Evolution of Functional Diversification within Quasispecies

Abstract

According to quasispecies theory, high mutation rates limit the amount of information genomes can store (Eigen’s Paradox), whereas genomes with higher degrees of neutrality may be selected even at the expenses of higher replication rates (the “survival of the flattest” effect). Introducing a complex genotype to phenotype map, such as RNA folding, epitomizes such effect because of the existence of neutral networks and their exploitation by evolution, affecting both population structure and genome composition. We reexamine these classical results in the light of an RNA-based system that can evolve its own ecology. Contrary to expectations, we find that quasispecies evolving at high mutation rates are steep and characterized by one master sequence. Importantly, the analysis of the system and the characterization of the evolved quasispecies reveal the emergence of functionalities as phenotypes of nonreplicating genotypes, whose presence is crucial for the overall viability and stability of the system. In other words, the master sequence codes for the information of the entire ecosystem, whereas the decoding happens, stochastically, through mutations. We show that this solution quickly outcompetes strategies based on genomes with a high degree of neutrality. In conclusion, individually coded but ecosystem-based diversity evolves and persists indefinitely close to the Information Threshold.

Evolving Living Technologies—Insights from the EvoEvo Project

“…Although the high mutation rate occasionally leads to fitness loss through loss of the master sequence, the increased evolvability enables the population to quickly recover, hence keeping the fitness of the best individuals at the same level (and sometimes at a higher level) than that of the wild-type individuals that evolved under a constant mutation rate. Similar results have been obtained in another class of model studied in the course of the project: by evolving RNA sequences, E. S. Colizzi [8] has shown that the RNA structure (the equivalent of the genome structure in the RNA model) is selected such that the population contains an efficient proportion of mutated sequences and such that the mutants help the master sequence to thrive.…”

Second‐order cooperation: Cooperative offspring as a living public good arising from second‐order selection on non‐cooperative individuals

“…Put differently, when considering the long-term reproductive success of a lineage, the genotype of the ancestor contains more information than its phenotype alone, because it determines its position on the mutational landscape. This question of mutational landscape location as a selectable property of an individual is a much broader one, central to, for example, the theory of quasispecies (Eigen and Schuster 1977;Nowak 1992;Wilke et al 2001;Colizzi and Hogeweg 2014).…”

“…Our work also directly connects with studies considering division of labor, another kind of social interactions between individuals. For example, Colizzi and Hogeweg (2014) present an RNA-world model where one "master sequence" encodes all the information necessary to produce (via mutations) a full ecosystem. Similarly, Rainey and Kerr (2010) suggest that non-cooperative types may be seen as a germline, whose ability to form a cooperative soma can be subject to selection.…”

“…At first glance, the stable coexistence model may seem more plausible, at least under a reasonably low mutation rate, because it does not require second-order selection increasing the rate of a very specific mutation (contingency loci). However, the continuous generation model, exemplified by Colizzi and Hogeweg (2014), may be more relevant when mutation rates are very high, population bottlenecks are frequent, or selective sweeps are fast, making the coexistence of several lineages highly unlikely. All of the examples of stable coexistence we cited above indeed happened in Lenski's Long-Term Experimental Evolution (LTEE) lines, whose experimental setup is characterized by weak bottlenecks (1 in 100 each day, which corresponds to about 5 × 10 5 founder individuals).…”

“…While such phenomena have not been much studied in the context of cooperation, there are multiple examples of selection acting on the ability to mutate toward a particular phenotype due to selection increasing the likeliness of a particular mutation (contingency loci) (Silverman 1979;Abraham et al 1985;Moxon et al 1994Moxon et al , 2006Anjuwon-Foster and Tamayo 2017). Going one step further, Colizzi and Hogeweg (2014) raise the question of the ability of a single sequence to encode an ecosystem, where second-order selection would shape the mutational landscape such that different likely mutants could stably interact together. In summary, it has been recognized for a long time that natural selection can act on the probability of a genotype to mutate toward a sequence giving an alternative phenotype, in a similar way that it can act to produce regulatory networks permitting switching between different phenotypes without changing the genotype.…”