Shift-inducible [transgenerational] increase in recombination rate as an evolving strategy in a periodic environment: a numerical model

Abstract

Numerous empirical studies have witnessed a plastic increase in meiotic recombination rate in organisms experiencing physiological stress due to unfavourable environmental conditions. Yet, it is not clear enough which characteristics of an ecological factor (intensity, duration, variability, etc.) make it stressogenic and therefore recombinogenic for an organism. Several previous theoretical models proceeded from the assumption that organisms increase their recombination rate when the environment becomes more severe, and demonstrated the evolutionary advantage of such recombination strategy. Here we explore another stress-associated recombination strategy, implying a reversible increase in recombination rate each time when the environment alternates. We allow such plastic changes in the organisms, grown in an environment different from that of their parents, and, optionally, also in their offspring. We show that such shift-inducible recombination is always favoured over intermediate constant optimal recombination. Besides, it sometimes outcompetes also zero and free optimal constant recombination, therefore making selection on recombination less polarized. Shift-inducible strategies with a longer, transgenerational plastic effect, are favoured under slightly stronger selection and longer period. These results hold for both panmixia and partial selfing, although selfing makes the dynamics of recombination modifier alleles faster. Our results suggest that epigenetic factors, presumably underlying the environmental plasticity of recombination, may play an important evolutionary role.

Modeling the evolution of recombination plasticity: A prospective review

“…Specifically, the stressful state was defined either as the rare (atypical) one 93 (Figure 2A) or as one with a stronger selection 94 (Figure 2B). In contrast to the two above‐mentioned studies, in which a certain environmental state (either rare or severe) was considered to be stressful, a recent model explored a symmetric environment – with the two alternating states differing only in terms of selection direction 95 . Under such a scenario, none of the states could be defined as stressful, and the recombination rate increased after each selection reversal, that is, in individuals who arose in an environment differing from that of their parents (Figure 2C).…”

“…This trade-off explains why plastic recombination was often rejected in selection regimes where the optimal constant strategy implied either zero or free recombination. 94,95 It probably also explains the disadvantage of plastic recombination in extremely asymmetric environments. 94 However, a remarkable finding is that the benefits of recombination plasticity quite often do outbalance the nonoptimality costs.…”

“…This may be rarity (recombination rates increase in an atypical environment), severity (recombination rates increase in an environment with strong selection), or lability (recombination rates increase after the environment alternates). Each of these options was previously modeled [93][94][95] (see Figure 2) but never in a single setting. Second, it is not clear which fitness components or genotype peculiarities modulate the recombination rate and its plasticity.…”

“…). There exist some indications that the fate (advantage/disadvantage) of a plastic strategy is more determined by the range of the population's mean fitness over the environmental period 95 . Another open question is to what extent the evolution of recombination plasticity is affected by patterns of environmental fluctuations.…”

“…However, only Zhuchenko et al reported a simultaneous group-selection advantage 93 ; in two other models, plastic recombination increased the population's mean fitness but could nevertheless be sometimes rejected. 94,95 It is not clear whether this discrepancy reflects profound peculiarities of the considered forms of recombination plasticity, or originates just from different selected systems (two loci directly affecting fitness in ref. 93…”