Multi-tiered analyses of honey bees that resist or succumb to parasitic mites and viruses

Avoiding the tragedies of parasite tolerance in Darwinian beekeeping

Bee declines have been partly attributed to the impacts of invasive or emerging parasite outbreaks. For western honeybees, Apis mellifera, major losses are associated with the virus-vectoring mite, Varroa destructor. In response, beekeepers have focused breeding efforts aimed at conferring resistance to this key parasite. One method of many is survival-based beekeeping where colonies that survive despite significant Varroa infestations produce subsequent colonies. We argue that this 'hands-off' approach will not always lead to Varroa resistance evolving but rather tolerance. Tolerance minimizes host fitness costs of parasitism without reducing parasite abundance, whereas resistance either prevents parasitism outright or keeps parasitism intensity low. With clear epidemiological distinctions, and as honeybee disease dynamics impact other wild bees owing to shared pathogens, we discuss why tolerance outcomes in honeybee breeding have important implications for wider pollinator health. Crucially, we argue that unintentional selection for tolerance will not only lead to more spillover from honeybees but may also select for pathogens that are more virulent in wild bees leading to 'tragedies of tolerance'. These tragedies can be avoided through successful breeding regimes that specifically select for low Varroa. We emphasize how insights from evolutionary ecology can be applied in ecologically responsible honeybee management.

Quantitative trait loci mapping for survival of virus infection and virus levels in honey bees

Israeli acute paralysis virus (IAPV) is a highly virulent, Varroa-vectored virus that is of global concern for honey bee health. Little is known about the genetic basis of honey bees to withstand infection with IAPV or other viruses. We set up and analyzed a backcross between preselected honey bee colonies of low and high IAPV susceptibility to identify quantitative trait loci (QTL) associated with IAPV susceptibility. Experimentally inoculated adult worker bees were surveyed for survival and selectively sampled for QTL analysis based on SNPs identified by whole-genome resequencing and composite interval mapping. Additionally, natural titers of other viruses were quantified in the abdomen of these workers via qPCR and also used for QTL mapping. In addition to the full dataset, we analyzed distinct subpopulations of susceptible and non-susceptible workers separately. These subpopulations are distinguished by a single, suggestive QTL on chromosome 6, but we identified numerous other QTL for different abdominal virus titers, particularly in the subpopulation that was not susceptible to IAPV. The pronounced QTL differences between the susceptible and non-susceptible subpopulations indicate either an interaction between IAPV infection and the bees' interaction with other viruses or heterogeneity among workers of a single cohort that manifests itself as IAPV susceptibility and results in distinct subgroups that differ in their interaction with other viruses. Furthermore, our results indicate that low susceptibility of honey bees to viruses can be caused by both, virus tolerance and virus resistance. QTL were partially overlapping among different viruses, indicating a mixture of shared and specific processes that control viruses. Some functional candidate genes are located in the QTL intervals, but their genomic co-localization with numerous genes of unknown function delegates any definite characterization of the underlying molecular mechanisms to future studies.

Transcriptomic Responses Underlying the High Virulence of Black Queen Cell Virus and Sacbrood Virus following a Change in Their Mode of Transmission in Honey Bees (Apis mellifera)

BACKGROUND

Over the last two decades, honey bees (Apis mellifera) have suffered high rates of colony losses that have been attributed to a variety of factors, chief among which are viral pathogens, such as deformed wing virus (DWV), whose virulence has increased because of vector-based transmission by the invasive, ectoparasitic varroa mite (Varroa destructor). A shift in the experimental mode of transmission of the black queen cell virus (BQCV) and sacbrood virus (SBV) from fecal/food-oral (direct horizontal) to vector-mediated (indirect horizontal) transmission also results in high virulence and viral titers in pupal and adult honey bees. Agricultural pesticides represent another factor that acts independently or in interaction with pathogens, and they are also thought to cause colony loss. Understanding the molecular mechanisms underlying the higher virulence following a vector-based mode of transmission provides deeper insight into honey bee colony losses, as does determining whether or not host-pathogen interactions are modulated by exposure to pesticides.

METHODS

Through an experimental design with controlled laboratory, we investigated the effects of the modes of transmission of BQCV and SBV (feeding vs. vector-mediated via injection) alone or in combination with chronic exposure to sublethal and field-realistic concentrations of flupyradifurone (FPF), a novel agricultural insecticide, on honey bee survival and transcription responses by using high-throughput RNA sequencing (RNA-seq) analysis.

RESULTS

Co-exposure to viruses via feeding (VF) or injection (VI) and FPF insecticide had no statistically significant interactive effect on their survival compared to, respectively, VF or VI treatments alone. Transcriptomic analysis revealed a distinct difference in the gene expression profiles of bees inoculated with viruses via injection (VI) and exposed to FPF insecticide (VI+FPF). The number of differentially expressed genes (DEGs) at log2 (fold-change) > 2.0 in VI bees (136 genes) or/and VI+FPF insecticide (282 genes) was very high compared to that of VF bees (8 genes) or the VF+FPF insecticide treatment (15 genes). Of these DEGs, the expression in VI and VI+FPF bees of some immune-related genes, such as those for antimicrobial peptides, Ago2, and Dicer, was induced. In short, several genes encoding odorant binding proteins, chemosensory proteins, odor receptors, honey bee venom peptides, and vitellogenin were downregulated in VI and VI+FPF bees.

CONCLUSIONS

Given the importance of these suppressed genes in honey bees' innate immunity, eicosanoid biosynthesis, and olfactory associative function, their inhibition because of the change in the mode of infection with BQCV and SBV to vector-mediated transmission (injection into haemocoel) could explain the high virulence observed in these viruses when they were experimentally injected into hosts. These changes may help explain why other viruses, such as DWV, represent such a threat to colony survival when transmitted by varroa mites.

Signatures of Adaptation, Constraints, and Potential Redundancy in the Canonical Immune Genes of a Key Pollinator

All organisms require an immune system to recognize, differentiate, and defend against pathogens. From an evolutionary perspective, immune systems evolve under strong selective pressures exerted by fast-evolving pathogens. However, the functional diversity of the immune system means that different immune components and their associated genes may evolve under varying forms of selection. Insect pollinators, which provide essential ecosystem services, are an important system in which to understand how selection has shaped immune gene evolution as their populations are experiencing declines with pathogens highlighted as a potential contributing factor. To improve our understanding of the genetic variation found in the immune genes of an essential pollinator, we performed whole-genome resequencing of wild-caught Bombus terrestris males. We first assessed nucleotide diversity and extended haplotype homozygosity for canonical immune genes finding the strongest signatures of positive selection acting on genes involved in pathogen recognition and antiviral defense, possibly driven by growing pathogen spread in wild populations. We also identified immune genes evolving under strong purifying selection, highlighting potential constraints on the bumblebee immune system. Lastly, we highlight the potential loss of function alleles present in the immune genes of wild-caught haploid males, suggesting that such genes are potentially less essential for development and survival and represent redundancy in the gene repertoire of the bumblebee immune system. Collectively, our analysis provides novel insights into the recent evolutionary history of the immune system of a key pollinator, highlighting targets of selection, constraints to adaptation, and potential redundancy.

Viral species differentially influence macronutrient preferences based on honey bee genotype

Food quantity and macronutrients contribute to honey bee health and colony survival by mediating immune responses. We determined if this held true for bees injected with chronic bee paralysis virus (CBPV) and deformed wing virus (DWV), two common honey bee ssRNA viruses. Pollen-substitute diet and syrup consumption rates and macronutrient preferences of two Varroa-resistant stocks (Pol-Line and Russian bees) were compared to Varroa-susceptible Italian bees. Bee stocks varied in consumption, where Italian bees consumed more than Pol-Line and Russian bees. However, the protein: lipid (P:L) ratios of diet consumed by the Italian and Russian bees was greater than that of the Pol-Line bees. Treatment had different effects on consumption based on the virus injected. CBPV was positively correlated with syrup consumption, while DWV was not correlated with consumption. P:L ratios of consumed diet were significantly impacted by the interaction of bee stock and treatment, with the trends differing between CBPV and DWV. Variation in macronutrient preferences based on viral species may indicate differences in energetic costs associated with immune responses to infections impacting different systems. Further, virus species interacted with bee genotype, indicating different mechanisms of viral resistance or tolerance among honey bee genotypes.

Honey Bee Genetic Stock Determines Deformed Wing Virus Symptom Severity but not Viral Load or Dissemination Following Pupal Exposure

Honey bees exposed to Varroa mites incur substantial physical damage in addition to potential exposure to vectored viruses such as Deformed wing virus (DWV) that exists as three master variants (DWV-A, DWV-B, and DWV-C) and recombinants. Although mite-resistant bees have been primarily bred to mitigate the impacts of Varroa mites, mite resistance may be associated with increased tolerance or resistance to the vectored viruses. The goal of our study is to determine if five honey bee stocks (Carniolan, Italian, Pol-Line, Russian, and Saskatraz) differ in their resistance or tolerance to DWV based on prior breeding for mite resistance. We injected white-eyed pupae with a sublethal dose (105) of DWV or exposed them to mites and then evaluated DWV levels and dissemination and morphological symptoms upon adult emergence. While we found no evidence of DWV resistance across stocks (i.e., similar rates of viral replication and dissemination), we observed that some stocks exhibited reduced symptom severity suggestive of differential tolerance. However, DWV tolerance was not consistent across mite-resistant stocks as Russian bees were most tolerant, while Pol-Line exhibited the most severe symptoms. DWV variants A and B exhibited differential dissemination patterns that interacted significantly with the treatment group but not bee stock. Furthermore, elevated DWV-B levels reduced adult emergence time, while both DWV variants were associated with symptom likelihood and severity. These data indicate that the genetic differences underlying bee resistance to Varroa mites are not necessarily correlated with DWV tolerance and may interact differentially with DWV variants, highlighting the need for further work on mechanisms of tolerance and bee stock-specific physiological interactions with pathogen variants.

Virome Analysis Reveals Diverse and Divergent RNA Viruses in Wild Insect Pollinators in Beijing, China

Insect pollinators provide major pollination services for wild plants and crops. Honeybee viruses can cause serious damage to honeybee colonies. However, viruses of other wild pollinating insects have yet to be fully explored. In the present study, we used RNA sequencing to investigate the viral diversity of 50 species of wild pollinating insects. A total of 3 pathogenic honeybee viruses, 8 previously reported viruses, and 26 novel viruses were identified in sequenced samples. Among these, 7 novel viruses were shown to be closely related to honeybee pathogenic viruses, and 4 were determined to have potential pathogenicity for their hosts. The viruses detected in wild insect pollinators were mainly from the order Picornavirales and the families Orthomyxoviridae, Sinhaliviridae, Rhabdoviridae, and Flaviviridae. Our study expanded the species range of known insect pollinator viruses, contributing to future efforts to protect economic honeybees and wild pollinating insects.

Context-Dependent Viral Transgenerational Immune Priming in Honey Bees (Hymenoptera: Apidae)

Transgenerational immune priming is the process of increased resistance to infection in offspring due to parental pathogen exposure. Honey bees (Apis mellifera L. (Hymenoptera: Apidae)) are hosts to multiple pathogens, and this complex immune function could help protect against overwhelming infection. Honey bees have demonstrated transgenerational immune priming for the bacterial pathogen Paenibacillus larvae; however, evidence for viral transgenerational immune priming is lacking across insects in general. Here we test for the presence of transgenerational immune priming in honey bees with Deformed wing virus (DWV) by injecting pupae from DWV-exposed queens and measuring virus titer and immune gene expression. Our data suggest that there is evidence for viral transgenerational immune priming in honey bees, but it is highly context-dependent based on route of maternal exposure and potentially host genetics or epigenetic factors.