Skip to main content
PMC home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
letter
. 2019 Jun 12;22(8):1306–1315. doi: 10.1111/ele.13323

Knock‐on community impacts of a novel vector: spillover of emerging DWV‐B from Varroa‐infested honeybees to wild bumblebees

Robyn Manley 1,2,✉, Ben Temperton 2, Toby Doyle 1, Daisy Gates 3, Sophie Hedges 1, Michael Boots 4, Lena Wilfert 1,5
Editor: Hillary Young
PMCID: PMC6852581  PMID: 31190366

Abstract

Novel transmission routes can directly impact the evolutionary ecology of infectious diseases, with potentially dramatic effect on host populations and knock‐on effects on the wider host community. The invasion of Varroa destructor, an ectoparasitic viral vector in Western honeybees, provides a unique opportunity to examine how a novel vector affects disease epidemiology in a host community. This specialist honeybee mite vectors deformed wing virus (DWV), an important re‐emerging honeybee pathogen that also infects wild bumblebees. Comparing island honeybee and wild bumblebee populations with and without V. destructor, we show that V. destructor drives DWV prevalence and titre in honeybees and sympatric bumblebees. Viral genotypes are shared across hosts, with the potentially more virulent DWV‐B overtaking DWV‐A in prevalence in a current epidemic. This demonstrates disease emergence across a host community driven by the acquisition of a specialist novel transmission route in one host, with dramatic community level knock‐on effects.

Keywords: Bumblebee, community, deformed wing virus, honeybee, indirect disease emergence, spillover, Varroa destructor, vector, virus

Introduction

Emerging infectious diseases are an ever‐present threat to human, animal and plant populations. Disease emergence can be driven by ecological, evolutionary and anthropogenic factors (Woolhouse et al. 2005) with the acquisition of novel transmission routes by existing pathogens increasingly recognised as an important driver (Jones et al. 2008). As exemplified by the dramatic emergence of Bovine Spongiform Encephalitis (BSE) via contaminated food (Wilesmith et al. 1988), novel transmission routes can have a profound impact on the ecology, epidemiology and evolution of infectious diseases, increasing disease prevalence and creating new selection pressures on pathogens such as those apparent in human blood–borne diseases through intravenous drug use (Mathers et al. 2008) and blood transfusions (Raghwani et al. 2012). Vector‐borne transmission can drastically increase transmission rate and disease prevalence, by transmitting higher numbers of infectious particles and by‐passing infection barriers encountered during direct transmission (Ryabov et al. 2014). Transmission by a novel vector may also lead to the evolution of increased virulence (i.e. the harm suffered by the host) both because of higher infectious doses and because vector‐mediated transmission can relax selection on the trade‐off between transmission and virulence (Day 2002).

The risks associated with disease emergence due to novel transmission routes and the spread of disease vectors have been documented (Kilpatrick & Randolph 2012; Carpenter et al. 2013; Mysterud et al. 2017), but we lack empirical studies to understand how novel transmission routes impact on multi‐host pathogen epidemiology. The existence of multi‐host RNA viruses across pollinator species and the recent, rapid spread of a specialist viral vector, the honeybee mite Varroa destructor, provide a unique system to empirically study the impact of a novel vector on multi‐host pathogens. V. destructor jumped hosts from the Asian honeybee (Apis cerana) to the Western honeybee (Apis mellifera) in the early 1900s (Wilfert et al. 2016). It provided a new route of virus transmission in A. mellifera, directly transmitting viruses into the haemocoel (Ramsey et al. 2019), leading to dramatic increases in prevalence and viral load, particularly for deformed wing virus (DWV), a single positive‐stranded RNA virus (Martin et al. 2012; Mondet et al. 2014). DWV is a viral complex consisting of three characterised variants, DWV‐A, DWV‐B (also known as VDV‐1) and DWV‐C (Mordecai et al. 2015), with only DWV‐A and DWV‐B previously found in UK and French bee populations (McMahon et al. 2015; Wilfert et al. 2016). Wilfert et al. (2016) argued that the anthropogenic movement of A. mellifera is the source of a globally emerging DWV epidemic, likely driven by the concurrent spread of the V. destructor mite. Increased DWV loads in V. destructor‐infected hives are associated with colony losses due to increased overwinter mortality (Highfield et al. 2009; Berthoud et al. 2010; Genersch et al. 2010; Dainat et al. 2012). This threat to beekeeping is of grave public concern: honeybees not only produce honey and are important for our cultural heritage (Potts et al. 2016) but also pollinate wildflowers and crops.

In addition to the direct effects of the new transmission route on disease in honeybees, there is the potential for indirect disease emergence into the wider community. While V. destructor exclusively parasitises honeybees, DWV is a multi‐host virus prevalent across wild bee populations (Evison et al. 2012; Levitt et al. 2013; Fürst et al. 2014; McMahon et al. 2015), pathogenic to B. terrestris as well as honeybees (Fürst et al. 2014). The shared use of flowers is a known route of disease transmission within and between pollinator species (Durrer & Schmid‐Hempel 1994; Graystock et al. 2015; Manley et al. 2015). While V. destructor (hereafter referred to as ‘Varroa' for simplicity) has invaded the entire European mainland, several Varroa‐free island refuges remain off the British Isles and French coast. This creates a natural experiment to study how the acquisition of a novel specialist transmission route affects multi‐host pathogen epidemiology and ecology. We compared the prevalence, viral load and genotype of DWV in honeybees and bumblebees on these Varroa‐free refuges, with matched Varroa‐present mainland and island sites to show that a change to host–parasite ecology in a single host, in this case the invasion of a novel viral vector in honeybees, has community‐level effects on transmission dynamics in a multi‐host pathogen system. Through extensive sequencing and phylogenetic analysis of viral genotypes across pollinator hosts and sites, we find support for DWV spillover from honeybees to bumblebees and identify DWV‐B as the current dominant viral genotype in the UK and France that has experienced a recent bottleneck and subsequent exponential expansion, which may be driven by Varroa‐mediated transmission.

Materials and methods

We collected 355 A. mellifera, 281 B. pascuorum, 640 B. terrestris and 38 B. lucorum (differentiating between B. terrestris/lucorum via an mtDNA length polymorphism; Table S1) from 12 sites across England and France, between June and August 2015 (Table S2). The sites comprised four Varroa‐free islands (Ushant, L'Hostis 2017; Alderney, Isle of Man and St Mary's of the Isles of Scilly, FERA 2005); three Varroa‐present islands and five Varroa‐present mainland sites (FERA 2005) (Figure S1; Table S2). Bees were collected from a 1 × 1 km area while foraging on flowers, kept alive individually on wet ice, before transfer to −190° C on the day of collection (or within 48 h for Belle‐Ile and Jersey, as it was not possible to transport a dry shipper to these islands). All samples were subsequently stored at −80° C.

We macerated each bee gut individually in 200 μL of insect ringer solution to allow screening for gut parasites. DNA was extracted from 35 μL of gut solution using Chelex® 100 resin (Biorad). For RNA extractions, we used 80 μL of this gut solution as well as half the head and thorax of individuals (bisected laterally), using 1.3 mL Trizol© (Invitrogen, Carlsbad, CA, USA) for homogenisation in a tissue lyser following the manufacturer's instruction, using bromo‐chloropropane for phase separation. The RNA was resuspended in 400 µL nuclease‐free water. We converted 2 µL of RNA into first‐strand cDNA using GoScript™ Reverse Transcriptase and random hexamer primers, using RNasin® to prevent RNA degradation. To determine the prevalence of DWV‐A, DWV‐B and N. bombi and N. ceranae, cDNA and DNA was diluted 1:10 prior to PCR. We identified positive samples by PCR using GoTaq® DNA Polymerase (Table S1). To confirm the multiplex Nosema PCR bands were species‐specific, we sequenced samples of each (Table S1). We collected microsatellite data of all bumblebee samples to assess if relatedness affected disease prevalence (methods S1).

All DWV‐A‐positive samples (N = 94) and 10 randomly chosen DWV‐B‐positive samples per site/species (or total number where sample size was below 10) (N = 184) were analysed by qRT‐PCR. We measured nucleic acid quality (all samples had a 260/280 nm ratio between 1.8 and 2.1 (Nanodrop™ 2000 spectrophotometer)) and concentration (Qubit™ Fluorometer) for each individual. We synthesised cDNA from 400 ng of RNA template using GoScript™ Reverse Transcriptase and diluted it 1:10. Duplicate reactions were performed for each sample on a Stratagene machine (Mx3005P) using GoTaq® qPCR Master mix for dye‐based detection (Promega, Table S3). Two no‐template negative samples containing RNase‐free water were run per plate. Absolute quantification was calculated using duplicate eight‐point standard curves of plasmid DNA in a 1 : 10 serial dilution on each plate (Method S2). DWV is a positive strand virus whose negative strand is only present during virus replication; thus, the detection of the negative strand is a strong indicator of a true infection (de Miranda & Genersch 2010). We conducted reverse transcription with tagged virus‐specific forward primers to target the negative strand exclusively (Table S3). We adapted the qPCR assays above using tagged primers to detect replicating virus across our samples: we tested all positive Bombus samples (DWV‐A n = 18; DWV‐B n = 49) and a randomly chosen subset of 10 virus‐positive Apis samples for each virus, for the presence of the negative strand (Methods S3, Table S3).

Following Wilfert et al. (2016), all individuals identified as positive for DWV‐A or DWV‐B by the prevalence PCR were assayed by PCR for four genomic fragments; L‐protein (lp), vp3, helicase and RNA‐dependent RNA polymerase (RdRp) (Table S4), purified using Exonuclease 1 and Antarctic phosphatase (NEB) and sequenced using Big Dye Terminator v3.1 (Applied Biosystems). We manually inspected sequences in Geneious® (v.6.8); only high‐quality (< three ambiguous base pairs), non‐heterozygous sequences of a fragment‐specific minimum length were included in further sequence analysis. Not all fragments from all samples were amplified successfully (of 294 DWV‐B positives, Nlp = 116 (329 bp), Nvp3 = 195 (240 bp), Nhelicase = 144 (239 bp), Nrdrp = 142 (294 bp)); thus, distinct data sets were used for each genomic fragment, optimising information by maximising the alignment length while keeping as many samples as possible (Table S5). Note that there were too few DWV‐A sequences to pursue population genetic and phylogenetic analyses for this viral genotype. The sequences are deposited in Genbank (MG264907‐MG265503). We created alignments using Geneious® (v.6.8) by mapping the sequences to DWV‐A and DWV‐B reference sequences (NC_004830 and NC‐006494), assigning each sequence to a viral genotype. To maximise the genetic information, we concatenated the DWV‐B sequences from all samples that amplified across all four fragments (n = 58, 1108 bp). For recombination analyses, see Method S4a.

We carried out phylogenetic analysis for individual and concatenated DWV‐B fragment alignments (Table S5) in Beast 1.8 following methods described in Wilfert et al. (2016); full details in methods S5 and Table S6. We produced Maximum Clade Credibility (MCC) trees (TreeAnnotator (v1.8.4)) to infer host ancestral state probabilities. Phylogenetic trees were produced for each alignment using MrBayes 3.2.6. We produced a median joining phylogenetic network for the concatenated fragment using PopArt (v.1.7). Using DNASPv5.10.1 (Librardo & Rozas 2009), we calculated Tajima's D, Kst (Hudson et al. 1992) and the nearest neighbour statistic SNN (Hudson 2000).

We pooled 1000 ng of RNA from 30 A. mellifera and 60 B. terrestris, from two Varroa‐free island sites (Ushant and the Isle of Man) and their paired Varroa‐present mainland sites (Le Conquet and Liverpool), to create eight populations for SMRT sequencing. As B. terrestris was rare on Ushant, only 13 individuals were in this pool. Full‐length cDNA libraries were prepared using Clontech SMARTer PCR cDNA Synthesis Kit and the BluePippin System. The PacBio Template Prep Kit was used to generate SMRTbell™ libraries, which were sequenced on the PacBio System by Exeter Sequencing Service. Non‐chimeric reads from each pool were mapped against their respective host species genomes using BWA (Li & Durbin 2009) (v. 0.7.12) with the following parameters: ‘bwa mem ‐x pacbio' to remove host‐derived sequences. Remaining reads were mapped against all sequenced bee RNA viruses and 23 novel bumblebee viruses (Pascall et al., 2018) (Table S7). Reads mapping to the genomes of DWV‐A, B and C were extracted for recombination analysis (methods S4b).

All statistical analyses were carried out in RStudio (v0.99.896) (R Core Team 2018). B. lucorum samples were excluded from prevalence analyses because of low sample size (n = 38), as was the single B. pascuorum from Quiberon. True prevalence with 95% confidence intervals was calculated to account for assay efficiency and sensitivity, conservatively set at 95% (Reiczigel et al. 2010) using the R library epiR v.0.9‐82 (Stevenson 2018) and the function epi.prev; confidence intervals are calculated based on Blaker (2000). To examine if disease prevalence was affected by Varroa presence, we used generalised linear mixed models (GLMMs) with DWV‐B and Nosema spp (N. bombi and N. ceranae) prevalence tested in separate models, with binomial error distribution and logit link function, using the lme4 package (v1.1–12) (Bates & Sarkar 2006). Full models included three‐way interactions between the fixed effects Varroa presence, species (a factor with three levels: A. mellifera, B. terrestris and B. pascuorum) and island/mainland location, with latitude and sunshine hour duration as additional fixed effects; field site and individual were included as random effects (individual was added to account for overdispersion in the model (Harrison 2014)). Latitude and sunshine hours provided a proxy for favourable disease transmission conditions (Fürst et al. 2014); sunshine hours were calculated as the mean sunshine hours from monthly data between March and July 2015 collected from MET office data (pers comms) and Meteo France (http://www.meteofrance.com/climat/france). The minimum adequate model was identified through the comparison of models using anova and removal of non‐significant terms. To investigate if viral load was affected by Varroa presence, we ran GLMMs with Gamma error distribution and reciprocal link function. Viral load data varied across orders of magnitude from 103 to 1010; thus, these data were log transformed. Varroa‐free refugia only exist on islands; thus, it was necessary to also sample on Varroa‐present islands, as well as paired Varroa‐present mainland sites, to test a possible island effect on disease prevalence. To test for the effect of an island location, we ran models on reduced data sets (1) comparing island sites with and without Varroa and (2) comparing Varroa‐present islands and mainland sites.

Results

Bombus and Apis individuals were predominantly infected with DWV‐B (true prevalence: DWV‐A = 1.93% (95% CI 0.5–3.6%) and DWV B = 19.70% (CI 17.2–22.3), test of proportions χ2 = 128.541, P < 0.001). We detected virus replication of DWV‐A in 11.1% (n = 18) and DWV‐B in 2.04% (n = 49) of all virus‐positive Bombus samples; and replication of DWV‐A in 30% (n = 10) and DWV‐B in 10% (n = 10) of a subsample of virus‐positive Apis samples. The negative‐strand assays are highly conservative and can only confirm replication occurs in principle, rather than quantify or exclude it. DWV‐A was notably absent from samples collected from Varroa‐free sites apart from one B. terrestris individual on the Isle of Man (Figure S2). PacBio single molecule RNAseq data confirmed the prevalence of DWV‐B in PCR results was not an artefact of potential primer bias. Of 20,578 non‐chimeric PacBio reads greater than 1000 bp, 20,560 (99.9%) mapped to DWV‐B, with only 18 mapping to DWV‐A and none to DWV‐C. We therefore focus analysis on the more prevalent DWV‐B.

Prevalence screens show DWV‐B to be predominantly a honeybee virus whose prevalence is increased in both honeybees and bumblebees by the presence of Varroa (GLMM: estimate ± SE = 3.04 ± 1.14, P = 0.008, Table S8). Across our 12 populations (Fig. 1), DWV‐B was highly prevalent in honeybees with an average of 45.1% (n = 355) positive for the virus, with lower prevalence of 9.5% in B. pascuorum (n = 280) and 10.0% in B. terrestris (n = 641) (Fig. 2). Varroa presence is a significant predictor of the prevalence of DWV‐B for all species, predicting a c. 7‐fold increase in A. mellifera and c. 19‐fold increase in both B. terrestris and B. pascuorum (Table S10). There was no evidence that honeybees and bumblebee species respond differently to Varroa presence, as the interaction between species and Varroa did not contribute to the model fit (model comparison anova: χ2 = 3.962, P = 0.14), and was thus removed from the model. Importantly, given that all Varroa‐free sites in this study are islands, there was no evidence that island/mainland location influenced DWV‐B prevalence (GLMM: estimate ± SE = −0.59 ± 1.03, P = 0.57, Table S8), even though there is a significant interaction between island and host species (Table S8). Furthermore, comparing between Varroa‐present islands and mainland sites, island/mainland location has no effect on DWV‐B prevalence (GLMM: estimate ± SE = −0.15 ± 0.92, P = 0.87). When comparing island sites with and without Varroa, the significant effect of Varroa on DWV‐B prevalence is confirmed (GLMM: estimate ± SE = 3.31 ± 1.51, P < 0.028). Latitude and sunshine hours were not significant and excluded from all final models (Table S8). Wild bumblebee colony density was similar across all sites, with an average of 1.18 and 1.15 individuals per colony per site for B. terrestris and B. pascuorum respectively (χ2 = 2.162, P = 0.34; Kruskal–Wallis rank sum test) (Table S9).

Figure 1.

Figure 1

Open in a new tab

The prevalence of DWV‐B mapped by pollinator genus (a) A. mellifera, (b) Bombus spp.), location and Varroa presence/absence. Varroa‐free sites are black, Varroa‐present islands are light purple and Varroa‐present mainland sites are fuchsia pink. Prevalence is indicated by the size of the circles as shown in the key.

Figure 2.

Figure 2

Open in a new tab

The prevalence (with 95% confidence intervals) of DWV‐B (a) Nosema spp. (b) by host species and Varroa‐free (light grey), Varroa‐present island (dark grey) and Varroa‐present mainland sites (white).

To further confirm that the presence of Varroa itself, rather than a higher degree of isolation of Varroa‐free refugia, is driving the increase in DWV, we investigated the prevalence of microsporidian Nosema parasites (N. bombi and N. ceranae). Varroa presence did not influence Nosema prevalence (GLMM: estimate ± SE −1.16 ± 0.82, P = 0.16). There was no evidence that island/mainland location influenced Nosema prevalence (GLMM: estimate ± SE = −1.37 ± 0.80, P = 0.09). The prevalence of these parasites was affected by host species (Fig. 2) and duration of sunshine hours (Table S11), with the emerging N. ceranae only found in honeybees (Fig. S3).

Quantitative analysis of viral titres confirmed that while honeybees overall show higher titres, the presence of Varroa not only significantly increases the mean titre of DWV‐B in honeybees but also in bumblebees (Fig. 3; Table S8): For A. mellifera the predicted mean DWV‐B viral load is 1.5 × 104 viral particles per bee on Varroa‐free sites, compared to 4.2 × 106 in Varroa‐present sites (P < 0.001). In both bumblebee species, the titre also increased by more than one order of magnitude in the presence of Varroa (from 3.7 × 103 to 1.7 × 105, and 2.4 × 103 to 7.1 × 104 viral particles per bee, for B. terrestris and B. pascuorum respectively).

Figure 3.

Figure 3

Open in a new tab

DWV B viral load (with 95% confidence intervals) by host species and Varroa‐free (light grey), Varroa‐present island (dark grey) and Varroa‐present mainland sites (white).

Testing the genetic diversity and structure of DWV‐B among species, we find that the same genotypes circulate in honeybees and bumblebees, with no evidence for population differentiation within the concatenated genomic fragments based on the proportion of genetic variation between species (Kstconcat = 0.005, P > 0.05). There is, however, evidence for geographic population structure (Kstconcat = 0.245, P < 0.001); samples that are genetic nearest neighbours often come from the same population (Snnconcat = 0.657, P < 0.001), as can be seen in a Bayesian phylogenetic tree (Fig. 4). The results for each individual fragment corroborate the results of the concatenated fragments data (Table S12 and Figure S4). Successful sequencing of DWV‐B‐positive samples from Varroa‐free sites was minimal because of low titre (1 for the lp fragment and 4 for the Vp3 fragment) and was thus excluded from the concatenated data set; however, analysing this small data set, we found no evidence for population differentiation between Varroa‐free and Varroa‐positive sites (KstVp3 = 0.0001, P > 0.05), as supported by the lp and Vp3 MrBayes trees (Figure S4). Phylodynamic reconstruction in Beast 1.8 (Drummond et al. 2012) identified A. mellifera as the ancestral host of DWV‐B in these populations for the concatenated fragments (state probabilities: A. mellifera = 99.9%, B. terrestris = 6.666 × 10−4; Figure S5), reflecting the true prevalence of this virus in the sampled populations.

Figure 4.

Figure 4

Open in a new tab

Bayesian phylogeny of DWV‐B. DWV‐B sequences isolated from A. mellifera and B. terrestris. Sequences are comprised of four concatenated fragments of DWV‐B (n = 58, 1108 bp)). The tip labels are coloured by geographic location (see key); host species are A. mellifera unless indicated by an asterisk (B. terrestris).

A sliding 50 bp window analysis identified few putative recombination events within the PacBio reads that mapped to a DWV genome: Of a total of 20,578 reads, 61 contained windows that had a best hit to more than one DWV genome. When a cut‐off minimum of three consecutive windows (150 bp) was applied, only five reads (all isolated from A. mellifera at the Varroa‐present Liverpool site) were identified as putative recombinants. Only three of these showed a consistent transition from one DWV genome to another with the point of recombination within the rdrp gene (Figure S6). Using RDP4 (Martin et al. 2015) (methods S4a), we also found evidence for recombination via Sanger sequencing in the lp‐gene region, with 13 of 116 sequences showing evidence of recombination using multiple algorithms in RDP4.

The low genetic diversity of the DWV‐B phylogeny (π = 0.0051, with 71 polymorphic sites of 1108 sites of the concatenated fragments, examined over 58 sequences, (Table S5) and a star‐shaped network of sequence similarity (Fig. 5), suggests a recent bottleneck and subsequent exponential expansion. This result is supported by phylogenies for the concatenated fragments (Fig. 4 and S5). The low levels of population structure enabled us to combine sequences from across populations to investigate the past demography of the virus. We found a large excess of rare variants compared with the neutral model, suggestive of an expanding population after a bottleneck (Tajima's D for DWV‐B concatenated fragments = −2.276, P < 0.001, 95% CI: −1.62 to 1.92). This is supported for each individual fragment across populations (Tajima's D for lp = −2.187, vp3 = −2.475, helicase = −2.255, rdrp = −2.498; P < 0.001). Phylodynamic reconstruction also supports this recent expansion: the most recent common ancestor for the concatenated fragments dates back only 6 years (mean root height for concatenated fragments = 5.7 (95% HPD 2.0–10.3), which corroborates findings for individual fragments (Table S13). All DWV‐B individual and concatenated fragments showed exponential growth, with doubling rates estimated to be less than a year (doubling rate for concatenated fragment = 0.73 years (95% HPD 0.39–2.39). We find evidence of a strong expansion of DWV‐B, with honeybees in Varroa‐present locations showing an average DWV‐B prevalence of 64.8%, compared to DWV‐A at 29.2% (n = 233).

Figure 5.

Figure 5

Open in a new tab

Median joining phylogenetic network of concatenated DWV‐B sequences (n = 58), showing a star‐shaped network as expected following a rapid expansion. The colours represent sampling location, the size of the node represents the number of samples with the same sequence and the black dots on branches show the number of mutations between nodes.

Discussion

By taking advantage of the natural experiment created by the spread of the novel viral vector Varroa in honeybees, we have shown clear knock‐on community epidemiological impacts that result from ‘novel transmission route' disease emergence in a single host. The comparison of Varroa‐free and matched Varroa‐present sites demonstrates clearly that Varroa increases the prevalence and titre of DWV in its host, the honeybee, and that this causes spillover into wild bumblebees. There is higher prevalence and viraemia in bumblebees in the presence of Varroa, even though the mite only infests honeybees. Genetic analysis confirms that the same genotypes of virus are circulating in all the bees, which is consistent with DWV spilling over from the honeybees. In direct comparison, the prevalence of Nosema species (N. ceranae and N. bombi), an emerging bee disease (Fürst et al. 2014) whose oral–faecal transmission is not linked to Varroa, is unaffected by the presence of this honeybee ectoparasite. These results support reports from Hawaii of increased DWV prevalence in two sympatric pollinators (Polistes wasp species and solitary bee Ceratina smaragdula) (Santamaria et al. 2018) and a honeybee predator, Vespula pensylvanica (Loope et al. 2019) following the invasion of V. destructor (Martin et al. 2012).

Data are scarce on DWV prevalence, titre and diversity preceding the arrival of Varroa into western honeybee populations, especially in relation to wild bees. However, low‐level and benign DWV infections likely circulated in A. mellifera (de Miranda & Genersch 2010; Genersch & Aubert 2010) and are documented from Hawaii (Martin et al. 2012) and New Zealand (Mondet et al. 2014). Phylogenetic analyses suggest that the current DWV epidemic is a re‐emergence of the virus driven by anthropogenic movement of honeybees, coinciding with the invasion of a novel vector, the Varroa mite (Wilfert et al. 2016). The primary mechanism behind the dramatic effect of Varroa on DWV re‐emergence is likely the increase in titre and transmission events through the direct injection of viruses into the honeybee haemocoel while feeding on the fat body (Ramsey et al. 2019). It is also possible that the physical feeding activity of the mite itself (Kuster et al. 2014) or immunosuppression of the bee (Nazzi et al. 2012) could cause the increase in DWV. In addition, it has been suggested that Varroa drives selection on DWV leading to loss of variant diversity, and resulting in the dominance of a single master variant (Martin et al. 2012). In contrast to Martin et al. (2012), we did not find greater viral genotype diversity without Varroa (based on our PacBio viral sequence mapping results, which confirm the predominance of DWV‐B).

A surprising finding of our study is that DWV‐B is the dominant variant, rather than the globally distributed DWV‐A variant implicated in the current worldwide DWV epidemic (Martin et al. 2012; Wilfert et al. 2016). Phylogenetic analyses suggest that DWV‐B emerged in our European populations within the last decade (since 2009) and expanded exponentially after this genetic bottleneck; this result is supported by a significant excess of rare mutations in these populations. A series of surveys across similar locations provides further support for a recent exponential spread of DWV‐B: in 2009, DWV‐A dominated (Wilfert et al. 2016); in 2011, DWV‐B prevalence was high but equal to that of DWV‐A (McMahon et al. 2016); and in 2015 (present study), we find that DWV‐B is dominant. It is possible that this change in genotype prevalence is caused by Varroa‐mediated transmission exerting strong selection on DWV (Martin et al. 2012; Gisder et al. 2018), favouring DWV‐B. Evidence from laboratory competition assays simulating transmission via Varroa show that DWV‐B replicates to higher titres than DWV‐A in honeybee adults (at 9 days post‐infection) (McMahon et al. 2016) and in eclosing bees (6–7 days after inoculation as pupae) (Tehel et al. 2019). Here, we find both DWV‐A and DWV‐B to be more prevalent with higher intensities where Varroa is present, indeed with DWV‐A being almost absent from Varroa‐free sites, and thus, our data do not per se support the hypothesis that the acquisition of Varroa has caused this change in the prevalence of viral genotypes. We also find little genetic variation across the DWV‐B populations, with no population structure by host species or Varroa presence, and only modest population structure by location. However, pathogens can spread ahead of their vector if a host can carry, replicate and transmit viruses, potentially obscuring any role of Varroa‐mediated selection in the field. As our populations have endured over 20 years of Varroa infestation (FERA 2005), we might therefore find a similar diversity and the prevalence of viral variants on Varroa‐free islands through spillover via trade, travel, deliberate and accidental transportation, and possibly migration, of infected competent pollinator hosts across these highly connected locations over time.

Varroa‐mediated transmission has also been suggested to select for DWV‐A/B recombinants, with the non‐structural proteins including the replication machinery typically provided by DWV‐A and the capsid genes typically provided by DWV‐B (Ryabov et al. 2014), a pattern also found in apiaries in the UK (Moore et al. 2011; Wang et al. 2013) and Israel (Zioni et al. 2011). However, our results provide little support for this pattern via Pacbio or Sanger sequencing. We found no evidence of previously reported recombination within the N‐terminal region of the helicase gene (Dalmon et al. 2017). Instead, we found limited evidence of recombination within the rdrp and lp genes, similar to reports from the UK (Moore et al. 2011; Wang et al. 2013). While these data cannot systematically address the question of genome‐wide recombination, our results give little evidence for an important role of recombination in the current spread of DWV.

Interspecific transmission clearly occurs because the same genotypes are found across all bee species. We find DWV‐B to be far more prevalent than DWV‐A in bumblebees across our samples, which is particularly concerning in light of McMahon et al. (2016) demonstrating in laboratory studies that DWV‐B is a more virulent genotype than DWV‐A in adult A. mellifera, with unknown effects on Bombus species. Field studies also show that DWV‐B infection is linked to overwinter hive mortality (Natsopoulou et al. 2017). Furthermore, DWV‐B has recently been shown to have dramatically increased in prevalence in the USA, from only 2.7% of colonies screened in 2010 (n = 75) to 66% of apiaries screened (n = 603) in 2016 (Ryabov et al. 2017).

We present compelling evidence, in line with other studies (Fürst et al. 2014; Wilfert et al. 2016), that A. mellifera is both the ancestral and reservoir host for DWV. Significantly higher prevalence of both DWV variants in A. mellifera compared to bumblebees found in this study is consistent with this hypothesis. The prevalence of DWV in A. mellifera has been linked to prevalence in bumblebees, strongly suggesting spillover between managed honeybees and wild pollinator populations (Fürst et al. 2014). Importantly, we show that DWV‐A and B cause true infection in both honeybees and bumblebees by the detection of the virus negative strand, which is only present during replication for a positive‐sense RNA virus (Fürst et al. 2014; Radzevičiùte et al. 2017). Furthermore, the high viral loads across bumblebees, specifically in DWV‐B, combined with the effect of Varroa presence increasing DWV viral load in bumblebees, suggest that we are detecting true DWV infections in bumblebee hosts. However, we cannot confirm if higher viral loads in bumblebee hosts are simply the result of spillover of higher viral loads also recorded in sympatric honeybees, or if the recently emerged DWV‐B variant that dominates in our samples is better able to replicate to high levels in bumblebee hosts. As the same genotypes are found in Varroa‐free sites at lower prevalence and titre, spillover is the likely explanation.

Infectious diseases, and their interactions with other anthropogenic drivers of species declines, are a concern for the sustainability of wild and managed pollinator populations (Goulson et al. 2015). The levels of Varroa infestations correlate with DWV viral titres in honeybees (Nazzi et al. 2012) and therefore their potential to transmit the virus to wild pollinators. Thorough Varroa and pathogen control by beekeepers is essential for the protection of wild pollinators from disease. It further highlights the importance of establishing vector‐free refugia both for this pollinator parasite, and for multi‐host pathogen systems in general, for maintaining future biodiversity.

The spread of disease vectors through global change poses the risk of disease emergence. The introduction of avian malaria and its mosquito vector to Hawaii, for example, has led to a dramatic reduction in abundance and diversity of Hawaiian land birds (van Riper et al. 1986). Vector range expansions, facilitated by climate change, pose a significant risk to humans, wildlife and plants alike, as illustrated by the increasing spread of arboviruses such as Bluetongue in Europe in line with the range expansion of their vectors (Purse et al. 2005). This work has so far focussed on systems where all potential hosts acquire the vector. Here, however, we show that a specialist vector on one host species can change disease epidemiology throughout a multi‐host pathogen community. While the emergence of novel vectors may be a rare event, global change and direct human interference may frequently lead to the establishment of novel transmission routes within and between species, from the use of blood transfusions to alterations in animal food chains. Our work demonstrates that seemingly isolated changes within one host may impact epidemiology at a community level with important knock‐on effects due to spillover.

Author contributions

The study was designed by LW and RM, with input from MB. RM led the fieldwork, with contribution from DG. RM, DG, SH and TD performed lab work. BT contributed bioinformatics analysis of SMRT data. RM and LW analysed the data. The paper was written jointly by RM, MB and LW, with input from all authors.

Supporting information

 

Click here for additional data file. (2.5MB, docx)

Acknowledgements

This work was funded by a Royal Society Dorothy Hodgkin Fellowship and a CB Dennis Trust grant to LW, and a NERC PhD stipend and a Genetics Society Travel grant to RM. For field, lab and statistical assistance, we thank Emma Davey, Martin and Eva Jones, Meriel Anderson, Lewis Campbell, Ciara McGinley, Megan Griffiths and Tom Jenkins. The authors thank the Exeter Sequencing Service for help with Pacbio sequencing and Ken Haynes' lab for hosting RM.

The Sanger sequences that support the findings of this study have been deposited in GenBank with virus accession codes MG264907‐MG265503 and Nosema accession codes MK942707‐MK942712; SMRT reads have been archived in NCBI's Sequence Read Archive with BioProject accession number PRJNA542789. Prevalence and qPCR data that support the findings will be available from the Dryad Digital Repository: https://doi.org/10.5061/dryad.70jt240.

Data Availability Statement

The Sanger sequences that support the findings of this study have been deposited in GenBank with virus accession codes MG264907‐MG265503 and Nosema accession codes MK942707‐MK942712; SMRT reads have been archived in NCBI's Sequence Read Archive with BioProject accession number PRJNA542789. Prevalence and qPCR data that support the findings will be available from the Dryad Digital Repository: https://doi.org/10.5061/dryad.70jt240.

References

  1. Bates, D. & Sarkar, D. (2006). lme4: Linear mixed models using S4 classes. R package version 0.99875‐8 DOI: http://CRAN.R-project.org/doc/Rnews/.
  2. Berthoud, H. , Imdorf, A. , Haueter, M. , Radloff, S. & Neumann, P. (2010). Virus infections and winter losses of honey bee colonies (Apis mellifera). J. Apic. Res., 49, 60–65. 10.3896/ibra.1.49.1.08. [DOI] [Google Scholar]
  3. Blaker, H. (2000). Confidence curves and improved exact confidence intervals for discrete distributions. Canadian Journal of Statistics, 28, 783–798. [Google Scholar]
  4. Carpenter, S. , Groschup, M.H. , Garros, C. , Felippe‐Bauer, M.L. & Purse, B.V. (2013). Culicoides biting midges, arboviruses and public health in Europe. Antiviral Res., 100, 102–113. 10.1016/j.antiviral.2013.07.020. [DOI] [PubMed] [Google Scholar]
  5. Dainat, B. , Evans, D.J. , Chen, Y. , Gauthier, L. & Neumann, P. (2012). Dead or alive: Deformed wing virus and Varroa destructor reduce the life span of winter honeybees. Appl. Environ. Microbiol., 78, 981–987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dalmon, A. , Desbiez, C. , Coulon, M. , Thomasson, M. , Le Conte, Y. , Alaux, C. , et al. (2017). Evidence for positive selection and recombination hotspots in Deformed wing virus (DWV). Sci. Rep., 7, 1–12. 10.1038/srep41045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Day, T. (2002). The evolution of virulence in vector‐borne and directly transmitted parasites. Theor. Popul. Biol., 62, 199–213. 10.1006/tpbi.2002.1595. [DOI] [PubMed] [Google Scholar]
  8. Drummond, A.J. , Suchard, M.A. , Xie, D. & Rambaut, A. (2012). Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol., 29, 1969–1973. 10.1093/molbev/mss075. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Durrer, S. & Schmid‐Hempel, P. (1994). Shared use of flowers leads to horizontal pathogen transmission. Proc. R. Soc. Lond. B, 258, 299–302. [Google Scholar]
  10. Evison, S.E.F. , Roberts, K.E. , Laurenson, L. , Pietravalle, S. , Hui, J. , Biesmeijer, J.C. , et al. (2012). Pervasiveness of parasites in pollinators. PLoS ONE, 7, 1–7. 10.1371/journal.pone.0030641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. FERA (2005). Managing Varroa. The Food and Environment. Research Agency, UK. [Google Scholar]
  12. Fürst, M.A. , McMahon, D.P. , Osborne, J.L. , Paxton, R.J. & Brown, M.J.F. (2014). Disease associations between honeybees and bumblebees as a threat to wild pollinators. Nature, 506, 364–366. 10.1038/nature12977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Genersch, E. & Aubert, M. (2010). Emerging and re‐emerging viruses of the honey bee (Apis mellifera L.). Vet. Res., 41, 1–20. 10.1051/vetres/2010027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Genersch, E. , von der Ohe, W. , Kaatz, H. , Schroeder, A. , Otten, C. , Büchler, R. , et al. (2010). The German bee monitoring project: a long term study to understand periodically high winter losses of honey bee colonies. Apidologie, 41, 332–352. 10.1051/apido/2010014. [DOI] [Google Scholar]
  15. Gisder, S. , Möckel, N. , Eisenhardt, D. & Genersch, E. (2018). In vivo evolution of viral virulence: switching of deformed wing virus between hosts results in virulence changes and sequence shifts. Environ. Microbiol., 20, 4612–4628. [DOI] [PubMed] [Google Scholar]
  16. Goulson, D. , Nicholls, E. , Botías, C. & Rotheray, E.L. (2015). Bee declines driven by combined stress from parasites, pesticides, and lack of flowers. Science, 347, 1255957 10.1126/science.1255957. [DOI] [PubMed] [Google Scholar]
  17. Graystock, P. , Goulson, D. & Hughes, W.O.H. (2015). Parasites in bloom: flowers aid dispersal and transmission of pollinator parasites within and between bee species. Proc. R. Soc. B, 282, 1–7. 10.1098/rspb.2015.1371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Harrison, X.A. (2014). Using observation‐level random effects to model overdispersion in count data in ecology and evolution. Peer J, 2, 1–19. 10.7717/peerj.616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Highfield, A.C. , Nagar, A.E. , Mackinder, L.C.M. , Noël, L.M.L.J. , Hall, M.J. , Martin, S.J. , et al. (2009). Deformed wing virus implicated in overwintering honeybee colony losses. Appl. Environ. Microbiol., 75, 7212–7220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hudson, R.R. (2000). A new statistic for detecting genetic differentiation. Genetics, 155, 2011–2104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hudson, R.R. , Boos, D.D. & Kaplan, N.L. (1992). A statistical test for detecting geographic subdivision. Mol. Biol. Evol., 9, 138–151. [DOI] [PubMed] [Google Scholar]
  22. Jones, K.E. , Patel, N.G. , Levy, M.A. , Storeygard, A. , Balk, D. , Gittleman, J.L. , et al. (2008). Global trends in emerging infectious diseases. Nature, 451, 990–994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kilpatrick, A.M. & Randolph, S.E. (2012). Drivers, dynamics, and control of emerging vector‐borne zoonotic diseases. Lancet, 380, 1946–1955. 10.1016/S0140-6736(12)61151-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kuster, R.D. , Boncristiani, H.F. & Rueppell, O. (2014). Immunogene and vital transcript dynamics during parasitic Varroa destructor mite infection of developing honeybee (Apis mellifera) pupae. The Journal of Experimental Biology, 217, 1710–1718. [DOI] [PubMed] [Google Scholar]
  25. Levitt, A.L. , Singh, R. , Cox‐Foster, D.L. , Rajotte, E. , Hoover, K. , Ostiguy, N. , et al. (2013). Cross‐species transmission of honey bee viruses in associated arthropods. Virus. Res., 176, 232–240. [DOI] [PubMed] [Google Scholar]
  26. L'Hostis, M. (2017). Situation sanitaire et surveillance vis‐à‐vis de Varroa destructor sur l'île d'Ouessant. Bulletin Epidémiologique santé animale et alimentation, 81, 1–5. [Google Scholar]
  27. Li, H. & Durbin, R. (2009). Fast and accurate short read alignment with Burrows‐Wheeler Transform. Bioinformatics, 25, 1754–1760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Librardo, P. & Rozas, J. (2009). DnaSP v5: A software for comprehensive analysis of DNA polymorphism data. Bioinformatics, 25, 1451–1452. [DOI] [PubMed] [Google Scholar]
  29. Loope, K.J. , Baty, J.W. , Lester, P.J. & Rankin, E.E. (2019). Pathogen shifts in a honeybee predator following the arrival of the Varroa mite. Proc. R. Soc. B, 286, 1–9. 10.1098/rspb.2018.2499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Manley, R. , Boots, M. & Wilfert, L. (2015). Emerging viral disease risk to pollinating insects: ecological, evolutionary and anthropogenic factors. J. Appl. Ecol., 52, 331–340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Martin, S.J. , Highfield, A.C. , Brettell, L. , Villalobos, E.M. , Budge, G.E. , Powell, M. , et al. (2012). Global honey bee viral landscape altered by a parasitic mite. Science, 336, 1304–1306. 10.1126/science.1220941. [DOI] [PubMed] [Google Scholar]
  32. Martin, D.P. , Murrell, B. , Golden, M. , Khoosal, A. & Muhire, B. (2015). RDP4: Detection and analysis of recombination patterns in virus genomes. Virus . Evolution, 1, vev003 10.1093/ve/vev003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mathers, B.M. , Degenhardt, L. , Phillips, B. , Wiessing, L. , Hickman, M. , Strathdee, S.A. , et al. (2008). Global epidemiology of injecting drug use and HIV among people who inject drugs. Lancet, 372, 1733–1745. [DOI] [PubMed] [Google Scholar]
  34. McMahon, D.P. , Fürst, M.A. , Caspar, J. , Theodorou, P. , Brown, M.J.F. & Paxton, R.J. (2015). A sting in the spit: widespread cross‐infection of multiple RNA viruses across wild and managed bees. J. Anim. Ecol., 84, 615–624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. McMahon, D.P. , Natsopoulou, M.E. , Doublet, V. , Fürst, M.A. , Weging, S. , Brown, M.J.F. , et al. (2016). Elevated virulence of an emerging viral genotype as a driver of honeybee loss. Proc. R. Soc. B, 283, 20160811 10.1098/rspb.2016.0811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. de Miranda, J.R. & Genersch, E. (2010). Deformed Wing Virus. J. Invertebr. Pathol., 103, S48–S61. [DOI] [PubMed] [Google Scholar]
  37. Mondet, F. , De Miranda, J.R. , Kretzschmar, A. , Le Conte, Y. & Mercer, A.R. (2014). On the front line: Quantitative virus dynamics in honeybee (Apis mellifera L.) colonies along a new expansion front of the parasite Varroa destructor . PLoS Pathog., 10, e1004323 10.1371/journal.ppat.1004323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Moore, J.D. , Jironkin, A. , Chandler, D. , Burroughs, N. , Evans, D.J. & Ryabov, E. (2011). Recombinants between Deformed wing virus and Varroa destructor virus‐1 may prevail in Varroa destructor‐infested honeybee colonies. J. Gen. Virol., 92, 156–161. 10.1099/vir.0.025965-0. [DOI] [PubMed] [Google Scholar]
  39. Mordecai, G.J. , Wilfert, L. , Martin, S.J.M. , Jones, I.M. & Schroeder, D.C. (2015). Diversity in a honey bee pathogen: first report of a third master variant of the Deformed wing virus quasispecies. International Society for Microbial Ecology, 10, 1264–1273. 10.1038/ismej.2015.178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mysterud, A. , Jore, S. & POsteras, O. and Viljugrein, H. (2017). Emergence of tick‐borne diseases at northern latitudes in Europe: a comparative approach. Sci. Rep., 7, 1–12. 10.1038/s41598-017-15742-6.# Reply# [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Natsopoulou, M.E. , McMahon, D.P. , Doublet, V. , Frey, E. , Rosenkranz, P. & Paxton, R. (2017). The virulent, emerging genotype B of Deformed wing virus is closely linked to overwinter honeybee worker loss. Sci. Rep., 7, 1–9. 10.1038/s41598-017-05596-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Nazzi, F. , Brown, S.P. , Annoscia, D. , Del Piccolo, F. , Di Prisco, G. , Varricchio, P. , et al. (2012). Synergistic parasite‐pathogen interactions mediated by host immunity can drive the collapse of honeybee colonies. PLoS Pathog., 8, e1002735 10.1371/journal.ppat.1002735. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Pascall, D.J. , Tinsley, M.C. , Obbard, D.J. & Wilfert, L. (2018). Host evolutionary history predicts virus prevalence across bumblebee species. BioRxiv, 1–56. 10.1101/498717. [DOI] [Google Scholar]
  44. Potts, S.G. , Imperatriz‐Fonseca, V. , Ngo, H.T. , Aizen, M.A. , Biesmeijer, J.C. , Breeze, T.D. , et al. (2016). Safeguarding pollinators and their values to human well‐being. Nature, 540, 220–229. 10.1038/nature20588. [DOI] [PubMed] [Google Scholar]
  45. Purse, B.V. , Mellor, P.S. , Rogers, D.J. , Samuel, A.R. , Mertens, P.P.C. & Baylis, M. (2005). Climate change and the recent emergence of bluetongue in Europe. Nat. Rev. Microbiol., 3, 171–181. 10.1038/nrmicro1090. [DOI] [PubMed] [Google Scholar]
  46. R Core Team (2018) R: A language and environment for statistical computing. In: R Foundation for Statistical Computing Vienna, Austria. [Google Scholar]
  47. Radzevičiùte, R. , Theodorou, P. , Husemann, M. , Japoshvili, G. , Kirkitadze, G. , Zhusupbaeva, A. , et al. (2017). Replication of honey bee‐associated RNA viruses across multiple bee species in apple orchards of Georgia, Germany and Kyrgyzstan. J. Invertebr. Pathol., 146, 14–23. [DOI] [PubMed] [Google Scholar]
  48. Raghwani, J. , Thomas, X.V. , Koekkoek, S. , Schinkel, J. , Molenkamp, R. , van de Laar, T.J. , et al. (2012). Origin and evolution of the unique Hepatitis C virus circulating recombinant form 2k/1b. J. Virol., 86, 2212–2220. 10.1128/JVI.06184-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Ramsey, S.D. , Ochoa, R. , Bauchan, G. , Gulbronson, C. , Mowery, J.D. , Cohen, A. , et al. (2019). Varroa destructor feeds primarily on honey bee fat body tissue and not hemolymph. PNAS, 116, 1792–1801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Reiczigel, J. , Foldi, J. & Ozsvari, L. (2010). Exact confidence limits for prevalence of a disease with imperfect diagnostic test. Epidemiol. Infect., 138, 1674–1678. [DOI] [PubMed] [Google Scholar]
  51. van Riper, C. , van Riper, S.G. , Goff, M.L. & Laird, M. (1986). The epizootiology and ecological significance of malaria in Hawaiian land birds. Ecol. Monogr., 56, 327–344. 10.2307/1942550. [DOI] [Google Scholar]
  52. Ryabov, E.V. , Woods, R.D. , Fannon, J.M. , Moore, J.D. , Bull, J.C. , Chandler, D. , et al. (2014). A virulent strain of Deformed wing virus (DWV) of honeybees (Apis mellifera) prevails after Varroa destructor‐mediated, or In Vitro, transmission. PLoS Pathog., 10, e10004230 10.1371/journal.ppat.1004230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Ryabov, E.V. , Childers, A.K. , Chen, Y. , Madella, S. , Nessa, A. , vanEngelsdorp, D. , et al. (2017). Recent spread of Varroa destructor virus‐1, a honey bee pathogen, in the United States. Sci. Rep., 7, 17447 10.1038/s41598-017-17802-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Santamaria, J. , Villalobus, E.M. , Brettell, L. , Nikaido, S. , Graham, J.R. & Martin, S.J. (2018). Evidence of Varroa‐mediated deformed wing virus spillover in Hawaii. J. Invertebr. Pathol., 151, 126–130. [DOI] [PubMed] [Google Scholar]
  55. Stevenson, M. (2018). epiR. In: Tools for the analysis of epidemiological data. CRAN.
  56. Tehel, A. , Vu, Q. , Bigot, D. , Gogol‐Döring, A. , Koch, P. , Jenkins, C. , et al. (2019). The two prevalent genotypes of an emerging infectious disease, Deformed wing virus, cause equally low pupal mortlaity and equally high wing deformities in host honey bees. Viruses, 11, 114 10.3390/v11020114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Wang, H. , Xie, J. , Shreeve, T.G. , Ma, J. , Pallett, D.W. , King, L.A. , et al. (2013). Sequence recombination and conservation of Varroa destructor Virus‐1 and Deformed wing virus in field collected honey bees (Apis mellifera). PLoS ONE, 8, e74508 10.1371/journal.pone.0074508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Wilesmith, J.W. , Wells, G.A.H. , Cranwell, M.P. & Ryan, J.B.M. (1988). Bovine spongiform encephalopathy: epidemiological studies. Veterinary Record, 123, 638–644. [PubMed] [Google Scholar]
  59. Wilfert, L. , Long, G. , Leggett, H.C. , Schmid‐Hempel, P. , Butlin, R. , Martin, S.J.M. , et al. (2016). Deformed wing virus is a recent global epidemic in honeybees driven by Varroa mites. Science, 351, 594–597. [DOI] [PubMed] [Google Scholar]
  60. Woolhouse, M.E.J. , Haydon, D.T. & Antia, R. (2005). Emerging pathogens: the epidemiology and evolution of species jumps. Trends Ecol. Evol., 20, 238–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zioni, N. , Soroker, V. & Chejanovsky, N. (2011). Replication of Varroa destructor virus 1 (VDV‐1) and a Varroa destructor virus 1–deformed wing virus recombinant (VDV‐1–DWV) in the head of the honey bee. Virology, 417, 106–112. 10.1016/j.virol.2011.05.009. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

 

Click here for additional data file. (2.5MB, docx)

Data Availability Statement

The Sanger sequences that support the findings of this study have been deposited in GenBank with virus accession codes MG264907‐MG265503 and Nosema accession codes MK942707‐MK942712; SMRT reads have been archived in NCBI's Sequence Read Archive with BioProject accession number PRJNA542789. Prevalence and qPCR data that support the findings will be available from the Dryad Digital Repository: https://doi.org/10.5061/dryad.70jt240.

Articles from Ecology Letters are provided here courtesy of Wiley

RESOURCES