Bee-havioral epigenetics

A great debate right now that is very relevant to nutrition is to what extent genes vs. environment (or genes in the context of certain environments) influence health and behavior.  The following is an extreme example that cannot be extrapolated to humans, but highlights an interesting co-evolution with dietary constituent(s) that is critical to a species’ survival.

Two years ago, Kucharski et al. provided an explanation for why the honey bee, Apis mellifera is able to develop into a queen or worker bee, whose characteristics are very different (e.g. fertile and larger vs. nonfertile and active etc., very different behaviorally), by having essentially identical DNA.  By turning down DNA methyltransferase 3 (DNMT3) with RNA interference, they were able to create bees with the characteristics of a queen.  Naturally, this happens with a longer intake of royal jelly, suggesting that something in royal jelly influences the activity of the methylation enzyme DNMT3.  Worker bees are the result of similar progeny being removed of the jelly at a different developmental phase.

In this recent study (open access) (published today- yay PLoS press list!) Lyko (along with Kucharski and others) examined methylation of brain tissue from these bees and verified previous research showing only a small part of the bee genome is methylated (and that part is highly conserved), in contrast to mammalian genomes.  There was significantly different methylation of 561 genes between queens and workers (and differences in drones as well).  Methylation is restricted to CpG dinucleotides (which it is not in humans) in the bee.  A number of other technical but seemingly novel findings were described, but I wanted to summarize the discussion:

Phenyl butyrate in royal jelly is a known deacetylase inhibitor so the authors suggest the jelly could function as both an epigenetic regulator and source of nutrients (see wikipedia page for nutrient profile- but please don’t buy it as a dietary supplement!).  The authors draw comparisons from rodent studies that have found epigenetic mechanisms in their nervous system as well, emphasizing their conservation.   Of importance in the bee model; there is no methylation of the gene promoters- instead on the bodies:

…rather than switching the genes on and off by promoter methylation, the intragenic methylation in  Apis operates as a modulator of gene activities. As a result the entire topology of a complex brain   network can be reprogrammed by  subtle adjustments of many genes that act additively to produce a given phenotype [38]. Such adjustable   DNA methylation levels generating variability in the transcriptional output of methylated genes could underlie genetically inherited propensity to pheno-typic variability in accord with the recently proposed model of stochastic epigenetic variations as a heritable force of evolutionary change [39].

We humans have a lot more methylation, and usually at different sites on the gene, if I understand correctly.  The true value of this research, as suggested by Maleszka in the press release:

“This study represents a giant step towards answering one of the big questions in the nature-nurture debate, because it shows how the outside world is linked to DNA via diet, and how environmental inputs can transiently modify our genetic hardware,” he said.

“Similar studies are impossible to do on human brains, so the humble honey bees are the pioneers in this fascinating area.”


Lyko F, Foret S, Kucharski R, Wolf S, Falckenhayn C, & Maleszka R (2010). The Honey Bee Epigenomes: Differential Methylation of Brain DNA in Queens and Workers PLoS Biology : 10.1371/journal.pbio.1000506