Rates of single and multiple virus infection in the honey bee , Apis mellifera

Colony collapse disorder has increased interest in the role of viruses in honey bee health. In this study, we determine if management practices affect the rates of infection for six common viruses by comparing the rates of infection between migratory, queen breeder, and hobbyist beekeepers. The study found that migratory beekeepers had higher rates of infection for Israeli acute paralysis virus and Kashmir bee virus. Migratory beekeepers also had higher rates of infection by more than one virus within a single colony. These results highlight the importance of management practices, perhaps including housing bees at high density, for the prevention of viral infection. Received 14 September 2016 Received in revised form 14 December 2016 Accepted 20 December 2016

Honey bees are infected by at least 18 viruses (reviewed in CHEN;SIEDE, 2007;RUNCKEL et al., 2011), six of which have widespread distributions (GENERSCH; AUBERT, 2010).Although viruses have not consistently been a major threat to bee health in the past, this changed with the arrival of Varroa destructor, a parasitic mite that is an efficient vector of viruses (BOWEN-WALKER et al., 1999;CHEN et al., 2004a, b;EMSEN et al., 2015;MARTIN et al., 2012).Viral loads, possibly elevated by mites, could be particularly acute in commercial beekeeping operations as these involve housing bees at high densities.
This study explores the presence and prevalence of virus infections in honey bees across California, which has the highest densities of colonies in the US.It examines rates of infection for six virusesdeformed wing virus (DWV), black queen cell virus (BQCV), sacbrood virus (SBV), Kashmir bee virus (KBV), acute bee paralysis virus (ABPV), and Israeli acute paralysis virus (IAPV)in three distinct honey bee operations: migratory beekeepers, queen breeders, and hobbyists.It was predicted that the strongest levels of viral infection would be found among migratory beekeepers, as their bees are under the most stress and are exposed to the highest concentrations of bees coming from all over the country for almond pollination.

MATERIALS AND METHODS
Nine migratory beekeepers, ten hobbyist beekeepers, and three queen breeders from across the state of California participated in the study.Migratory beekeepers were defined as having at least several hundred colonies and making a significant portion of their income from either pollination or honey production.Queen breeders were defined as having large numbers of colonies for the production of queens and package bees.Hobbyists had less than 10 colonies, usually in a large backyard or on a small farm.From each migratory beekeeper and queen breeder, we chose 10 colonies from which to sample.For hobbyist beekeepers, who typically only keep a handful of colonies, 3-5 colonies were sampled.For the migratory beekeepers and queen breeders, colonies were split between two apiaries (5 colonies each).Colonies were chosen at random within apiaries.In total, 155 colonies were sampled, of which 88 were from migratory beekeepers, 30 from queen breeders, and 37 from hobbyists.Initially, foragers from the entrance of each hive were sampled.Bees were collected using an insect vacuum and were quickly transferred to 50 mL centrifuge tubes.Then, bees from the brood nest were sampled using the same method as for foragers.Both groups of bees were immediately put on dry ice for transportation to the laboratory, where they were stored at -80 °C until testing.After collection, each hive was evaluated by making a visual inspection of the following traits: amount of adhering bees on the combs (population size), number of brood combs and nectar combs (colony size), and brood pattern.Visual inspections, which are commonly conducted, are useful for identifying obviously weak and dying colonies.
Each colony was tested for six viruses using reverse transcription polymerase chain reaction (RT-PCR).Five bees from each colony were ground to a fine powder in liquid nitrogen and 75 mg of powder were then used for extraction.Total ribonucleic acid (RNA) was extracted using Trizol (Thermo Fisher Scientific Inc).cDNA was generated using Promega's GoScript Reverse Transcription System, following the manufacturer's instructions (Promega Corporation, Madison, WI -USA).RT-PCR was then carried out using a thermocycler (Eppendorf Mastercycler Pro S, Eppendorf AG, USA), programs and primers, as described in Singh et al. (2011).A positive control was used for each PCR run, together with two negative controls (water and no enzyme).PCR products were run on 1% agarose gels and stained in 0.5 ug/mL ethidium bromide before being observed under ultraviolet (UV) light.For final verification, PCR products were purified using Qiagen's QIAquick Gel Extraction Kit (Qiagen Inc, Valencia, CA) and sequenced in both directions using Sanger sequencing method.Nucleotide sequences were compared to published sequences in GenBank.For each sample, the corresponding virus taken as the best hit was that with coverage greater than 95% for all samples.One-way analysis of variance (ANOVA), ordinal logistic regression, and binary logistic regression were used as statistical tests to study the influence of management practices on colony health characteristics, and on the nature and frequency of virus infections.

DISCUSSION
The results of the visual inspections were consistent with our expectations.Our assessment of brood quality, which can be inferred by the compactness of the pattern, showed no significant differences between operation types.In general, although there was significant variation between colonies in different operation types, it was not particularly strong and was not likely to cause poor colony health due to a general lack of management.
It was found that migratory beekeepers had higher rates of infection for two viruses, IAPV and KBV, and higher rates of multiple viral infections.IAPV was one of the first viruses thought to be associated with CCD (COX-FOSTER et al., 2007).A significant association is found in this study with migratory beekeeping practices.KBV has also recently been found to be associated with CCD ( VAN ENGELSDORP et al., 2009).The pathogenicity of this virus is unknown; however, its lower rates of infection overall (60% for IAPV compared to 22% for KBV) make it less likely to be a culprit for major losses than IAPV.
With respect to multiple infections, this study confirms that migratory beekeepers have higher rates than stationary beekeepers.This is a significant result in that one possibility for the rise in viral loads across the US could be due to queen breeders, who are producing a higher percentage of the country's bees than ever before, and shipping sick bees throughout the country.This does not seem to be the case, however, given the relatively low rates of virus infection in queen breeding operations.
Given that the bees kept by some migratory beekeepers are periodically stored in holding yards with thousands of colonies from around the country, while queen breeders and hobbyists keep their bees in relative isolation, there is a clear mechanism that could cause this pattern.Of course, this hypothesis is speculative at this point, and future work will need to look carefully at rates of infection in migratory bees before and after the pollination season.Studies that look at the fate of newly founded (and relatively healthy) colonies before and after pollination will be particularly useful.

Table 1 -
Results of visual inspections of colony state.

Table 2 -
Rates of virus infection in different types of beekeeping operations.Multiple and range refer to the mean number and range of different viruses infecting a single colony.