🔍 Search ARV papers · 📘 Full review PDF · 📂 Browse corpus
Apis rhabdovirus was first described in 2017. Two genotypes, ARV-1 and ARV-2, were identified in Australian honey bee surveys (Remnant et al. 2017), and in the same year Levin et al. independently confirmed and extended the finding, detecting ARV-1 in honey bees (Apis mellifera), bumble bees (Bombus impatiens) and Varroa destructor mites from the United States and Israel (Levin 2017). Before this, the bee viruses shown to be shared across species and vectored by Varroa were all positive-sense RNA viruses; ARV was the first negative-sense, enveloped virus shown to circulate this way (Levin 2017).
The naming is unsettled. Because the abbreviation “ARV” was already assigned to Adelaide River virus, and because the virus occurs in non-Apis hosts, Levin et al. proposed the alternative “bee rhabdovirus” (BRV-1, BRV-2) (Levin 2017); Galbraith et al. independently reached the same conclusion after detecting it in both A. mellifera and Bombus impatiens (Galbraith 2018). Both names (ARV-1/ARV-2 and BRV-1/BRV-2) remain in active use across the literature, which can complicate data searches.
ARV belongs to the family Rhabdoviridae (order Mononegavirales; phylum Negarnaviricota) — the same family as rabies and vesicular stomatitis virus. Virions are enveloped and bullet-shaped, reported at 100–430 nm long and 45–100 nm in diameter (Levin 2017). The genome is a single, negative-sense RNA strand carrying the five genes typical of rhabdoviruses in the canonical N–P–M–G–L organization: nucleoprotein (N), phosphoprotein (P), matrix (M), glycoprotein (G) and the large polymerase (L) (Galbraith 2018). Because the genome is negative-sense, it cannot act directly as mRNA — it must first be transcribed by a virion-packaged RNA-dependent RNA polymerase, unlike the positive-sense iflaviruses and dicistroviruses that dominate the bee-virus literature.
Phylogenetically, ARV-1 and ARV-2 form a monophyletic group with Farmington virus, a rhabdovirus of birds: their L proteins share roughly 30% (ARV-1) and 23% (ARV-2) amino-acid identity with Farmington virus (Levin 2017). The L protein is large (~2,143 aa), and across honey bee, bumble bee and Varroa isolates the ARV-1 proteins were essentially identical (99.9–100% identity) (Levin 2017). The genus has since expanded beyond the original two genotypes: nationwide sequencing in China described additional members provisionally named Apis rhabdovirus 3, 4 and 5 (Li 2023), and ARV-5 was subsequently recorded in Korea (Kwon 2023).
ARV has been detected in managed honey bees (A. mellifera), the eastern honey bee (A. cerana), bumble bees (Bombus impatiens), several wild/solitary bees, and consistently in Varroa destructor mites (Levin 2017; Galbraith 2018; Schoonvaere 2018; Li 2023). Which host it primarily replicates in is the central open question. Levin et al. presented evidence that ARV-1 actively replicates in both bees and mites, detecting the positive-sense (replicative) RNA strand in both (Levin 2017).
More recent small-RNA work shifts the emphasis toward the mite. Damayo et al. found ARV-1 and ARV-2 in almost all Varroa samples but at low prevalence in honey bees, and showed an abundant antiviral small-interfering-RNA response (a 24-nt antisense peak) against both genotypes inside mites — a signature of active replication — leading them to argue that V. destructor may be the primary host rather than the bee (Damayo 2023). A bee-virus review similarly judged the mite a “genuine host” because the viral small-RNA profiles differ between mites and bees (Yañez 2020). Crucially, ARV-1 also occurs in Bombus impatiens, which is not parasitized by Varroa — so the virus does not strictly require the mite and is likely also shared directly among co-foraging bees (Levin 2017).
ARV has a near-global footprint, reported from North America, the Middle East, Europe, Asia, South America, Africa and the South Pacific (Thaduri 2018; Li 2023). Reported prevalence varies widely with host, region, method and season. In the discovery work, 21 of 104 Israeli colonies (~20%) were ARV-1-positive, with per-apiary rates of 28.1%, 26.5% and 7.9%; within positive colonies the virus was found in only 4.8% of individual bees but 76.2% of mites, and titres reached 10⁷–10⁸ genomic copies per individual bee or mite (Levin 2017).
Subsequent surveys reinforce the wide distribution: ARV-1 and ARV-2 were detected in all seven sampled regions of China (Li 2023); across 15 European countries ARV-1 was found in 26% (n=24) and ARV-2 in 9% (n=8) of samples (Sircoulomb 2025); ARV-1 appeared in 25 of the New Zealand samples, occasionally exceeding a third of the total viral load (Lester 2022); and first-record detections were reported for Korea (as ARV-5; Kwon 2023), southern Brazil (da Silva 2023) and western Canada (Lee 2023). The prevailing pattern is high prevalence in Varroa mites and lower, patchier prevalence in bees.
Active replication has been inferred by two complementary methods: detection of the positive-sense replicative RNA strand (Levin 2017) and the presence of a characteristic 24-nt antisense small-RNA (vsiRNA) peak generated by the host RNAi response (Damayo 2023). ARV shows clear seasonal dynamics. In Swedish colonies ARV-1 titres rose from summer into autumn and peaked the following spring, increasing by roughly an order of magnitude over the season (Thaduri 2018; Thaduri 2021). A Belgian nanopore study detected ARV-1 and ARV-2 specifically in autumn (October) and only in nurse bees, not foragers — suggesting infection before adult emergence (Van Herzele 2024).
A recurrent observation is that ARV-1 and ARV-2 tend to travel together: a Czech virome study reported that bee rhabdovirus 1 and 2 were always present simultaneously, with BRV-1 more abundant than BRV-2, and present in only one of three replicate samples per hive — the first description of this paired-occurrence relationship (Kadlečková 2022). ARV is typically found as one member of a multi-virus community alongside the negative-sense Varroa orthomyxovirus-1 and the mite-associated VDV-2/VDV-3/VDV-5 viruses (Levin 2019; Li 2023; Damayo 2023).
There is, as yet, no demonstrated disease caused by ARV. It has been detected in honey bees with deformed wings in the United States, but no causal link has been shown, and the discovering authors explicitly stated that whether ARV-1 causes symptoms “remains to be determined” and that controlled studies are needed (Levin 2017). It is routinely found in apparently healthy colonies (Kadlečková 2022). The only suggestive signals are indirect: a honey bee line bred for low Varroa population growth carried significantly lower ARV-1/ARV-2 levels than a high-growth line, prompting the authors to call for investigation of ARV-2’s potential pathogenicity and effects on the bee central nervous system (Morfin 2023). On current evidence ARV is best described as a widespread, actively circulating, but biologically uncharacterized virus.
Scientific Reports · 2018 · 58 citations
Objective. Screened A. mellifera and 11 other bee species across 9 countries / 4 continents with a novel virus-discovery pipeline.
Findings:
Identified ARV-1 (family Rhabdoviridae) as one of the first negative-sense ssRNA viruses in bees, detecting it in two species (A. mellifera and Bombus impatiens) from the US (Pennsylvania) at >99% identity.
Recovered the five canonical rhabdovirus genes — L, G, N, P and M — from assembled contigs.
Given occurrence in two bee species, recommended the name Bee Rhabdovirus (BRV).
Viruses · 2019 · 39 citations
Objective. RNA metagenomics of Western/Eastern honey bee subspecies and their Varroa mites; discovery of Varroa orthomyxovirus-1 (VOV-1).
Findings:
ARV-1/BRV-1 was present in all Varroa libraries, showing the mite as a consistent reservoir; distribution differed across bee vs mite libraries.
In one A. cerana sample the dominant viral component was ARV-2, illustrating genotype-specific abundance differences.
Placed ARV within an emerging group of negative-sense bee/mite viruses (with VOV-1, VDV-2, VDV-3), raising the question of how these co-infecting viruses interact with major pathogens (DWV, ABPV, IAPV, CBPV).
Microbiome · 2023 · 27 citations
Objective. Meta-transcriptomic sequencing of ~2,000 honey bee and mite samples across China (2016–2019).
Findings:
ARV-1 and ARV-2 detected in all seven sampled Chinese regions; first evidence of ARV-1/ARV-2 (with VOV-1, VDV-2, VDV-3/5, BMLV) in China.
ARV-1 among viruses found in >75% of mite libraries, reinforcing the mite-reservoir pattern.
Expanded the genomic record: complete ARV-1 genomes from 22→29 and ARV-2 from 2→6.
Described three novel negative-sense rhabdoviruses — Apis rhabdovirus 3, 4 and 5 — extending the genus beyond ARV-1/ARV-2.
Antisense-strand (replication-intermediate) detection suggested ARV-1/ARV-2 can replicate in both honey bees and mites, whereas VDV-2/VDV-3/5 were mite-restricted.
Journal of Virology · 2023 · 26 citations
Objective. Used small-RNA (vsiRNA) sequencing of individual mites to define which viruses actually replicate in V. destructor vs the bee.
Findings:
ARV-1 and ARV-2 were present in all but one Varroa sample (the exception from South Africa), among the most prevalent mite-associated viruses.
Both genotypes elicited an abundant antiviral vsiRNA response in mites with a 24-nt antisense peak — a signature of active replication within V. destructor.
Negative-sense viruses (ARV-1/-2, VDV-2/-3/-5) showed a strong antisense read bias (60–99%).
Concluded that, given high mite prevalence and low bee prevalence, V. destructor is likely the primary host for these −ssRNA rhabdoviruses rather than A. mellifera.
Noted additional rhabdovirus diversity in Varroa jacobsoni and in A. mellifera/A. cerana in China.
Frontiers in Microbiology · 2017 · 25 citations
Objective. First characterization of ARV-1 distribution, prevalence, replication and infectivity across pollinators and Varroa, using metagenomics + RT-PCR/RT-qPCR.
Findings:
Detected ARV-1 in A. mellifera, Bombus impatiens and Varroa destructor from the US and Israel — first report of ARV-1 in a bumble bee and of rhabdovirus replication in honey bees.
Confirmed active replication in both bees and mites by detecting the positive-sense (replicative) RNA strand.
Titres reached 10⁷–10⁸ viral genomic copies per individual bee or mite.
Screened 104 Israeli colonies: 21 (~20%) ARV-1-positive; per-apiary 28.1% / 26.5% / 7.9%.
Within positive colonies: virus in 4.8% of individual bees vs 76.2% of mites; amplified in 7/10 honey bees and 1/10 bumble bees.
Genome/virology: enveloped bullet-shaped virion (100–430 x 45–100 nm); ARV-1/ARV-2 L proteins ~30%/23% identity to Farmington virus of birds, forming a monophyletic group; ARV-1 isolates 99.9–100% identical across hosts.
Because Bombus is not parasitized by Varroa, argued ARV does not require the mite and is shared among co-foraging bees; proposed the name bee rhabdovirus (BRV-1).
Stated plainly that whether ARV causes symptoms remains undetermined and controlled studies are needed.
PLoS ONE · 2018 · 25 citations
Objective. RNA-seq + RT-qPCR comparison of the mite-resistant Gotland population vs susceptible colonies across the 2009–2010 season.
Findings:
Identified ARV-1 and obtained a near-complete genome — first record of ARV-1 in Swedish bees; all known ARV-1 sequences were >99% identical.
ARV-1 titres were consistently higher in mite-susceptible than mite-resistant colonies, rising from summer to autumn 2009 and peaking in spring 2010 (differences significant in Aug and Oct 2009).
Strand-distribution evidence (uneven +/− read coverage) supported replication of ARV-1 in bees.
Confirmed near-global distribution: noted ARV-1 detections across North America, Europe, Middle East, Africa and the South Pacific.
mSystems · 2022 · 16 citations
Objective. Replicated virome sampling of healthy, Varroa-controlled colonies in Czechia at season's end.
Findings:
Reported the first description of a close relationship between two bee rhabdoviruses: BRV-1 and BRV-2 were always present together.
In every positive sample BRV-1 was more abundant than BRV-2.
The rhabdovirus pair appeared in only one of three replicate nine-bee samples per hive — highlighting strongly uneven within-hive distribution.
Found ARV in apparently healthy colonies with no disease signs, underscoring the absence of an obvious pathology.
Veterinary Research · 2024 · 3 citations
Objective. Year-long third-generation nanopore monitoring of honey bee haemolymph for pathogens not in standard screens.
Findings:
Detected ARV-1 and ARV-2 (plus AmFV and BMLV) — viruses usually missed by routine screening.
ARV-1/ARV-2 appeared specifically in autumn (October) and only in nurse bees, not foragers, implying infection before adult emergence.
Interpreted nurse-only detection as either pre-foraging mortality or successful clearance of the virus before foraging age.
Emphasized how little is known about ARV given its very recent identification.
Insects · 2020 · 132 citations
Objective. Synthesized the global diversity and distribution of A. mellifera viruses, including HTS-era discoveries.
Findings:
Frontiers in Microbiology · 2020 · 91 citations
Objective. Reviewed transmission routes of bee viruses across Apis and non-Apis hosts.
Findings:
Summarized that V. destructor was proposed as a biological vector for ARV-1/BRV-1.
Argued the mite is a genuine host for ARV-1 and ARV-2, because the viral small-RNA profiles in mites differ from those in bees.
Noted ARV-1/BRV-1 has been found in Bombus sp., and that vertical transmission of ARV remains unknown.
Viruses · 2019 · 69 citations
Objective. DWV-A vs DWV-B pupal infection experiment; virome of inocula reported.
Findings:
Frontiers in Microbiology · 2018 · 62 citations
Objective. Metatranscriptomic survey of wild Andrena, Bombus and Osmia species in Belgium.
Findings:
Found rhabdovirus-related sequences in wild bees; an Andrena cineraria RdRp had 36–42% identity to rhabdoviruses including ARV-1 and ARV-2.
Cited evidence that both bee rhabdoviruses elicit an antiviral immune response in honey bees — indirect support for genuine infection.
Frontiers in Microbiology · 2020 · 38 citations
Objective. Experimental DWV-genotype infections in Varroa/DWV-naïve Australian pupae; incidentally screened co-present viruses.
Findings:
A DWV-B inoculum contained low amounts of ARV-1 and ARV-2 (and LSV).
When pupae were injected, ARV-1 and ARV-2 were NOT transmitted/detected in recipients — evidence that injection alone did not establish ARV infection in naïve pupae.
Viruses · 2021 · 22 citations
Objective. Virus diversity/sharing among co-foraging bees; discovery of AnBV-1.
Findings:
Scientific Reports · 2022 · 19 citations
Objective. Characterized the viral community of bees and mites in New Zealand (single introduced Varroa haplotype).
Findings:
ARV-1 detected in 25 samples — usually as few transcripts, but exceeding one-third of total viral load in four samples.
ARV-2 found in nine samples but only at very low levels (≤0.02% of total viral load).
PNAS · 2023 · 19 citations
Objective. Global DWV origin/host-switch analysis; ARV among screened viruses.
Findings:
Front. Cell. Infect. Microbiol. · 2023 · 14 citations
Objective. Virome of live vs dead bees after Korea's 2021 winter losses.
Findings:
Viruses · 2023 · 11 citations
Objective. Metagenomic bee+plant virus monitoring at blueberry farms in Canada (BC and ON).
Findings:
Veterinary Sciences · 2022 · 10 citations
Objective. Methodological review of molecular detection of bee pathogens.
Findings:
Insects · 2020 · 10 citations
Objective. Effect of the almond compound amygdalin on the colony microbiome/virome.
Findings:
This hub is a curated synthesis of representative and most-cited studies — not an exhaustive catalogue. The full ARV corpus is searchable here.