NC State University

Current Projects

  1. Effects of genetic diversity on honey bee colony phenotype
  2. Assessing the mating 'health' of honey bee queens
  3. Molecular mechanisms of honey bee mating
  4. Potential roles of nutrition and genotype on Colony Collapse Disorder (CCD)
  5. Competition among honey bee queens during the requeening process
  6. Non-chemical control of varroa mites
  7. Mapping putative feral honey bee populations

Our lab studies the behavioral ecology of insect societies, with a primary focus on the proximate and ultimate mechanisms of honey bee queen behavior. In doing so, we attempt to address questions of basic science that have practical relevance. Our philosophy is to integrate a general understanding of bee biology to help improve overall colony health and productivity; in an era when the honey bee population is being severely impacted by any number of factors, we feel that it is incumbent upon the honey bee scientific community to become more proactive in asking questions that address not just basic (long-term) or applied (short-term) questions, but both.

Effects of genetic diversity on honey bee colony phenotype

Our lab has a protracted history of investigations into the adaptive benefits of intracolony genetic diversity. Specifically, we have investigated the fitness advantages of queen polyandry (mating with multiple males) and its consequences on colony phenotype. Increased genetic diversity makes it less likely that any particular trait is overly prevalent within a colony. The fathers of the workers, a queen's mates, carry different alleles that vary with respect to many traits. Because of the haplodiploid genome of honey bees, a queen that mates once produces genetically similar workers that all carry the same alleles from their father. If, by chance, his alleles are unfavorable—for example, susceptible to a particular disease—then all of the workers will be of low genetic quality and the colony would be impacted severely under adverse conditions. A queen that mates multiply, however, produces genetically diverse workers that carry different alleles from their collective fathers. By doing so, a queen reduces the risk that all of her worker offspring will be of low genetic quality, increasing the probability that the colony, as a whole, will be diverse enough to overcome a wide range of ecological conditions. We have empirically tested several non-mutually exclusive hypotheses that have shown the adaptive benefits of genetic diversity to minimize the impact of parasites and pathogens (Tarpy, 2003; Tarpy and Seeley, 2006; Seeley and Tarpy, 2007) and homozygosity at the complementary sex determination (csd) locus (Tarpy and Page, 2001, 2002).

Assessing the mating 'health' of honey bee queens

As an extension of this theoretical work, one of our newest research paradigms is to investigate mating number of naturally mated honey bee queens. There is both anecdotal and empirical evidence to suggest that many queens produced in the U.S. may not be adequately mated. Consequently, we are investigating the current status of the "mating health" of honey bee queens by looking at the physical quality, insemination success, and mating numbers of a large sample of queen bees. To measure mating number, we use PCR using microsatellite markers to genotype workers in colonies and infer the mating numbers of individual queens (Tarpy and Page, 2000; Tarpy and Nielsen, 2002; Tarpy et al., 2004b). This is a large-scale project that will enable us perform some very powerful investigations into the evolution of extreme polyandry in honey bees, but it will also ascertain whether or not a lack of genetic diversity within honey bee colonies may partially explain the global ill-health of commercial beehives.

Molecular mechanisms of honey bee mating

Together with Dr. Christina Grozinger's lab (now at Penn State), we have another research project to investigate the regulation of honey bee reproduction by looking at the genomic and physiological changes in queen bees during mating. We are using techniques such as microarray gene-expression, classical behavioral observation, and GC-MS to determine how virgin queens (who are receptive to mating) transition to laying queens (who never mate again in their entire lifetimes). Comparisons of virgin, partially mated, and fully mated queens have revealed that physiological changes in ovary maturation and pheromone profiles correlate with gene-expression changes in the ovaries, while changes in behavior correlate with gene-expression changes in the brain (Kocher et al., 2008). We have also shown that queens inseminated with semen from a single drone produce significantly different pheromone blends that are less attractive than those from queens inseminated with semen from 10 drones (Richard et al., 2007).

Potential roles of nutrition and genotype on Colony Collapse Disorder (CCD)

The accelerated losses of honey bee colonies in the fall and winter of 2006-07 resulted in the bringing together of a working group to explore the causes of a condition we have termed colony collapse disorder, or CCD. This disorder is characterized by the rapid loss of adult worker bees from affected colonies with a resultant weak or dying colony with excess brood. The CCD working group is focused on three main areas. First, the potential influence of nutritional health on colonies, either as a potential cause of CCD or an effect of another factor. Linked to colony health is the genotype of colonies, such that there may be potential genotype-environment interactions that may be important in the incidence of the disorder. Second, the possible influence of pathology on CCD-affected colonies, either a more virulent version of an existing parasite or potentially a currently unknown disease. Third, the possible influence of environmental contaminants, either from external sources (e.g., pesticides), internal sources (e.g., acaricides used to control varroa mites), or some combination of sources. We are concentrating on the first of the above potential causes, namely the role of nutritional stress and genotype on the incidence and prevalence of CCD. We have quantified whole-bee protein levels as a proxy for colony nutritional stress. This measure will help us determine if CCD is associated with poor colony nutrition, either as a potential cause (e.g., stressed colonies are more susceptible to pathogen attack) or as a potential effect (e.g., collapsing colonies are less able to acquire sufficient forage to maintain proper colony health and function). Finding such an association will prompt hypothesis testing of colony nutrition to determine cause versus effect and possible mitigating management practices. We are also currently genotyping colonies using microsatellite markers to determine if intracolony genetic diversity has any bearing on the disorder.

Competition among honey bee queens during the requeening process

We also have an extended interest in the dramatic worker-queen interactions during queen production in honey bee colonies. When honey bee colonies rear new queens, they raise many but then leave the queens fight to the death for only one to reclaim the nest. The fitness discrepancy between "winning" and "losing" queens cannot be greater: losing queens die (i.e., zero fitness), whereas winning queens not only survive, they inherit an established, successful nest with all the workers and inherent resources. Clearly, natural selection ought to be acting very strongly on any behavioral mechanism that governs which queen(s) survives in order to bias the process in favor of certain queens. The question is, how? We have investigated the potential roles that worker may play in biasing the outcomes of competitive duels (Tarpy and Fletcher, 1998; Tarpy et al., 2000; Gilley et al., 2003; Gilley and Tarpy, 2005), the physical advantages of individual rival queens (Tarpy and Mayer, in review) including the fighting tactics that they may use against each other (Tarpy and Fletcher, 2003), and how selection may act on colonies at the individual- and colony levels (sometimes in opposing directions; Hatch et al., 1999; Tarpy and Gilley, 2004; Tarpy et al., 2004a). As such, this rich behavioral paradigm affords opportunities to answer questions about levels of selection, kinship theory, reproductive skew, and classic behavioral competition.

Non-chemical control of varroa mites

Varroa mites (Varroa destructor) preferentially parasitize drone brood (developing males) over worker brood because their development time is longer compared to workers, enabling the female mite, or foundress, an opportunity to produce a greater number of daughter offspring. Because of this increase in reproductive potential, mites have developed the ability to recognize and seek out drone brood based on olfactory, gustatory, and mechanical cues. This differential attraction to drone brood makes it a promising target for mite control efforts, and previous studies have shown significant control by removing drone frames and freezing them (along with their mites). We are testing a new variation on this technique by "rescuing" the drones after removing the mites from them. This manipulation of drone brood could potentially benefit the honey bees at the colony level as well as the population level. The colony itself will benefit by decreasing infestation of Varroa destructor and encouraging overall colony health and productivity. On a population level, it is possible that surviving drones that live to sexual maturity may possess some element of resistance to parasitism by the Varroa mite. This research is being conducted by Holly Wantuch, a Masters student in our program.

Mapping putative feral honey bee populations

Feral honey bee populations once flourished in North America. Upon the arrival and establishment of a parasitic brood mite, Varroa destructor, feral honey bee populations were largely decimated. While it has been hypothesized that these once-thriving feral populations served as a genetic reservoir of diversity for managed honey bee populations in the United States, no empirical evidence is available to support this claim, to date. In response to growing concerns about levels of genetic diversity in managed honey bee populations in the United States, plans to import additional honey bee germplasm are currently being addressed. This leads to the question of the status of feral populations. Research and anecdotal evidence suggest that the feral population may be recovering (see for a current map). We are currently investigating the status of feral honey bee populations in the U.S. by re-sampling non-managed honey bee populations in the southeastern United States to determine its genetic composition and to finally address whether feral populations are truly "survivor stock" or simply recent "escaped swarms". Results from this study will reveal the amount of genetic diversity available within U.S. honey bee populations and will allow us to determine the extent of genetic loss. This research is being conducted by Dr. Debbie Delaney, a USDA Postdoctoral Fellow in our program.