ARTICLE IN BRIEF
Socially unresponsive bees were found to share overlapping gene sets with humans with autistic spectrum disorder.
Scientists have found significant overlap between the genes in the brains of honey bees demonstrating socially unresponsive behavior and lists of genes that have been found to have links to autism spectrum disorders (ASD) in humans.
The findings show a high conservation of genes linked to social behavior, offering further evidence of a strong genetic underpinning for ASD, according to the authors of the July 31 report in the Proceedings of the National Academies of the Sciences.
“The social systems likely evolved independently in bees and humans,” said Gene Robinson, PhD, the lead author and director of the Carl R. Woese Institute for Genomic Biology at the University of Illinois, where he uses the Western honey bee to understand the evolution and mechanisms of human behavior. “But our paper suggests even if you have independent evolutions of social behavior, those independent evolutions can draw upon similar genes.”
Independent experts said that a honey bee model could be useful, but had concerns about the human gene lists used for comparison and cautioned against concluding that these genes are specific to autism.
In the study, researchers exposed 2,460 bees — broken up into 246 groups of 10, seven-day old honey bee nest mates — to two kinds of social stimuli. One was a social challenge in which an unrelated bee was introduced to the group, which can provoke an aggressive response. The other stimulus was a social opportunity in which bees were given a chance to rear a bee larva as a queen, which can prompt alloparental care.
The researchers then focused on three behavior groups: the “guards,” which consistently responded only to the territorial threat and comprised 7.7 percent of those studied; the “nurses,” which consistently responded only to the rearing opportunity and comprised 9.3 percent; and the “unresponsive” group that didn't respond to either stimulus and comprised 14 percent.
The other bees, for the most part, had weak responses to one stimulus or the other. A very small percentage consistently responded to both stimuli, although these bees were not evaluated further because it is highly unusual behavior, since bees have a division of labor not given to both duties.
Many genes — 1,057 — were differentially expressed between the guard, nurse and unresponsive groups. Then the researchers compared this list to human lists of ASD-associated genes. To do this, they used a test they developed, called Orthoverlap, which allows for comparison of gene lists that have been developed independently and that cover long evolutionary distances, using weighting to account for the strength of the overlap within groups of genes.
They found significant overlaps with two independent gene lists developed through expression studies using brain tissue gathered at autopsy. Then they turned to the gene variants associated with ASD that came from the Simons Foundation Autism Research Initiative (SFARI), which includes at least some variants that are assumed to have a causal link to ASD behaviors. They found significant overlap with this gene list as well.
“The SFARI database is divided into seven sets of genes that vary in the confidence level with which they are related to ASD,” researchers wrote. “When comparing the bee DEG [database of essential genes] list to the highest confidence SFARI sets, the overlap was statistically significant. When weaker confidence SFARI sets were included, statistical significance weakened proportionately.”
Comparisons of the bee gene list and gene lists related to other disease types, including schizophrenia and depression, didn't uncover any significant overlap.
Dr. Robinson acknowledged the complexity of ASD, saying it “encompasses many traits, not just unresponsiveness to social stimuli,” which was assessed in this study.
“Bees are not little people and people are not big bees, so there's a gulf, a world of difference between them. We're not trying to trivialize ASD in any way at all.”
Nonetheless, he said, honey bees have value in that “these are very social animals. They live in very intricate societies, all of their activities are influenced by the activities of other individuals in their society. They represent one of the pinnacles of social evolution.
“As such, they present an interesting opportunity to be able to study the root of social behavior. And with genomics, we now have an amazing opportunity to do this in a way that we never could before.”
The last common ancestor for honey bees and humans is thought to be a marine flatworm that lived 600 million years ago and had a very simple nervous system, no centralized brain and no social life. So bee society and human society, it would follow, evolved independently.
“These genes for some reason, which we don't yet understand, are involved in elaborating the brains' systems for social behavior, and that evolution can repeatedly call upon those systems for social behavior,” Dr. Robinson said.
The findings also offer an avenue for further research, he said, since genes previously implicated in autism “now are getting this kind of second endorsement, if you will, from an unlikely source, the honey bee.”
“Those genes now become of particular interest, to find out what do they do and how might they be contributing to social responsiveness in general,” he said.
Stephen Scherer, PhD, director of the Centre for Applied Genomics at the Hospital for Sick Children in Toronto, said, “I think the model is a good one... to explore general themes common between honey bee and human in sociobiology.”
But he cautioned that the human gene lists that were used, other than the SFARI list, have not been validated, although they contain some of the same genes — these are “not bad data, they just need to be replicated by another group,” he said.
He said the model could be used to look at conserved genes that are not on the SFARI list or other lists, including the genes identified in work from his laboratory and presented in a recent paper.
“These could be interesting genes not yet found in humans,” he said.
“The most interesting finding for me,” he said, “is that there could be so many genes involved in socio-behavior.” But he said it isn't surprising that they would be conserved across evolution. In a 2015 paper, his lab found that social behavior is the most significant signal found among different mutation carriers.
He said a bee model would have to be used cautiously.
“They aren't humans or exposed to the complexities of inputs during development,” he said. “I like the shared biology, but it needs to be interpreted carefully.”
Shafali Spurling Jeste, MD, assistant professor of psychiatry and neurology and a behavioral child neurologist specializing in autism at the University of California, Los Angeles, said the findings “speak to the robust biology of social behavior and that there are clearly certain genetic predispositions and genetic factors that contribute to social impairment.”
“The cautionary tale I would tell is that what they didn't look for... are genes that are related to intellectual disability or global developmental delay or other true neurodevelopmental syndromes, not psychiatric. Because I would actually imagine that there's a tremendous amount of overlap. We need to be careful when we think about how specific these genes are to social behavior.”
A preclinical model using honey bees could have value in helping develop treatment, she said. “Our goal is not to completely recapitulate autism in a mouse or in a bee,” she said. “Once we have clear robust behaviors that we can measure, then we can start asking whether we can modulate those with treatment before we test them in humans.”