Apocephalus borealis, a new threat to honeybees?

If you’re not a long time reader of Biofortified, you might not know that I study parasitoid wasp physiology when I’m not armchair-quarterbacking head louse treatment studies. It’s not often that parasitoids end up on the news, so when they do I get super-excited. I’ll be writing this from the perspective as a parasitoid biologist. A shorter article which does an excellent job of tackling misconceptions about this paper discussed here can be found at Biodiversity in Focus.

Frank likes bees too.

When I was an undergraduate, I spent about a year or so working as a beekeeper. It was a fun job, and I learned all sorts of fun facts about bees. By this time I had been interested in parasitoids for nearly a decade and a half, having raised parasitic wasps out of caterpillars since I was five. Naturally, I attempted to see if there were any parasitoids which attacked Apis mellifera but I always ended up empty handed and disappointed. This always confused me because there were parasitoids which attacked ants, termites and caterpillars living in ant nests. I never understood why parasitoids had never been documented attacking honeybees.
This changed earlier this week, when a description of a parasitoid fly which attacks bees was published in PLOS ONE: A New Threat to Honey Bees, the Parasitic Phorid Fly Apocephalus borealis by Core et al. Unfortunately, the authors tried way too hard to connect the fly to Colony Collapse Disorder, but I’ll discuss that later. First…

What are Phorid flies? What are parasitoids? Why do we care?

Figure 1 from Core et. al 2012. I labeled the identifying characteristic of the family Phoridae with a black arrow and the ovipositor which is convergent with a wasp stinger with a red arrow.

Parasitoids are insects which are parasitic as larvae, free living as adults and which kill their hosts after development is complete. Most parasitoids are either flies or wasps, with wasps being the best studied. They’re important to agriculture because they’re good at regulating the populations of their hosts by killing them in large numbers. My studies revolve around how these insects evade the immune system, which gives us a springboard to learn more about how insect immunity works on a biochemical level.
Parasitoids in the family Phoridae are particularly interesting. Most species, such as the very common Megaselia scalaris, are actually scavengers but some species have made the leap to parasitism. The paper lists a particularly great example of parasitoid phorids, the decapitating flies which are used as fire ant biocontrol. These flies can be identified by a bunch of scrunched up veins on their wings, labeled by a black arrow in the first picture.
Figure 1 B from Core et. al 2012. The fly is visible laying it’s eggs into the abdomen of the bee to the left of the picture.

Parasitoids go through standard holometabolous development of egg, larvae, pupa and adult. The adults lay eggs either on or inside their hosts. The larvae develop within the hosts, pupate, and then hatch into adults. Each of these stages requires particular adaptations, and parasitoid wasps and flies use completely different strategies. Larvae, for example, must evade the immune system. Wasps tend to suppress the immune system through venom or polydnaviruses. Flies, on the other hand, tend to hide in tissues which aren’t easily accessible to the immune system. They also have a tendency to hijack the immune system in some rather impressive ways like using melanization machinery to build snorkels to keep the parasitoid larvae supplied with air.
Sometimes, however, parasitoid flies and wasps solve similar problems with similar solutions. This parasitoid oviposits inside the host like the wasp, but uses a hard spike on it’s fleshy ovipositor to help it insert it’s eggs into the bee. I labeled this with a red arrow in picture A.
In the picture above, you can see the fly laying it’s eggs into the abdomen of the bee. From here, the eggs hatch and hatch into larvae which feed on the bee’s tissues. After this the larvae emerge in a particularly gruesome and characteristic fashion, between the head and thorax. The larvae emerge from under the ‘chin’ of the bee after the bee leaves the colony at night.
Fly larvae emerging from honeybee host, highlighted with red arrows.

This is a particlularly interesting example of a host shift. Apocephalus borealis is a specialist on bees and wasps, so a shift to Apis mellifera makes sense because it has a similar immune system. Honeybees hail from Africa, whereas this parasitoid is uniquely American so this is a definite example of a native parasitoid infecting and adapting to a new host.

What about Colony Collapse Disorder?

Despite the good job the authors did documenting the development of this parasitoid inside the honeybee, they lose my enthusiasm when they get to the colony collapse stuff. I really think they did some good work on natural history in this paper, that is they did some good work on looking at how the flies develop in the bees in labs. However, there are some things which weren’t very well fleshed out in this paper which mostly pertain to CCD.
This portion of their paper to me seemed like an attempt to connect this fly to CCD and this part of their work could have been done better in my view. This is a criticism that could be made of any paper, but given what I think they were trying to show I think they should have done some work quantifying the numbers of disease organisms present in the flies. In their defense, I think the main purpose of this paper was the natural history work on this parasitoid. The CCD work appears to be done as almost an afterthought, but given the high profile of the paper I thought this warranted it’s own dissection.

To investigate internal hive behavior and possible infections within a hive, we kept an observation hive in a laboratory near our primary study hive. Samples taken from the observation hive in June 2010 confirmed infection with A. borealis. Rates of infection varied between June 2010 and December 2010 (Mean = 25% Range = 12%–38%) peaking over the sample period in November at 38%. In September, the number of bees in the hive declined and we observed phorid pupae and empty pupal casings among dead bees at the bottom of the hive, indicating emergence of adult phorids within the hive and the potential for A. borealis to multiply within a hive and infect a queen.

Observation hives, at least the ones I’m familiar with, consist of a portion of beeswax housed in a plexiglass cage. Most bees are kept in very different environments which generally shield them from light. I’m unsure how this would change the bee behavior and their ability to clean their hive, so let’s overlook the fact this was observed in a single hive kept differently than how most bees are housed. There’s a bigger issue here.
At my university, I am the curator of the insect zoo. We used to have some pretty big problems with Megaselia scalaris in our roach colonies. M. scalaris is a very common phorid fly that develops on dead and dying insects. Weakened hives aren’t able to fight off scavengers, so it’s possible in my view that Megaselia could have been in the hive after the populations declined because the large number of fresh dead bees would be the perfect environment where this scavenger could develop. The authors didn’t explicitly explain how they differentiated Apocephalus puparia from Megaselia puparia, and I think this is a fatal flaw in their work. These are two very similar looking Phorids with ecological habits that couldn’t really be more different. Megaselia does not develop as a parasitoid, and would thus pose no threat to the bees.  I think this is a rather important oversight and I wouldn’t trust their conclusions without further explanation of how they differentiated between the two. Put bluntly, I don’t think this piece of data should have been in this paper without that information.
Secondly, they did a bunch of tests looking for bee pathogens in the Phorids. They looked for genetic material, correctly noting this didn’t necessarily indicate that the pathogens were growing inside the flies. Quite frankly, I think it would be very strange if a parasitoid which fed on tissues of a bee didn’t consume any of it’s pathogens even if those pathogens didn’t infect the fly. Many science writers have been confused by this result, and many articles give the impression that the authors thought the flies vectored the pathogens. I find that doubtful but I won’t completely rule it out. Either way, I need to point out that it doesn’t come anywhere near proving it because there is no data indicating the pathogens were growing inside the flies. They also pointed out a correlation between Phorid emergence and the point of the year when colonies collapse, but demonstrating causation needs more data. The authors explicitly stated all of this, but some writers didn’t realize this.
Third, the majority of the data dealing with how the bees act when infected with the flies appears to have been conducted on a very specific building of the campus of San Francisco State University. That’s well and good, but urban beekeepers are a very specific subset of beekeepers and this data might not be relevant to most beekeepers, and thus irrelevant to the hives most important to agriculture. I think they need to do a lot more work on behavior before we can make any solid conclusions on how parasitized bees act in the wild before emergence of the parasitoids.
I don’t blame them for trying to connect this parasitoid to CCD, because that’s the hot topic right now in bee biology. However, I don’t think this fly really has much to do with the collapse of honeybee colonies. Despite this, this is still a really important find because this fly could shed more light on CCD, but not in the direct way the researchers imply in this paper. This is an insect which invades the bee and evades it’s defenses. Figuring out how the fly evades the bee’s defenses could shed light on how the bee’s immune system works. Figuring out how the bee’s immune system works might help us figure out how whatever pathogen actually does cause CCD is also able to evade the defense mechanisms of the bees. Once we fully understand the bee’s defense mechanisms, we can then think about potential interventions based on this data.
Edit 9/29/2010 12:43 AM
I had originally said that some of the work in this paper was poorly performed, but I could have been more explicit about why I thought this was. Looking back on this article, I felt I was perhaps a bit more harsh than I should have been. A lot of the work in this paper was actually performed quite well in my view, but I feel like they should have paid more attention to the work they performed on the pathogens. There could be good reasons for this…a lack of proper equipment in the lab for instance, or it could be a bias caused in part by my background in molecular biology. The authors were very measured in their claims made in the paper, and have been in their quotes in the popular media, more so than I implied in this article. So let me explain why I was a bit nonplussed by the pathogen work they did in this paper…which was mostly pushed into a supplemental figure I might add. In my mind, I was writing this section more to reply to claims made in the popular media on this paper and I feel like I ended up unintentionally trashing the entire paper for something that was relatively minor.
If one looks at figure S2, you can see the results of this work. The sample sizes were low, perhaps as a result of not being able to get samples. They performed this work using a microarray, a technique that involves fixing DNA fragments to chips, labeling a sample and then looking to see if complimentary sequences bind to the sequences immobilized on the chip. This is good for the type of present/absent work done in this paper.
My main issue comes down to how they performed the tests. They ground up whole larvae and adults to perform the microarray tests. I work with organisms which are about the size of many phorids I’ve seen, and I’ve dissected tissues out of them to try to perform some minor experiments. To me, this wasn’t particularly difficult although I can see why it might be for some depending on dexterity and tools available. It would have been very easy to dissect out the guts of the parasitoids and perform these tests on tissues of the parasitoid that wouldn’t contact food. This would have given a much better answer on whether or not the pathogens could *possibly* infect these flies. After all, if the virus shows up in the body wall then that means that it made it into the phorid instead of being digested in the gut.
They could have also tested the bees they raised the flies out of for pathogens. If the pathogens were lost in the flies, but present in the bees then this may have hinted at vector/host relationships. The sample sizes were pretty low, which is why they probably didn’t put a whole lot of effort into the tests and shoved them in a supplementary figure. This is a question for further research.
Given that the popular media tends to pick up any honeybee pathogen as the ONE TRUE CAUSE FOR COLONY COLLAPSE, I would have paid far more attention to this work and done something more than a present/absent test as done in this paper. A role for this fly in colony collapse disorder is not at all out of the question, but given the complexity of the system I wouldn’t immediately assume that this fly is a major player. A lot of hard work needs to be done to determine whether or not it is…and the people from this lab will undoubtedly be at the forefront of this work.
And now… some gratuitous parasitoid videos. These are particularly impressive examples of physical grace and biochemical warfare from some of the most incredible critters in the world.

A special thanks to Bug Girl for pointing out the Biodiversity In Focus article.
Core, A., Runckel, C., Ivers, J., Quock, C., Siapno, T., DeNault, S., Brown, B., DeRisi, J., Smith, C., & Hafernik, J. (2012). A New Threat to Honey Bees, the Parasitic Phorid Fly Apocephalus borealis. PLoS ONE, 7 (1) DOI: 10.1371/journal.pone.0029639


  1. Hi Joe,
    Thats a nice read. good job. I was intrigued by the paper and to be frank, their link to CCD seemed like a good story… at least it explained why you were not seeing the honey bees…:) but after reading it in depth and thanks to your points here…I know better..
    Good job again

  2. There should be a concern and more awareness about the challenges our American Honeybees and Bumblebees are having not only with the Varroa Mite, but the Apocephalus Borealis parasite as well. We need to be concerned about what chemicals we use in our own gardens and agriculture. Thank you for bring this story to the forefront. Good read and good story.

  3. “By this time I had been interested in parasitoids for nearly a decade and a half, having raised parasitic wasps out of caterpillars since I was five.”
    Joe, I can only assume that your favourite movie growing up was “Alien.”

  4. You know, Richard, I’ve always had a thing for parasitoid themed B-movies. Aliens is awesome because it’s merely a biocontrol project from the pest’s point of view. There’s a movie called ‘Ticks’ which is pretty good. The life cycle of the critters from the ‘Tremors’ movies are also pretty cool, with multiple young developing inside the first large animal and then going through a fairly significant morphological change in the process.
    I…know of way too many movies on this subject 🙂

  5. Pingback: Brainfood
  6. Nice read. Good information and reasoning that perhaps will eventually lead to a fuller understanding by researchers studying CCD through this particular fly.

  7. After the USDA/ARS distribution of several species of phorid fly like Pseudacteon Tricuspis which started almost 20 years ago, Apocephalus borealis pops into the picture to starts to infect honey bees. Is there a connection between the seeding of phorid flies to control fire ants and the crossover of Apocephalus to honey bees? I wonder? Is the road to hell really paved with good intentions?

Comments are closed.