SP600125)–injected oysters, the mRNA transcripts of CgIL-17-3, CgIL-17-4, CgIL-17-5 CgIL-17-6, Cg-Defh1, Cg-Defh2, and CgMolluscidin were downregulated significantly after V. splendidus stimulation, which were 0.03-, 0.07-, 0.17-, 0.04-, 0.00-, 0.14-, and 0.35-fold (p < 0.01) of that in the SSW-injected group, respectively (Figs. 9C, 9D, 10).

DM9CP has been reported in several animal phyla but not in plants and mammals since it was first identified in D. melanogaster (9). It is a new type of PRR to recognize specific ligands, pathogenic bacteria, and mediate the immune responses, including phagocytosis (8), agglutination (13), encapsulation (14), hemagglutination (10), and other immune reactions (17).

There are two forms of DM9CP in domain architecture, one containing only DM9 domain repeats, and the other one containing DM9 domain and other functional domains. Most of the reported DM9CPs possess only DM9 domains. For example, AgPRS1 from A. gambiae (18), CgDM9CP-1, -2, -3, and -4 from C. gigas (8, 13–15), and FgDM9-1 and -2 from F. gigantica (10, 47) only contain two DM9 domains, whereas AgPfs47Rec from A. gambia (17) contains four DM9 domains. Some DM9CPs contain both DM9 and other domains, and most of them are reported in fish and Arthropoda. For instance, Natterin from fish contains the N-terminal DM9 domain and C-terminal Clostridium epsilon toxin/Bacillus mosquitocidal toxin (ETX/MTX2) domain (19, 20). FAMeT from the desert locust, Schistocerca gregaria, contains an N-terminal FAMeT domain and C-terminal DM9 domain (48, 49). Crystal structure analysis of CgDM9CP-1 revealed a unique homodimer in which each protomer was composed of two DM9 domains related by a pseudo 2-fold axis (8, 21). Meanwhile, seven β-sheets were identified in each DM9 domain in CgDM9CP-1. In the current study, the tertiary structure of CgDM9CP-5 was found to be a homodimer by homology modeling. It consisted of 14 β-sheets, and each DM9 domain was composed of 7 β-sheets. Sequence alignment and phylogenetic tree analysis revealed that CgDM9CP-5 shared higher similarity with the other DM9CPs identified from oysters (Fig. 1C), and it was clustered in the invertebrate branch in the unrooted phylogenetic tree (Fig. 1D).

DM9CPs have been reported to recognize pathogenic microorganisms and PAMPs in the first line of the immune system as PRRs. For instance, rCgDM9CP-1, -2, -3, and -4 were able to bind a variety of microorganisms including Gram-negative and Gram-positive bacteria and fungi (8). In the current study, rCgDM9CP-5 exhibited binding affinity to Gram-negative bacteria (V. splendidus, E. coli), Gram-positive bacteria (S. aureus), and fungi (P. pastoris), indicating that CgDM9CP-5 also functioned as a PRR to bind microorganisms. Meanwhile, rCgDM9CP-5 was also found to bind LPS, PGN, and d-mannose, which was similar to the binding activity of CgDM9CP-1 (8). As LPS, PGN, and mannose are the main components of the cell wall of Gram-negative bacteria, Gram-positive bacteria, and fungi, respectively, it is speculated that CgDM9CP-5 may recognize microorganisms by binding to LPS, PGN, and d-mannose. It was reported that the key amino acids of D22A and K43A in CgDM9CP-1 determined its binding activity to d-mannose, LPS, and PGN (8). The same amino acids of Asp²⁴ and Lys⁴³ were also identified in CgDM9CP-5, which might facilitate its binding activity to d-mannose, LPS, and PGN.

DM9CPs have been reported to distribute in important immune tissues. For example, AgPRS1 was found to be mainly distributed in the salivary glands and midgut (18), and FgDM9-1 was located in the parenchyma of the parasite with higher abundance in the insoluble crude extract (47). CgDM9CP-1 and CgDM9CP-4 were reported to be highly distributed in the haemocytes of oysters (8, 15), whereas CgDM9CP-2 and CgDM9CP-3 were mainly distributed in the gill tissue (13, 14). In the current study, CgDM9CP-5 mRNA was found to be highly expressed in the gill tissue of oysters, and the CgDM9CP-5 protein was mainly distributed in the membrane of gill cells. The expression level of CgDM9CP-5 in gill mucus was higher than that in mantle mucus. These results indicated that DM9CP-5 might play an important role in gill mucus and gill cells. Furthermore, accumulating evidence has indicated that DM9CPs could be induced by the stimulations of pathogens and PAMPs. For example, the expressions of DM9CP (CG16775) in D. melanogaster and AgPRS1 in the midgut and salivary gland of A. gambia were upregulated postinfection (16, 18). In oysters, the expression levels of CgDM9CP-2 and -3 in gill tissue were upregulated after V. splendidus or mannose stimulation (13, 14). It is speculated that CgDM9CP-5 is a secreted recognition receptor that plays significant roles in gill mucus, which recognizes extracellular V. splendidus and then transmits the signals into cells. A β integrin was previously identified from C. gigas with LPS binding activity, whose expression can be induced by LPS rather than PGN stimulation (27). In the current study, CgDM9CP-5 and CgIntegrin were found to be expressed in the gills, and their expressions were upregulated after V. splendidus stimulation. At the same time, pull-down and BLI assays revealed that CgDM9CP-5 was able to bind CgIntegrin in vitro. A Co-IP experiment further verified the interaction between CgDM9CP-5 and CgIntegrin in vivo. The results suggested that CgDM9CP-5 was able to recognize and bind pathogenic bacteria in the gill mucus, and then regulated the subsequent immune response by interacting with CgIntegrin in gill tissue.

It has been reported that DM9CPs are able to mediate various downstream immune response processes. Three DM9CPs (CG3884, CG10527, and CG13321) from D. melanogaster were found to interact with other proteins to form functional complexes involved in phagocytosis of microbial pathogens (22). CgDM9CP-1 in oysters could be translocated into the cytoplasm and colocalized with the engulfed microbes during haemocyte phagocytosis (8). In F. gigantica, FgDM9-1 could agglutinate Gram-positive and Gram-negative bacteria, and it caused hemagglutination across all ABO blood group phenotypes (10). In the current study, CgDM9CP-5 was also found to modulate the release of immune effectors. When the expressions of CgDM9CP-5 or CgIntegrin were interfered with RNAi, the expression levels of CgIL-17s (CgIL-17-3, -4, -5, and -6) and Cgdefensins (Cg-Defh1, Cg-Defh2, and CgMolluscidin) were downregulated after V. splendidus stimulation. At the same time, the expression of Integrin decreased when the expression of CgDM9CP-5 was interfered with RNAi. These results indicated that CgDM9CP-5 was able to interact with CgIntegrin to mediate the release of CgIL-17s and Cgdefensins. Usually, the classic PRRs, such as Dectin-1 and Dectin-2, are able to induce the production of cytokines such as IL-1b, IL-6, IL-17, IL-23, and TNF-α by recruiting Syk and triggering the downstream NF-κB and MAPK pathways (50–52). For example, Dectin-1 triggers the release of macrophage IL-6 and TNF through coactivation of their downstream Syk-JNK-AP-1 signaling complex in cooperation with CR3 (53). Lectin from Rhizoctonia bataticola induces cytokine production in human PBMCs and Th1/Th2 cells via the p38 MAPK and STAT-5 signaling pathways (54). TLR4 promotes IL-6 expression in liver tissue by activating the JNK signaling pathway (55). CgCLec-HTM is associated with CgSyk to transfer immune signals into the intracellular ERK-Rel pathway to induce CgIL-17 and CgTNF production (56). In the current study, when the expressions of CgDM9CP-5 and CgIntegrin were interfered respectively by RNAi, the phosphorylation of kinases P38 and JNK of the MAPK pathway was also reduced significantly. In addition, when CgIntegrin was blocked by anti-CgIntegrin Ab, the phosphorylation levels of MAPK kinases P38 and JNK decreased, and the expression levels of CgIL-17s and Cgdefensins also decreased significantly. It was further confirmed that CgIntegrin could modulate MAPK signaling and the expression of CgIL-17s and Cgdefensins. These results indicated that CgDM9CP-5 was able to interact with CgIntegrin to regulate the release of proinflammatory factors and defensins through the MAPK pathway.

In conclusion, a DM9CP (CgDM9CP-5) was identified as a PRR, which could be significantly induced in gill tissue and gill mucus of C. gigas by the stimulations of V. splendidus and d-mannose. It exhibited activity to bind LPS, d-mannose, PGN, poly(I:C), as well as various microbes. CgDM9CP-5 was able to interact with CgIntegrin in vitro and in vivo. When CgDM9CP-5 and CgIntegrin were knocked down respectively, the phosphorylation levels of JNK and P38 in the MAPK pathway, as well as the expressions of proinflammatory factors and defensins, were downregulated. The expression of Integrin decreased when the expression of CgDM9CP-5 was interfered with by RNAi. Moreover, the phosphorylation levels of JNK and P38 and the expressions of CgIL-17s and Cgdefensins all decreased after the CgIntegrin was blocked by CgIntegrin Abs. After P38 and JNK were inhibited respectively by their inhibitors, the expressions of CgIL-17s and Cgdefensins were downregulated. These results indicated that CgDM9CP-5 mediated the CgDM9CP-5-Integrin-MAPK pathway to regulate the release of proinflammatory factors and defensins from the gill tissue in oysters (Fig. 10).

We are grateful to all of the laboratory members for technical advice and helpful discussions.

The authors have no financial conflicts of interest.