Journal Club - Balancing selection shapes density-dependent foraging behaviour

Post date: Dec 12, 2016 4:39:24 PM

People have long been curious about how genetics affects behavior. Recent studies have found many interesting cases, such as in mice, where different sets of genes control burrow length and whether the mice build an escape tube. In "Balancing selection shapes density-dependent foraging behaviour," Joshua Greene and colleagues investigate the genetic basis of insensitivity to certain pheromones in the roundworm Caenorhabditis elegans. Through some beautiful work, they show that two reasonably common alleles of the gene srx-43 confer different pheromone responses. On a behavioral level, this manifests itself as a difference in the worms' propensity to stay put or roam about as they exhaust a food source. Depending on the feeding environment, either allele can be advantageous.

In the most fascinating twist, this allelic difference is part of a highly diverged haplotype that is 30 kilobases long, with the srx-43 alleles differing at 20% of their bases! The authors suggest this results from balancing selection - if two alleles are maintained without recombining in an interbreeding population, over time they will accumulate different mutations. What's interesting to me is that for this to work, there needs to be very little recombination between the diverged alleles when they encounter each other in a diploid. C. elegans has impressively low recombination rates and is mostly self-fertilizing (i.e., inbred), and, moreover, sequence divergence itself will block recombination, but, given the sheer number of generations required to accumulate such high divergence, I think something more is required. There could be selection against recombination in this area (say, due to pairs of suppressor mutations that cause lethality if separated), or a recombination-blocking inversion. My favorite possibility is that one of the two alleles could be an introgression from another species of roundworm. This species would likely be currently unknown to us, as Greene checked that the two alleles are more similar to each other than to sequences from the closest known relatives of C. elegans.