Constrained Genetic Architecture Promotes Cooperation

Abstract

Cooperation in nature often has direct costs but only indirect benefits. Kin and group selection theories comprehensively address its evolutionary origins, but our knowledge of the precise genetic mechanisms that prevent cheater invasion and maintain cooperation is incomplete. Here we review our published work on cooperation in Aevol, an agent-based, in silico genomic platform used to evolve and study populations of digital organisms that compete, reproduce, and cooperate by secreting a public good. Motivated by the observation of phenotypically identical individuals who had radically different evolutionary fates, we recorded and compared gene locations, effectively performing bio-inspired genomics analyses of our digital organisms. We found that the association between metabolic and secretion genes (promoter sharing, overlap via frame shift or sense-antisense encoding) was characteristic for populations with robust, stable cooperation. Such architecture arose de novo during the evolution of cooperation, but only when producing the public good was costly. Effectively, cooperation evolved to be protected and maintained through constrained, entangled genetic architecture. Beyond confirming the importance of second-order selection, we uncover a novel genetic mechanism for the evolution and maintenance of cooperation. Our results suggest a method to limit the evolutionary potential of synthetically engineered organisms, in order to reduce the change or loss of synthetic gene circuits, a major issue in synthetic biology. Background and Introduction The evolution of cooperation in microbial populations is a fascinating, rich and controversial evolutionary problem (West et al. 2006). Evolutionary explanations of cooperation are constantly being improved and refined by a host of mathematical tools such as game theory and meta-population models (Lehmann and Keller 2006). However, these methods typically do not distinguish between genotypes and phenotypes and are unable to investigate the structure of genomes that encode the cooperative traits. Microbial studies have already hinted at potential mechanisms by which genetic architecture can affect cooperation, including gene coregulation and pleiotropy (Foster et al. 2004; Dandekar et al. 2012). Here we review our published work (Frenoy et al. 2013) that examines two specific types of genomic architecture: (1) operon sharing, when secretion and metabolism genes are on the same operon and share a promoter and a terminator, and (2) overlap, the base-pair sharing between metabolic and secretion genes possible when genes are in different reading frames or on different DNA strands. Well described in viruses, overlaps in bacteria are existent but rare. Similar to operons, they are thought to be caused by strong maximum genome size constraints or selection for co-regulation (Normark et al. 1983). However, overlaps have not themselves been studied as an evolutionary constraint, which we do here, in the context of cooperation.