Limits to Adaptation in Partially Selfing Species

Abstract

In outcrossing populations, "Haldane's sieve" states that recessive beneficial alleles are less likely to fix than dominant ones, because they are less exposed to selection when rare. In contrast, selfing organisms are not subject to Haldane's sieve and are more likely to fix recessive types than outcrossers, as selfing rapidly creates homozygotes, increasing overall selection acting on mutations. However, longer homozygous tracts in selfers also reduce the ability of recombination to create new genotypes. It is unclear how these two effects influence overall adaptation rates in partially selfing organisms. Here, we calculate the fixation probability of beneficial alleles if there is an existing selective sweep in the population. We consider both the potential loss of the second beneficial mutation if it has a weaker advantage than the first one, and the possible replacement of the initial allele if the second mutant is fitter. Overall, loss of weaker adaptive alleles during a first selective sweep has a larger impact on preventing fixation of both mutations in highly selfing organisms. Furthermore, the presence of linked mutations has two opposing effects on Haldane's sieve. First, recessive mutants are disproportionally likely to be lost in outcrossers, so it is likelier that dominant mutations will fix. Second, with elevated rates of adaptive mutation, selective interference annuls the advantage in selfing organisms of not suffering from Haldane's sieve; outcrossing organisms are more able to fix weak beneficial mutations of any dominance value. Overall, weakened recombination effects can greatly limit adaptation in selfing organisms.KEYWORDS adaptation; selection interference; dominance; self-fertilization; recombination S ELF-FERTILIZATION-reproduction where both gametes arise from the same parent-frequently evolves from outcrossing species in nature. Self-fertilization is widespread in angiosperms (Igic and Kohn 2006), some groups of animals (Jarne and Auld 2006), and fungi (Billiard et al. 2011;Gioti et al. 2012). It confers an initial benefit to an individual's fecundity, including up to a 50% transmission advantage (Fisher 1941) and reproductive assurance under mate limitation (Baker 1955(Baker , 1967Pannell et al. 2015). Both factors should allow selfing organisms to rapidly spread upon invasion of new habitats, unless countered by high levels of inbreeding depression (Lande and Schemske 1985). However, empirical studies usually find that selfing lineages are a "dead end," since back transitions to outcrossing are rare, and high extinction rates have been inferred from comparative studies of related selfing-outcrossing taxa (Igic et al. 2008;Goldberg et al. 2010;Wright and Barrett 2010;Wright et al. 2013).Self-fertilization has therefore been posited to be detrimental in the long term. For an organism with selfing rate s, the population has an inbreeding rate F ¼ s=ð2 2 sÞ; equivalent to Wright's (1951) F IS statistic. The effective population size N e is reduced by a factor of at least 1=ð1 þ...

Sexual selection and inbreeding: Two efficient ways to limit the accumulation of deleterious mutations

“…In an initially outcrossing population exposed to regular selfing, purging is therefore expected to slow down as the recessive semi‐lethals are eliminated and the small‐effect alleles persist. Their persistence may also be facilitated by the decrease in the effective size due to both homozygosity and selective interferences under self‐fertilization (Hartfield and Glemin ), which is confirmed here by the lower genetic diversity of the S lines compared to all others. This persistence might explain that inbreeding depression still persisted at G50.…”

“…However, it is not clear whether purging by inbreeding would be efficient at preventing decreases in population fitness over a larger number of generations and with higher selfing rates (i.e., nearly obligate selfing instead of the average selfing rate of 50% we imposed). Selective interference and reduced effective size may eventually limit the efficiency of purging (and of selection in general) in predominant selfers (Lande and Porcher ; Hartfield and Glemin ; Noël et al. ), making them evolutionary dead‐ends (Goldberg et al.…”

Effects of partial selfing on the equilibrium genetic variance, mutation load, and inbreeding depression under stabilizing selection

“…Due to the lower genetic diversity of self‐fertilizing populations, it has been suggested that they should be less able to adapt to a changing environment (e.g., Stebbins ; Williams ; Takebayashi and Morrell ). In the absence of epistasis, existing models indeed predict that selfing populations should have lower rates of adaptation than outcrossing ones (Glémin and Ronfort ; Hartfield and Glémin ). When compensatory effects between mutations are possible, however, a substantial amount of genetic variance may be hidden by genetic associations between loci in highly selfing populations (Lande and Porcher , the present study).…”

Effect of partial selfing and polygenic selection on establishment in a new habitat

“…When loci are tightly linked, the effective rate of recombination and effective population size are significantly reduced, even in populations with not very close to 1, rendering selection against deleterious alleles ineffective (Kamran‐Disfani and Agrawal ). By the same token, linkage generates Hill–Robertson interference between adaptive alleles, which reduces the advantage selfers experience during adaptation from co‐dominant or even mildly recessive alleles (Hartfield and Glémin ). Finally, linkage increases hitchhiking of deleterious variants with adaptive alleles, especially in highly selfing populations, which reduces the fixation probability of even slightly recessive adaptive alleles (Hartfield and Glémin ; Kamran‐Disfani and Agrawal ).…”