Does Mate Limitation in Self-Incompatible Species Promote the Evolution of Selfing? The Case of Leavenworthia Alabamica

Abstract

Genetic diversity at the S-locus controlling self-incompatibility (SI) is often high because of negative frequency-dependent selection.In species with highly patchy spatial distributions, genetic drift can overwhelm balancing selection and cause stochastic loss of Salleles. Natural selection may favor the breakdown of SI in populations with few S-alleles because low S-allele diversity constrains the seed production of self-incompatible plants. We estimated S-allele diversity, effective population sizes, and migration rates in Leavenworthia alabamica, a self-incompatible mustard species restricted to discrete habitat patches in rocky glades. Patterns of polymorphism were investigated at the S-locus and 15 neutral microsatellites in three large and three small populations with 100-fold variation in glade size. Populations on larger glades maintained more S-alleles, but all populations were estimated to harbor at least 20 S-alleles, and mate availabilities typically exceeded 0.80, which is consistent with little mate limitation in nature.Estimates of the effective size (N e ) in each population ranged from 600 to 1600, and estimated rates of migration (m) ranged from 3 × 10 −4 to nearly 1 × 10 −3 . According to theoretical models, there is limited opportunity for genetic drift to reduce S-allele diversity in populations with these attributes. Although pollinators or resources limit seed production in small glades, limited S-allele diversity does not appear to be a factor promoting the incipient breakdown of SI in populations of this species that were studied. K E Y W O R D S :Baker's Law, effective population size, frequency-dependent selection, migration, S-alleles, self-compatibility.

Insights Gained From 50 Years of Studying the Evolution of Self-Compatibility in Leavenworthia (Brassicaceae)

“…The partially selfing a2 population is fixed for an S-allele (purple dots) which is also found at a low frequency in a geographically close SI population. Phenotypic characterization of the S-alleles in selfing populations suggests that they both disrupt the pollen component of SI (Busch et al 2011) outcrossing floral morphology, yet self-fertilizes substantially (s m = 0.523), such that appreciable levels of selfing do not require floral adaptations typically found in selfing populations (Busch et al 2010b). Examination of the genetic basis of SC has revealed that both of the S-alleles found in the selfing populations of L. alabamica have a dysfunctional pollen component of SI, suggesting mutations in the SCR gene, as would be expected in theory (Uyenoyama et al 2001).…”

“…Recent work has made progress on understanding how and why this evolutionary transition occurred in L. alabamica (Busch et al 2010b(Busch et al , 2011. This species has co-occurring SI and SC populations, and an S-linked marker permits the study of S-alleles (Busch et al 2008;Fig.…”

Evolution of theS-Locus Region in Arabidopsis Relatives

“…One hypothesis for the labile nature of SI is that the transition from outcrossing to selfing can provide a selective advantage at a given time and place (Stebbins, 1957;Busch et al, 2010;Goldberg et al, 2010;Wright and Barrett, 2010). A single inactivated S-locus haplotype was fixed in C. rubella, representing the key step in the recent transition to selfing for this species (Guo et al, 2009).…”

Environmental requirements trump genetic factors in explaining narrow endemism in two imperiled Florida sunflowers

“…Population inbreeding coefficients (F IS ) for all populations of H. carnosus or P. tenuifolius were either not significantly different from zero or significantly lower than zero, indicating that there was less inbreeding than expected (Table 1). This lack of inbreeding may be due to the combination of sporophytic self-incompatibility with generally small population sizes in these species, where limited numbers of S-alleles skew successful matings strongly toward pairings of the least related individuals (Busch et al 2010;Charlesworth and Charlesworth 1987). This may have long-term Unlike the narrow endemics, four populations of H. radula had inbreeding coefficients significantly higher than zero, indicating higher levels of inbreeding than expected, and both the species grand mean and pooled species-level inbreeding coefficients were also significantly higher than (Table 1).…”