Methylation turns a wannabe bumblebee into a queen

Bumblebees in an established hive live a high school rom-com social life. At the center table is the queen bee, surrounded by males that she created. On the outskirts sit the worker-bee nerds, all female, doomed to never get a date or form a hive of their own. But as in the movie Mean Girls, when the queen bee goes down, sometimes a worker has a chance to claw her way toward the throne. Which wannabee rises to the top is more than personality or fashion sense. It’s a matter of epigenetics.

In the rigid social structure of a bumblebee (Bombus terrestris) hive, the queen does the breeding and the workers do the working. But when a queen goes missing, starts making males and new queens, or even simply gets old, some of the worker females rise up in revolt. They become more aggressive, their ovaries develop and they start laying eggs to produce their own males. They will never be a true queen bee, but laying their own eggs puts them in competition.

Entomologist Harindra Amarasinghe and colleagues at the University of Leicester in England were interested in what makes some workers go reproduction rogue. Their results, published February 18 in the Proceedings of the Royal Society B, show that removing chemical tags on a worker bee’s DNA leads to aggressive bees that are more likely to lay eggs of their own. The findings show an important role for epigenetics in bumblebees, and suggest that whether a gene is turned off or on depends on whether it comes from the father or the mother. 

Amarasinghe and colleagues first looked at what happened when a queen bee was taken away from the hive. Sure enough, some of the workers developed into aggressive reproductive usurpers. When the scientists compared the workers that started reproducing with those that didn’t, they found that the two groups had different patterns of gene methylation. Epigenetic changes, such as chemical tags called methyl groups attached to a gene, determine whether and how much of a gene is made into a protein. In this case, the different patterns of methylation were associated with a more aggressive, reproducing worker bee.

To determine whether methylation changes were actually causing the difference in bee behavior, the scientists used the drug decitabine. This drug is normally used to treat acute myeloid leukemia, and it works by inhibiting the enzyme that puts methyl groups on DNA. When the leaderless hives received decitabine in their nectar, the worker bees became much more likely to develop ovaries and lay eggs. So changing methylation patterns in the bumblebees directly affected their role in the hive.

Gene Robinson, a geneticist at the University of Illinois at Urbana-Champaign, says that the paper “opens a new line of study on the role of epigenetics in controlling reproductive conflict” in bumblebees. David Haig, a biologist at Harvard University, agrees. “These are nice results,” he says, “and they show there’s a methylation system present and that genetic methylation is associated with differences in behavior and development.” 

Evidence for a methylation system in bumblebees could be a sign of something deeper going on, suggests Eamonn Mallon, an entomologist at the University of Leicester and the lead author on the study. He thinks a phenomenon called genomic imprinting may be at play. Genomic imprinting is when a gene is turned on based on which parent it came from. For example, a matrigene is a gene that is expressed only when you receive it from your mother. You will get two copies, one from mom and one from dad, but only mom’s is going to get genetic play. In Angelman’s syndrome a child receives two copies of genes on chromosome 15 from his or her father, and a single inactivated copy of those genes from his or her mother. Chromosome 15 is matrilineal, so the mother’s gene gets expressed no matter what, and neither of the father’s copies is expressed. The result is severe intellectual and developmental disability.

While scientists have known that genomic imprinting occurs in humans, there is no direct evidence for it in bumblebees. Mallon hypothesizes that the methylation effects his group saw are signs of genomic imprinting, but he still has to prove it. “There is nothing that directly shows that these changes in behavior are different depending on the parental origin of the gene,” Haig explains, “so that’s something that still needs to be demonstrated.”

Mallon acknowledges that he and his colleagues have work ahead of them. “What you have to do,” he explains, “is find a gene that has only one copy expressed.” Then you need to perform a series of crosses between two bee colonies, where all bees in colony A are gene type A (say, long fuzz), and all bees in colony B are gene type B (short fuzz). When you cross them all together you will get one group that is AB and one group that is BA. If the gene does not show genomic imprinting, whichever gene is dominant will show, so the two groups will have the same fuzz length. But if the two groups show different fuzz lengths, it means that genomic imprinting is taking place. They are the same genetically, but one gene in each group is imprinted and not expressed. It’s an elegant experiment. “But,” Mallon says, “that is a couple years’ work ahead of me.”

Robinson notes that genomic imprinting can’t be confirmed until the full genome of the bumblebee is available, but says that this paper “represents an important first step.” Future work could tell scientists a lot more about how genetic and epigenetic changes control social behavior, and predict what makes a worker reach for queenly status.