The impacts of invertebrate RNA virus population dynamics on virulence and infection outcomes are poorly understood. Deformed wing virus (DWV), the main viral pathogen of honey bees, negatively impacts bee health, which can lead to colony death. Despite previous reports on the reduction of DWV diversity following the arrival of the parasitic mite Varroa destructor, the key DWV vector, we found high genetic diversity of DWV in infested United States honey bee colonies. Phylogenetic analysis showed that divergent US DWV genotypes are of monophyletic origin and were likely generated as a result of diversification after a genetic bottleneck. To investigate the population dynamics of this divergent DWV, we designed a series of novel infectious cDNA clones corresponding to coexisting DWV genotypes, thereby devising a reverse-genetics system for an invertebrate RNA virus quasispecies. Equal replication rates were observed for all clone-derived DWV variants in single infections. Surprisingly, individual clones replicated to the same high levels as their mixtures and even the parental highly diverse natural DWV population, suggesting that complementation between genotypes was not required to replicate to high levels. Mixed clone–derived infections showed a lack of strong competitive exclusion, suggesting that the DWV genotypes were adapted to coexist. Mutational and recombination events were observed across clone progeny, providing new insights into the forces that drive and constrain virus diversification. Accordingly, our results suggest that Varroa influences DWV dynamics by causing an initial selective sweep, which is followed by virus diversification fueled by negative frequency-dependent selection for new genotypes. We suggest that this selection might reflect the ability of rare lineages to evade host defenses, specifically antiviral RNA interference (RNAi). In support of this hypothesis, we show that RNAi induced against one DWV strain is less effective against an alternate strain from the same population.

Honey bees, the primary managed insect pollinator, suffer considerable losses due to Deformed wing virus (DWV), an RNA virus vectored by the mite Varroa destructor . Mite vectoring has resulted in the emergence of virulent DWV variants. The basis for such changes in DWV is poorly understood. Most importantly, it remains unclear whether replication of DWV occurs in the mite. In this study, we exposed Varroa mites to DWV type A via feeding on artificially infected honey bees. A significant, 357-fold increase in DWV load was observed in these mites after 2 days. However, after 8 additional days of passage on honey bee pupae with low viral loads, the DWV load dropped by 29-fold. This decrease significantly reduced the mites’ ability to transmit DWV to honey bees. Notably, negative-strand DWV RNA, which could indicate viral replication, was detected only in mites collected from pupae with high DWV levels but not in the passaged mites. We also found that Varroa mites contain honey bee mRNAs, consistent with the acquisition of honey bee cells which would additionally contain DWV replication complexes with negative-strand DWV RNA. We propose that transmission of DWV type A by Varroa mites occurs in a non-propagative manner.

We developed a honey bee RNA-virus vector based on the genome of a picorna-like Deformed wing virus (DWV), the main viral pathogen of the honey bee (Apis mellifera). To test the potential of DWV to be utilized as a vector, the 717 nt sequence coding for the enhanced green fluorescent protein (eGFP), flanked by the peptides targeted by viral protease, was inserted into an infectious cDNA clone of DWV in-frame between the leader protein and the virus structural protein VP2 genes. The in vitro RNA transcripts from egfp-tagged DWV cDNA clones were infectious when injected into honey bee pupae. Stable DWV particles containing genomic RNA of the recovered DWV with egfp inserts were produced, as evidenced by cesium chloride density gradient centrifugation. These particles were infectious to honey bee pupae when injected intra-abdominally. Fluorescent microscopy showed GFP expression in the infected cells and Western blot analysis demonstrated accumulation of free eGFP rather than its fusions with DWV leader protein (LP) and/or viral protein (VP) 2. Analysis of the progeny egfp-tagged DWV showed gradual accumulation of genome deletions for egfp, providing estimates for the rate of loss of a non-essential gene an insect RNA virus genome during natural infection.