C-type Lectin Binds to β-Integrin to Promote Hemocytic Phagocytosis in an Invertebrate*

Background: Our previous study found shrimp C-type lectin (FcLec4) facilitated bacterial clearance in vivo. Results: Native FcLec4 promoted hemocytic phagocytosis. The FcLec4 receptor was identified as β-integrin. Conclusion: FcLec4 functions as an opsonin, and β-integrin determines its opsonic activity. Significance: This is the first study to show the structural basis of an opsonin in crustaceans. Phagocytosis is a conserved cellular response among metazoans. Opsonins are some molecules that label targets to increase their susceptibility to phagocytosis. Opsonins are usually captured by receptors on the surface of phagocytes. Our previous study found the C-type lectin FcLec4 from Chinese white shrimp Fenneropenaeus chinensis might function as an opsonin to facilitate bacterial clearance. In the present study we purified the native FcLec4 protein and confirmed its opsonic activity in the near relation, kuruma shrimp Marsupenaeus japonicus. The possible receptor of FcLec4 was identified as β-integrin by panning a T7 phage display library of shrimp hemocytes and then confirmed by co-immunoprecipitation assay. We further proved that the interaction between FcLec4 and β-integrin did not rely on the carbohydrate recognition domain but on the N terminus of FcLec4. In addition, inhibition of FcLec4 expression using RNAi delayed bacterial clearance, and β-integrin knockdown suppressed the opsonic activity of FcLec4. This study is the first to show the direct interaction between an opsonin and its receptor in crustaceans. Our study provides new insights into invertebrate phagocytosis and the functions of C-type lectins.

Mammalian macrophages express a series of phagocytic receptors, including Fc␥ receptors, complement receptors, scavenger receptors, and integrins (8). These receptors bind to specific opsonins. For example, Fc␥ receptors are involved in the uptake of IgG-opsonized particles, whereas complement receptors bind to complement components (4,9). These receptors induce cytoskeleton reorganization, which drives phagocytosis (10).
Despite the availability of studies regarding phagocytosis, however, information regarding phagocytosis and opsonization in invertebrates is limited, and few molecules have been proven to facilitate phagocytosis. Some of these molecules include direct transmembrane receptors, such as eater, scavenger receptor, Down syndrome cell adhesion molecules, and integrins (11)(12)(13)(14). Mosquito aTEP-1, a soluble protein, was the first molecule found to promote phagocytosis of Gram-negative bacteria as a complement-like opsonin by binding to bacteria according to its thioester bonds (15). Some other recognition receptors, such as ficolin-like proteins, could also promote phagocytosis (16,17). Although many studies on the intracellular events of phagocytosis have been conducted (18 -20), the opsonin receptors have yet to be determined.
We are interested in a family of important pattern recognition receptors, C-type lectins, particularly the soluble members, because they are abundantly expressed and effectively discriminate non-self carbohydrates. Many members of this family are involved in phagocytosis, including mannose binding lectin (MBL), which induces the deposition of complement C3 (21) or directly functions as an opsonin (22). C-type lectins are mostly soluble and significantly more abundant in invertebrates than in vertebrates (23). Several lectins are assumed to be opsonins, but direct evidence of this assumption is unavailable (24,25). We previously reported that a novel lectin, FcLec4, from the Chinese white shrimp Fenneropenaeus chinensis could bind to the pathogenic bacterium Vibrio anguillarum and facilitate bacterial clearance in vivo. We thus speculated that FcLec4 functioned as an opsonin during clearance (26). In the present study we further analyzed the opsonic mechanism of FcLec4. Given that a homolog of FcLec4 was found in kuruma shrimp Marsupenaeus japonicus, the in vivo study was performed in kuruma shrimp because they are easier to access and cultivate under laboratory conditions than Chinese white shrimp. Using the purified native protein, we confirmed the opsonic activity of FcLec4 in vivo. The hemocytic receptor was identified as ␤-integrin, and the structural basis of protein interaction was investigated. Our results provide new insights into invertebrate hemocytic phagocytosis and expand the knowledge on C-type lectins as opsonins.

EXPERIMENTAL PROCEDURES
Animals-Healthy Chinese white shrimp F. chinensis (ϳ15 g) were obtained from an offshore area in Qingdao, China, deepfrozen in liquid nitrogen, and stored at Ϫ80°C. Healthy kuruma shrimp M. japonicus (ϳ5 g) were obtained from a shrimp farm in Rizhao, Shandong, China, cultured in our laboratory in artificial seawater at 25°C, and fed daily with a commercial diet. A permit to obtain the shrimp was approved ((2009)-HY-158320) by the Shandong Provincial General Office of Ocean and Fishery. Animal experiments were performed in accordance with the rules set by Shandong Provincial Experimental Animal Center under permit SYXK (Shandong) 2013 0001.
Bacterial Challenge and Sample Collection-Kuruma shrimp were challenged with 2 ϫ 10 7 cfu V. anguillarum SDMCC210204 and 50 l of PBS (140 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 2 mM KH 2 PO 4 , pH 7.4) as a control. At 12 h post-challenge, total RNA was extracted from the gills and stomach using Unizol (Biostar, Shanghai, China), and the cDNA was synthesized using a RevertAid First Strand cDNA Synthesis kit (Fermentas, Burlington, Canada) in accordance with the manufacturer's instructions. Protein samples were prepared by homogenizing the gills and stomach of the shrimp in PBS supplemented with 5 mM EDTA and 1 mM phenylmethanesulfonyl fluoride (PMSF) and then centrifugation at 14,000 ϫ g for 15 min to obtain the supernatant.
Western Blot Analysis-Protein samples from the gills and stomach of Chinese white shrimp and kuruma shrimp (50 g) were separated using 12.5% SDS-PAGE. The proteins were then transferred onto a nitrocellulose membrane followed by blocking with 3% nonfat milk dissolved in Tris-buffered saline (10 mM Tris-HCl, 150 mM NaCl, pH 7.5). The membrane was incubated overnight with FcLec4 antiserum (1:200 dilution in blocking buffer) followed by coating with the peroxidase-conjugated goat anti-rabbit IgG. After incubation with the membrane for 3 h, unbound IgG was washed out, and the target protein was visualized by 4-chloro-1-naphthol oxidation.
cDNA Cloning and Bioinformatics Assay-Based on a fragment sequence that showed high similarity to FcLec4 obtained from a transcriptomic analysis of kuruma shrimp, a pair of primers, GSF1 and GSR1 (Table 1), was designed to amplify the 3Ј and 5Ј ends of the cDNA with the 3Ј anchor R and the 5Ј primer, respectively. Another pair of primers, GSF2 and GSR2 (Table 1), was used to amplify the full-length cDNA to confirm accuracy. The full-length cDNA of the resultant gene from kuruma shrimp was highly similar to that of FcLec4 determined by basic local alignment search tool (BLAST) analysis and thus designated as hFcLec4. The FcLec4 and hFcLec4 sequences were aligned using ClustalX.
hFcLec4 Expression Profile Analysis-The hFcLec4 expression pattern was determined using semi-quantitative RT-PCR with ␤-actin as an internal reference (primers ␤-actin F and R are listed in Table 1). The PCR procedure was set as follows: initial temperature of 94°C for 3 min and 30 cycles of 94°C for 15 s, 54°C for 15 s, and 72°C for 15 s. The PCR products were Shrimp C-type Lectin Binds to ␤-Integrin separated by 1.5% agarose electrophoresis and stained with ethidium bromide. Native Protein Purification-The antibody against FcLec4 was initially purified. Briefly, the antiserum was diluted in binding/wash buffer (0.15 M NaCl, 20 mM Na 2 HPO 4 , pH 8.0) and added to Protein A resin (GenScript, Nanjing, China). After washing thoroughly with binding/wash buffer, the antibody was eluted with elution buffer (0.1 M glycine, pH 2.5) and immediately neutralized to pH 7.4 with neutralization buffer (1 M Tris-HCl, pH 8.5). The purified antibody was dialyzed overnight in coupling buffer (0.1 M NaHCO 3 , 0.5 M NaCl, pH 8.3) at 4°C and then coupled with 60 mg of CNBr-activated Sepharose 4B (Amersham Biosciences) with rotation at 25°C for 1 h. The resin was washed five times with coupling buffer and then equilibrated in 0.1 M Tris-HCl, pH 8.0, at 25°C for 2 h. After four times of alternate washing with acetic acid buffer (0.1 M sodium acetate, 0.5 M NaCl, pH 4.0) and Tris buffer (0.1 M Tris-HCl, 0.5 M NaCl, pH 8.0), the resin was equilibrated in Tris buffer. The homogenization of Chinese white shrimp gills (10 mg of tissue homogenized in 10 ml of Tris buffer and then centrifuged at 14,000 ϫ g for 20 min) was then mixed with the resin and subjected to gentle rotation overnight at 4°C. After five washings with Tris buffer, the native protein was eluted with elution buffer and neutralized with neutralization buffer. The purified protein was confirmed by both SDS-PAGE (silver staining) and Western blot analysis. The protein concentration was determined using a Gene-Quant spectrophotometer (Amersham Biosciences).
Bacteria Binding Assay-The purified native FcLec4 (nFcLec4, 10 g) was incubated with four strains of bacteria (ϳ2 ϫ 10 8 cfu each): V. anguillarum SDMCC210204, Escherichia coli DH5␣, Staphylococcus aureus AS1.89, and Micrococcus luteus SDMCC210058. The mixture was subjected to gentle rotation at 28°C for 1 h. Bacterial pellets were obtained by centrifugation at 5000 ϫ g for 5 min, washed 3 times with Tris buffer, pH 8.0, and then analyzed by Western blot using FcLec4 antiserum. Bacteria that had not been incubated with nFcLec4 were used as controls.
Clearance Assay-The bacterial clearance test was performed as previously described (26) with some modifications. Four bacterial suspensions (optical density (A 600 ) ϭ 0.2) were first prepared. nFcLec4 or recombinant FcLec4 (rFcLec4, 20 g) was then incubated with 500 l of the suspension. After incubation for 1 h with slow rotation at 28°C, the bacteria were collected by centrifugation, washed 3 times with Tris buffer, and then suspended in 1 ml of PBS. 50 l of the suspended bacteria in PBS were injected into the shrimp (ϳ6 -7 g each), and the hemolymph was collected 30 min after injection. The number of residual bacteria was determined by plating the samples onto Luria broth (LB) agar plates (3% NaCl for V. anguillarum). For each group, three to four shrimp were used, and each experiment was repeated three times. The results are presented as the mean Ϯ S.D. Uncoated bacteria were used as controls. For the recombinant protein, another shrimp lectin, SL11, was also used as a control. The data were subjected to Student's t test, and p Ͻ 0.05 was considered statistically significant.
Phagocytosis Assay-Overnight cultures of V. anguillarum were heat-killed at 72°C for 20 min and dissolved in 0.1 M NaHCO 3 , pH 9.0. The sample was labeled with FITC (Sigma) at a final concentration of 1 mg/ml at 28°C for 1 h. The labeled bacteria were washed 5 times with PBS to remove the free FITC and then coated with FcLec4. V. anguillarum (ϳ10 7 cfu) were injected into the shrimp. Hemolymph was collected 30 min after injection, and the hemocytes were isolated by centrifugation for 800 ϫ g for 5 min at 4°C. The hemocytes were then washed with PBS and dropped onto poly-L-lysine-coated glass slides. After 30 min the cells were stained with propidium iodide (Sigma) for 1 min. After washing with PBS, trypan blue (2 mg/ml) was added to quench the non-phagocytosed bacteria for 20 min. The slides were then washed 5 times with PBS, and phagocytosis was observed under a fluorescence microscope (Olympus BX51, Tokyo, Japan). 600 cells were counted to determine the phagocytic percentage and phagocytic index, which were calculated as the number of phagocytic cells divided by the total number of cells and the number of bacteria phagocytosed divided by the total number of cells, respectively. The phagocytosis assay was performed three times. The data were subjected to Student's t test, and differences with p Ͻ 0.05 were considered statistically significant.
Hemocyte Binding Assay-To detect the binding of FcLec4 to the surface of the hemocytes, immunocytochemistry was performed. nFcLec4 (20 g) was injected into the shrimp. The hemocytes were collected 30 min after injection, washed 2 times with PBS containing reduced glutathione (2 mg/ml), spread onto slides, and then fixed with 4% paraformaldehyde in PBS at 4°C for 1 h. The slides were washed 3 times with PBS, blocked with 2% BSA in PBS for 1 h at 37°C, and then incubated overnight with 1:100 diluted FcLec4 antiserum at 4°C. After washing 5 times with PBS, the slides were incubated with 1:1000 diluted goat anti-rabbit-ALEXA 488 (Molecular Probes) at 37°C for 1 h. The slides were again washed 5 times with PBS, and the nuclei of the hemocytes were stained with DAPI for 10 min. After washing three times with PBS, the slides were observed under a fluorescence microscope. A recombinant crayfish lectin with its corresponding antiserum was used as a negative control (27).
Panning of the T7 Phage Display Library-Panning was performed based on our previous report (28) with some modifications. Briefly, 40 g of purified FcLec4 was added to a 96-well plate and incubated at 4°C overnight. The unbound protein was removed by washing with PBS twice, and 10 l of T7 phage display library derived from Chinese white shrimp hemocytes was added to the wells. Incubation was performed overnight. After washing 3 times with TBST (50 mM Tris-HCl, 0.5 M NaCl, 0.02% Tween 20, pH 7.5), FcLec4-bound phages were eluted with 200 l of 1% SDS and then centrifuged at 6000 ϫ g for 5 min. 10 l of the supernatant was inoculated into 1 ml of BLT5403 cell culture (A 600 ϭ 0.5) and cultured at 37°C for 2 h. The culture was centrifuged, and the supernatant was plated on an agarose plate containing the BLT5403 cells. The plates were cultured at 37°C for 3 h. Extraction buffer (100 mM NaCl, 6 mM MgSO 4 , 20 mM Tris-HCl, pH 8.0) was added to the wells, and the plates were incubated at 4°C overnight. The extraction buffer was then collected for the next panning. After three repeats, single plaques appeared, and the 10 longest (Ͼ500 bp) fragments were sequenced.
␤-Integrin Antibody Preparation-A fragment encoding the outer membrane part of M. japonicus ␤-integrin was amplified with specific primers (integrinExF and integrinExR, Table 1) and then ligated into pET32a(ϩ) vector (Novagen). The recombinant plasmid was transformed into E. coli Rosseta (DE3) cells. Protein expression was induced with 0.2 mM isopropyl-␤-Dthiogalactopyranoside at 37°C for 4 h. The recombinant protein was purified by affinity chromatography with Ni-NTA 2 beads (Novagen). After dialysis in PBS for 24 h, the protein was concentrated to 3 mg/ml using an ultra-centrifugal filter (Millipore). Equal volumes (1 ml) of the protein sample and complete Freund adjuvant (Sigma) were mixed thoroughly and injected into a rabbit. The injection was repeated after 25 days with incomplete adjuvant. The antiserum was collected 7 days after injection, and the antibody was purified.
Co-immunoprecipitation-To confirm the interaction between FcLec4 and ␤-integrin, co-immunoprecipitation was performed. Shrimp gills were homogenized in radioimmune precipitation assay buffer (50 mM Tris-HCl, 150 mM NaCl, 0.1% SDS, 0.5% Nonidet P-40, 1 mM EDTA, 0.5 mM PMSF, pH 8.0) and then centrifuged at 14,000 ϫ g for 20 min. The supernatant was pre-cleared with 30 l of Protein A beads (GenScript) at 4°C for 40 min with agitation. The mixture was then centrifuged at 12,000 ϫ g for 10 min. The supernatant was incubated with 5 g of FcLec4 or ␤-integrin antibodies overnight at 4°C with rotation. The suspension was then incubated with 20 l of Protein A beads for 1 h at 4°C under agitation. The beads were collected by centrifugation at 2000 ϫ g for 3 min and then washed 4 times with PBS for 10 min. The washed beads were resuspended in 30 l of SDS-PAGE sample buffer and boiled for 10 min. Resultant samples were then separated by SDS-PAGE and analyzed by Western blot using FcLec4 and ␤-integrin antibodies. IgG purified from naïve rabbit was used as a control.
Expression and Purification of Truncated FcLec4-To study the interaction of FcLec4 with integrin, different truncated mutations of FcLec4 were expressed in E. coli BL21 (DE3) cells using specific primers (Table 1, Fig. 7B) similar to the expression of full-length FcLec4 with the pET30a(ϩ) vector (Novagen) (26). The proteins were expressed as inclusion bodies, denatured in 8 M urea, and purified with Ni-NTA beads. The proteins were then refolded in a buffer containing 50 mM Tris-HCl, 50 mM NaCl, 0.5 mM EDTA, 1% glycine, and 10% glycerol at pH 7.9 and dialyzed in PBS overnight. The refolded purified proteins were used for the following pulldown assay or to coat bacteria for the clearance test.
Pulldown Assay-Shrimp gill extract was prepared by homogenizing the tissue in PBS and then centrifuging at 14,000 ϫ g for 20 min. The supernatant (500 l) was incubated with recombinant truncated proteins (2 g) overnight with rotation at 4°C. Charged Ni-NTA beads were added to the suspension, and the mixture was incubated for 1 h. The beads were collected by centrifugation at 2000 ϫ g for 3 min and washed with PBS three times. The bound protein was eluted with 250 mM imidazole in 50 mM Tris-HCl and then subjected to Western blot using the antibody against ␤-integrin.
RNAi-Primers (hFcLec4iF, hFcLec4iR, integriniF, and integriniR, Table 1), each linked to the T7 promoter, were used to amplify the hFcLec4 and ␤-integrin as templates for dsRNA synthesis. The transcription mixture (50 l) contained 8 g of template DNA, 2.4 l of each NTP (100 mM), 60 units of T7 RNA polymerase, 80 units of RNase inhibitor, and 20 l of 5ϫ transcription buffer. The mixture was incubated at 37°C for 8 h, after which 8 units of DNase I were added to remove the template DNA. After phenol/chloroform extraction and ethanol precipitation, the dsRNA was dissolved in RNase-free water. A total of 15 g of dsRNA was injected into the shrimp, and the mRNA and protein samples from the hemocytes and gills were isolated after 24 h. The RNAi efficiency was determined by RT-PCR and Western blot analysis using dsGFP RNA as a control. The clearance rates of the bacteria (uncoated or coated with rFcLec4) were then evaluated after hFcLec4 and ␤-integrin expression knockdown (24 h after dsRNA injection).

RESULTS
FcLec4 Homolog Existed in Kuruma Shrimp-FcLec4 was a novel shrimp lectin that was similar to insect C-type lectins. It was ubiquitously distributed in the gills, stomach, hepatopancreas, and intestine of Chinese white shrimp (26). We found that the FcLec4 antiserum could recognize a single band from the homogenates of the gills and stomach of kuruma shrimp (Fig. 1A), thereby suggesting the presence of an FcLec4 homolog in this specie. We obtained its cDNA sequence by transcriptomic analysis and RACE and designated it as hFcLec4 (GenBank TM ID KC505420). The two proteins were 94% identical (Fig. 1B). We then checked the hFcLec4 mRNA expression pattern. This gene was widely expressed in several tissues, such as the stomach, gills, hemocytes, and intestines (Fig. 1C), and it was up-regulated by V. anguillarum challenge 12 h after infection (Fig. 1D). These findings were similar to the expression profiles of FcLec4 in Chinese white shrimp, which suggested that a homolog of FcLec4 was present in kuruma shrimp, and it would be reliable to study the function of FcLec4 in kuruma shrimp.
Native FcLec4 Bound to Bacteria and Promotes Bacterial Clearance-We previously reported that rFcLec4 could facilitate bacterial clearance in Chinese white shrimp. To further ensure the reliability to study FcLec4 in kuruma shrimp, we checked whether rFcLec4 showed a similar function in this specie. As shown in Fig. 2A, rFcLec4 promoted bacterial clearance in kuruma shrimp, whereas SL11, another shrimp C-type lectin, did not contribute to bacterial clearance.
Using the FcLec4 antibody, we purified nFcLec4 from the gills of Chinese white shrimp (Fig. 2B). The native protein could bind to four kinds of bacteria, including Gram-negative and Gram-positive bacteria (Figs. 2C and 3A). Moreover, coating the bacteria with nFcLec4 also facilitated their clearance in kuruma shrimp in vivo (Figs. 2D, and 3, B-D).
The effect of FcLec4/hFcLec4 was confirmed by RNAi. It was shown in Fig. 4A that both mRNA (left panel)  (right panel) were suppressed by hFcLec4 dsRNA injection. The clearances of uncoated (left panel) and FcLec4-coated (right panel) bacteria were both delayed in the hFcLec4-silenced shrimp (Fig. 4B). FcLec4/hFcLec4 Functioned as an Opsonin-To investigate the mechanism by which FcLec4/hFcLec4 promoted bacterial clearance, FITC-labeled V. anguillarum was coated with nFcLec4 and then injected into kuruma shrimp. Hemocytes were separated to determine the induction of phagocytosis. The results showed that FcLec4 coating significantly promoted hemocytic phagocytosis of bacteria (Fig. 5). The phagocytic percentage (left panel) and phagocytic index (right panel) of the coated group were significantly higher than those in the uncoated group. These results indicated that FcLec4/hFcLec4 functioned as an opsonin to promote hemocytic phagocytosis.
nFcLec4 Bound to Hemocyte Surface to Promote Phagocytosis-Given that many opsonins bind to phagocyte receptors, we determined whether or not FcLec4/hFcLec4 could bind to hemocyte surfaces. As shown in Fig. 6, when nFcLec4 was injected into shrimp, the proteins were detected on the surface of shrimp hemocytes 30 min after injection. By contrast, another C-type lectin, rPcLec2, showed no binding activity. These results suggested that shrimp hemocytes expressed some FcLec4/hFcLec4 receptors on their surfaces and that the interaction of FcLec4/hFcLec4 with these receptors might initiate opsonic activity.
FcLec4/FcLec4 Bound to ␤-Integrin through the N Terminus-To determine the FcLec4/hFcLec4 receptors, we panned a T7 phage library derived from the hemocytes of Chinese white shrimp using nFcLec4 as the target (28). After three rounds of panning, single plaques were selected for PCR amplification and sequencing. Two of the 10 sequenced fragments were identified as ␤-integrin based on BLAST analysis ( Table 2). The full-length of ␤-integrin cDNA from kuruma shrimp was also obtained from our transcriptomic sequence data (GenBank TM ID KC527624). Given that integrin was involved in phagocytosis, we focused on this protein. Using a co-immunoprecipitation assay, we found that ␤-integrin could bind to both FcLec4 in Chinese white shrimp and hFcLec4 in kuruma shrimp (Fig. 7A). These results indicated that ␤-integrin was an FcLec4/hFcLec4 receptor.
We next expressed the different truncated proteins of FcLe4 in E. coli (Fig. 7B) to determine whether the interaction was dependent on the C-type lectin-like domain or the N or C terminus. The proteins were used in a pulldown assay to determine their integrin binding ability. As shown in Fig. 7C, only the two truncated proteins containing the N terminus of FcLec4 (F1R1 and F1R2) physically interacted with ␤-integrin. Moreover, we found that the opsonic effect of FcLec4 also relied on the N terminus (Fig. 7D). The host cleared bacteria coated with F1R1 and F1R2 more rapidly than bacteria coated with F2R1 and F2R2. These results suggested that the N terminus of FcLec4 determined the interaction with ␤-integrin and that such interactions initiated the opsonic activity of FcLec4.
RNAi of ␤-Integrin Inhibited the Opsonic Effect of FcLec4-To verify the function of ␤-integrin-FcLec4/hFcLec4 interaction during bacterial phagocytosis, ␤-integrin expression was suppressed using RNAi at both mRNA (upper panel) and protein (lower panel) levels (Fig. 8A). In this case FcLec4 was not detected on the surface of hemocytes that had been pretreated with ␤-integrin dsRNA (Fig. 8B). In addition, no significant    difference was found between the clearance of bacteria coated with F1R2 (with N terminus) and that of bacteria coated F2R1 (without N terminus) (Fig. 8C). Moreover, both the phagocytic percentage (Fig. 8D) and phagocytic index (Fig. 8E) showed no significant difference between the F1R2-coated and F2R1coated groups. These results showed that suppression of ␤-integrin expression inhibited the opsonic effect of FcLec4 and further confirmed that the N terminus of FcLec4-derived opsonization relied on ␤-integrin.

DISCUSSION
Integrins, which are assembled by two distinct ␣ and ␤ subunits, are important transmembrane receptors. Twenty-four integrin chains consisting of 18 ␣ and 8 ␤ subunits have been discovered in human, whereas only 5 ␣ and 2 ␤ subunits have been discovered in fruit fly (29). Both ␣ and ␤ subunits contain extracellular and cytoplasmic domains. These domains facilitate the contribution of integrins in many pathological and developmental processes by linking cells with the environment (30).
The lack of genomic data hinders definitive identification of the number of integrin subunits in shrimp. Nevertheless, the integrin subunit identified in the present study shared high similarity with many ␤ subunits (BLAST result, data not shown), thereby suggesting that the integrin found in the current study is a member of the ␤ family. Some ␤-integrins in invertebrates participate in phagocytosis (31)(32)(33). For example, the Drosophila integrin ␤v subunit acts as a recognition receptor for bacterial cell wall components to promote the phagocytosis of S. aureus (14); this subunit is also a receptor for the phagocytosis of apoptotic cells by embryo hemocytes (34). However, all of these observations involve the direct recognition of targets by integrins. Although some reports have shown that vertebrate integrins could capture opsonins (35), no such indirect recognition by integrin was found in invertebrates. The functions of some opsonins have been described, but the mechanism of their linkage with phagocytotic cells is unclear (36,37). Our study is the first to demonstrate the interaction between the transmembrane receptor integrin with an opsonin, a C-type lectin, in invertebrates.
Some members of the C-type lectin family, particularly the soluble members, which are more like to be pattern recognition receptors other than transmembrane signal transductors, function as opsonins in different ways, and an increasing number of cell surface receptors have been identified from immune cells (38). For example, the opsonic effect of mannose-binding lectin depends on cC1qR, which is a cell surface calreticulin (39). Furthermore, CD14 and CD35 are both mannose-binding lectin receptors (40,41). Although some reports in vertebrates show that C-type lectins interact with integrins, these interactions are not involved in phagocytosis (42,43). Our results on  FcLec4/hFcLec4-integrin interaction and function expand the current knowledge of soluble C-type lectins as opsonins.
Our previous study showed that FcLec4 opsonization is mediated by its C-type lectin domain, wherein FcLec4 binds to peptidoglycan on the surface of bacteria (26). However, the interaction of FcLec4 with integrin depends on its N terminus, instead of the C-type lectin domain. Integrins bind to extracellular ligands generally through the von Willebrand factor type/␤A domain, which contains the DXSXS, DDL, and SDL motifs as core regions for interaction (44 -46). Most ligands are recognized according to the RGD-like motif they harbor (47). However, the N terminus of FcLec4 does not contain such a motif. Considering an increasing number of novel motifs besides RGD, VKC, and GLPEER, for instance, have been identified as integrin target sites (48,49), FcLec4 may interact with integrin in a non-classical manner. Determination of the specific integrin-recognizing motif in FcLec4 deserved further study.
The FcLec4/hFcLec4-integrin interaction was found in both Chinese white shrimp and kuruma shrimp, which belong to the same family but different genera. Therefore, the lectin-integrin mechanism is probably conserved at least among the family Penaeidae. Our previous study found that FcLec4 is a novel shrimp C-type lectin similar to insect lectins but not to other crustacean lectins (26). This observation further indicates that the protein and its homologs may be critical for phagocytosis not only in shrimp but also in many arthropods.
In summary, the present study demonstrated a novel mechanism for an invertebrate C-type lectin to participate in phagocytosis. FcLec4/hFcLec4 generally detected invading bacterial pathogens via the carbohydrate recognition domain (i.e. C-type lectin domain). Upon recognition, binding between the N terminus of the lectin and ␤-integrin leads to cytoskeletal reorganization, which induces phagosomes formation to ingest the invading bacteria (Fig. 8F). The promotion of phagocytosis by the interaction between C-type lectins and ␤-integrin contributes to our understanding of the function and structural basis of other pattern recognition receptors. FcLec4 (20 g) was injected into shrimp that had been treated by dsGFP or ds␤integrin. Hemocytes were collected 30 min later and detected by immunocytochemistry with antibody of FcLec4. The opsonic activity of FcLec4 (clearance rate) (C), phagocytosis percentage (D), and phagocytosis index (E) were inhibited when the integrin expression is suppressed. V. anguillarum coated with F1R2 or with F2R1 was injected into the shrimp after integrin knockdown. The residual bacteria in hemolymph were counted 30 min later. The clearance test was performed using three to four animals in each group. Also FITC-labeled bacteria was coated by F1R2 or F2R1 and injected into the integrin-silenced shrimp. The number of phagocytic hemocytes and bacteria were counted to calculate the phagocytosis percentage and the phagocytosis index. Data indicate the mean Ϯ S.D. of three independent experiments. F, potential mechanism of the opsonic activity of FcLec4/hFcLec4. FcLec4/hFcLec4 relies on its carbohydrate recognition domain to recognize bacterial pathogens that invade into hemolymph and recruits them onto the hemocyte surface by binding the hemocytic receptor ␤-integrin through its N terminus. ␤-Integrin clustering triggers the cytoskeletal reorganization for membrane fusion to form the phagosome for internalizing the opsonized bacteria. CTLD, C-type lectin-like domain.