Social immunity has ancient roots in insect evolution. Ants urgently carry their dead to refuse piles, termites prefer burials and honey bees drag their dead and diseased brood out to their doorstep. This serves the same purpose for insects as quarantine did for humans in the smallpox era: in a cramped, crowded place, disease spreads like wildfire unless the infected individuals are removed. In previous issues of the American Bee Journal, Randy Oliver and Keith Delaplane did a fantastic job at introducing social immunity in honey bees, so by now we all know why Varroa destructor-sensitive hygiene (VSH) and hygienic behavior are beneficial traits. What I’d like to do in this article is start answering ‘how?’ How do the bees determine who needs to be removed and who doesn’t? What are they detecting? Are the mechanisms evolutionarily conserved? These topics have gained a lot of traction among scientists and with more experiments, including one I’m in the process of publishing, the gaps are finally starting to be filled.

Social immunity strategies exhibited by honey bees, ants and termites [1]
First, a brief history: E. O. Wilson, a famous American entomologist, coined the term ‘necrophoresis’ to describe how ants systematically isolate their dead from the nest. As far back as the 1950s he discovered that applying a dab of oleic acid to inanimate objects would stimulate the other ants to carry the objects off to their midden.2 Later, he found this holds true even when the substance is applied to live ants – despite them being very much alive and kicking – and since then oleic acid has been a well-accepted “necromone,” or pheromone inducing necrophoresis. Oleic acid is quite viscous (think: olive oil) so its odor doesn’t travel through the air, but ants are often bumping shoulders and touching antennae so the oily necromone is still readily spread. This is now widely accepted as the “death cue” hypothesis – that viscous compounds signaling the insect’s death trigger its removal to the midden. However, this doesn’t explain how the behavior is still triggered even for individuals which haven’t been dead long enough for the death cue to develop. This led to a competing “vital sign” hypothesis, based on the idea that a change in a recognizable vital sign can occur at a faster time scale than the production of a death cue. We still don’t know which idea is more correct, but like most things, it is probably some combination of the two.

Bees have it a little tougher than ants, though, since there’s a physical barrier of the cap separating the workers from the brood they would like to inspect. Without the benefit of direct contact, researchers have speculated that the odors stimulating hygienic behavior (the bee version of necrophoresis) must be more volatile than the oily substance found in ants and termites. Swanson et al., under the supervision of Marla Spivak (the professor who bred and distributed the Minnesota hygienic line) set the stage in 2009 by identifying odors emitted from chalkbrood that induced hygienic behavior.3 There have been many other investigations into disease-specific odors in honey bees, but this was the first one to show that the odors actually induced the behavior. The compounds they identified fit the bill: they were very volatile and should have no problem permeating the brood cap and alerting the workers above. However, not all diseases are so straight-forward.

Very recently Mondet et al. found that varroa-infested brood which was targeted for hygienic behavior emitted higher amounts of brood ester pheromone (BEP) compared to those that were not targeted.4 As far as I know, this was the first indication that deviations in brood pheromone profiles, rather than some exotic disease-specific cue, might be contributing to social immunity. Although it hadn’t been proposed before, this idea is intuitive: colony functions are finely controlled via pheromone signals, so why should hygienic behavior be any different? In my own research, I have gathered evidence suggesting that BEP is not the only pheromone involved.

I have recently been on a mission to identify potential hygienic behavior-stimulating odors from freeze killed brood (FKB). FKB might be less biologically interesting than studying an actual disease; however, since it’s the main method for hygienic behavior selective breeding, it’s very likely to release reproducible hygienic behavior-inducing odors. And since bees selected in this way are also able to target real diseases (including chalkbrood and varroa), it’s possible there are similarities between the odors emitted from FKB and those emitted by the various diseases. In my own research, which is under review at Scientific Reports (and publicly available at, I found that one other brood pheromone (beta ocimene) and oleic acid (read: ant necromone) was emitted from FKB. However, when we tested how well these odors stimulated the bees’ antennae, we found that only β-ocimene produced dose-dependent antenna responses whereas oleic acid didn’t stimulate them at all.

Beta ocimene is particularly intriguing because this pheromone is already known to increase worker visits to the emanating brood cells,5 so by recruiting attention to capped cells which shouldn’t otherwise need it, the pheromone could be quite an effective alert signal. As the second brood pheromone to be associated with hygienic behavior, this supports the idea that there is a broader pattern of pheromones regulating social immunity. This may not be true for all diseases, since we’ve seen previously that it’s not necessarily the case with chalkbrood.3 I suspect it may not be the case for American foulbrood either, which produces such obviously ghastly odors that additional pheromone cues are unlikely to be necessary. But for the subtler disease states (varroa infestation, FKB, and I suspect virus infection as well), pheromone cues could be key.

Oleic acid, however, has been giving me grief. On one hand, there are several strong lines of evidence pointing to it being a good hygienic behavior-inducer: 1) we already know it serves that very function in other social insects,2 and 2) I found that it’s abundantly released from FKB of all ages and across diverse colonies. Being evolutionarily conserved in ant and termite species, it would certainly make sense if oleic acid is involved in honey bee behaviors as well. But it gets better: we know from previous work that bees depend on odorant binding proteins (OBPs) to detect odors, and hygienic bees have significantly more of certain OBPs compared to unselected bees.6,7 These OBPs are thought to transport odor molecules from the antenna pores to the olfactory neurons, which is necessary for signal recognition in the brain. And get this – oleic acid is one of the strongest known ligands for those OBPs.6 Yet on the other hand, it does not appear to stimulate their antennae in our experiments, presumably because it has such low volatility.

The idea that non-volatile compounds could be triggering hygienic behavior in honey bees goes against conventional wisdom, but we are not the first ones to find non-volatiles associated with disease states.  For example, a compound by the code name “p32” is found in higher amounts in varroa-infested cells targeted for removal.8 There are reasonable explanations for why oleic acid didn’t stimulate the bees’ antennae, too: the temperature of the hive is ~ 23 °F (10 ºC) warmer than the room where we conducted the experiments, which would make the effective concentration of airborne oleic acid higher, or perhaps we needed to use larger amounts to be realistic.

This brings me full circle, back to the old “death cue” vs. “vital sign” hypothesis. In our experiments with FKB, we identified one death cue (oleic acid) and one vital sign (β-ocimene). Others have identified a death cue (p32) and vital sign (BEP) in varroa-infested brood as well. Now, this is only a hunch which has not been tested yet, but with these observations it’s hard not to think that the death cue and vital sign mechanisms are working together (Figure 2). Here is a hypothetical sequence of events: first, the pheromone vital sign calls out to the hygienic workers, triggering their attention; next, upon uncapping and closer inspection via direct contact with the brood, the non-volatile death cue becomes detectable much like it is among ants and termites. With the bee season getting underway, we are looking forward to testing these ideas and hopefully put to rest the death cue vs. vital sign debate.

Fig2 (1)

Here is a hypothetical sequence of events: first, the pheromone vital sign calls out to the hygienic workers, triggering their attention; next, upon uncapping and closer inspection via direct contact with the brood, the non-volatile death cue becomes detectable much like it is among ants and termites. With the bee season getting underway, we are looking forward to testing these ideas and hopefully put to rest the death cue vs. vital sign debate.

This article appeared in the April 2017 issue of American Bee Journal.


  1. Sun Q and Zhou X. “Corpse management in social insects.” International journal of biological sciences. 9(3):313-21. (2013).
  2. Wilson EO, Durlach NI and Roth LM. “Chemical releaser of necrophoric behavior in ants.” 65(8) 108-14. (1958).
  3. Swanson JA, Torto B, Kells SA, Mesce KA, Tumlinson JH and Spivak M. “Odorants that induce hygienic behavior in honeybees: Identification of volatile compounds in chalkbrood-infected honeybee larvae.” Journal of chemical ecology. 35(9): 1108-16. (2009).
  4. Mondet F, Kim SH, de Miranda JR, Beslay D, Le Conte Y, Mercer AR. “Specific cues associated with honey bee social defence against Varroa destructor infested brood.” 6: 25444. (2016).
  5. He XJ, Zhang XC, Jiang WJ, Barron AB, Zhang JH and Zheng ZJ. “Starving honey bee (Apis mellifera) larvae signal pheromonally to worker bees.” Scientific reports. 6: 22359. (2016).
  6. Guarna MM, Melathopoulos AP, Huxter E, Iovinella I, Parker R, Stoynov N, Tam A, Moon KM, Chan QWT, Pelosi P, White R, Pernal SF and Foster LJ. “A search for protein biomarkers links olfactory signal transduction to social immunity.” BMC Genomics. 16: 63. (2015).
  7. Hu H, Bienefeld K, Wegener J, Zautke F, Hao Y, Feng M, Han B, Fang Y, Wubie AJ and Li J. “Proteome analysis of the hemolymph, mushroom body and antenna provides novel insight into honeybee resistance against Varroa Journal of proteome research. 15: 2841-54. (2016).
  8. Wagoner K and Reuppell O. “Brood chemical associated with common honey bee stressors and hygienic response.” Poster presentation at The North American beekeeping conference and tradeshow. Galveston TX. (2017).