The Interaction of Bluetongue Virus VP6 and Genomic RNA Is Essential for Genome Packaging

Abstract

The genomes of the Reoviridae, including the animal pathogen bluetongue virus (BTV), are multisegmented double-stranded RNA (dsRNA). During replication, single-stranded (ss) positive-sense RNA segments are packaged into the assembling virus capsid, triggering genomic dsRNA synthesis. However, exactly how this packaging event occurs is not clear. A minor capsid protein, VP6, unique for the orbiviruses, has been proposed to be involved in the RNA-packaging process. In this study, we sought to characterize the RNA binding activity of VP6 and its functional relevance. A novel proteomic approach was utilized to map the ssRNA/dsRNA binding sites of a purified recombinant protein and the genomic dsRNA binding sites of the capsid-associated VP6. The data revealed that each VP6 protein has multiple distinct RNA-binding regions and that only one region is shared between recombinant and capsid-associated VP6. A combination of targeted mutagenesis and reverse genetics identified the RNA-binding region that is essential for virus replication. Using an in vitro RNA-binding competition assay, a unique cell-free assembly assay, and an in vivo single-cycle replication assay, it was possible to identify a motif within the shared binding region that binds BTV ssRNA preferentially in a manner consistent with specific RNA recruitment during capsid assembly. These data highlight the critical roles that this unique protein plays in orbivirus genome packaging and replication.IMPORTANCE Genome packaging is a critical stage during virus replication. For viruses with segmented genomes, the genome segments need to be correctly packaged into a newly formed capsid. However, the detailed mechanism of this packaging is unclear. Here we focus on VP6, a minor viral protein of bluetongue virus, which is critical for genome packaging. We used multiple approaches, including a robust RNA-protein fingerprinting assay, to map the ssRNA binding sites of recombinant VP6 and the genomic dsRNA binding sites of capsid-associated VP6. By these means, together with virological and biochemical methods, we identify the viral RNA-packaging motif of a segmented dsRNA virus for the first time.

Keywords: RNA-protein interaction; double-stranded RNA virus; genome packaging.

FIG 1

RNA packaging in viral particles relies on VP6. (A) A virus with truncated VP6 was used to infect BSR or BSR-VP6 cells. Twelve hours later, the viral particles were harvested and purified, and the positive (+) and negative (–) strands, representing genomic ssRNA and dsRNA, respectively, were measured by qRT-PCR. (B) BTV ssRNAs in the cell lysate from the experiment for which results are shown in panel A were measured by qRT-PCR. (C) A CFA assay was performed either by the original method (control), in the absence of VP6, or with VP6 added at a later time point. The assembled complex was purified and packaged, and viral RNA was measured by qRT-PCR. Mean values ± standard deviations are shown (n = 3).

FIG 2

Regions of the BTV VP6 that contact RNA. The yellow bar represents the VP6 protein, with the vertical dashes denoting the positions of positively and negatively charged amino acids. The colored lines represent peptides identified in the RCAP analyses performed with recombinant VP6 protein (reVP6) and either single-stranded or double-stranded RNAs and with VP6 from purified cores. Details of the peptides are presented in Table 1. The three regions in reVP6 that contact RNAs are named Re1 to Re3. Regions in the core-associated VP6 that contact packaged genomic RNAs are named Ca1 to Ca4. Mutations were introduced on 13 specific charged sites (top), and the sites at which mutations had lethal effects are marked with asterisks.

FIG 3

Plaque formations of WT and mutant viruses.

FIG 4

Mutant VP6 is expressed similarly to wild-type (WT) VP6. Four mutants (KK_246-7AA, KK_246-7EE, RK_257-8AA, and RK_257-8EE) and WT VP6 encoding S9 RNAs were used to transfect BSR cells. Protein expression and localization were monitored by immunofluorescence staining and confocal microscopy. Green, VP6; blue, Hoechst staining.

FIG 5

Conformational analysis of mutant reVP6. (A) Five mutant reVP6 proteins were expressed in E. coli and analyzed by SDS-PAGE together with wild-type (WT) VP6, followed by Coomassie blue staining. The sizes on the protein ladder are indicated. (B) CD spectra of VP6 mutants normalized to those of wild-type VP6. Each mutant and wild-type VP6 is indicated by a different color.

FIG 6

VP6 shows a BTV RNA binding preference. (A) RK_208-9EE mutant, RK_257-8EE mutant, or WT VP6 was incubated with ³²P-labeled BTV S10 ssRNA for interaction. The shifting of the VP6-RNA complex was observed by EMSA using a 0.8% agarose gel and TBE buffer before analysis with s phosphorimager. (B) WT or RK_257-8EE reVP6 was bound to Ni beads and was incubated with BTV ssRNA S10 in the absence or presence of a 100-fold quantity of tRNA (+tRNA). The bound RNA was then eluted and quantified by qRT-PCR. Mean values ± standard deviations are shown (n = 3).

FIG 7

Mutations in the Ca2/Re3 region destroy BTV RNA preferential binding. A competition assay was performed using 1 μg of WT (A), RK_257-8AA (B), KK_246-7AA (C), or RK_208-9EE (D) reVP6 and 0.1 μg of ³²P-labeled BTV S10. Xenopus elongation factor (Xef) mRNA was added in the quantities indicated. The size of free S10 RNA is indicated.

FIG 8

Mutation on VP6 prohibits genomic RNA packaging. A CFA assay was performed in the presence of WT VP6 (control) or RK_257-8EE VP6 or in the absence of VP6. The assembled complex was purified and the packaged viral RNA measured by qRT-PCR. Mean values ± standard deviations are shown (n = 3).

FIG 9

A VP6 mutant virus is deficient at producing the dsRNA genome. The VP6 RK_257-8EE mutant virus was used to infect BSR cells (RK₂₅₇-₈EE) (red line) or BSR-VP6 cells (WT) (blue line). Total cytoplasmic RNA was harvested at 0, 2, 4, and 6 hpi. The positive-strand (+) and negative-strand (–) RNA, representing genomic ssRNA (left) and dsRNA (right), respectively, were measured by qRT-PCR. Mean values ± standard deviations are shown (n = 3).