This study was brought into force in severe accordance with the ethical criterion and the objective of the “Regulations for the Administration of Affairs Concerning Experimental Animals” proclaimed by the State Science and Technology Commission of Shandong Province. This study was also allowed by the Committee of the Ethics on Animal Care and Experiments at the Ocean University of China. The flounders were anesthetized with ethyl 3-amino-benzoate-methanesulfonic acid (MS222) before killing prior the experiment.

Flounders (800 ± 40 g) were purchased from an aquafarm in Rizhao, Shandong Province, PR China. The fishes were communally reared in a pond including oxygen-rich and filtered seawater at 20.0 ± 1.0°C for seven days before the experiment.

The pathogenic E. tarda isolated from diseased flounder were cultured and inactivated by formalin as described before (26). Next the inactivated bacteria were adjusted to 1.0 × 10⁸ CFU/ml with 0.1 M phosphate buffered saline (PBS), and then mixed with Freund’s complete adjuvant (1:1, V/V). Each flounder was intraperitoneally injected with 200 μl emulsified E. tarda, and injected with the same volume of sterile PBS which was administrated as the negative control. After 2 weeks, the inactivated-E. tarda emulsified with Freund’s incomplete adjuvant (1:1, V/V) was given as a booster immunization. Blood samples were collected from the tail vein 35 days after the first immunization, and allowed to clot at room temperature for 1 h, then centrifuged at 4 °C at 8,000g for 15 min to obtain the anti-E. tarda serum.

In order to produce polyclonal antibodies against flounder IgM, the flounder IgM was purified from serum according to the procedure described previously with some modifications (27, 28). Briefly, the crude extract of IgM was isolated from the serum by salting out with 50% saturated ammonium sulfate and then purified using HiTrap^® Protein A column (GE Healthcare) by protein purification system (AKTA prime, Amersham). The purity of IgM was examined by SDS-PAGE, and its concentration was determined using the Bradford method and adjusted to 1.0 mg/ml using PBS. New Zealand white rabbits were then immunized with purified IgM to produce polyclonal antibodies according to previous procedure (29).

We first tested the characteristics of the prepared flounder anti-E. tarda serum. Briefly, wells of flat-bottom microplates (96-wells, Costar) were coated overnight with 100 μl/well of 1.0 × 10⁸ CFU/ml E. tarda at 4°C. The wells were washed three times with 0.1 M PBS, 0.1% Tween 20, and pH 7.4 (PBST) and then blocked with 3% BSA in PBS for 1 h at 37°C. After washing, the serum (1:100 diluted in PBS) sampled from vaccinated flounder was added 100 μl per well and incubated for 2 h at 28°C. Following washing, each well was added with 100 μl monoclonal antibody (mAb) 2D8 against flounder IgM (1:2,000 diluted in PBS), which was previously prepared by our lab (28). After 1 h incubation at 37°C, the goat-anti-mouse Ig-alkaline phosphatase conjugate (Sigma) diluted 1:5,000 in PBS was added and incubated for 45 min at 37°C. After the last washing, 100 μl 0.1% (w/v) p-nitrophenyl phosphate (pNPP, Sigma) in 50 mM carbonate-bicarbonate buffer (pH 9.8) containing 0.5 mM MgCl₂ was used for color development for 30 min at 25°C before being measured by an automatic enzyme-linked immunosorbent assay (ELISA) reader (TECAN, Männedorf, Switzerland). For indirect fluorescence assay (IFA), the diluted E. tarda were settled onto glass slides for 2 h and fixed with 4% (V/V) paraformaldehyde for 20 min at 20°C, followed by the prepared antiserum which was added to glass slides and incubated for 2 h at 28°C; the healthy flounder sera were used as control. Then mAbs 2D8 (1:2,000) was added after three washings with PBST for 1 h at 37°C. Next the slides were incubated with Alexa Fluor^® 649 goat anti-mouse IgG (Invitrogen, Molecular Probes) for 45 min at 37°C. After the last washing, the slides were fixed with 90% glycerin before observation under a fluorescence microscope (Invitrogen™ EVOS™ FL Auto 2). Next, the prepared flounder anti-E. tarda serum was used for the opsonization of the E. tarda.

Meanwhile, the properties of the prepared rabbit anti-IgM polyclonal antibodies were also characterized. In brief, for western blotting, the flounder serum was separated by SDS-PAGE and transferred to PVDF membrane (Merck Millipore, Germany). The membrane was blocked with PBS containing 3% bovine serum albumin (BSA), and then incubated with the prepared rabbit anti-IgM polyclonal antibodies (diluted at 1:2,000 in PBS) at 37°C for 1 h and healthy rabbit sera were used as negative control. After washing thrice with PBST, goat-anti-rabbit IgG-alkaline phosphatase conjugate (Merck Millipore) was used to detect antibodies binding at 37°C for 1 h. After washing, the reactive bands were visualized for 5 min in the substrate solution (100 mM NaCl,100 mM Tris, 5 mM MgCl₂, pH 9.5) containing NBT (Sigma) and BCIP (Sigma). The prepared rabbit anti-IgM polyclonal antibodies were used for blocking experiments later.

Complement in the prepared sera from healthy and vaccinated flounder were inactivated by water bath at 56°C for 30 min (30). The E. tarda was adjusted to the concentration of 1.0 × 10¹⁰ CFU/ml, and then incubated with complement-inactivated sera at a volume ratio of 1:10 for 5 h with gentle shaking at 28°C. After centrifugation at 10,000g for 5 min, the opsonized E. tarda was obtained and diluted with PBS to 1.0 × 10⁸ CFU/ml. Next, the healthy flounder sera and antiserum-opsonized E. tarda were labeled with fluorescein isothiocyanate (FITC) in preparation for subsequent phagocytosis experiments (5).

The leukocytes isolated by Percoll gradient from the three flounders were handled as three individual samples, which were then counted and diluted to 1.0 × 10⁷ cells/ml with L-15 medium. The 1,000 μl cell suspensions were incubated with antiserum-opsonized E. tarda for 1, 3 or 5 h at 22°C in 24-well culture plates, while the healthy flounder serum-opsonized E. tarda was used as negative control. The ratios of cells/bacteria were all set as 1:40. Next the non-adherent cells were collected and mixed with adherent cells loosened by the means of trypsin-EDTA. Non-ingested bacteria were removed by glucose cushion (3 ml PBS, pH 7.3, with 3% (W/V) BSA and 4.5% (W/V) D-glucose) and centrifuged at 100g for 10 min at 4°C. After ingestion, the isolated cells were suspended in 1,000 μl PBS containing 5% (V/V) newborn calf serum (NCS), followed by incubation with anti-IgM mAbs as primary antibody. Washing with PBS containing 5% (V/V) NCS, the cells were labeled with Alexa Fluor^® 649 goat anti-mouse IgG (Invitrogen, Molecular Probes). After the last washing, the phagocytosis rates of mIgM⁺ B lymphocyte were measured by a BD Accuri C6 Flow Cytometer (Becton Dickinson. USA). For determining cellular reactive oxygen species (ROS), mIgM⁺ B lymphocyte was stained by ROS Assay Kit (Beyotime) after being phagocytosed by antiserum-opsonized E. tarda for 2 h at 22°C, where healthy flounder serum-opsonized E. tarda was used as control.

In order to reversely illustrate the function of Fc fragment of flounder IgM in antibody-dependent cellular phagocytosis (ADCP), we used prepared rabbit anti-IgM antibodies to block the Fc fragment on the surface of antiserum-opsonized E. tarda. Briefly, the rabbit anti-IgM polyclonal antibodies (diluted in 1:1,000) were added to the flounder antiserum-opsonized E. tarda (1.0 × 10¹⁰ CFU/ml) then they reacted with each other at an incubator temperature of 37°C for 12 h, where the healthy flounder serum-opsonized E. tarda was used as negative control. Next the treated E. tarda was acquired by centrifugation at 10,000g for 5 min and adjusted to 1.0 × 10⁸ CFU/ml. The experimental and measuring methods of phagocytosis rates and production of cellular ROS in mIgM⁺ B lymphocyte were the same as the previous description. Finally, phagocytosis of E. tarda treated the mIgM⁺ B lymphocytes in different ways were observed by fluorescence microscopy (Olympus DP70, Tokyo, Japan) as described above.

The detailed methods for sorting mIgM⁺ B lymphocytes from peripheral blood were as previously described (7). The purity of mIgM⁺ B lymphocytes before and after sorting was observed by flow cytometry and also fluorescence microscopy. First, total RNA was extracted from selected mIgM⁺ B lymphocytes using the RNeasy Plus Mini kit (Qiagen, Germany), according to the manufacturer’s instructions. RNA quality was analyzed by Bioanalyzer 2100 (Agilent Technologies, USA). All the primers used in this study are listed in Table 1 . The specificity of the primers was confirmed by melting curves and the sequencing of amplicons. Real-time quantitative PCR (qRT-PCR) was performed in LightCycler^®480 II Real Time System (Roche, Switzerland) by SYBR GreenIMaster (Roche, Switzerland). As one of the most conserved genes in vertebrates, our previous studies have shown that 18s rRNA was the least affected by various stimuli in flounder, so 18S rRNA gene was used as the internal reference (7, 31). All data were associated with 18S rRNA gene and analyzed by 2^−ΔΔCt method. The differences between the treatment and the control group were analyzed to assess changes in gene expression. Syk, FcγRII, and FcγRIII mRNA levels in mIgM⁺ B lymphocyte were examined by qRT-PCR after incubation with antiserum-opsonized E. tarda for 2 h at 22°C, and the healthy flounder serum-opsonized E. tarda was used as negative control, as mentioned in RNA isolation. For blocking assay, the Fc fragment of IgM on the E. tarda was blocked by the prepared anti-IgM polyclonal antibodies, then the expression level of the three genes in mIgM⁺ B lymphocytes were also detected by qRT-PCR after incubation with antiserum-opsonized E. tarda for 2 h at 22°C, while the healthy flounder serum-opsonized E. tarda was used as negative control, as mentioned in RNA isolation.

Based on the genome sequence of flounder FcγRII (Genbank No. XM_020105965.1), specific primers (F: 5’-CGCGGATCCATGGGCCCAACGTGCAA-3’, BamH I; R: 5’-CCCAAGCTTTCATTGGTCGAACCTCAGAGAC-3’, Hind III) were designed to amplify the target gene. The recombinant FcγRII (rFcγRII) protein and the rabbit anti-rFcγRII polyclonal antibody were prepared and characterized as previously described (32). We treated mIgM⁺ B lymphocytes with prepared anti-rFcγRII polyclonal antibody to block FcγRII on their membrane surface. Briefly, Percoll gradient isolated peripheral blood leukocytes (PBLs) were adjusted to 1.0 × 10⁷ cell/ml with PBS, and incubated with anti-rFcγRII antibodies (20:1, V/V) for 1 h to block the FcγRII. Then, the antiserum-opsonized E. tarda were incubated with the blocked leukocytes for 3 h. The unblocked leukocytes were used as control. After washing, mIgM⁺ B lymphocytes were labeled with prepared anti-IgM mAb and Alexa Fluor^® 649 goat anti-mouse IgG, and their phagocytosis rates were analyzed by flow cytometry. The FcγRII-blocked mIgM⁺ B lymphocytes were sorted after phagocytosis as previously described, and the ADCP-related and phagolysosome-related genes in them were detected by qRT-PCR.

Statistical analysis was carried out by Statistical Product and Service Solution (SPSS) software (version 20.0; IBM Corp., USA), and statistical significance was measured by independent-sample t-tests. The results are shown as mean ± S.D., and differences were regarded as significant at P <0.05.

According to SDS-PAGE, the purified IgM had two main bands with molecular weights of approximately 72.0 kDa and 26.0 kDa, corresponding to the predicted heavy and light chains of flounder IgM, respectively, and almost didn’t have other protein bands ( Figure 1A , lane 1). ELISA results showed that the titers of the flounder antiserum and rabbit anti-IgM polyclonal antibodies were 1/80 and 1/160,000, respectively (data not shown). Western blotting showed that the prepared rabbit anti-IgM polyclonal antibodies were able to specifically recognize the heavy and light chains of flounder sera IgM ( Figure 1A , lane 3). IFA results showed that the surfaces of all E. tarda bacteria showed strong green fluorescence signal, indicating successful FITC labeling. Furthermore, the flounder antiserum could fully adhere to the surface of E. tarda, while no specific binding was observed using the healthy flounder sera ( Figure 1B ).

Flow cytometry showed the phagocytosis rates of antiserum-opsonized E. tarda in peripheral blood mIgM⁺ B lymphocytes at 1, 3 or 5 h were 51.1 ± 0.8%, 63.0 ± 0.4%, and 77.5 ± 0.9%, respectively, which were significantly higher than that of control groups with 40.2 ± 0.7%, 50.9 ± 0.6%, and 63.8 ± 0.8%, respectively ( Figure 2 ). After the Fc fragment of opsonizing IgM on the surface of E. tarda was blocked with prepared antibodies, the phagocytosis rates of opsonized E. tarda in peripheral blood mIgM⁺ B lymphocytes at 1, 2 or 3 h were 31.3 ± 1.0%, 39.9 ± 0.8%, and 49.1 ± 0.6%, respectively, which were significantly lower than that of unblocked groups with 45.8 ± 0.7%, 51.9 ± 0.5%, and 69.2 ± 0.6%, respectively ( Figure 3 ). Fluorescence microscopy showed that the average number of bacteria within each mIgM⁺ B lymphocyte was greater than 5 after phagocytosis of antiserum-opsonized E. tarda, while the control group was less than 5, and the proportion of phagocytosed mIgM⁺ (pha⁺/mIgM⁺) B lymphocytes was also significantly higher than that of the control group. When the Fc fragment of osponizing IgM on the surface of E. tarda was blocked, the number of phagocytosed E. tarda in mIgM⁺ B lymphocyte was significantly lower than that in the opsonization group, and the proportion of pha⁺/mIgM⁺ B lymphocytes was also obviously decreased compared with that of the opsonization group ( Figure 4 ).

Flow cytometry showed that 32.1 ± 0.7% of mIgM⁺ B lymphocytes had higher intracellular ROS levels after phagocytosis of antiserum-opsonized E. tarda, which were significantly higher than that of control group with 23.0 ± 0.5%. In addition, when the Fc fragment of opsonizing IgM on the surface of E. tarda was blocked, 14.5 ± 0.4% of mIgM⁺ B lymphocytes had high intracellular ROS levels after phagocytosing the treated E. tarda, which were significantly lower than that of opsonization group with 23.9 ± 0.6% ( Figure 5 ).

The purity of the magnetic bead sorted and unsorted mIgM+ B lymphocytes form peripheral blood leucocytes was detected by flow cytometry to be 98.4% and 21.4%, respectively, which was fairly consistent with the observation by fluorescence microscopy, and could be used for subsequent experiment ( Figures 6A, B ). The qRT-PCR results showed that the expression levels of FcγRII and Syk in mIgM⁺ B lymphocytes were progressively upregulated with the prolongation of phagocytosis time in the opsonization group, whereas FcγRIII was progressively downregulated ( Figure 6C ). When the Fc fragment of opsonizing IgM on the surface of E. tarda was blocked, the expression levels of FcγRII, FcγRIII, and Syk in mIgM⁺ B lymphocytes did not vary with phagocytosis time ( Figure 6D ).

Flounder FcγRII protein consisted of a low complexity (residues 9–19), two IG domains (residues 29–109, 117–191), and a transmembrane region (residues 202–224) ( Figure 7A ). The FcγRII was expressed in E. coli (DE3) using the pET-32a plasmid system and SDS-PAGE analysis showed a distinct 49.6 kDa band after isopropyl β-D-1-thiogalactoside (IPTG) induction ( Figure 7B , lane 2). After performing Ni²⁺ affinity chromatography purification and stepwise dialysis refolding methods, high purity rFcγRII protein was obtained ( Figure 7B , lane 3). Moreover, western blotting showed that the prepared rabbit anti-rFcγRII antibodies could recognize a 31.7 kDa molecule of PBLs, which was consistent with the theoretical molecular weight of flounder FcγRII ( Figure 7B , lane 4).

Fluorescence microscopy showed that flounder FcγRII was expressed on mIgM⁺ B lymphocytes membranes and also on other types of mIgM^- leukocytes membranes ( Figure 8A ). When FcγRII of mIgM⁺ B lymphocytes was blocked, the phagocytosis rate of antiserum-opsonized E. tarda was 39.0 ± 0.5%, which was significantly lower than that of unblocked group with 54.0 ± 0.8% ( Figure 8B ). After FcγRII blocking, the qRT-PCR results showed that compared with the unblocked group, the expression levels of FcγRII, Syk, and phagolysosome-related genes were found to be significantly downregulated, whereas FcγRIII showed no significant differences in mIgM⁺ B lymphocytes post phagocytosis ( Figure 9 ).