Bumblebees and honeybees share diseases and the outcome is not a good one.

lyme regis february 2014
Bumblebee on Rosemary on Lyme Regis sea front (February 22nd 2014) (photo by Hazel Strange)

I recently read Dave Goulson’s excellent book “A Sting in the Tale” and learnt a lot about bumblebees. Although I was aware of the global trade in honeybees, I hadn’t realised that there was an equivalent trade in bumblebees. Some crops such as tomatoes and peppers require buzz-pollination, the rapid vibration of the flower. Bumblebees do this very well and are now used extensively by commercial growers of tomatoes and other crops. To supply the demand for these useful insects there are at least thirty factories producing bumblebees for shipping all over the world. The numbers are staggering with European factories producing up to a million nests per year. This is big business with huge financial rewards but keeping so many insects together in one place risks the rapid spread of disease unless stringent hygiene precautions are observed. To complicate matters, commercially-reared bumblebees are fed pollen from honeybees so that they are potentially exposed to all the diseases that affect honeybees.

But what about wild bumblebees? What happens when a wild bumblebee forages at a flower that has already been visited by a honeybee? Are the bumblebees exposed to honeybee diseases and what might the consequences be?



Last week’s Nature magazine carried an article addressing this issue. The team who carried out the work were from Royal Holloway London, Queen’s University Belfast and Exeter University. They showed that some honeybee diseases are indeed a problem for wild bumblebees and could be causing a decline in these wild pollinators. They studied two diseases: the fungal parasite Nosema which weakens honeybee colonies, and Deformed Wing Virus (DWV) which causes abnormalities in the wings and abdomen of infected honeybees as well as severely curtailing their lifespan.

The starting point for the work was to test whether these honeybee diseases could actually infect bumblebees. The researchers inoculated bumblebees (B.terrestris) with DWV or Nosema, and found that bumblebees were indeed susceptible to infection by both diseases. In the case of DWV, infection led to reduced survival of B.terrestris workers. For Nosema, although it could infect bumblebees, this did not reduce their lifespan.

Having established that the two honeybee diseases could infect bumblebees, the researchers examined the incidence of the two diseases. They performed a large scale study on the prevalence of DWV and Nosema in honeybees and bumblebees across 26 sites in Great Britain and the Isle of Man. DWV was found in 36% of honeybees tested and in 11% of bumblebees tested. For many of the infected bumblebees, the virus was active showing that the bumblebees were not simply acting as carriers. Nosema was less prevalent being found in 9% of honeybees and 7% of bumblebees. When the geographical distribution was analysed, there was some evidence for clustering, indicating disease hotspots. Hotspots for DWV were found in the south west and east of Great Britain and for Nosema in the south east. By analysing the distribution of the two diseases they were also able to show that the prevalence of DWV in honeybees influenced the prevalence of DWV in bumblebees, implying local transmission between the two insects. Local transmission was confirmed by analysing the form (nucleotide sequence) of the virus present in the two types of bee collected from the same site.

Honeybees have a high prevalence of DWV, a consequence of infestation of colonies by Varroa mites. The most obvious conclusion from this new work is that honeybee DWV is spreading to wild bumblebees. This probably occurs when the two types of bee forage in the same environment. Because DWV infection of bumblebees reduces their lifespan, the spread of this pathogen could be contributing to the decline in bumblebee numbers.

Both honeybees and bumblebees are important pollinators and need to be maintained. Their loss would have immense financial implications. This research shows that disease control in honeybee populations, for example through the efforts of beekeepers, has important implications for the health of other pollinators as well.

15 thoughts on “Bumblebees and honeybees share diseases and the outcome is not a good one.”

  1. I do wonder about the upsurge in popularity of bee keeping. Overall are the bee keepers as well informed (or better) nowadays or were the smaller number of dedicated bee keepers in the past better informed and aware of controlling the diseases?


    1. I have read that some people worry that the influx of new beekeepers could mean that the increase in the number of hives might be unstable. There is probably more than one reason for this, one might be lack of knowledge about handling diseases, another could be that with all the diseases it’s all a bit daunting and some give up. I feel slightly nervous about commenting and it would be good if an experienced beekeeper gave their view. Any comments?


      1. I’m not convinced that beekeepers in the past all knew more about treating diseases. For instance, the ‘Frow’ remedy was used in the UK to treat for Acarine mites, even though it was a toxic combination of petrol, nitro benzene and other substances. Scientific understanding of bee behaviour and diseases was not nearly so advanced as now 100, 50 or even 30 years ago.

        However I think it is the case that bees now have a worse environment to live in (not as much forage as previously, fewer nesting sites) and also more diseases and pests to contend with, because humans have spread these around the world through our activities.


  2. Philip, I enjoy your writing. This piece and the recent one on bumblebees and insecticides need some connections, however! There is a growing body of evidence that low levels of insecticide exposure – usually the neonics are the real problems – contribute to immune deficiencies in the bees. My two favorite studies that demonstrate a synergy between pathogen and pesticide are:

    1. Aufauvre J, Biron DG, Vidau C, Fontbonne R, Roudel M, et al. (2012) Parasite-insecticide interactions: a case study of Nosema ceranae and fipronil synergy on honeybee. Sci. Rep. 2: 326.

    2. Di Prisco G, Cavaliere V, Annoscia D, Varricchio P, Caprio E, et al. (2013)
    Neonicotinoid clothianidin adversely affects insect immunity and promotes replication of a viral pathogen in honeybees PNAS2013, doi:10.1073/pnas.1314923110

    The first study looks at honeybees subject to 1ppb fipronil and relatively low levels of nosema spore infections. Either stressor alone does not generate mortality much above the control but the combination kills 80% of the bees in 20 days.
    Di Prisco looks at DWV replication in the presence of several insecticides. With the neonics imidacloprid or clothianidin at levels of 1ppb and 10ppb, the DWV would replicate in the adult bees. In controls or with an organophosphate insecticide, the virus did not replicate in adult bees.
    The levels of contamination in both of these studies that cause problems with the pathogens is much lower than is usually considered problematic for the bees – yet it is very much a problem when the viruses or nosema spores are present.
    With Di Prisco they only followed the progress of the infection for three days, yet there is evidence that with the neonics – time is not your friend.

    Now look again at the study about transmission of disease from honeybees to bumblebees. Is the real problem that we have bees that succumb to low levels of infection that they would normally be able to fend off? I certainly wonder.
    Is colony collapse the perfect storm of pathogens and neonics? Why not?
    The problem with the neonics is that they continue to act in the insect long after the initial exposure has past. My guess, based upon extrapolation of the time-dependence observed in a few studies, is that the neonics have some detrimental effects in the 0.2 ppb range. If true – many bees will experience these levels much of the time. When the right pathogen comes along, the bees succumb because of the immune deficiency caused by the pesticide.
    How does that work? Neonics are supposed to be neurotoxins. I don’t know. But rigth now we are running a giant wolrdwide experiment on the effects of low levels of neonics on everything.


  3. I was just looking at some “ag maps” of the UK, and it seems that the east and south east regions are also the most high intensity ag regions, but I don’t know my UK geography that well, but wonder if the ultimate cuase of the correlation between DWV in honeybees and bumblebees is related to general levels of insecticide contaminationin the regions. Is there any data on, for instance, pesticides in waterways throughout the UK?


    1. Hi Gary, thanks for your two insightful comments which gave me a lot to think about. I apologise for being a bit slow to respond but I wanted to have a proper look at the papers you mentioned.
      In the first comment you were talking about how bees might be more susceptible to diseases when treated with different chemical pesticides/fungicides. The two papers you cited provide evidence for this idea and I would also add the paper by Pettis et al PLOS One 2013 where they showed how pollen collected from foraging bees became contaminated by fungicide and the fungicide level correlated with Nosema infection.
      The Di Prisco paper provides a mechanism whereby neonics modulate the bee’s immune response so that they are more susceptible to DWV infection. My only concern about this paper is that many of their results are obtained after topical application of chemicals. This means applying the chemical in a drop of acetone on to the thorax. They do all the necessary controls but it seems a little harsh and calls for repetition. They do perform some experiments where they put chemicals in the feeding solutions but here it looks like imidacloprid is much less effective and I think statistically non significant. In my view, more work is required before we can be sure about this.
      I do agree, however, that bees are living in a chemical soup and this may be weakening them. For the neonics, the sub-lethal effects on behaviour are also very important. My view about CCD is that it is a perfect storm as you say but I would add to the pathogens and chemicals the effects in the US of commercial beekeeping methods.
      Your second comment was about the prevalence of pesticide use in the UK in relation to the occurrence of DWV and Nosema in honeybees and bumblebees. You were wondering whether in areas of intensive agriculture the bees might be weakened by chemicals and this might account for the prevalence of the two diseases. I don’t know of any data on the amounts of chemicals used in different parts of the UK although the East is a cereal growing region which would probably imply high insecticide use. The bees could be weakened by this effect and so become more susceptible to disease but the paper I was discussing was more concerned about bee to bee transmission. I think the important piece of data is where they sequence the virus from honeybees and bumblebees collected from the same area. If the occurrence of DWV in the two kinds of bee was due to general weakening then you might expect the sequence to vary in a random manner. In fact the virus has a similar sequence from honeybees and bumblebees collected together suggesting that it has moved from one bee to another. I hope I have understood this correctly, the Figure in the paper is quite obscure and the jargon dreadful!
      Please let me know what you think.


      1. Hi Philip,
        In terms of the Fürst paper, I seem to have lost my free access to it, but I don’t have any disagreement with your interpretation of their results. Indeed, they make a case that there is transmission from honeybees to bumblebees. But I would rather ask how that can happen in the first place. At least in honeybees, DWV is known to be associated with varroa, presumably because it take the blood sucking mite to pass the virus between the bees, especially at the larval stage. Commonly the bees with DWV clearly have deformed wings when the hatch. I’m not an expert on this, but conventional wisdom is that without varroa, DWV would not be a problem. However, there have recently been antidotal reports that DWV is becoming a problem without the varroa mite vector.
        When I first encountered varroa in my backyard beekeeping operation, I first just wanted to see what would happen, so I let the varroa run its course. By the time the colony went down, the number of mites was truly astounding. There was some DWV. Now, colonies will go down with a much lower mite load. DWV is, if anything more of a problem even at lower mite infestation levels. So it is not just the number of mite bites that are encouraging the DWV.
        I’m all for the “it’s everything” theory of CCD, but I think it is more of some things than others! My fear is that the tests we use to evaluate the relative toxicity of environmental pollutants are inadequate to detect 1) toxins that accumulate over time, 2)synergies with pathogens. As a result, some chemicals have been approved that never should have. Time will tell…
        Big beekeeping in the US involves trucking bees all over and certainly could be a stressor. However, it is not just the big boys that are having trouble.
        I understand your hesitation about the topical application of insecticides in the Di Prisco paper, but I think the oral results are also very significant because the levels tested are “field relevant” and the duration of the trials was short (1 to 3 days). The test levels in the Di Prisco paper were 0.1, 1, and 10 ppb insecticide in syrup. With clothianidin there was a positive effect at 0.1ppb, imidacloprid at 1 ppb. Typical LC50 for honeybees is around 100 ppb for 10-day mortality, so these are very low sublethal exposures that are changing the insect’s immune response. (Treated crops yield 0.1 to 10 ppb in nectar and pollen – 1 to 5 ppb is probably most typical).
        I looked at the time-dependent toxicity scaling to imidacloprid, and find that it is “worse-than-cumulative.” http://squashpractice.wordpress.com/2013/05/07/time-dependent-toxicity-of-imidacloprid-in-bees-and-ants/ The fact that Di Prisco observed replication of virus after such a short time period means there was neither much time for there to be accumulation of toxin nor much time for the viral replication in the first place. From my work, this begs the question about what happens at 0.1, or 0.01 ppb or less if you allow time for the residual levels in feed to accumulate on the insect’s NChRs for a couple of weeks.
        A good example of the time-factor when dealing with pesticide-pathogen synergy is the following paper:
        1. Aufauvre J, Biron DG, Vidau C, Fontbonne R, Roudel M, et al. (2012) Parasite-insecticide interactions: a case study of Nosema ceranae and fipronil synergy on honeybee. Sci. Rep. 2: 326.
        The key point is that nothing much shows up until you wait a couple of weeks, but given time, relatively low sublethal doses of both nosema spores and fipronil were deadly.
        Finally, there are a few troubling toxicity studies that have been called into question because they could not be replicated. Specifically, Suchail 2001, found concentrations of imidacloprid in syrup in the 0.1 ppb range was killing bees. Schmuck 2003, repeated similar experiments and was unable to verify the Suchail results, but it is curious to read the Schmuck paper because at one of their four test sites, they also had unexplained high mortality, above the controls, with very low concentrations of imidacloprid. http://squashpractice.wordpress.com/2013/04/01/resolving-the-imidacloprid-paradox-and-the-ccd-connection/ It is my contention that when the right pathogen(s) (probably viruses) are present, then even minute quantities of these insecticides appear lethal because of immune suppression.
        This model, of immune suppression by the neonics, could easily explain why it has been so hard to place a finger on the cause of CCD. DiPrisco’s paper is the nail in the coffin here, mostly because the implicated insecticide levels are so low.
        Another paper…
        2. Cornman RS, Tarpy DR, Chen Y, Jeffreys L, Lopez D, et al. (2012) Pathogen Webs in Collapsing Honey Bee Colonies. PLoS ONE 7(8): e43562. doi:10.1371/journal.pone.0043562
        has a very nice figure that shows what happens in CCD colonies. It sure looks like immune system failure to me.
        Note: neonic levels are difficult to detect below 1ppb. I think some labs can go down to 0.1 ppb now. For many years, the LOD was ~5ppb. The LOD required to rule out a presence of < 0.01ppb, which could easily accumulate to levels that Di Prisco shows are problematic, is not presently available to my knowledge.
        Big question I have – Why do neonics do this, but not the OP insecticide tested? My suspicion is that excitotoxic effects influence the immune function. OP insecticides at residual levels don’t do much damage because the intermediate enzyme AChE function is still largely intact at residual pesticide concentration levels.

        A skeptical response is appreciated… I need to know what is wrong with my argument!



  4. Hi Gary,
    Thanks again for your insightful comments, and again I must apologise for my slow response, I had to finish off some other work before I could get back to the bee problem. Here are my comments:
    1. From what you say, honeybees have become more susceptible to DWV, how widely has this been observed by other beekeepers?
    2. I agree completely with you that the testing of chemicals on the environment has been inadequate and for the neonics the time dependent and sub lethal effects have not been factored in. This was a cornerstone of the EU case for banning the three neonics last year
    3. Di Prisco used very low, sub-lethal concentrations of pesticides and found effects on immunity. I don’t dispute the clonthianidin effect but the P value for imidacloprid in the feeding experiments is 0.055 which according to most statistical measures is non-significant.
    4. What is the experimental evidence that the neonics act as irreversible blockers of nAChRs?
    5. You cite the Aufavre paper on fipronil as an example of the time factor when dealing with pesticide/pathogen synergy but fipronil is not a neonic; it has a different mechanism of action and as far as I can see it is not irreversible.
    6. I agree that if the neonics affect immune function then this could be very important for CCD. I would really like to see Di Prisco replicated to be sure of the importance
    7. Related to the previous point, I wanted to address your question about lack of effect of organophosphate (OP) insecticide on immune function in the Di Prisco paper. The OP should inhibit acetylcholinesterase and therefore acetylcholine should build up and over-stimulate the nAChRs, or at least that’s the conventional wisdom. The neonics are direct acting acetylcholine analogues, so both the neonics and OP should produce the same effect unless the assumed mechanisms are incorrect. Perhaps as you say, the levels of OP used are low and acetylcholinesterase action is still intact. But if that is the case then Di Prisco has not tested the OP correctly and that result must be set aside.

    I shall be interested to hear your response, I think it’s very important we get to the bottom of all of this.
    Best wishes


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s