RING-H2 protein WSSV249 from white spot syndrome virus sequesters a shrimp ubiquitin-conjugating enzyme, PvUbc, for viral pathogenesis

Abstract

Modification of proteins by ubiquitin is essential for numerous cellular processes. The RING-H2 finger motif has been implicated in ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Four proteins, WSSV199, WSSV222, WSSV249, and WSSV403, from white spot syndrome virus (WSSV) contain the RING-H2 motif. Here we report that WSSV249 physically interacts with a shrimp ubiquitin-conjugating enzyme, PvUbc, and mediates ubiquitination through its RING-H2 motif in the presence of E1 and PvUbc. Mutations of the putative zinc coordination residues in the RING-H2 domain of WSSV249, however, ablate ubiquitination efficiency. In addition, the RING-H2 domain of WSSV249 is capable of ubiquitination with UbcH1, UbcH2, UbcH5a, UbcH5b, UbcH5c, UbcH6, and UbcH10, respectively, exhibiting a low degree of E2 specificity. Significantly, the expression of WSSV249 and PvUbc increased during infection, as revealed by real-time PCR. Furthermore, in situ hybridization showed that WSSV249 and PvUbc display similar expression patterns in infected shrimps, and immunofluorescence and immunohistochemistry assays showed an increase of PvUbc in infected shrimp cells. These results suggest that the RING-H2 protein WSSV249 from WSSV may function as an E3 ligase via sequestration of PvUbc for viral pathogenesis in shrimp.

FIG. 1.

Structural motifs of WSSV249 and time course study of WSSV249 and PvUbc expression. (A) Structural motifs of WSSV249. Domains I (amino acids 144 to 156), II (amino acids 207 to 219), and V (amino acids 692 to 704), respectively, represent EF-hand calcium binding domains. Domain IV (amino acids 454 to 639) represents a repeat region. Domain III (amino acids 310 to 357) represents the position of the RING finger and is indicated with the locations of the first and last zinc-coordinating cysteine residues. In the lower row, the numbers refer to the positions of eight zinc-coordinating residues, and the mutated zinc-coordinating residues for functional analysis are indicated with an asterisk. (B) Time course study of WSSV249 and PvUbc by RT-PCR. The transcription of WSSV249 was detected as early as 3 h postinfection, while low expression of PvUbc was detected in noninfected samples (mock). β-Actin acted as the housekeeping control gene. (-), negative control; M, mock (noninfected sample); 3, 6, 12, 24, and 48, hours after WSSV infection. (C) Time course study of WSSV249 and PvUbc expression by real-time PCR. Expression is expressed as number of copies/reaction as determined by comparison with plasmid controls. Error bars represent the standard error of the mean taken from three independent experiments.

FIG. 2.

WSSV249-RING-H2 interacted with shrimp proteins PvUbc, TAFII p250, and snoRNP in a yeast two-hybrid assay. Specific interaction between the RING-H2 domain of WSSV249 and PvUbc, TAFII p250, and snoRNP in yeast AH109 cells were detected on a plate lacking tryptophan, leucine, and histidine. Plasmids pGBKT7-53 plus pGADT7-T were indicated as the positive control, while pGBKT7-Lam plus pGADT7-T were used as the negative control. Plasmids pGBKT7 plus pGADT7-cDNA and pGBKT7-RING plus pGADT7 were also employed as controls.

FIG. 3.

Sequence analysis of PvUbc. (A) Complete nucleotide and amino acid sequence of PvUbc. The coding sequence is represented in capital letters. (B) Alignment of PvUbc and its orthologs. Identical amino acids are indicated with an asterisk. Included are DmUbc (Drosophila melanogaster; AAS93730), DrUbc (Danio rerio; NP_001002072), HsUbc (Homo sapiens; NP_003338), and MmUbc (Mus musculus; NP_033482). The shrimp E2 PvUbc is 76.0% identical to DmUbc, 82.5% identical to DrUbc, and 81.2% identical to HsUbc and MmUbc. The alignment was performed by the Clustal method in MegAlign.

FIG. 4.

Ubiquitination assay with WSSV249-RING-H2 and PvUbc. (A) Expression and purification of GST-RING and GST-PvUBc fusion proteins. The GST protein was loaded as a control. The numbers indicate the size of the marker protein. (B) RING-H2 domain of WSSV249 mediating ubiquitination is dependent on PvUbc. Ubiquitination assays were carried out to evaluate the E3 ligase activity of the WSSV249 RING-H2 domain in the presence or absence of ubiquitin-activating enzyme (E1), PvUbc (E2), or GST. The samples were resolved on an SDS polyacrylamide gel and detected using antiubiquitin antibody in Western blots. (C) RING-H2 domain of WSSV249 mediating ubiquitination is dependent on PvUbc when detected with anti-GST antibody in Western blots. (D) An intact RING-H2 domain of WSSV249 is necessary for E3 ligase activity. The wild-type (WT) (lane 1) and mutated versions (C313S, C333S, H335Y, and C341S) of WSSV-249-RING-H2 (lanes 2 to 5) were evaluated for mediation of their own ubiquitination in a similar ubiquitination assay.

FIG. 5.

WSSV-RING-H2 exhibits a low degree of E2 specificity. GST-fused WSSV249 RING-H2 was evaluated for ubiquitination by incubation with ATP, ubiquitin, E1, and either one of the E2 enzymes (UbcH1, UbcH2, UbcH3, UbcH5a, UbcH5b, UbcH5c, UbcH6, UbcH9, and UbcH10), followed by resolving it on a 10% SDS gel and immunoblotting with anti-Ub antibody. For UbcH1 (lane 2), UbcH2 (lane 4), UbcH5a (lane 10), UbcH5b (lane 6), UbcH6 (lane 16), and UbcH10 (lane 18), polyubiquitination bands were detected.

FIG. 6.

Detection of WSSV249 and PvUbc mRNAs in the experimentally infected shrimps by in situ hybridization. The infected Penaeus vannamei was sampled 72 h postinfection. In the subcuticular epithelium of the exoskeleton, strong expression of WSSV249 (E) and PvUbc (G) was detected only in infected shrimps, as indicated by arrows. Expression was also detected in the stomach lining cells of the infected shrimps for both WSSV249 (M) and PvUbc (O). In hemocytes present in the connective tissues, strong expression of WSSV249 (U) and PvUbc (W) was detected. No expression for WSSV249 and PvUbc was observed in noninfected shrimps (A and C from the subcuticular epithelium, I and K from the stomach lining cells, Q and S from hemocytes). DIG-labeled sense RNA probes WSSV249-sense and PvUbc-sense were employed as negative controls; neither displayed any positive signal in noninfected or infected shrimps (B, F, D, and H from the subcuticular epithelium, J, N, L, and P from the stomach lining cells, R, V, T, and X from hemocytes). Scale bar = 30 μm.

FIG. 7.

Detection of PvUbc in primary cell cultures by IFA. Primary cell cultures of lymphoid organ tissue were inoculated with WSSV (D) or medium (C) (as the control) and detected by anti-PvUbc by fluorescence microscopy (magnification, ×200). Primary cell cultures in (A) and (B) under light were used as controls. The arrows indicated the positive signals for PvUbc.

FIG. 8.

Increased expression of PvUbc in infected shrimp cells by immunohistochemistry assay. Sample slides from noninfected shrimps and infected shrimps were detected with preimmune rabbit serum as a negative control (A, from noninfected shrimps; B, from infected shrimps) or the anti-PvUbc (C, from noninfected shrimps; D, from infected shrimps). Arrows indicate the positive reaction of PvUbc in the stomach epithelial cells. Scale bar = 20 μm.