Four Major Envelope Proteins of White Spot Syndrome Virus Bind To Form a Complex

Abstract

Early events in white spot syndrome virus (WSSV) morphogenesis, particularly the formation of viral membranes, are poorly understood. The major envelope proteins of WSSV are VP28, VP26, VP24, and VP19. Our previous results indicated that VP28 interacts with VP26 and VP24. In the present study, we used coimmunoprecipitation assays and pull-down assays to confirm that the four major proteins in the WSSV envelope can form a multiprotein complex. Yeast two-hybrid assays were also used to test for interactions among the four proteins. In summary, three pairwise protein interactions (VP19-VP28, VP19-VP24, and VP24-VP26) and one self-association (VP24-VP24) were identified for the first time.

White spot syndrome virus (WSSV), the only member of the novel virus genus Whispovirus and family Nimaviridae (14, 15), is a double-stranded DNA, rod-shaped, enveloped virus (25, 26). The virus is an important pathogen of cultured peneid shrimp, but it can also infect many other species of crustaceans (2, 5, 7, 10, 13). So far, the WSSV genome has been completely sequenced for three isolates (6, 21, 30), and more than 50 virus-encoded proteins have been identified as structural proteins (12, 19, 29). However, early events in WSSV morphogenesis, particularly regarding the formation of viral membranes, are poorly understood.

Accumulated knowledge of viral assembly, especially for enveloped viruses, has revealed that complexes of viral proteins play a vital role in viral envelope formation. Szajner et al. found that a complex of seven vaccinia virus proteins is required for the association of membranes and viroplasm to form immature virions (16). In addition, immunobased experiments suggest that the envelope proteins ODV-E18 and ODV-EC27 from Autographa californica multiple nucleopolyhedrovirus produce a heterodimer called ODV-E35, a component of viral particles derived from occlusion bodies (1). ODV-E35 may stabilize occlusion-derived virus (ODV) structure by cross-linking envelope and nucleocapsid proteins.

VP28, VP26, VP24, and VP19 are four major envelope proteins of WSSV, and they have no known homology to structural proteins of other viruses. VP28 was reported to be located on the virion surface and to be involved in attachment to and penetration into shrimp cells (31). VP26 and VP24 were originally considered to be nucleocapsid proteins (22, 23), but subsequent studies demonstrated that both could be regarded as envelope proteins (17, 27, 28, 32) or tegument proteins (18). Although VP19 was reported to be an envelope protein (22), its exact functions in WSSV assembly and infection are unclear.

Our previous results indicated that VP28 interacts with both VP26 and VP24 (28, 29). Recently, we found that only one peak with a molecular mass of more than 220 kDa (MW-GF-200; Sigma) was observed when the envelope fraction of WSSV was analyzed by Sephacryl S-200 gel filtration (GE Healthcare) (data not shown). Therefore, we speculate that most of the envelope proteins of WSSV may form a complex. To prove this hypothesis, we employed a coimmunoprecipitation assay using an anti-VP28 mononclonal antibody (VP28-MAb) and a viral envelope fraction. Highly purified WSSV virions, as well as viral envelope and nucleocapsid fractions, were prepared according to previously described procedures (29). For coimmunoprecipitation, 100 μl of a solution containing 0.5 mg/ml of envelope protein fractions in TNM buffer (50 mM Tris-HCl, 0.15 M NaCl, 5 mM MgCl₂ [pH 7.5]) was mixed with 10 μg of VP28-MAb or control anti-His mononclonal antibody (His-MAb) (GE Healthcare) and incubated overnight at 4°C. Immune complexes were captured with 30 μl of protein A-Sepharose (GE Healthcare) for 3 h at 4°C, washed four times by centrifugation with 0.5 ml of wash buffer (20 mM Tris-HCl, 0.5% Triton X-100, 0.5 M NaCl [pH 7.5]), boiled in Laemmli buffer for 5 min, subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and visualized by Coomassie blue staining. As shown in Fig. 1, bands corresponding to VP28, VP26, VP24, and VP19 were coprecipitated with VP28-MAb, while the control His-MAb could not immunoprecipitate these proteins.

FIG. 1.

Identification of protein complex by coimmunoprecipitation. VP28-MAb or control His-MAb was incubated with (+) or without (−) the WSSV envelope protein fraction (EP). The bound proteins were dissociated, separated by SDS-PAGE, and stained with Coomassie brilliant blue. Masses in kilodaltons are indicated on the left, and four viral envelope proteins, VP28, VP26, VP24, and VP19, along with immunoglobulin G heavy (IgG H) and light (IgG L) chains, are indicated on the right.

At this stage, we could not be sure whether the coprecipitated proteins interacted individually or as part of a large complex with VP28. However, the coimmunoprecipitation assay with the polyclonal antiserum against VP26 and VP24 failed to obtain excellent results (data not shown), which may have been due to the low specificity and sensitivity of the polyclonal antibody or to the fact that the target protein bands were covered by those of the light chains of immunoglobulin G. Fortunately, evidence for the complex was obtained by in vitro pull-down assay with VP19 as the target protein.

Since pairwise protein interactions of VP28-VP26 and VP28-VP24 were demonstrated previously, we reasoned that interaction of VP19 with either of these pairs should result in the formation of a complex containing all four. The entire vp19 gene was amplified from the genomic DNA of WSSV with the forward primer 5′-ACGGATCCATGGCCACCACGACTAACAC-3′ and the reverse primer 5′-GCAAGCTTTTACTGCCTCCTCTTGGGGT-3′ (BamH I and HindIII restriction sites underlined). The PCR product was digested with BamH I and HindIII and cloned into pMAL-c2X vector (New England Biolabs), and the recombinant plasmid was transformed into Escherichia coli BL21(DE3). The bacterial cultures were induced with 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) for 30 h at 18°C and then harvested. The soluble protein formed by the fusion of maltose binding protein (MBP) with VP19 (MBP-VP19) was immobilized on a column of amylose resin and purified according to the manufacturer's recommendations (New England Biolabs). To perform the MBP pull-down assay, MBP-VP19 immobilized on amylose resin was mixed with the viral envelope fraction and incubated in binding buffer (20 mM Tris-HCl, 0.05% Triton X-100, 0.15 M NaCl [pH 7.5]) for 2 h at 4°C. The resin was extensively washed with washing buffer (20 mM Tris-HCl, 0.05% Triton X-100, 0.5 M NaCl [pH 7.5]). The bound proteins were eluted with a 10-mM maltose solution and resolved by 12% SDS-PAGE. To assess nonspecific binding, parallel control reactions were conducted using equal molar amounts of MBP protein instead of MBP-VP19 protein. The results showed that MBP-VP19 could pull down VP28, VP26, VP24, and VP19 (Fig. 2) from the viral envelope fraction, which confirmed that the four major envelope proteins could form a multiprotein complex.

FIG. 2.

Identification of protein complex by pull-down assays. Purified MBP-VP19 or MBP resins were incubated with (+) or without (−) the WSSV envelope protein fraction (EP). After extensive washing, the bound proteins were dissociated and separated by SDS-PAGE gels and stained with Coomassie brilliant blue. Masses in kilodaltons are indicated on the left, and four viral envelope proteins, VP28, VP26, VP24, and VP19, are indicated on the right.

The yeast two-hybrid system (Stratagene) was used to extend our analysis of protein-protein interactions in the four-protein complex. Full-length vp28, vp26, vp24, and vp19 genes were PCR amplified from the genomic DNA of WSSV. The PCR products were digested with EcoRI and SalI and cloned, respectively, into pBD-GAL4 Cam vector as bait and pAD-GAL4-2.1 vector as prey. To test interactions, various combinations of bait and prey plasmids were transformed into the Saccharomyces cerevisiae host strain YRG-2, using the lithium acetate method. Meanwhile, the plasmids pLamin C/pAD-MUT and pBD-MUT/pAD-MUT (supplied with the kit) were also cotransformed to YRG-2 as negative and positive controls, respectively. The transformants were first selected for expression of the HIS3 reporter gene by plating on synthetic dropout minimal medium deficient in histidine, leucine, and tryptophan. Putative positive colonies which grew on the selective medium were assayed for expression of the lacZ reporter gene by the detection of β-galactosidase activity with a solution containing an X-Gal (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside) substrate. The colonies that turned blue were verified as true positives. This assay identified positive interactions that included VP28-VP24, VP28-VP26, VP26-VP26, VP24-VP26, VP24-VP24, and VP19-VP24 (each former protein inserted in the bait vector) (Table 1; see Fig. S1 and S2 in the supplemental material). The negative control could not grow on the plate deficient in histidine, leucine, and tryptophan, whereas the positive control could grow. No envelope protein could autoactivate when inserted in the bait vector (Table 1).

TABLE 1.

Interactions among four major envelope proteins identified by yeast two-hybrid experiments^a

Bait protein	Activity with indicated target protein:
	VP28	VP26	Vp24	VP19
VP28	−	+	+++	−
VP26	−	+	−	−
Vp24	−	++	+	−
VP19	−	−	++	−

^aInteractions were initially assessed using a qualitative in vivo plate assay (at day 3), with the readout being the expression of the reporter gene HIS3. No growth (−) indicates a negative interaction. Putative positive colonies were assayed for expression of the lacZ reporter gene by the detection of β-galactosidase activity, with + indicating a low level of activity, ++ indicating normal activity, and +++ indicating a high level of activity.

The same in vitro pull-down assay using MBP-VP19 and bacterially expressed viral proteins was performed to confirm the pairwise protein interactions among the four envelope proteins. All procedures were the same as described for the MBP pull-down assay except that the prey proteins were changed into His-tagged fusion proteins rVP28, tnVP26 (amino acids 36 to 204), and tnVP24 (amino acids 26 to 208), all of which were expressed and harvested as previously described (27-29). Among them, rVP28 was full length, while tnVP26 and tnVP24 were truncated by removal of the N-terminal hydrophobic region due to the poor solubility of full-length VP26 and VP24 expressed in E. coli. As shown in Fig. 3, rVP28 or tnVP24 bound to MBP-VP19 (Fig. 3, lanes 1 and 4) but not to MBP alone (Fig. 3, lanes 2 and 5). However, tnVP26 did not interact directly with MBP-VP19 (Fig. 3, lane 7). We also immobilized tnVP24 or rVP28 on Ni-nitrilotriacetic acid Sepharose beads (Qiagen); unfortunately, they failed to pull down MBP-VP19. It is possible that the VP19-interacting regions in tnVP24 or rVP28 may be close to the His tag that binds to resin. If so, the steric hindrance from the resin could have prevented MBP-VP19 from reaching these regions. Furthermore, we found that the VP19-VP28 interaction identified by the immobilized MBP-VP19 pull-down assay could not be confirmed by the yeast two-hybrid assay. We cannot entirely rule out the possibility of a false negative, since in some cases, physiological protein-protein interactions are not detected by two-hybrid assays due to false negatives that may arise from poor or unstable expression, improper folding, failure of nuclear localization, and steric hindrance of the two fusion proteins (8, 9, 20).

FIG. 3.

Interaction of rVP28, tnVP24, or tnVP26 with MBP-VP19 analyzed by pull-down experiments. Purified recombinant protein rVP28, tnVP24, or tnVP26 was incubated with resins containing bound MBP or MBP-VP19 and then washed extensively. The bound proteins were dissociated and separated by SDS-polyacrylamide electrophoresis gels and stained with Coomassie brilliant blue. M, protein molecular mass marker (kDa). Arrows indicate rVP28, tnVP24, and tnVP26.

Other viral structural proteins have been shown to be able to interact with at least one of the four major proteins. These interactions include that of WSV010 with VP24 (4), VP51A with VP26 (3), and VP38 with VP24 (11). This means that at least three more proteins are involved in the complex. Moreover, Xie and Yang first reported the association between VP26 and the viral nucleocapsid (27), and Wan et al. subsequently reported that VP26 linked the envelope and nucleocapsid by binding with VP51, a nucleocapsid protein (24). It is reasonable to suggest that the four major proteins form an underlying framework for the WSSV envelope and that other low-abundance proteins bind to the framework by interacting with at least one of its constituent proteins. This structure may embed in or anchor to the lipid membrane of WSSV (33) and link to the viral nucleocapsid through VP26.

In conclusion, we found that the four major envelope proteins of WSSV can form a multiprotein complex. Moreover, the interactions among the four proteins were also analyzed. In summary, five pairwise protein interactions and two self-associations were identified in our study. Three of the pairwise interactions and one self-association were not identified previously. Some interactions, such as those of VP19-VP28 and VP24-VP26, need to be confirmed by further study, as do the oligomerizations of VP24 that were revealed by yeast two-hybrid assay. The characterization of this multimember complex will shed light on the WSSV structural organization and raise further questions about its origin and pathway of assembly.