Weak Epistasis May Drive Adaptation in Recombining Bacteria

Abstract

The impact of epistasis on the evolution of multi-locus traits depends on recombination. While sexually reproducing eukaryotes recombine so frequently that epistasis between polymorphisms is not considered to play a large role in short-term adaptation, many bacteria also recombine, some to the degree that their populations are described as "panmictic" or "freely recombining." However, whether this recombination is sufficient to limit the ability of selection to act on epistatic contributions to fitness is unknown. We quantify homologous recombination in five bacterial pathogens and use these parameter estimates in a multilocus model of bacterial evolution with additive and epistatic effects. We find that even for highly recombining species (e.g., Streptococcus pneumoniae or Helicobacter pylori), selection on weak interactions between distant mutations is nearly as efficient as for an asexual species, likely because homologous recombination typically transfers only short segments. However, for strong epistasis, bacterial recombination accelerates selection, with the dynamics dependent on the amount of recombination and the number of loci. Epistasis may thus play an important role in both the short- and long-term adaptive evolution of bacteria, and, unlike in eukaryotes, is not limited to strong effect sizes, closely linked loci, or other conditions that limit the impact of recombination.

Keywords: approximate Bayesian computation; bacteria; epistasis; homologous recombination; multilocus selection.

Figure 1

Observed pairwise compatibility vs. distance. (A) Patterns of PC (green) vary among the species included in this study. Since PC measures the compatibility of two SNPs with a single phylogeny, these data indicate that SNPs > ∼1kb apart have distinct phylogenetic histories from recombination, with the exception of S. aureus which exhibits linkage. (B) Simulated patterns of PC vs. distance (brown) using parameter estimates from Table 1 fit observed data well. We note that the sensitivity of PC to sample size (n) makes these patterns not directly comparable across species, and that the product of effective population size and recombination (or ρ = 2Nr) affects PC.

Figure 2

Higher recombining bacteria may have greater selection responses when multilocus epistatic traits are controlled by many loci (L = 10). (A) Selection responses of simulations parameterized by bacterial recombination rates, relative to an asexual control (R/R_asex), for increasing pairwise epistatic effects (Ns_i) when additive effects per locus are weak (Ns_a = 1; left) or strong (Ns_a = 10; right). (B) LD, as measured by the mean D’ across all locus pairs. For each plot, the mean of 2000 simulations is shown for each parameter set, and selection responses were calculated after 0.2 N generations of selection. Simulations were parameterized with values for S. aureus (dark blue), C. jejuni (light blue), S. pneumoniae (orange), N. gonorrheae (light red), H. pylori (dark red), and a eukaryote (black).

Figure 3

Bacterial recombination rates do not hinder short-term responses to selection for multilocus epistatic traits composed of few loci (L = 3). (A) Selection responses of simulations parameterized by bacterial recombination rates, relative to an asexual control (R/R_asex), for increasing pairwise epistatic selection effects (Ns_i) when additive effects per locus are weak (Ns_a = 1; left) or strong (Ns_a = 10; right). (B) LD, as measured by the mean D’ across all locus pairs. For each plot, the mean of 2000 simulations is shown for each parameter set, and selection responses were calculated after 0.2 N generations of selection. Colors follow the same scheme as Figure 2.

Figure 4

The effect of starting conditions on relative responses to selection. Shown are selection responses relative to an asexual control (R/R_asex) when recombination is suppressed during selection, illustrating how levels of haplotype diversity from mutation-recombination-drift balance affect selection responses, as opposed to the recombination that occurs during selection. Here, relative selection responses are measured after 0.2 N generations for a 10-locus trait (A) or a three-locus trait (B). For each plot, the mean of 2000 simulations is shown for each parameter set. Colors follow the same scheme as Figure 2. Epistatic effects per locus pair are shown on the x-axis, and additive effects per locus increase from the left column to the right column, with Ns_a = 1 or 10 for (A) or Ns_a = 3.33 or 33.3 for (B).