Healthy penaeid shrimps (L. vannamei) with approximate size of 10–14 cm in length and 15–20 g in weight from Shantou Huaxun Aquatic Product Corporation were obtained and reared in 25-L seawater tanks at 25°C. Air was continuously supplied using an electric pump.

Hemolymph was taken directly from the pericardial sinus using a sterile tube and then allowed to clot overnight at 4°C. The serum was separated after centrifuging at 3,000 g for 20 min and kept at −20°C until analysis. All animal experiments were conducted in accordance with the recommendations set forth in the Animal Ethics Procedures and Guidelines of the People’s Republic of China.

HMC purification was performed by gel-filtration chromatography and anion-exchange chromatography as previously described with some modifications (5). Briefly, 2 ml of L. vannamei serum was loaded onto a Sephadex G-100 column, and then, the column was washed with Tris–HCl buffer (0.05 M, pH 8.0) at a flowrate of 1 ml/min until absorbance at 280 nm reached baseline. The eluted proteins were concentrated by ultrafiltration centrifuge tube. Then, the concentrated proteins (8 ml) were loaded onto a DEAE-cellulose (DE-52) column, equilibrated with 0.01 M pH 7.5 phosphate-buffered saline (PBS) buffer. Elution was performed with 0.2 M NaCl at a flowrate of 1 ml/min. Eluted proteins from the main peak (potential HMC) was determined by a modified Bradford assay (Bio-Rad, USA) and stored at −20°C until analysis.

Identification of the potential HMC was carried out using one-dimensional sodium sulfate–polyacrylamide gel electrophoresis (1-DE) and its immunoblotting. In brief, 1-DE was performed using a 5% stacking gel (pH 6.8) and a 10% separating gel (pH 8.9) in Tris–glycine buffer (pH 8.3). Then, the gel was transferred to a polyvinylidene fluoride (PVDF) membrane with a semi-dry transfer apparatus according to the manufacturer’s instructions. The membrane was blocked for 1 h with 5% skim milk in Tris-buffered saline (TBS) (20 mM Tris, 0.15 M NaCl, pH 7.4) at room temperature, then incubated with rabbit anti-shrimp HMC antisera (1:1,000 dilution) and goat anti-rabbit immunoglobulin G (IgG)–horseradish peroxidase (HRP) (1:3,000 dilution) antibodies at room temperature for 40 min and 1 h, respectively. Finally, the membrane was washed and developed with substrate (3′3-diminobenzidine, DAB) until optimum color developed.

To investigate the diversity of HMC, two-dimensional polyacrylamide gel electrophoresis (2-DE) and its immunoblotting were performed as our previous descriptions with some modifications (36). Briefly, a total of 30 µg of HMC in rehydration buffer [containing 7 M urea, 2 M thiourea, 4% CHAPS, 65 mM dithiothreitol (DTT), 0.2% bio-lyte, and 0.001% bromophenol blue, pH 3–10] was used to rehydrate the IPG strip (7 cm, pH 4.7–5.9 Bio-Rad, Hercules, CA) for 12 h. The isoelectric focusing (IEF) was performed at a constant temperature of 20°C using a continuous increase in voltage (up to 4,000 V) until reaching 35,000 Vh. Prior to the second dimension, the focused IPG was incubated for 15 min in an equilibration buffer containing 20% w/v glycerol, 2% sodium dodecyl sulfate (SDS), 0.375 M Tris–HCl (pH 8.8), 2% DTT, then further equilibrated for 15 min in a similar buffer in which 2% DTT was replaced with 2.5% of iodoacetamide. The strip was placed onto the top of a 12% 1-DE gel. Low-melting point agarose was used to cover the IPG strip and filter paper. Separation of proteins was carried out in the same conditions as described above for 1-DE. Following 2-DE, immunoblotting analysis was carried out as the same descriptions in Section 2.3 for the identification of HMC. Gels were stained with colloidal Coomassie, and spot volumes were compared using the image analysis software PDQuest Analysis Software (Bio-Rad).

To identify the proteins reacted with rabbit anti-shrimp HMC antibodies in the 2-DE map, matrix-assisted laser desorption ionization–time of flight–time of flight (MALDI-TOF-TOF) mass spectrometry analysis was further performed as our previous descriptions (26). The spots were excised from 2-DE gels, and the gel plug was digested with trypsin; then, 0.5 µl of the peptide mixture was mixed with the matrix a-cyano-4-hydroxycinnamic acid (1:1) and spotted onto a stainless steel MALDI plate. MS spectra were obtained using the ABI 4700 Proteomics Analyzer MALDI-TOF-TOF mass spectrometer (Applied Biosystems, Foster City, CA) operating in a result-dependent acquisition mode. Peptide mass maps were acquired in reflection mode (1-keV accelerating voltage) with 1,000 laser shots per spectrum. Six external standards (mass standard kit for the 4700 Proteomics Analyzer calibration mixture, Part Number 4333604, Applied Biosystems, Foster City, CA) were used to calibrate each spectrum to a mass accuracy within 50 ppm. Selected peptide masses were submitted to Mascot (http://www.matrixscience.com/cgi/search_form.pl?FORMVER=2&SEARCH=PMF) for NCBInr databases search.

For the pathogens challenge tests, two representative bacterial, namely, Streptococcus agalactiae (Gram-positive bacterium) and Vibrio parahaemolyticus (Gram-negative bacterium) were selected. The shrimps were inoculated intramuscularly by using 1-ml syringes in the second abdominal segment with 50 μl of bacterial inoculum (10⁷–10⁸ CFU/ml). The treated animals were then returned to the tanks at room temperature. Hemolymph was collected at 0, 6, 12, and 24 hpi. The following procedures for serum preparation and 2-DE analysis were performed as described in Sections 2.1 and 2.4.

To purify the HMC fractions binding with pathogens, bacterial pull-down assay was performed as described previously (37). In brief, five kinds of strains including S. agalactiae, V. parahaemolyticus, Vibrio alginolyticus, Vibrio fluvialis, and Escherichia coli K12 were used for incubation with L. vannamei serum in vitro. The bacterial cells were harvested at OD₆₀₀ of 1.0 and washed three times with 0.9% (w/v) saline. Then, the supernatant was boiled for 10 min and harvested by centrifugation at 5,000 g for 10 min. Three hundred microliters of serum was mixed with 300 µl bacterial solution (1.0 × 10⁹ CFU/ml), incubated at room temperature for 2 h, and the sample was centrifuged at 5,000 g for 10 min at 4°C. The pellet was washed three times with 0.9% saline and then resuspended in 1 M Tris–HCl (pH 8.0) containing 0.8 M NaCl at 37°C for 2 h. The supernatant incubated with S. agalactiae, V. parahaemolyticus, V. alginolyticus, V. fluvialis, and E. coli K12, and mixed five bacteria were collected and named as HMC-Sa, HMC-Vp, HMC-Va, HMC-Vf, HMC-Ec, and HMC-Mix, respectively. After determination with a modified Bradford assay (Bio-Rad, USA), these HMC isomers were identified by 1-DE, 2-DE, and their immunoblotting as described in Sections 2.3 and 2.4.

Agglutinative activities of six kinds of pathogens-binding HMC fractions, namely, HMC-Sa, HMC-Vp, HMC-Va, HMC-Vf, HMC-Ec, and HMC-Mix, were performed as our previous descriptions (38). Briefly, these bacteria were cultured in broth medium or Luria–Bertani medium overnight at 28 or 37°C (E. coli K12). The cells were harvested, washed, and diluted to 10⁸ CFU/ml in TBS-Ca²⁺ (0.05 M Tris, 0.75% NaCl, 0.05 M CaCl₂). Agglutination tests using the five bacteria and six specimens were performed at 37°C for 30 min. The proteins were twofold diluted in TBS-Ca²⁺ and added into 10 μl of each bacterium. Agglutination was observed in a light microscope and scored as positive (+) or negative (−) compared to a control placing the corresponding bacteria only in the TBS-Ca²⁺ buffer. Agglutinative titer was defined as the highest dilution of the test samples when the agglutination was appeared.

Antibacterial activity was assessed by the numbers of bacterial colonies grown on the Petri dish according to our previous descriptions (5). V. alginolyticus and V. fluvialis were cultured in broth medium at 28°C for 12–24 h, then diluted to 10³–10⁴ CFU/ml with sterile PBS (0.01 M, pH 7.4). One hundred microliters of bacterial suspension was taken out and mixed with an equal volume of filtered HMC-Sa, HMC-Vp, HMC-Va, HMC-Vf, HMC-Ec, and HMC-Mix (experimental groups) or sterile 0.01 M pH 7.4 PBS (control group) at 37°C for 2 h. Fifty microliters of mixed solution was taken out and incubated on solid broth medium by the spread plate method at 37°C for 12 h, then counted the colony of the experiment group (A₁) and control group (A₀). The antibacterial activity was calculated as follows: antibacterial rate = (A₀ − A₁)/A₀ × 100%. All samples were prepared in triplicate, and digital photomicrographs were taken with an Olympus BH-2 microscope.

Lectin-blotting analysis of HMC purified by gel-filtration chromatography and anion-exchange chromatography and the six kinds of pathogens-binding HMC fractions was carried out as previously described with some modifications (8). For 1D or 2D lectin blotting, 30 µg of HMC was separated by 1-DE or 2-DE and then transferred to a PVDF membrane as above descriptions in Sections 2.3 and 2.4. The membrane was blocked for 2.5 h with 5% bovine serum albumin (BSA) in TBS (20 mM Tris, 0.5 M NaCl, pH 7.4) at room temperature, then incubated with biotinylated lectin, viz., 1:1,000 dilution of concanavalin A (ConA), 1:100 dilution of peanut agglutinin (PNA), 1:500 dilution of ulex europaeus agglutinin (UEA), or 1:1,000 dilution dolichos biflorus agglutinin (DBA), for 1 h and avidin-peroxidase for 40 min at 37°C, respectively. Finally, the membrane was washed and developed with substrate (3′3-diminobenzidine, DAB) until optimum color developed. For Dot-lectin blotting, 1.5 μl of the six kinds of pathogens-binding HMC fractions (0.025, 0.05, or 0.5 mg/ml) was spotted onto a nitrocellulose (NC) membrane, which has been cut into desired size, soaked in TBS for 5 min, and then allowed to dry, using a regulatory pipe, respectively. After drying, the NC membrane was blocked with 2% polyvinylpyrrolidone (PVP, Sigma) for 1.5 h at room temperature. The following procedures for incubating with four biotinylated lectins and avidin-peroxidase and stained with DAB were performed as above described.

The carbohydrate content measurement of HMC-Va, HMC-Vf, and HMC-Mix were determined by the colorimetric method as previously described with some modification (5). Briefly, 0, 100, 200, 300, and 400 μl of 0.01 M standard glucose solution were added to tubes; then, the duplicate distillated water was complemented to the final solution volume of 500 μl. Three hundred microliters of 6% (m/v) phenol and 1.5 ml sulfuric acid were added rapidly. After incubation at room temperature for 25 min, the absorbance was detected at 490 nm. Standard curve was constructed using sugar content as ordinate and the absorbance as abscissa. The total glycan content of samples were calculated according to the standard curve after detecting the absorbance as above using 200 μl of 250 μg/ml HMC-Va, HMC-Vf, and HMC-Mix, respectively.

Deglycosylation of two HMC fractions was performed with O-glycosidase or N-glycosidase (New England Biolabs, USA) according to the manufacturer’s instructions and previous description (38). Briefly, HMC-Vp or HMC-Vf was deglycosylated under the following conditions: 5 μl of 10× G7 buffer, 3 μl of O-glycosidase or N-glycosidase, 5 μl of neuraminidase (only for O-glycosidase), then added HMC fractions (300 μg/ml for each) up to 50 μl; the reaction solution above was incubated 37°C for 4 h, respectively. The deglycosylated HMC fractions (HMC-Vp or HMC-Vf) by O-glycosidase and N-glycosidase were collected and named as dO-HMC-Vp and dO-HMC-Vf, and dN-HMC-Vp and dN-HMC-Vf, respectively. Deglycosylated HMC fractions (non-treated HMC fractions as control) were further processed by agglutinative activity assays as above described in Section 2.8.

In this study, data are presented as mean ± standard deviation (SD). Statistical significance across groups was analyzed using one-way ANOVA analysis; differences were considered to be significant at p < 0.05 and extremely significant at p < 0.01.

For a global assessment of HMC diversity in shrimp, HMC was first isolated from the hemolymph of L. vannamei by gel-filtration chromatography and anion-exchange chromatography and then identified with 1-DE and immunoblotting. As shown in Figure 1A , two bands at molecular weights approximately 75 and 77 kDa could react specifically with anti-shrimp HMC antibodies, suggesting that a good separation of the HMC has been achieved. Then, the purified HMC was analyzed by 2-DE (pH 4.7–5.9) and immunoblotting. The pattern of spots obtained from purified HMC is shown in Figure 1B . Over 40 distinct protein spots (mainly ranging from 25 to 77 kDa) were detected in 2-DE gels after Coomassie blue staining followed by analysis with the PDQuest software version 8.0. Immunoblotting showed a similar protein profile, indicating that all protein spots were HMC ( Figure 1C ). Furthermore, these spots were excised from gels and subjected to MALDI-TOF-TOF analysis. In total, 34 spots were successfully identified as L. vannamei HMC ( Table 1 ), of which 6 spots (spots 1–6) and 3 spots (spots 7–9) were confirmed as HMC subunit with 75 and 77 kDa, respectively. Notably, 25 spots (spots 10–34), ranging from 25 to 66 kDa, also showed homology with HMC, suggesting that these may be HMC fragments or modified proteins. These results indicate that L. vannamei HMC might possess high molecular diversity at the protein level.

In order to examine whether different HMC isomers responded to pathogen stimulation, 2-DE analysis was carried out to compare the variation in the nine protein spots in HMC of L. vannamei treated with S. agalactiae or V. parahaemolyticus. As shown in Figure 2A , the protein expression of HMC spots increased or decreased to some extent at different time points during the 24-h time period. After treatment with S. agalactiae for 6–24 h, spots 1–4 were upregulated, while spots 5 and 7 were downregulated compared with 0 h control group. However, spots 5–7 were upregulated after treatment with V. parahemolyticus for 6–24 h, while spots 4, 8, and 9 were downregulated at least at one time point post V. parahemolyticus treatment compared with the 0-h control group. Notably, a comparative analysis of the two gels indicated that spot 2 was upregulated about 2.5-, 2.5-, and 1.9-fold after stimulation with S. agalactiae at 6, 12, and 24 h, respectively, whereas no significant difference was observed in spot 2 during the entire 24-h period of infection by V. parahemolyticus ( Figures 2B, C ). Thus, these results indicated that L. vannamei HMC diversity might be related to shrimp’s resistance to specific bacteria in vivo.

To further determine the relationship between HMC isomers and its resistance to different pathogens in vitro, six HMC fractions bound directly to mixed bacteria (HMC-Mix), S. agalactiae (HMC-Sa), V. parahaemolyticus (HMC-Vp), V. alginolyticus (HMC-Va), V. fluvialis (HMC-Vf), and E. coli K12 (HMC-Ec) were purified by pull-down. Results of the 1-DE and 1-D immunoblotting (1-D IB) showed two similar bands, which were identified as 77 and 75 kDa HMC ( Figure 3A and Table 1 ). Comparative analysis of 2-DE and 2-D immunoblotting (2-D IB) profiles indicated that there were significant differences in HMC combined with different pathogenic bacteria. For example, the protein spots of HMC-Sa were mainly concentrated in the middle, while HMC-Vp and HMC-Vf were the opposite ( Figure 3B ). Thus, these results indicated that L. vannamei HMC diversity might be related to shrimp’s resistance to specific bacteria in vivo.

Next, agglutinative and antibacterial activities were further performed to compare the immunological properties of the six HMC fractions. As shown in Table 2 , agglutination could be observed when the six HMC fractions were separately bound with all of the five examined bacteria, showing agglutinative activities of 0.15–1.17 μg/ml. It is worth emphasizing that HMC-Sa, HMC-Vp, HMC-Va, HMC-Vf, and HMC-Ec possessed the strongest agglutinative activities against S. agalactiae, V. parahaemolyticus, V. alginolyticus, V. fluvialis, and E. coli K12, respectively, whose agglutinative titer was two- or fourfold as compared to the other four bacteria. Similarly, the six HMC fractions also showed a different degree of antibacterial activities against V. alginolyticus or V. fluvialis, respectively ( Figure 4A ). Higher antimicrobial activity was detected in HMC-Va against V. alginolyticus, and HMC-Vf and HMC-Ec against the two tested bacteria, showing inhibition of about 54.9–83.3%, but lower in HMC-Sa and HMC-Vp against V. alginolyticus. However, no activity was detected in HMC-Vp and HMC-Va to V. fluvialis and HMC-Mix against V. alginolyticus and V. fluvialis under the same conditions ( Figure 4A ). Similar results were found with counted colony method in Figure 4B . These results indicate that the specific bacterial resistance of HMC depends on co-effect of HMC isomers.

Since glycosylation is one of the most important post-translational modifications for immune molecules to regulate their function (39), we hypothesize that glycosylation difference may be also present in HMC isomers. To confirm the hypothesis, 1-D and 2-D lectin blotting was first applied. Lectins including concanavalin A (ConA, recognizing α-d-mannose > α-d-glucose), peanut agglutinin [PNA, recognizing β-D-gal-(1,3)-D- acetyl galactosamine], ulex europaeus agglutinin 1 (UEA, recognizing α-L-fucose), and dolichos biflorus agglutinin (DBA, recognizing N-acetyl-d-galactosamine) were used to blot the type of glycans in L. vannamei HMC purified by gel-filtration chromatography and anion-exchange chromatography. As expected, two bands at molecular weights approximately 75 and 77 kDa were observed, which could react with four lectins to different degrees ( Figure 5A ). For 2-D immunoblotting analysis, nine spots were detected ( Figure 5B ), and for 2-D lectin-blotting analysis, five to nine spots of HMC could bind specifically to the four examined lectins. Specifically, PNA, UEA, and DBA can recognize nine, nine, and eight HMC spots, respectively. However, ConA only recognized five HMC spots ( Figure 5B ). This indicates that different spots of HMC had different glycosylation modification. Next, dot lectin blotting and total glycan measurement were further performed to characterize glycosylation diversity of the above bacteria-binding HMC fractions and HMC-Mix as a control. As shown in Figure 5C , the six HMC fractions could bind to the four lectins, but the degree of binding was different. Among these, HMC-Va, HMC-Vp, HMC-Sa, HMC-Vf, and HMC-Ec could react with four lectins (ConA, PNA, UEA, and DBA), three lectins (ConA, PNA, and UEA), two lectins (ConA and PNA), one lectin (UEA), and one lectin (ConA), respectively, whereas HMC-Mix showed the lowest reaction to the four lectins compared with the other five HMC fractions. Consistently, the total glycan content of HMC-Va and HMC-Vf exhibited very significant difference (p < 0.01) compared with that of HMC-Mix. The HMC-Va carbohydrate content was 6.38%, which was about 2.6- and 5.4-fold to HMC-Vf and HMC-Mix ( Figure 5D ). These results suggest that glycosylation diversity might exist in HMC and HMC fractions bound to different bacteria.

To examine the relationship between immunological function and glycosylation modification of HMC, agglutination of V. parahaemolyticus and V. fluvialis was performed. We found that deglycosylated HMC fractions, dO-HMC-Vp and dO-HMC-Vf, by O-glycosidase, subsequently led to a general significant decrease in agglutinative activities to five kinds of pathogens, even completely abolished, whereas no effect was observed for those of deglycosylated HMC fractions, dN-HMC-Vp and dN-HMC-Vf, by N-glycosidase ( Table 3 and Figure 5E ). These results suggest that O-linked carbohydrates play a crucial role in agglutinative activities to five kinds of pathogens.