Recent research on honey bee virus loads and vectors helps create a multiple feedback model to predict colony survival. Of course, varroa as a vector for virus is the key process, but I believe this framework accounts for the impact of sub-lethal pesticide contamination, and the impact of vast aggregation of colonies in the almond pollination migratory system. Both these risk factors affect virus transmission as detailed by the new research.
Several genetic adaptations to Varroa are known. These include the Minnesota Hygienic traits, Varroa Sensitive Hygiene (VSH - Baton Rouge), Africanized Grooming behavior, Africanized Queen supersedure and swarm frequency and possibly small-cell regression.
Maria Spivak and her associates have been breeding the Minnesota hygienic trait since the 1990's and of course have commercially released the bloodline. A recent review of honey bee genetics makes the crucial point that **all** bees are hygienic, in that a undertaker caste removes diseased bees, larvae and pupa. The paper notes the key benefit of the Minnesota bloodline is the hygienic behavior is very rapid, and proposes to rename the trait "Rapid Hygienic" to emphasize this point. The velocity of the trait is based on the interaction of about 7 genes (Spivak), and 2012 research (Tsuruda, 2012) finds 39 genes on two chromosomes with distinct forms in the VSH (Baton Rouge) bloodline bees. Spivak has an important popular Bee Journal paper (2009)(http://beelab.umn.edu/prod/groups/cfans/@pub/@cfans/@bees/documents/asset/cfans_asset_317498.pdf
) that notes hygienic behavior expresses in wild, uncontrolled outcrosses with decreasing frequency and effectiveness. Spivak has localized one expression of the trait to a portion of the bee's brain that learns and processes antenna odor perception -- she can turn the trait on and off by dosing the bee brain with signaling compounds or neuron suppressors.
Grooming behavior is separate trait - the tendency of (Africanized) bees to immediately remove phoretic mites by chewing their legs. Very recent research has established that this is a single gene expression. (Arechavaleta-Velasco ME, 2012 http://www.plosone.org/article/fetchObject.action?uri=info%3Adoi%2F10.1371%2Fjournal.pone.0047269&representation=PDF
). Tim Ivey and the Indiana beekeepers have posted images of public assays of chewing behavior, so the trait can be adopted by mostly European bloodline bees.
A 2011 paper by Laurent Gauthier ( http://www.plosone.org/article/fetchObject.action?uri=info%3Adoi%2F10.1371%2Fjournal.pone.0016217&representation=PDF
) finds that queens with high loads of DWV virus suffer from ovary degeneration. The DWV (and similar virii) form "crystalline arrays" in the ovaries completely disrupting them and causing loss of egg-laying ability. Gauthier makes the important observation that the enormous food demands of queens means they are "filters" that absorb enormous virus and pesticide loads -- far more likely to be affected than workers.
Rajwinder Singh ( http://www.plosone.org/article/fetchObject.action?uri=info%3Adoi%2F10.1371%2Fjournal.pone.0014357&representation=PDF
) has recently reported that DWV and similar virii are directly transported into the hive from attachment on pollen grains, and infected pollen grains are a major mode of transport from high DWV colonies to other bees, and even other species.
Yang in Taiwan University (2012) establishes that sub-lethal levels of Neonicantinoids pesticides affect larvae neuron development, specifically olfactory learning. http://www.plosone.org/article/fetchObject.action?uri=info%3Adoi%2F10.1371%2Fjournal.pone.0049472&representation=PDF
I believe much of this research can be integrated to illustrate varroa-virus-colony survival interaction.
Yang (and others have replicated this) localizes neo-nic impacts to olfactory learning. Olfactory learning (antenna detection of DWV larvae to brain processing to removal) is how hygienic behavior (slow or rapid / normal or enhanced) is accomplished, so the implication based on these two independent research is sub-lethal doses of pesticides interrupt hygienic mite suppression,
The discovery that intercolony transmission of DWV is via pollen sources that multiple populations gather implies that industrial-scale migratory beekeeping marked by enormous aggregations of (chronically infected) colonies are a risk factor in building virus levels.
Colonies with high DWV loads in my yards dwindle even after intervention and suppression of mites and visibly deformed workers. The impact of DWV on queen ovaries (viz Gauthier) confirms this observation. DWV irruptions damage the queen's fecundity. A DWV affected colony should be re-queened, the colony mites are only symptomatic of prematurely geriatric queen.
A primary adaption in my region by wild-crossing bees to Varroa is nearly constant supersedure and swarming. This is associated with Africanized bee traits, but I think anecdotal the prevalence of "my nuc is swarming" posts on this forum indicate that bee behavior even in northern areas is adopting the African swarm-early-and-often pattern. An important controlled trial on using Baton Rouge-bred VSH Bees in commercial colonies found 75% of original queens in surviving hives were replaced in a single season. (Danka, 2012 http://dx.doi.org/10.1603/EC11286
). I have thought of swarming as a "brood break" mechanism, but the discovery of crystalline virus particles in queen's ovaries mean nearly constant queen replacement is vital to colony strength.
Short of the CCD markers (foragers dying off-site, leaving behind a queen and brood), the research indicates that neo-nic pesticides can potentiate varroa-virus decline by interrupting hygienic behavior. In other words, the best genetics in the world cannot perform if the brain is too "foggy" to act.
The Spivak and Danka papers should be read together by the various advocates of "treatment-free" hobby beekeeping. Bee's are obligate out-crossers-- queens **must** breed with drones expressing other-than-inbred sex alleles. This means wild out-crossing quickly homogenizes the local breeding group, even while multiple, diverse genetics are maintained within the hive (workers are the offspring of 15-20 separate fathers).
Out-crossing both promotes (perhaps paradoxically) diversity and stability of the colony populations. This is adaptive for honeybees, as they force nectar sources to evolve to the bee -- rather than the bee adapting to rapidly evolving flower forms (as in Orchid Bees, etc). Out-crossing means bees quickly revert to the "norm" of the local population, whether they are **treated or not**. The expense and tragedy of allowing hobbyists to sacrifice 90% of their colonies in the belief they are improving genetics is a misreading of bee breeding systems.
Bees are adapting to Varroa (and Varroa is likely adapting to honey bees). They are doing this a population level, irrespective of local hobbyist colonies, (unless the colonies are surrounded by 10 kilometers of sterile cornfield and then inbreeding depression kicks in). The noted rapid adaption to Tracheal mites in the '90's (and tracheal evolution to less virulence) is evidence of population-level adaptation unmediated by active treatment and/or non-treatment. Selective breeding for foul-brood resistance was initiated in the 1930's in the Midwest, I haven't seen papers that evaluate local adaptation or propagation of those traits in the intervening decades.
Hygienic behavior (Minnesota or Baton Rouge) is a complex multi-gene expression. Mite chewing and rapid swarming (the African responses) are single gene or dominant behaviors through growth potential. The likely low-energy adaptations are going to be these simple changes - as has been observed throughout Latin America.
Selection in breeding programs is made by quantifying the hygienic response (viz. Spivak liquid nitrogen assay) and breeding only those lines. This is very different from the live-and-let-die approach by treatment-free propagandists. Luck and mischance will result is the constant loss of perfectly adapted colonies, and is accompanied by the continual reversion to the widespread "norm". Bee genotypes that are markedly distinct (Iberian, Cyprus, Crete and such) are typically the product of extreme isolation on islands or behind mountain ranges that allows a single founder population to escape the mainland homogenization (and these local races are nearly lost in our modern rapidly mobile epoch).