Selective Breeding for Low and High Varroa destructor Growth in Honey Bee (Apis mellifera) Colonies: Initial Results of Two Generations.

After two years of bidirectional selection for low and high rates of Varroa destructor population growth (LVG and HVG, respectively) in honey bee (Apis mellifera) colonies in Ontario, Canada, significant differences between the two genotypes were observed. LVG colonies had V. destructor population increases over the summer of 1.7 fold compared to 9.6 fold for HVG colonies by Generation 2. Additionally, HVG colonies had significantly higher mite infestation rates in adult bees compared to LVG colonies for both selected generations. DWV prevalence and levels were significantly higher in HVG colonies than in LVG colonies in Generation 1 but not in Generation 2. Winter mortality rates of Generation 1 colonies were significantly different at 26% and 14% for the HVG and LVG genotypes, respectively. The results of this study thus far indicate that selection for LVG may result in colonies with lower V. destructor infestation rates, lower prevalence, and levels of DWV and higher colony winter survivorship. Future work will focus on determining what mechanisms are responsible for the genotypic differences, estimating genetic parameters, and molecular analyses of the genotypes to identify candidate genes associated with resistance to V. destructor and DWV that could potentially be used for marker-assisted selection.

Evaluation of Dry and Wet Formulations of Oxalic Acid, Thymol, and Oregano Oil for Varroa Mite (Acari: Varroidae) Control in Honey Bee (Hymenoptera: Apidae) Colonies.

The efficacy and safety of dry and wet formulations of three nonsynthetic compounds, oxalic acid (OA), thymol (T), and oregano oil (OO), for the control of Varroa destructor Anderson and Trueman infestations in honey bee (Apis mellifera Linnaeus) colonies were determined. The treatments were OA in dust, OA diluted in glycerin solvent embedded in a towel, T in dust, T in glycerin solvent and towel, OO in dry microcapsules, OO in glycerin solvent and towel, and the control. The treatments were applied weekly for 4 wk during the fall season. The rates of acaricide efficacy, weekly mite fall, bee mortality, colony survivorship, and strength, were determined for each of the treatments. All formulations, with the exemption of OO microcapsules, were effective at controlling infestations of V. destructor. The most effective formulations were T dust (96.6%), T glycerin (92.4%), and OA glycerin (79%). More than 85% of the mites were killed during the first 2 wk of treatment with T formulations, compared to less than 30% for the OA glycerin formulation. The lowest efficacy rate was for OO microcapsules (21.3%), and the only treatment that significantly increased bee mortality was OA glycerin. The rates of winter colony survival and honey bee populations were related to the varroacidal efficacy of the formulations. The implications of these findings are discussed.

Seasonality of Nosema ceranae Infections and Their Relationship with Honey Bee Populations, Food Stores, and Survivorship in a North American Region.

Nosema ceranae is an emerging pathogen of the western honey bee (Apis mellifera L.), and thus its seasonality and impact on bee colonies is not sufficiently documented for North America. This study was conducted to determine the infection intensity, prevalence, and viability of N. ceranae in >200 honey bee colonies during spring, summer, and fall, in a North American region. We also determined the relationship of N. ceranae infections with colony populations, food stores, bee survivorship, and overwinter colony mortality. The highest rates of N. ceranae infection, prevalence, and spore viability were found in the spring and summer, while the lowest were recorded in the fall. N. ceranae spore viability was significantly correlated with its prevalence and infection intensity in bees. Threshold to high levels of N. ceranae infections (>1,000,000 spores/bee) were significantly associated with reduced bee populations and food stores in colonies. Furthermore, worker bee survivorship was significantly reduced by N. ceranae infections, although no association between N. ceranae and winter colony mortality was found. It is concluded that N. ceranae infections are highest in spring and summer and may be detrimental to honey bee populations and colony productivity. Our results support the notion that treatment is justified when infections of N. ceranae exceed 1,000,000 spores/bee.

Differential Gene Expression Associated with Honey Bee Grooming Behavior in Response to Varroa Mites.

Honey bee (Apis mellifera) grooming behavior is an important mechanism of resistance against the parasitic mite Varroa destructor. This research was conducted to study associations between grooming behavior and the expression of selected immune, neural, detoxification, developmental and health-related genes. Individual bees tested in a laboratory assay for various levels of grooming behavior in response to V. destructor were also analyzed for gene expression. Intense groomers (IG) were most efficient in that they needed significantly less time to start grooming and fewer grooming attempts to successfully remove mites from their bodies than did light groomers (LG). In addition, the relative abundance of the neurexin-1 mRNA, was significantly higher in IG than in LG, no groomers (NG) or control (bees without mite). The abundance of poly U binding factor kd 68 and cytochrome p450 mRNAs were significantly higher in IG than in control bees. The abundance of hymenoptaecin mRNA was significantly higher in IG than in NG, but it was not different from that of control bees. The abundance of vitellogenin mRNA was not changed by grooming activity. However, the abundance of blue cheese mRNA was significantly reduced in IG compared to LG or NG, but not to control bees. Efficient removal of mites by IG correlated with different gene expression patterns in bees. These results suggest that the level of grooming behavior may be related to the expression pattern of vital honey bee genes. Neurexin-1, in particular, might be useful as a bio-marker for behavioral traits in bees.

Higher prevalence and levels of Nosema ceranae than Nosema apis infections in Canadian honey bee colonies.

This study was conducted to determine the prevalence and infection levels of the microsporidia fungi Nosema apis and/or Nosema ceranae in honey bee colonies of two Canadian provinces. Three surveys were conducted in the springs of 2008, 2010 and 2012 and PCR identification of Nosema species were performed in samples from 169 and 181 Ontario colonies and from 76 Alberta colonies that tested positive to Nosema spp. Infection levels of positive colonies were determined by microscopy and analyzed by Nosema spp. Results showed that N. ceranae was the dominant species in all three surveys (prevalence range of 41-91 vs. 4-34 % for N. apis), whereas mixed infections were less frequent than single infections (5-25 %). Infection levels of colonies parasitized by N. ceranae were three to five times higher than those of colonies parasitized by N. apis in the three surveys whereas mixed infections showed the highest spore counts. This is the first field study demonstrating significantly higher infection levels in colonies parasitized with either N. ceranae only or with both, N. ceranae and N. apis, than in colonies parasitized with N. apis only. Taken together, these results suggest that N. ceranae may be more virulent and better adapted than N. apis in cold climates such as Canadian environments.

Lower virus infections in Varroa destructor-infested and uninfested brood and adult honey bees (Apis mellifera) of a low mite population growth colony compared to a high mite population growth colony.

A comparison was made of the prevalence and relative quantification of deformed wing virus (DWV), Israeli acute paralysis virus (IAPV), black queen cell virus (BQCV), Kashmir bee virus (KBV), acute bee paralysis virus (ABPV) and sac brood virus (SBV) in brood and adult honey bees (Apis mellifera) from colonies selected for high (HMP) and low (LMP) Varroa destructor mite population growth. Two viruses, ABPV and SBV, were never detected. For adults without mite infestation, DWV, IAPV, BQCV and KBV were detected in the HMP colony; however, only BQCV was detected in the LMP colony but at similar levels as in the HMP colony. With mite infestation, the four viruses were detected in adults of the HMP colony but all at higher amounts than in the LMP colony. For brood without mite infestation, DWV and IAPV were detected in the HMP colony, but no viruses were detected in the LMP colony. With mite infestation of brood, the four viruses were detected in the HMP colony, but only DWV and IAPV were detected and at lower amounts in the LMP colony. An epidemiological explanation for these results is that pre-experiment differences in virus presence and levels existed between the HMP and LMP colonies. It is also possible that low V. destructor population growth in the LMP colony resulted in the bees being less exposed to the mite and thus less likely to have virus infections. LMP and HMP bees may have also differed in susceptibility to virus infection.

Genotypic variability and relationships between mite infestation levels, mite damage, grooming intensity, and removal of Varroa destructor mites in selected strains of worker honey bees (Apis mellifera L.).

The objective of this study was to demonstrate genotypic variability and analyze the relationships between the infestation levels of the parasitic mite Varroa destructor in honey bee (Apis mellifera) colonies, the rate of damage of fallen mites, and the intensity with which bees of different genotypes groom themselves to remove mites from their bodies. Sets of paired genotypes that are presumably susceptible and resistant to the varroa mite were compared at the colony level for number of mites falling on sticky papers and for proportion of damaged mites. They were also compared at the individual level for intensity of grooming and mite removal success. Bees from the "resistant" colonies had lower mite population rates (up to 15 fold) and higher percentages of damaged mites (up to 9 fold) than bees from the "susceptible" genotypes. At the individual level, bees from the "resistant" genotypes performed significantly more instances of intense grooming (up to 4 fold), and a significantly higher number of mites were dislodged from the bees' bodies by intense grooming than by light grooming (up to 7 fold) in all genotypes. The odds of mite removal were high and significant for all "resistant" genotypes when compared with the "susceptible" genotypes. The results of this study strongly suggest that grooming behavior and the intensity with which bees perform it, is an important component in the resistance of some honey bee genotypes to the growth of varroa mite populations. The implications of these results are discussed.