Synthesised from the abstracts and full texts of the most-cited and representative papers within a harvested corpus of 1,482 records (1,206 open access). This is a curated review of landmark findings, not an exhaustive catalogue of every study. Generated 2026-06-12.
🔍 Search DWV papers · 📘 Full review PDF · 📂 Browse corpus
Deformed wing virus (DWV) is a positive-sense, single-stranded RNA virus of the family Iflaviridae and is now the most prevalent and intensively studied viral pathogen of the honey bee (Apis mellifera). For most of the 20th century DWV was an obscure, largely harmless covert infection. Its transformation into a global killer was driven almost entirely by one event: the host switch and worldwide spread of the ectoparasitic mite Varroa destructor, which transmits DWV directly into developing bee pupae and amplifies it to lethal titres (Martin 2012; Wilfert 2016). DWV is the leading viral factor in overwintering colony losses, occurs in at least three genotypes (DWV-A, DWV-B, DWV-C) that recombine readily, suppresses bee immunity in a mutualistic loop with the mite, and has spilled over into bumblebees, wasps and dozens of other arthropods. Unlike most bee viruses, its causal pathology — wing deformity, shortened abdomen, early death — is experimentally established. This document synthesises the published findings by theme and then catalogues the landmark studies.
DWV is a member of the genus Iflavirus, family Iflaviridae — small, non-enveloped viruses with a monopartite, positive-sense RNA genome of ~9–11 kb that infect arthropods, mostly insects (Valles 2017). The DWV genome is 10,140 nucleotides with a single large open reading frame encoding a ~328-kDa polyprotein, flanked by a 1,144-nt 5′ leader and a 317-nt 3′ region with a poly(A) tail (Lanzi 2006). The polyprotein yields three major capsid proteins (VP1 ~44 kDa, VP2 ~32 kDa, VP3 ~28 kDa) at its N-terminus and the non-structural enzymes — an RNA helicase, a chymotrypsin-like 3C protease and an RNA-dependent RNA polymerase — at its C-terminus (Lanzi 2006). Because it is positive-sense, the genome functions directly as mRNA. Its close relatives include Kakugo virus and Varroa destructor virus-1 (VDV-1, now recognised as DWV-B), which share ~84% genome identity and replicate in the mite (Ongus 2004).
In the absence of Varroa, DWV persists as a low-level covert infection that causes no visible symptoms and little measurable harm (de Miranda 2010). The arrival of Varroa destructor changed everything: the mite is not merely a mechanical carrier but a biological vector in which DWV replicates before being injected into pupae (Bowen-Walker 1999; Gisder 2009). The clearest evidence is a natural experiment in Hawaii, where Varroa's arrival raised DWV prevalence from ~10% to 100% of bees, drove a millionfold increase in viral titre, and collapsed DWV genetic diversity to a single dominant strain (Martin 2012). At the global scale, phylogeographic analysis shows DWV is a recent, human-driven pandemic spreading from European A. mellifera via trade and colony movement, with Varroa providing the direct transmission route (Wilfert 2016). Mites capable of inducing overt, deformed-wing infections carry 10¹⁰–10¹² viral genome equivalents, versus a maximum of ~10⁸ in mites that cannot (Gisder 2009).
DWV is unusual among bee viruses in having an experimentally demonstrated causal pathology. When the mite transmits DWV to pupae, infected bees emerge with shrivelled, deformed wings, a bloated and shortened abdomen, and discolouration; they are non-viable and die soon after emergence (de Miranda 2010). Deformed bees carry roughly 10⁶ times more virus than mite-parasitised but normal-winged bees (Yang 2005), and in crippled bees the virus is present throughout the body including the head, whereas asymptomatic bees show it only in thorax/abdomen (Yue 2005). The route and dose of infection determine the outcome: injecting pupae produces overt, deformed-wing disease with an overt-infection dose around 2,500 genome equivalents, and injecting adults with >10⁷ equivalents also yields overt infection, whereas the same dose fed orally produces only covert infection (Möckel 2011). This dose- and route-dependence explains why mite vectoring — which delivers virus directly to the haemocoel — is so much more pathogenic than natural oral/food-borne transmission.
DWV exists as a quasispecies — a swarm of related variants — resolved into master variants. DWV-A was the originally described genotype; DWV-B (formerly VDV-1) is more virulent in adult bees and has been increasing in prevalence in Europe and North America (McMahon 2016). A third master variant, DWV-C, was identified by deep sequencing and estimated to have diverged from the others roughly three centuries ago (Mordecai 2016). The genotypes recombine extensively in vivo, and DWV/VDV-1 recombinants can accumulate to higher levels than either parent in both bees and mites — the genome appears modular, with the 5′-UTR, capsid and non-structural regions evolving semi-independently (Moore 2011; McMahon 2016). Critically, Varroa-mediated or in-vitro transmission selects for a near-clonal, virulent DWV variant out of the diverse covert population, whereas oral exposure preserves diversity (Ryabov 2014). This selection for virulent, transmissible variants is central to DWV's emergence.
DWV interacts with the bee immune system and with other stressors in ways that amplify its impact. Varroa parasitism suppresses bee immunity — down-regulating antimicrobial peptides and immune enzymes — which activates latent DWV replication (Yang 2005; Shen 2005). At the molecular level this is mediated by suppression of NF-κB immune signalling, converting a cryptic, vertically transmitted virus into a rapidly replicating killer late in the season (Nazzi 2012). The relationship between mite and virus is mutualistic: DWV's immunosuppression of the bee actually enhances Varroa reproduction, creating a reciprocal loop with escalating damage to colony health (Di Prisco 2016). External stressors feed in: the neonicotinoid insecticide clothianidin (and imidacloprid) suppresses NF-κB-controlled antiviral defences and promotes DWV replication in bees carrying covert infections (Di Prisco 2013), and poor nutrition raises viral loads (DeGrandi-Hoffman 2010).
DWV uses multiple, well-characterised transmission routes. Vector transmission by Varroa into pupae is the most pathogenic (Bowen-Walker 1999). Vertical transmission occurs from both queens and drones to offspring through eggs and DWV-positive semen — true covert infection maintained across generations (Chen 2006; Yue 2007). Oral/food-borne transmission occurs via virus-contaminated larval food, pollen and honey; viruses in pollen pellets are infective and can establish infection when fed to virus-free colonies (Singh 2010; Yue 2005). Horizontal transmission among adults occurs through phoretic mites and through nurse bees cannibalising infected pupae (Möckel 2011). The interplay of these routes lets DWV persist covertly in the absence of Varroa and erupt into overt disease in its presence.
Although named for honey bees, DWV is a multi-host pathogen. Field and experimental data show honey bees are the likely source of DWV spillover into wild bumblebees, which share the same DWV strains and can develop the same wing deformities (Fürst 2014; Genersch 2006). DWV prevalence in honey bees predicts prevalence in sympatric bumblebees (McMahon 2015). It has been detected — with evidence of active replication — in numerous non-Apis species: at least six in one survey (Levitt 2013), and about a third of sampled bumblebees and wasps in another (Evison 2012), though notably not in all pollinator groups, so it is not a universal pollinator parasite. The most recent review records DWV in 65 arthropod species spanning eight insect orders and three orders of Arachnida (Martin 2019), making it one of the most host-versatile insect viruses known and a recognised threat to wild pollinator conservation.
DWV is a leading viral driver of overwintering colony mortality. It is among the few pathogens that independently predict colony loss: DWV load and prevalence correlate with overwintering failure even in mite-controlled colonies (Highfield 2009), and DWV (with Varroa) shortens the lifespan of winter bees, providing a proximate mechanism for collapse (Dainat 2012). Prevalence is high and global — DWV was found in 97% of French apiaries in an early survey (Tentcheva 2004), and a recent review reports a minimum average of 55% of colonies/apiaries infected across 32 countries (Martin 2019). Control remains difficult and indirect: managing Varroa is the primary lever, alongside breeding tolerant bees and RNA-interference (RNAi) approaches. A notable advance engineered the bee gut symbiont Snodgrassella alvi to produce double-stranded RNA that activates the bee's RNAi defence and can also kill Varroa (Leonard 2020), pointing toward future symbiont-mediated therapeutics.
Science · 2012 · 430 citations
Objective. Used Varroa's arrival in Hawaii as a natural experiment to track DWV prevalence, load and diversity.
Findings:
Varroa raised DWV prevalence from ~10% to 100% of bees within populations.
Accompanied by a millionfold increase in viral titre.
Caused a massive collapse of DWV strain diversity to a single dominant variant.
Concluded the global spread of Varroa selected DWV variants that made it one of the most widespread, contagious insect viruses on Earth.
PNAS · 2013 · 422 citations
Objective. Tested whether neonicotinoid exposure undermines antiviral immunity and promotes DWV.
Findings:
Clothianidin negatively modulates NF-κB immune signalling in insects.
It up-regulates a leucine-rich-repeat inhibitor of NF-κB, reducing antiviral defences.
This promotes replication of DWV in bees bearing covert infections; imidacloprid does the same, chlorpyriphos does not.
Effect occurs at sublethal, field-relevant doses — implications for pesticide risk assessment.
Nature · 2014 · 411 citations
Objective. Infection experiments + landscape-scale field data on DWV spillover to bumblebees.
Findings:
DWV prevalence in honey bees and bumblebees is linked; sympatric individuals share the same DWV strains.
Honey bees have higher DWV prevalence, identifying Apis as the likely source of this emerging disease in wild pollinators.
Highlights the need for pathogen control in managed bees to protect declining wild pollinators.
J. Invertebrate Pathology · 2010 · 360 citations
Objective. Foundational review of DWV genetics, pathobiology and transmission.
Findings:
Without Varroa, DWV causes no visible symptoms or apparent fitness cost.
Varroa-mediated transmission to pupae causes pupal death and adults with deformed wings, bloated shortened abdomens and discolouration; such bees die soon after emergence.
Framed DWV within RNA-virus concepts of genetic variability, virulence evolution and disease-symptom development.
PNAS · 2005 · 328 citations
Objective. Tested the mechanism by which Varroa harms bees, via immunity and DWV replication.
Findings:
Varroa parasitism significantly suppressed expression of immunity-related genes (antimicrobial peptides and enzymes).
DWV titres were negatively correlated with immune-enzyme expression.
Deformed-wing bees carried ~10⁶ times more DWV than mite-infested but normal-winged bees.
Immune challenge dramatically boosted DWV titres, indicating immunosuppression activates latent infection.
Science · 2016 · 320 citations
Objective. Phylogeographic reconstruction of DWV's global spread and the role of Varroa and trade.
Findings:
DWV is globally distributed and recently spread from a common source, European Apis mellifera.
Shows epidemic growth driven by European and North American honey bee movement and trade.
Varroa provides the direct transmission route fuelling a worldwide, man-made epidemic.
PLoS Pathogens · 2012 · 294 citations
Objective. Integrated population and molecular analysis of Varroa-infested collapsing colonies.
Findings:
Varroa de-stabilises within-host DWV dynamics, turning a cryptic vertically transmitted virus into a rapidly replicating killer that peaks late in the season.
De-stabilisation is linked to immunosuppression via strong down-regulation of NF-κB.
Proposed NF-κB as a 'common currency' underlying multifactorial colony collapse.
PLoS Pathogens · 2014 · 250 citations
Objective. Tested whether Varroa amplifies/selects virulent DWV strains or suppresses host immunity.
Findings:
Varroa-free pupae carried low levels of a highly diverse DWV population.
Varroa-associated pupae either had low diverse DWV or high levels of a near-clonal virulent variant.
Varroa-mediated and in-vitro transmission selected the virulent variant from the diverse population.
All tested mites contained a diverse replicating DWV population implying the virulent variant.
J. Invertebrate Pathology · 1999 · 241 citations
Objective. Established Varroa as a DWV vector under field conditions.
Findings:
Varroa mites are highly effective vectors of DWV between bees.
Mites carried virus titres far in excess of their hosts', suggesting DWV may replicate within the mite.
DWV-positive bees showed wing deformity and/or died during pupation; a positive relationship between mite numbers and deformity/death.
Proposed a threshold concentration of DWV in pupae is required to cause pathology.
Journal of Virology · 2006 · 233 citations
Objective. Cloned and sequenced the DWV genome and mapped its proteins.
Findings:
Genome is 10,140 nt with a single ORF encoding a 328-kDa polyprotein, plus a 1,144-nt 5′ leader and 317-nt 3′ region with poly(A) tail.
Identified three capsid proteins VP1 (44 kDa), VP2 (32 kDa), VP3 (28 kDa) at the polyprotein N-terminus.
C-terminal non-structural region contains an RNA helicase, 3C-like protease and RdRp.
Genome organization and capsid place DWV in the genus Iflavirus.
Applied & Environmental Microbiology · 2012 · 219 citations
Objective. 6-month colony monitoring linking pathogens to winter-bee longevity.
Findings:
Workers in colonies that failed to overwinter had reduced lifespan, were more likely DWV-infected and carried higher DWV loads.
Varroa infestation and individual DWV infection were associated with reduced life expectancy.
Nosema ceranae and ABPV were NOT correlated with longevity — singling out Varroa+DWV as the parsimonious mechanism for colony loss.
J. General Virology · 2004 · 201 citations
Objective. Characterised VDV-1, the DWV relative now recognised as DWV-B.
Findings:
Described 27-nm virus-like particles replicating in mite tissue.
VDV-1 genome is 10,112 nt, ~84% identity to DWV and Kakugo virus, one large ORF (2,893-aa polyprotein).
Negative-strand RT-PCR confirmed VDV-1 and DWV both replicate in the mite.
Applied & Environmental Microbiology · 2009 · 196 citations
Objective. Year-long qPCR monitoring of DWV/ABPV/BQCV in mite-controlled colonies.
Findings:
Despite Varroa control, a significant correlation existed only between DWV load and overwintering colony loss.
Suggests DWV can act independently of heavy mite infestation to cause colony losses.
Positions DWV as a potentially major factor in overwintering mortality.
PNAS · 2016 · 180 citations
Objective. Tested the mechanistic and evolutionary basis of the Varroa-DWV association.
Findings:
DWV vectored by Varroa suppresses humoral and cellular immunity by interfering with NF-κB signalling.
This immunosuppression enhances Varroa reproduction.
Identified an unrecognised mutualistic symbiosis: a reciprocal loop with escalating negative effects on bee immunity and health.
J. General Virology · 2009 · 172 citations
Objective. Tested whether DWV replication in the mite determines overt disease.
Findings:
Mites able to induce overt (deformed-wing) infection contained 10¹⁰–10¹² DWV genome equivalents per mite.
Mites unable to induce crippled wings contained at most ~10⁸ equivalents.
Overt disease depends not just on transmission but on DWV replication and titre within the vectoring mite.
PLoS Pathogens · 2014 · 169 citations
Objective. Tracked viral landscapes as Varroa spread across New Zealand over a decade.
Findings:
DWV titres in bees kept increasing with Varroa-infestation history even as infestation rates dropped.
Linked to rising DWV titres in the mites themselves.
Suggests DWV levels in mites, boosted by replication, maintain the DWV epidemic after initial establishment.
Science · 2020 · 168 citations
Objective. Engineered the bee gut symbiont Snodgrassella alvi to deliver RNAi.
Findings:
Engineered S. alvi stably recolonised bees and produced dsRNA to trigger host RNAi.
Improved bee survival after viral challenge.
The same approach killed parasitic Varroa mites by triggering the mite's RNAi response — a route toward symbiont-mediated DWV/Varroa control.
Applied & Environmental Microbiology · 2006 · 165 citations
Objective. Examined queen tissues/feces and offspring for viruses including DWV.
Findings:
DWV (and other viruses) detected in queen feces and tissues including ovaries and spermatheca.
Viruses present in queens were detected in their eggs, larvae and adult workers — first evidence of vertical transmission in honey bee colonies.
Proc. Royal Society B · 2016 · 153 citations
Objective. Compared virulence of DWV genotypes A and B via lab experiments and field survey.
Findings:
DWV-B is more virulent than the established DWV-A and is widespread in the landscape.
Modelling showed colonies infected with DWV-B collapse sooner than those with DWV-A.
Revealed extensive genome-wide recombination between genotypes in vivo.
Emergence of DWV-B, including via recombination with DWV-A, has major ecological/economic implications.
Annual Review of Virology · 2019 · 149 citations
Objective. Comprehensive review of DWV epidemiology, host range, structure and evolution.
Findings:
DWV is the most prevalent honey bee virus: a minimum average 55% of colonies/apiaries infected across 32 countries.
Detected in 65 arthropod species spanning eight insect orders and three orders of Arachnida.
Reviews progress on capsid structure, expanding host range, genome evolution and recombination.
Sets goals: identify tissues where variants replicate and understand DWV impact in non-honeybee hosts.
Added 2026-06-23 from a scan of recent, lightly-cited papers — see Research Frontier for the full review and caveats. These are recent single studies; treat as leads, not settled fact.
This hub is a curated synthesis of representative and most-cited studies — not an exhaustive catalogue. The full DWV corpus is searchable here.