• Mann Lake Ltd.

Thanks to these sponsors, you can enjoy this website without annoying popup ads!
You can show your appreciation by clicking on their banners above to go directly to their websites.

Print Friendly, PDF & Email

Sick Bees – Part 8: Time for a Paradigm Shift!

First published in: American Bee Journal April 2011


Understanding Viruses. 2

Bee/Virus Coevolution. 2

Viral Quasispecies. 2

Regional Bee Stocks. 3

The Natural Situation Prior to Varroa. 5

What’s Changed?. 7

The Seasonal Progression of Virus Infections. 7

Enter Varroa. 9

Things are Different These Days. 12

Varroa and Viruses. 14

Holding Yards. 17

Beekeeper-Applied Miticides. 17

Varroa changed everything. 19

Practical Wrap Up. 19

A Paradigm Shift. 20

Suggested Reading and References. 23


 Sick Bees—Part 8

Time for a Paradigm Shift!

Randy Oliver


First published in American Bee Journal April 2011

Bee health issues completely changed with the invasion of the varroa mite. Beekeeping success these days is largely dependent upon managing the mite level in your hives. However, it’s not varroa that actually kills colonies; rather, it is the bee viruses, which is why I’ve been belaboring the subject. Mite management should be considered as only one facet in the overall context of virus management. I feel that it’s time for a paradigm shift in the way that we look at bee health issues!

Understanding Viruses

Just as the successful beekeeper these days must understand varroa biology, it helps greatly to understand virus biology, which is why I’ve been going into such great detail on the subject. Let’s return to the discussion…

Bee/Virus Coevolution

To understand the process of evolution, one must first disabuse himself of any misconception that evolution has any plan, goal, or rules of fairness. Look at it from a Zen perspective—it just happens, and no telling where it will go! Whatever beats the competition is favored; everything else is cast by the wayside without regret or empathy. Like Danny DeVito explained in the film “Other People’s Money,” no matter how exquisite and perfect the now outdated buggy whip, it will be abandoned if it can no longer compete in the “marketplace” of nature.

Honey bees are masters (or is that mistresses?) of adaptability. They exhibit the highest genetic recombination rate (of germ cells) of any know animal (Beye 2006), and queens go to great trouble to ensure that they mate with as wide a diversity of drones as possible. This great genetic diversity within the colony ensures that there are plenty of alleles (gene variations) present in a natural population to allow rapid adaptation (Pritchard 2010). The honey bee is always poised to adapt to whatever Nature throws at her!

Viral Quasispecies

And what Nature throws at bees are a mess of ever-shifting RNA viruses. All the bee RNA viruses are noted for their mutability and existence in multiple, constantly evolving forms. This rapid mutation and evolution is evident not only within the bee population as a whole, but even within a single hive, or within a single bee! Scientists use the term “virus swarm” to recognize the great variability within any bee virus species, and apply the descriptive term “quasispecies” when speaking of a particular virus (Lauring 2010).

Once an individual virion enters a bee cell and begins to replicate, a process takes place similar to the “Pass the Message” game played at parties, in which a message is whispered from one person to the next, often becoming bastardized in the process. RNA viruses use an intentionally sloppy replication process (they exhibit by far the highest mutation rate of any known life form), so that in any infected host they create a wide range of modified new forms of themselves. The result is that any single bee that cannot suppress the initial infection winds up being inhabited by a swarm of slightly different virus mutants, any of which might be more or less virulent, or better adapted to the immediate environmental conditions. Most new genetic combinations are not adaptive, but even so, from time to time the virus gets lucky!

There’s a chance that in any hive an extremely virulent virus form might evolve, but luckily such a virus (generally) quickly kills the individual bee (or perhaps the colony) before it can transmit to other bees. Remember that the population of a colony of bees mostly consists of up to a few dozen distinct half sister cohorts, each fathered by drones form different mothers. This genetic diversity serves to prevent the sort of rapid spread of a virus that you’d see if all the bees were very closely related (Tarpy 2003). Along the same line, each colony of bees differs in its overall susceptibility to any specific virus strain, so it takes a generic sort of virus mutation to be virulent in the bee population overall.

Practical Application: Multiple mating ensures that the “team” of sister groups within the colony is more likely to include some members with better disease resistance, janitorial skills, or the ability to produce and propagate an antiviral RNAi response for the benefit of the entire colony.

Regional Bee Stocks

One controversial hypothesis is that bees could use viruses to their competitive advantage, in a process called “plague culling” (Hunter 2010). Say a virulent new virus strain evolves, but instead of being killed by it, the bees are able to ramp up their immune response to it, recover, and then keep the virus in check. From that point on, they could use that virus as a potent weapon to take out the competition (other colonies that lack immunity to that specific virus) by spreading it via the drift of infected drones or by intentionally contaminating the nectar of visited flowers! Soon the only colonies left would be those which were able to carry that virus as a harmless inapparent infection.

There is of course a problem of carrying such a ticking time bomb—it might go off at any time that the host bee is stressed or infected by another synergistic virus! Burden (2005) cites cases where covert viral infections in insects exploded years later into epidemics that devastated the population.

What you wind up with in nature are bee populations micro adapted for specific regions, which may be as small as a particular valley. Within that region, the bee population may have atypical symbiotic bacteria (Evans 2006) and resistance mechanisms to specific virus strains. Of course, we beekeepers completely mess up these regional adaptations by shipping a limited genetic pool of queens cross country and by trucking bees (and their viruses) all over the place (not to mention importing them from other continents)!

Brown and Fries (2008) explain the natural evolutionary situation well:

Given the likely distribution and density of wild colonies, vertical [mother-to-daughter] transmission [of viruses]…to offspring swarms should be the dominant mode of transmission. If this is true then there should be strong selection at the colony-level for low virulence in viral pathogens, as highly virulent pathogens will either kill off their host colony prior to [swarming] (and thus have no opportunity for transmission) or will reduce their growth to the point that they are unable to swarm (with similar impacts on transmission). Given that maintenance of the virus in the honey bee population ultimately depends upon inter-colony transmission, vertical transmission as the dominant route should result in viruses that have little or no effect at the colony level. In fact, unless there is some horizontal [colony to colony] transmission among colonies, theory suggests that viruses should ultimately be expected to evolve to be commensals [causing no harm]. Both acute paralysis virus and slow paralysis virus, which prior to the advent of Varroa mites were never associated with disease, provide evidence for such a scenario.

Practical application: It is normally only when something either depresses normal bee immunocompetency (such as lack of pollen (Fig. 1), chilling, environmental toxins, or parasites) or favors the transmission of unusually virulent virus mutants (by the crowding of too many hives into an area) that virus epidemics rage through the bee population.

Figure 1.  This colony is under nutritional stress.  Note the lack of a band of pollen around the brood, and the spotty brood pattern.  A colony under nutritional stress cannot defend itself well against viruses and other parasites.


The Natural Situation Prior to Varroa

Allow me to first dispel a common notion—that there is any normal “natural order” of things. The only thing that is assured in nature is that things will change. There is nothing essentially stable about ecosystems—they are in constant flux due to shifts in the environment, the introduction of new species, and the never ending evolutionary changes in the players, especially with regard to pathogens. Ecosystems and the species within exist in a state of “dynamic equilibrium,” and can change substantially should a new factor enter the equation. It is only in terms of the short human lifespan that we can speak of “normal” or “stable.”

The mechanism that maintains such apparently stable systems is a dominance of negative feedback loops. In my proposed model for colony collapse, I focused on the positive feedback loops that may lead to the rapid depopulation of a hive. On the other hand, the apparent stability in natural systems almost guarantees that there are strong negative or stabilizing feedbacks in operation.

In my neck of the woods, the Gray Squirrel population swings up and down between booms in response to food availability and then decimation due to epidemics of mites and other parasites. Similarly, the Gray Fox population builds up over several years, until the foxes become so abundant that the host density favors the transmission of canine distemper, at which point the fox population crashes. One could say that the squirrels and foxes are stable members of the foothill ecosystem, but their actual populations are anything but stable! However, the negative feedback loops that either increase or decrease parasite transmission in these species tend to keep the populations within a certain range—neither too abundant (for long) nor eradicated to the point of extinction.

The major negative feedback on virus epidemics is host density. The main constraint upon virulent bee viruses in natural situations is the distance that bees will fly to rob deadouts. An epidemic of a virulent strain of virus will spread until it reaches the point where there are no more susceptible hives within robbing distance to transmit to. At that point, the surviving colonies that demonstrate resistance to that particular strain of virus will restock the population, with the deadout cavities being cleaned out by wax moth in the interim.

Unfortunately, managed bee populations generally remove the constraint of host density upon viruses—we tend to crowd as many hives into an area as possible, often to the point of stressing them nutritionally.

I previously wrote about the regular periodicity of colony collapse events due to Thai Sacbrood in Asia. Such periodic virus epidemics are common in human, animal, and plant populations (Fig. 2). The same goes for insect populations. Dr. Jennifer Cory (2005) gives an example in the case of the Western Tent Caterpillar, whose population collapses every 6-11 years due to a virus. She also notes that sublethal effects of the virus appear to regulate the caterpillar population—that is, the virus does not need to necessarily kill the host to affect it at the population level. She also brings up a point of interest to beekeepers: “that the diversity of the virus population plays a role in infection severity, such that mixed genotype infections are more virulent than any individual clone.” As we beekeepers bring large numbers of colonies together, we ensure that multiple genotypes of each virus strain have more chances to interact. I’ll return to this point when I discuss how beekeepers help to create virus epidemics.

Figure 2.  An example of recurrent outbreaks of a pest, in this case, the Gypsy Moth.  Virus epidemics may be periodic (occurring cyclically) or sporatically triggered by environmental events.  Source http://www.inspection.gc.ca

The 50-year adult human lifespan often does not allow us to recognize the changes in ecosystems over the years. With regard to bees, in the Big Picture, the European and African honey bee populations on all continents except Australia are in the process of adjusting to new parasites, notably the varroa mite. In the absence of human intervention, the natural bee population generally recovers to a new “equilibrium” within a decade, after establishing a workable host-parasite relationship.

In most managed populations of honey bees, our intervention to prevent the loss of apiary businesses finds us tenuously nursing along strains of bees that are unable to hold their own against the ubiquitous varroa mite and its associated viruses. The sooner we can replace those strains with parasite-adapted stock, the sooner beekeeping will again be easier.

What’s Changed?

Most of our current viruses were present prior to varroa, yet we didn’t see as many collapse events as we see these days—why’s that? The answer is due to a simple fact that all beekeepers should keep in mind: it is rarely to a virus’s advantage to actually kill a bee or the colony! In the mindless and mathematical process of natural selection, it is most adaptive for a virus to exist rather benignly in the bee population as an inapparent infection.

The reason for this is that as soon as a virus makes a bee feel sick, that bee stops sharing food with other bees (thus minimizing the transmission of that virulent strain of virus) and performs “altruistic self removal”—the process by which sick bees protect the mother colony by flying away to die. It is exactly this process (which is normally a negative feedback), that can cause the rapid depopulation of a hive when it snowballs into a positive feedback loop due to poor nutrition, multiple pathogens, chilling and/or environmental toxins.

It is generally to the virus’s interest to keep its host bee alive and kicking as long as possible, in order to increase the chance that infected bee will transmit that particular virus strain to either another uninfected bee or to another colony (Brown 2005).

Of the roughly twenty known bee viruses, only one instance comes to mind in which it might be to the virus’s advantage to kill an individual bee within a hive. That would be for Sacbrood to kill bee propupae, as the virions would then be spread when undertaker bees remove the dead larva. However, for most of the year Sacbrood exists mainly as a very common infection in adult bees, with no dead brood to be seen (but note that Sacbrood-infected workers exhibit shortened lifespans). This constraint does not apply, of course, to viruses which manage to kill entire colonies, and thereby get spread (along with their mite accomplices) to other hives via robbing.

In general, there is a sporadic and dynamic ebb and flow of the infection levels of the various bee viruses in the hive (this has been independently documented by both Dr. Joe DeRisi and Dave Wick). But most of the time, the viruses persist as inapparent infections without any visible symptoms.

The point to keep in mind that bee viruses don’t “want” to kill bees. They are most successful when they vertically transmit from parent colony to swarm without killing off the host! But it is not as simple as that, since the bee colony goes through seasonal growth phases, in which different evolutionary pressures are put upon the viruses. It will help the beekeeper to recognize this dynamic, in order to better understand why colonies crash in fall and winter.

The Seasonal Progression of Virus Infections

Let’s look at the dynamics of virus epidemics within a hive:

Phase 1—Late Winter Population Turnover. The winter break in the brood cycle sets back both mite and virus reproduction. At this point of the time, varroa levels are at their lowest. Sometime in December, the colony initiates broodrearing in order to start building up in order to be able to take advantage of the first pollen and nectar flows. This is the last hurrah for the long-lived “winter bees,” which will start dying en masse once they’ve raised a generation or two of brood (Harris 2009). There is a relatively rapid turnover of the hive population from the long-lived “winter bees” to a new generation of short-lived “summer bees” (Harris 2010). The colony at this time is very susceptible to both chill and nutritional stress, either of which can lead to virus problems.

Phase 2—Spring Buildup. The colony builds up rapidly in the spring and produces abundant brood and maintains a large proportion of nurse bees. During this period of rapid growth, it can simply “outrun” the mites and viruses due to the rapid turnover of the bee population.

Phase 3—The Flow: During an intense honey flow, the nectar processors fill the brood cells with nectar, thus shutting down the queen. In order to capitalize on the available nectar flow, the nurse bees then quickly transition into short-lived foragers. As a result, the hive population structure shifts to being heavy on foragers rather than nurses. At this time, the mites have less brood to parasitize, and thus the infestation level of the brood greatly increases. In other words, this is the turning point when the mites catch up with the bees. The result is that the current generation of brood, which is critical for the next phase, will consist of bees that were heavily parasitized.

Phase 4—Population Recovery. Once the flow is over, the colony must attempt to replace its population, which at that time will consist largely of worn-out foragers, yet there may be little sealed brood from which to draw recruits. This is the critical time when viruses can go rampant if mite levels are high.

Phase 5—Late Summer Pollen Dearth. This is the worst time for Western beekeepers, since without adequate pollen, the colonies cut back on broodrearing, and the nutritionally-stressed hives simply can’t hold their own against varroa and the viruses. This is when we typically start to see the initial signs of a brewing virus epidemic in the hive (sick pupae, adults with deformed wings).

Phase 6—Fall Population Turnover Once the fall pollen flow is over (due to seasonality or weather), the remaining summer population of bees flies off to die, leaving a smaller population of broodless long-lived “winter bees” to hold the fort until spring (Matilla 2007). It is critical for colony survival that these bees go into winter in healthy condition!

Phase 7—The Winter Cluster. Once broodrearing ceases, all mites must subsist in the phoretic stage on adult bees. They move about freely from bee to bee, and abandon dying bees, moving to fresh hosts (Bowen-Walker 1998). During this period of time, due to the combination of cold temperatures, and the constant transmission of virus strains throughout the hive, it is easy for an epidemic of a virulent virus strains to infect the majority of the hive population, especially as infected mites are forced to concentrate upon fewer and fewer surviving bees!

At each stage of this cycle, the evolutionary pressures on the bee parasites change. For the Big Three—varroa, viruses, and nosema—their levels go from benign to epidemic and then back down, each on its own schedule. Nosema ceranae infection typically peaks in May; varroa infestation goes from a low point in early spring, and peaks around September 1st.

The surprising thing is the viruses. Colonies, and even individual bees are often infected with at least three viruses simultaneously, but that does not necessarily mean that they are all reproducing (USDA 2010, p. A-4)—they may just sit tight until the bee is stressed, which then upsets the virus/immunity equilibrium. Each virus has its own seasonal peaks for prevalence (reviewed by Ribière 2008):

Chronic Bee Paralysis Virus (CBPV)

Sacbrood Virus

Spring and summer
Kashmir Bee Virus (KBV) Spring
Israeli Acute Paralysis Virus (IAPV) Unknown, but is genetically similar to ABPV and KBV
Deformed Wing Virus (DWV)

Acute Bee Paralysis Virus (ABPV)

Both track the varroa infestation level, so peaks are generally in late summer and fall
Black Queen Cell Virus (BQCV) Tracks nosema infection level
Insect Iridovirus (IIV) During cool, damp conditions

Note in the above table that in the absence of varroa, virus levels typically peaked in spring, when the bee population was growing most rapidly, and declined in fall, as the population dropped and older bees abandoned the hive. Why would this be? When the hive population is growing rapidly, and consists mostly of short-lived bees, then it is to the virus’s advantage to be more virulent. On the other hand, when there is little brood, when the older bees are flying off to die in the fall, or when the population is composed of long-lived “winter bees,” then less virulent virus strains are more successful. Prior to the mite, the seeds of virus epidemics sprouted during spring and summer, and then were naturally weeded out in fall and winter.

Enter Varroa

Direct Effects of Varroa

The host upon which varroa mites reproduce are the bee pupae. Workers parasitized as pupae never recover from their loss of hemolymph proteins; they begin foraging at an earlier age, and have a significantly reduced life span. Those handicaps placed upon individual bees put a real drag on overall colony growth and peak population.

A bee’s exoskeleton is its first line of defense against viruses and other pathogens. For the first time in their evolution, European honey bees now routinely have that very cuticle perforated by varroa feeding wounds. The trauma of the wound alone may be enough to “wake up” an inapparent virus infection in the bee (Anderson 1988). Diana Cox-Foster’s team (Shen 2005, Yang 2005) also brought to our attention that varroa mites actually inject an immune suppressor into the bee (likely in order to inhibit the normal rapid healing of the wound). This immune suppression may allow the latent viruses or nosema that are normally endemic in the bee to explode into full-blown infections resulting in actual disease.

Kanbar and Engels (2003) produced some stunning electron micrographs of the bacterial colonies that then become established in those wounds. What occurs to me is that if mite wounds are infected with bacteria, then the bees will need to maintain a metabolically-expensive immune response for the rest of their lives in order to keep the bacteria in check (Hain 2008).

This bacterial infection has other consequences. Yang (2005) suggests that: “The increased replication of DWV in honey bees needs two components, varroa mite parasitization and exposure to a bacterial factor. This microbial challenge may naturally exist, because bacterial colonies are found on the varroa feeding sites in some bee pupae.” The suppression of bacterial infection by antibiotics may help mite-infested colonies to survive. Note that “antibiotics” may include the essential oils and propolis, rather than manmade antibiotics, the overuse of which leads to antibiotic-resistant bacteria.

Imagine the arrival in your community of a crab-sized external bloodsucking parasite that infested, say, about one out of ten people, and left gaping half-inch wide, bacteria-infected, unhealing skin wounds, and that the parasites moved freely from person to person, spreading viruses and bacteria from one traumatized victim to the next. Sounds like a recipe for disastrous epidemics, huh? Well, that graphic scenario accurately describes a colony of bees suffering from a 10% varroa infestation (a level commonly seen in late summer).

Question: what would be the long-term effect on apiaries if parasitism by the foundress mite inevitably caused bee pupae to quickly die?

Hint: A mite typically creates a long-term feeding wound on the left underside of the abdomen of its favored host—a drone pupa; there’s generally only one shared wound per pupa, no matter the number of mites in the cell (Kanbar and Engels 2004, 2005). The single wound is likely an adaptive trait, since multiple wounds might kill the pupa, resulting in the mite(s) being trapped in the capped cell. Interestingly, the mite does not yet appear to be well adapted to worker brood, and often creates wounds in the thorax. Think about the answer to the above question…it’s in References.

Remember that the bees’ first line of defense against viruses is an intact integument. In the past few decades, bees in the U.S. have picked up three parasites that breach that barrier—varroa mites pierce the exoskeleton, tracheal mites the tracheal lining, and Nosema ceranae the basal cells of the gut. It’s small wonder that bees are under duress from viruses!

Aaronstein (2011) reports that “Bees parasitized as pupae (with normal wings) by varroa are approximately twice as susceptible to IAPV infections as non-parasitized bees.” Once a bee’s been bit, it appears to be a sitting duck for the next pathogen to come along!

To make matters worse, recent research indicates that miticides and other pesticides may induce the death of bee gut cells, make bees more susceptible to virus infection, or increase the reproduction of nosema (Gregorc 2011; Diana Cox-Foster, pers comm; Judy Wu, pers comm). This breaching of the normal bee gut defense against parasites must be factored in when we discuss bee health.

It’s not completely clear as to whether the bee viruses are merely vectored by varroa, or if they actually replicate in the mite as an alternate host. A number of researchers have found high virus titers in some mites. However, when Santillán-Galicia (2008) meticulously sectioned (sliced into thin pieces) mites, she only found negative DWV strands (negative strands are indicative of actual virus reproduction) in the mite guts, and nowhere else in the mites’ bodies. This finding suggests that perhaps DWV does not actually replicate in the mite, or maybe it does replicate, but only as a relatively harmless infection of the gut lining, similar to the way in which influenza viruses benignly infect the guts of their natural host—waterfowl. This finding is very intriguing, especially since the closely-related Varroa Destructor Virus appears to specifically be adapted to reproduce in the mite (see my previous article for the implications of DWV/VDV hybrids).

DWV and Bee Behavior

A number of viruses and other parasites have been shown to affect host behavior to cause the infected host to better transmit the parasite (human cold viruses cause us to sneeze, rabies virus induces infected animals to bite, other viruses initiate behaviors that increase the chance that a host will be eaten by a predator). Well, bee viruses are likely no different! A study by Shah (2009) found that DWV preferentially replicates in bee brains: “Therefore, it is possible that the virus affects the bees’ flight behavior, homing performance, and perception of odorants.” In the case of colony collapse or reduced honey production by colonies with high mite levels, the loss of bees might be directly attributable to disorientation due the replication of DWV in the antennal and optic lobes of their brains!

Varroa and Queens

Excessive queen loss has been problematic in recent years, yet research by various researchers indicates that it is neither the queens nor the producers to blame, except perhaps for the fact that all queens are infected by DWV, many at surprisingly high levels, relative to other viruses (Delaney 2010). Delaney also suggests that: “There is also an intriguing possibility that DWV may affect sperm production by drones, the ability of queens to adequately store sperm, or both.”

Dr. Frank Eischen recently presented intriguing data which suggests that queen losses are linked to varroa infestation levels. It appears that queens, all of which test positive for at least one virus, are prone to succumb when mite levels increase. This could be due to viruses being inadvertently passed via jelly from nurse bees to the queen, or perhaps by direct feeding of a mite upon the queen. I’ve searched the literature, and also asked several notable varroa researchers if there is data on whether varroa actually feed upon queen bees, but no one seems to have actually investigated!

Things are Different These Days

Our bees have always dealt with viruses. Even in varroa-free Australia, inapparent virus infection is common in pupae (Anderson 1988). In the absence of varroa, viruses generally infect hives with little obvious effect, except for the occasional flare up of a geographically-constrained event. There are historical records of such events from all continents of Sacbrood, Chronic Paralysis Virus, Kashmir Bee Virus, and Black Queen Cell Virus, but these epidemics were uncommon, and generally related to nutritional stress.

Before the arrival of varroa (Fig. 3), episodes of sudden colony collapse were not unusual (often called “Disappearing Disease” (Underwood 2007). Beekeeper Andy Nachbaur (1996) wrote about “Seasonal Affective Disorder” and “Bee Immune Deficiency” problems in commercial California operations. But when varroa arrived, things got a whole lot worse!

Figure 3.  Prior to varroa, beekeeping was much easier!  However, as far back as we have records, there have been episodes of collapse events that were likely caused by virus epidemics, generally initiated by nutritional stress and poor weather. Photo:  Apiary at Cogswell’s Sierra Madre Villa, Ca. 1886, Carleton E. Watkins.  Courtesy of the California History Room, California State Library, Sacramento.

Healthy bees have a hulluva strong natural resistance to getting sick from viruses by the normal route of infection—from eating contaminated jelly, honey, or pollen. Oral inoculation typically requires the ingestion of millions or billions of virus particles to initiate observable disease. However, if a varroa mite inadvertently injects as few as 100 virus particles directly into a pupa or adult bee when feeding, that bee may quickly sicken and die!  This why varroa is such a problem – it transmits viruses throughout the hive in a novel manner (Martin 2007).

Such vectoring of virus particles means that if a phoretic female mite feeds on an infected bee, that she will likely transfer that infection to the next pupae that she subsequently parasitizes, from which her offspring (and any other mites in the cell) then become infected (Di Prisco 2010). When those infected offspring then go into the phoretic stage, or switch from bee to bee in the winter cluster, they can rapidly serially transfer the most infective virus strains from bee to bee. This is a huge point! That the relatively quick “hopping” of mites from bee to bee, and their abandonment of dying bees, specifically selects for the most virulent strains of virus to be successfully transmitted. This is totally unlike the situation prior to varroa, when natural selection worked against highly infective viruses.

Practical consideration: Martin’s computer modeling (2001) demonstrated that the rate of mite transfer from bee to bee greatly affects the overall proportion of bees in a hive infected by virus. This brings up an interesting thought—any disturbance (trucking? smoking? alarm pheromone) or substance introduced into a hive (pesticides, sublethal doses of essential oils) that increases bee-to-bee mite movement without actually killing the mites could hypothetically increase the rate of virus transmission!

Varroa and Viruses

Varroa has affected the historical host/parasite relationship between bees and viruses. I’ve detailed a number of the direct effects of varroa upon individual bees (Fig. 4). But how about the colony- and population-level effects of the mite on bees?

Figure 4.  Mites chew a hole in the skin of a pupa, which initiates both bacterial and viral infections.  Bees parasitized as pupae will never develop into fully competent adults.  These photo is of mite-infested drone brood skewered on a cappings fork.

Parasitism by mites greatly increases the chance that pupae will become infected by more than one virus, and that those combinations of viruses will be subsequently transmitted throughout the hive via mites, as well as through the action of mid-aged “cleaner” bees as they chew away dying pupae. Again, such removal of virus-killed brood selects for the transmission of the most virulent strains of virus. This vectoring of viruses is an indirect effect of varroa, but perhaps the mite’s major impact upon beekeeping. The tenuous bee/virus equilibrium was upset when this highly-effective vector entered the picture. Instead of natural selection weeding out the most virulent strains of viruses, varroa actually favored them!

When varroa initially arrived in the U.S., the viruses had not yet adapted to using it as a vector, and you could see apparently healthy hives with as many as 20-50,000 mites (the same was reported from England, South Africa, and New Zealand). A yearly treatment with a “silver bullet” miticide would bring the mite population down enough for the colony to recover (we often marveled at the sight of mites too numerous to count on the stickyboards). A varroa infestation appeared to be relatively harmless those first years, so long as you knocked the mites back each fall (unfortunately, many beekeepers still try to stick to this strategy). But this was back when viruses were less common in hives, and those viruses had not yet evolved to exploit the mite for their purposes.

Practical Application: Simply knocking the mites back in fall is no longer adequate for successful colony wintering. Mite management must begin by midsummer at the latest.

Shortly after the arrival of varroa we started to notice the symptoms of Deformed Wing Virus (DWV). Varroa and DWV is a match made in Hell! Either species alone is a relatively benign parasite, mostly transmitted vertically from parent colony to daughter. But the two parasites in combination spread quickly when their joint infestations caused colonies to collapse and get robbed out, thus horizontally transmitting both parasites to new host colonies. As long as there are plenty of new hosts, the most virulent strains of varroa and DWV spread hand-in-hand like wildfire through the bee population–both the mite and the virus assisting each other.

When I wrote an article in 2007 on mite threshold levels, there were U.S. researchers suggesting that colonies could winter with as many as 3500 mites in fall! But at the same time European researchers were saying that 1000 mites in fall was the kiss of death. The difference was apparently due to two factors in the European hives:

1.The constant presence of varroa since the mid 1980’s meant that there was much greater overall presence of several viruses in hives, and

2.That due to the evolution of the mite-vectored viruses, there was an increase in the efficiency in transmission and in the virulence in those viruses.

In a few years, the U.S. situation caught up with that in Europe. The most observable change to beekeepers was the evolution of DWV, since we can visually see the symptom of deformed wings in those bees most severely infected as pupae. In the early years of varroa, we’d only see bees with deformed wings once mite infestation built up to high levels. But after several years, things began to change. We started seeing bees with deformed wings even when there weren’t many mites. DWV has evolved away its formerly obscure pre-mite status, and is now a ubiquitous and potentially devastating inhabitant of virtually every bee in every hive, causing winter mortality even in the near absence of mites (Highfield 2010).

As I explained in my last article, the various strains of DWV can hybridize and interact, and can cross with its sister virus, Varroa Destructor Virus (VDV-1) to form highly infective and virulent hybrids. We beekeepers in the U.S. have this to look forward to as VDV-1 or other strains evolve and spread rapidly across the country, helpfully assisted by migratory beekeepers.

Other viruses also seem to have evolved, especially the Kashmir Bee Virus (KBV) family, which includes Acute Bee Paralysis Virus and Israeli Acute Bee Paralysis Virus (ABPV and IAPV). Dr. Stephen Martin explained (2007) that rapid-acting ABPV can cause a colony to collapse very quickly (which we’ve also observed when we inoculate colonies with a good dose of IAPV). Di Prisco (2010) found that varroa actively transmits IAPV. DWV, however, tends to cause a colony to decline more slowly, generally collapsing during fall or winter when the normal decrease in broodrearing cuts back on the recruitment of fresh young workers to take the place of sick ones.

Generally it takes a minimum number of mites in a hive to kick start a virus epidemic. Stephen Martin’s 2001 model indicated that a mite population of 2000 could initiate an unstoppable epidemic of either DWV or ABPV, leading to eventual colony death sometime between fall and spring. Note that 2000 mites in a colony with 10 frames of bees (about 40,000 bees) would represent a 5% infestation of the total hive population, including the bees in the sealed brood. A recent study by Katie Lee (2010) found that roughly half the mites would be in the brood during the summer, so that 2000-mite infestation level would show as a 2½ % infestation of the adult bees (less than 3 mites per 100 bees, or 8 mites in a ½ cup sample of bees).

Practical application: most successful beekeepers already have realized this, but if you allow mite levels to climb beyond the 5% level in September, you are in serious risk of later losing that colony to a virus epidemic, no matter if you later knock down the mites with a treatment! That’s like closing the barn door after the horses are already loose—the virus epidemic has already taken on a life of its own, and the colony will have a hard time recovering. Coupled with poor late-summer nutrition, the sublethal effects of the miticide and some ag pesticides, the colony may have little chance at making it through the winter!

Practical Tip: As a general rule, do not allow mite infestations to ever exceed 2 mites per 100 adult bees (6 mites in an alcohol wash of ½ cup of bees). This threshold should be even lower in winter and spring, but might go a bit higher around September 1st.

As the virus epidemic spreads (generally invisibly) through the hive population in late summer, the workers’ lifespans are shortened, brood survivability drops, and the colony is not able to maintain the necessary recruitment rate of young workers to replace the aged, sick, and fallen. As Rosenkranz points out, colonies can handle mite and virus epidemics so long as the colony is in the rapid growth stage (note that few colonies die at this time of year). The problems start as the colony begins its natural population decline in late summer, when the mite and virus infection levels are increasing, just as the bee population is decreasing.

Practical application: Be aware that mite/virus problems begin when colonies slow down on broodrearing, such as during an intense honey flow, in late summer, or during pollen dearths. Management to promote broodrearing in late summer can help compensate for the negative effects of the mite/virus buildup at that time. Moving hives to better pasture or feeding supplemental protein can greatly improve colony health.

Holding Yards

In large holding yards, drift and robbing quickly homogenize infected bees and vectoring mites throughout the operation. Some viruses which infect the brain, such as DWV, appear to affect the bees’ ability to navigate, and may lead to increased drifting of infected bees into other hives (ditto for nosema, Kralj 2009). The process would then go into a positive-feedback loop that selects for the most virulent virus strains. If the rest of the colonies are stressed by poor nutrition, cold, and pesticides, they could also be drug down—even if mite counts are relatively low! Infected bees will quietly fly off to die, leaving emptied hives with little patches of young bees valiantly trying to save the queen! Sound familiar?

Practical application: Large holding yards are perfect breeding grounds for mite/virus/nosema epidemics.

Beekeeper-Applied Miticides

One of the major indirect effects of the varroa invasion is that, for the first time in history, beekeepers are intentionally dumping large quantities of synthetic pesticides into their hives. It now appears that the cure may have been as bad as the disease! Bees have had to deal with agricultural pesticides for many years, but the addition of high levels of miticides may have been the tipping point in suppressing overall colony immunocompetence not only against viruses, but also against nosema, and perhaps even varroa itself!

This is truly a case of the pot calling the kettle black! There is considerable irony in the fact that as beekeepers worldwide decry the putative harmful effects of agricultural pesticides to their bees, they in fact are typically the primary polluters of those very hives! As much as beekeepers wish to stick their heads in the sand, in every recent study of collapsing hives, from every country of the world with varroa, the presence of high levels of beekeeper-applied miticides is a primary suspect for causing poor colony health.

It is important to keep in mind that when a miticide or agricultural pesticide is tested for bee toxicity, it is generally tested on healthy, pesticide- and miticide-free bees. Unfortunately, in the real world, that pesticide will in actuality be one more addition to the toxic stew already existing in a hive, likely with synergistic negative effects upon bee immunocompetence and survival!

The reality is that any assessment of apiary health must take into account that the presence of these miticides in the combs likely negatively affects brood survival, queen longevity, and the ability of the colony to maintain an effective immune response to the ever present pathogens in the hive.

Practical application: there are effective, off-the-shelf “natural” miticides that leave insignificant residues in the combs (Fig. 5).

Figure 5.  My sons applying Apiguard thymol gel in Nevada in late summer.  I highly recommend placing it between the brood boxes in the middle of the cluster for best efficacy.  We have no problem with keeping mites in check by using resistant stock and “natural” treatments.


Varroa changed everything

Dr. Keith Delaplane (Bee World 2010) recently summed up the feelings of most bee researchers (including myself):

I have come to believe that Varroa commands the prominent place in the list of bee problems, to the point –I propose – of constituting the kingpin, the overarching preconditioning liability, the snowball that starts the avalanche. This bloodfeeding, nonnatural ectoparasite attacks bees at both the larval and adult life stages, shortening life span, altering behaviours, vectoring or activating a host of bee viruses, and suppressing immune systems. Moreover, the synthetic miticides used to control Varroa are themselves hazardous to the bees they are intended to protect.

I echo the summation of Peter Rosenkranz and coauthors when they said in a recent special issue, “No other pathogen [beside Varroa] has had a comparable impact on both beekeeping and honey bee research during the long history of apiculture.” In summary, it is simply noncontroversial among the world’s practicing bee scientists that Varroa destructor is problem #1. Thus, Varroa associated bee morbidity is an unholy mix of direct injury, mitevectored or activated pathogens, suppressed immune systems, and nontarget miticide effects.

Practical Wrap Up

I bring to the reader’s attention Dr. Delaplane’s use of the word “morbidity,” a recent buzzword in the bee literature. Beekeepers have traditionally been more concerned with bee or colony “mortality”—the death rate in a population. But the damage due to the varroa/virus duo is more insidious.

Brown and Fries (2008) explain:

We use mortality for the sake of simplicity, but note that many parasites cause morbidity (that is, they lower host fitness without causing mortality), and thus the absence of mortality in any given honey bee/virus system should not be taken to mean the absence of an impact of the virus on honey bee colonies. While using mortality as a metric may seem to be straightforward, the presence of covert infections…puts a serious wrinkle into the picture.

The term “morbidity” relates to the incidence of unhealthiness or disease in a population. Even when there is no apparent overt bee or colony mortality, one might notice slower colony build up, poor brood patterns, a higher incidence of nosema or chalkbrood, smaller hive populations, decreased honey production, increased queen supersedure or failure, or poorer wintering. Any of the above would be considered as examples of colony morbidity due to viruses.

Many long-time beekeepers have noticed that colonies just seem more “fragile” these days, less productive, and harder to get through the winter. Note how these symptoms reflect the bee/virus coevolution since the arrival of varroa. Prior to varroa, the bees and the viruses had reached an uneasy sort of “détente” or generally peaceful coexistence. This didn’t change immediately after the arrival of the mite, but went to hell as the viruses evolved and “caught up” with the altered transmission dynamics afforded with having the mite as a vector. It’s now the bees’ turn to catch up with this entirely new situation of having both varroa and the recently-evolved virus strains constantly present in the hive.

The resulting decreased health and fitness of managed hives means that they are often closer to a “tip point” from which they may succumb to a combination of additional stresses imposed by any of their traditional nemeses: poor nutrition, chill, parasites, or environmental toxins.

A Paradigm Shift

Granted, beekeepers who practice good varroa management are generally successful at keeping their colonies alive. But I feel that it is time for a paradigm shift in our understanding of the actual biological situation. What we are really dealing with is not varroa per se, but rather the impacts of the viruses, exacerbated by varroa. So I feel that it would be of benefit to beekeepers to shift our perspective from mite management to that of virus management.

Virus management depends upon maintaining colony immunocompetence, plus holding the virus vector/activator (varroa) in check at the appropriate times of the year. The following would be the main items to address in a good integrated pest management program against viruses:

Basic Immunocompetence

  1. Colony immunocompetence is largely a function of good protein nutrition. Provide good nutrition by either stocking fewer colonies in a yard, moving hives to good pasture, or by supplemental protein feeding.

3.Maintain genetic diversity, both within the hive, and within one’s operation. The less diversity, the greater the chance that a virus mutant will tear through your entire operation.

4. It’s been demonstrated that it’s easy to breed for virus resistance, and that such selection is as important as breeding for mite resistance.

Medication and Treatment

  1. We may find that certain natural antibiotics such as propolis or essential oils are helpful at controlling the bacteria that infect varroa wounds.
  2. The new siRNA product RemebeeTM may prove to be useful as an antiviral immunization.

Sublethal Toxins

  1. Another major aspect of colony immunocompetence is to minimize pesticide residues within the hive. A number of insecticides, fungicides, and pesticide adjuvants are clearly harmful to bees and brood, and appear to suppress immune function. Multiple pollination contracts on heavily-sprayed crops can leave colonies in stressed condition.
  2. That includes any persistent miticides! The major pesticide stress in hives is likely due to the very synthetic miticides applied by beekeepers. Switch to an Integrated Mite Management program which uses timely application of “natural treatments” (thymol, formic, oxalic, plant extracts) for varroa control.

Vector Management

  1. Use mite-resistant bee stock. This is the number one most effective part of virus management!
  2. Make sure that colonies start each season with very low mite levels. An ounce of prevention is worth a pound of cure!
  3. Regularly monitor mite infestation levels, and take action if they start to approach the 2% threshold.
  4. Minimize drift, keep smaller yards, avoid allowing hives to collapse and be robbed out.
  5. Keep an eye on your neighbors! Don’t allow a mite-ridden operation to load yours up when it crashes!


  1. Recognize when a virus epidemic starts brewing (generally in mid summer) and spot check brood for signs of pupal death or bees with deformed wings.
  2. August 15th. Put this critical treatment date on your calendar! Break the back of the incipient virus epidemic by removing the vector! You want to create a break in virus transmission prior to the colony raising the bees that will eventually form the winter cluster.
  3. Recognize that the most critical time for virus epidemics to take hold is when the colonies naturally decrease their populations (immediately after the main flow, or generally around September 1st). Take action to prevent the mite to bee ratio from rapidly shifting upward at this critical junction.
  4. Double check mite levels in late fall (Fig. 6). Realize that during winter, mites move from bee to bee within the cluster, and if there are virus issues, then they will spread. This can be a major problem, as there is little recruitment of fresh young bees during the winter to take the place of bees that fall ill.

Figure 6.  Comparing the efficacy of my fall mite treatments after a snowfall in the Sierra Foothills.  There is little broodrearing in November and December, so virtually all the mites will be moving about on the adult bees in the cluster, spreading viruses during this stressful time for a colony.

Recruitment of Replacement Bees

  1. Colonies deal with diseases by out breeding them! Sick bees must be replaced with healthy young bees. Give your colonies a fighting chance by stimulating broodrearing in late summer.

Nosema ceranae and Viruses

  1. I’m not yet clear whether nosema problems are a function of high virus levels, or vice versa. I’m getting mixed messages as to the economic benefit of treating for Nosema ceranae.
  2. Keep your eyes open for virulent viruses! Remove sick colonies, steer clear of operations with sick bees, avoid large holding yards.

If you can get your head around this paradigm shift to virus management, rather than simple mite management, then you can better make the most appropriate decisions regarding mite treatments and timing, nutritional supplementation, and other actions. The above recommendations are all based upon common-sense animal husbandry. Look carefully at your operation, and identify the critical times for action, and the best and most cost-effective opportunities for managing against viruses during the season.


I dedicate this article to the memory of British beekeeper Dave Cushman, who generously helped beekeepers worldwide (including myself) via correspondence, and through his excellent website http://www.dave-cushman.net/.

As ever, I am greatly indebted to the research assistance provided by my friend Peter Loring Borst. I also wish to thank Colorado beekeeper Al Summers for his suggestion to start using the term “immunocompetence.”

Suggested Reading and References

Answer to question: what would be the long-term effect on apiaries if parasitism by the foundress mite inevitably caused bee pupae to quickly die?

Varroa would shortly go extinct, since it could no longer reproduce! So there is a negative feedback that prevents the mite/virus combination from being too virulent to bee pupae!

The following suggested reading docs are all free downloads (thank you, authors):

Aubert, M (2008) Virology and the Honey Bee. (Google the title) This recent review contains a wealth of information—I highly recommend it for the serious beekeeper.

BRAVE (2005) Bee Research And Virus in Europe Proceedings

(Broken Link!) http://www.academie-medecine.fr/userfiles/file/rapports_thematiques/veterinaire/abeille_virus_europe_2006.pdf

Delaney, DA, JJ Keller, JR Caren, DR Tarpy (2010) The physical, insemination, and reproductive quality of honey bee queens (Apis mellifera L.). Apidologie 10.1051/apido/2010027

Nachbaur, A (1996) SAD & BAD Bees. http://www.beesource.com/point-of-view/andy-nachbaur/sad-bad-bees/

USDA 2010 Colony Collapse Disorder Progress Report. www.ars.usda.gov/is/br/ccd/ccdprogressreport2010.pdf

Other Citations

Aaronstein, K (2011) Cited in USDA 2010 Colony Collapse Disorder Progress Report.

Anderson, D and AJ Gibbs (1988) Inapparent virus infections and their interactions in pupae of the honey bee (Apis mellifera Linnaeus) in Australia. J Gen Virol 69: 1617-1625.

Beye, M, et al (2006) Exceptionally high levels of recombination across the honey bee genome. Genome Res. 16: 1339-1344.

Bowen-Walker, PL, A Gunn (1998) Inter-host transfer and survival of Varroa jacobsoni under simulated and natural winter conditions. JAR 37(3):199-204.

Brown, MJF and I Fries (2008) Evolutionary epidemiology of virus infections in honey bees. In Virology and the Honey Bee.

Burden, JP, et al (2005) Covert infection strategies of insect viruses. In BRAVE.

Cory, JS (2005) Resistance and virulence in insect-pathogen relationships: the baculovirus-Lepidoptera case. In BRAVE.

Delaplane, KS (2010) Varroa destructor: Back in Fashion. Bee World 87(4): 82-83.

Di Prisco, G, et al (2010) Varroa destructor is an effective vector of Israeli acute paralysis virus in the honey bee, Apis mellifera. J Gen Virol 92: 151-155.

Evans, JD and T-N Armstrong (2006) Antagonistic interactions between honey bee bacterial symbionts and implications for disease. BMC Ecology 6:4

Evans, JD, S Cros-Arteil, D Crauser and Y Le Conte (2008) Differential gene expression of the honey bee Apis mellifera associated with Varroa destructor infection. BMC Genomics 9:301

Gregorc, A and JD Ellis (2011) Cell death localization in situ in laboratory reared honey bee (Apis mellifera L.) larvae treated with pesticides. Pesticide Biochemistry and Physiology 99: 200–207.

Gregory PG, JD Evans, T Rinderer , L de Guzman (2005) Conditional immune-gene suppression of honey bees parasitized by Varroa mites. Journal of Insect Science 5: 7.

Haine, ER, Y Moret, MT Siva-Jothy, J Rolff (2008) Antimicrobial defense and persistent infection in insects. Science 322: 1257-1259.

Harris, LJ (2009) Development of honey bee colonies on the Northern Great Plains of North America during confinement to winter quarters. Journal of Apicultural Research and Bee World 48(2): 85-90.

Harris, LJ (2010) The effect of requeening in late July on honey bee colony development on the Northern Great Plains of North America after removal from an indoor winter storage facility. Journal of Apicultural Research and Bee World 49(2): 159-169.

Highfield, AC, A El Nagar, L Mackinder, ML Laure, J Noël, MJ Hall, SJ. Martin, and DC. Schroeder (2009) Deformed wing virus implicated in over-wintering honeybee colony losses. Appl. Environ. Microbiol 75(2): 7212-7220.

Hunter, P (2010) The missing link; Viruses revise evolutionary theory. EMBO reports VOL 11 | NO 1: 29-31.

Kanbar G and W Engels (2004) Number and position of wounds on honey bee (Apis mellifera) pupae infested with a single Varroa mite. Eur. J. Entomol. 101: 323–326.

Kanbar G and W Engels (2005) Communal use of integumental wounds in honey bee (Apis mellifera) pupae multiply infested by the ectoparasitic mite Varroa destructor. Genetics and Molecular Research 4 (3): 465-472.

Kanbar, G and W Engels (2003) Ultrastructure and bacterial infection of wounds in honey bee (Apis mellifera) pupae punctured by Varroa mites. Parasitol Res 90: 349–354.

Kralj, J and S Fuchs (2009) Nosema sp. influences flight behavior of infected honey bee (Apis mellifera) foragers. Apidologie 41(1): 21–28.

Lauring AS and R Andino (2010) Quasispecies theory and the behavior of RNA viruses. PLoS Pathog 6(7): e1001005. doi:10.1371/journal.ppat.1001005

Martin, SJ (2001) The role of Varroa and viral pathogens in the collapse of honeybee colonies: a modelling approach. Journal of Applied Ecology 38: 1082–1093.

Martin, SJ (2007) Quoted by Peter Edwards summarizing a presentation. (Broken Link!) http://www.stratford-upon-avon.freeserve.co.uk/Newsletters/May2007.htm

Matilla, HR and GW Otis (2007) Dwindling pollen resources trigger the transition to broodless populations of long-lived honeybees each autumn. Ecological Entomology 32: 496–505.

Navajas, M, et al (2008) Differential gene expression of the honey bee Apis mellifera associated with Varroa destructor infection. BMC Genomics, 9: 301

Neumann, P and NL Carreck (2010) Honey bee colony losses. Journal of Apicultural Research 49(1): 1-6

Pritchard, J and A Di Rienzo (2010) Adaptation – not by sweeps alone. Nature Reviews Genetics 11: 665-667.

Ribière, M, B Ball, and M FA Aubert (2008) Natural history and geographical distribution of honey bee viruses. In Virology and the Honey Bee.

Rosenkranz, P., et al. (2009) Biology and control of Varroa destructor. Journal of Invertebrate Pathology 103(1): S96-S119.

Santillán-Galicia, MT, R Carzaniga, BV Ball and PG Alderson (2008) Immunolocalization of deformed wing virus particles within the mite Varroa destructor. J Gen Virol 89: 1685-1689.

Shah, KS, EC Evans and MC Pizzorno (2009) Localization of deformed wing virus (DWV) in the brains of the honeybee, Apis mellifera Linnaeus. Virology Journal 6:182.

Shen, M, X Yang, D Cox-Foster and L Cui (2005) The role of varroa mites in infections of Kashmir bee virus (KBV) and deformed wing virus (DWV) in honey bees. Virology 342(1): 141-149.

Tarpy D.R. (2003) Genetic diversity within honey bee colonies prevents severe infections and promotes colony growth, Proc. R. Soc. Lond. B 270, 99-103.

Underwood, RM and D vanEngelsdorp (2007) Colony collapse disorder: Have we seen this before? http://ento.psu.edu/directory/duv2/underwood_vanEngelsdorp_2007.pdf

Yang, X and DL Cox-Foster (2005) Impact of an ectoparasite on the immunity and pathology of an invertebrate: Evidence for host immunosuppression and viral amplification. PNAS 102: 7470 – 7475.

Yang, X., and D Cox-Foster (2007) Effects of parasitization by Varroa destructor on survivorship and physiological traits of Apis mellifera in correlation with viral incidence and microbial challenge. Parasitology 134: 405–412.