Population genetic structure of Lemon Sharks in the Western Atlantic: is there evidence of gender-biased dispersal and differences between neutral and adaptive loci?

Abstract

Sampling difficulties and low genetic diversity have limited population genetic studies of large, vagile shark species. Through extensive sampling (580 individuals from 12 locations) and a multilocus approach involving two mitochondrial DNA loci (control region and ND2, composite sequence 1,730 bp) and eight microsatellites, I was able to delineate the population genetic structure of the lemon shark, Negaprion brevirostris, in the western Atlantic. I articulated two hypotheses to explain genetic structure in coastal sharks: (1) females exhibit natal philopatry to parturition sites, resulting in local population structure and (2) males are dispersive, resulting in high male-mediated gene flow. I predicted that maternally inherited mitochondrial genetic markers would be highly structured even on a local geographic scale, while bi-parentally inherited microsatellite markers would exhibit little to no genetic structure over the same range. Significant genetic structure was detected in the mitochondrial composite sequence of ND2 and control region (Φ_ST =0.293, p<0.000001), with at least seven distinct groups evident in the sampling area (East Peninsular Florida, West Peninsular Florida/Tiger Beach (Bahamas), Lower Florida Keys/Bimini (Bahamas)/Belize, Eleuthera (Bahamas), Louisiana, U.S. Virgin Islands and Brazil). Significant genetic structure was individually detected in both the coding ND2 (Φ_ST =0.293, p<0.000001) and the non-coding control region (Φ_ST =0.278, p<0.000001), but the ND2 gene was found to be an inappropriate locus to test for local adaptation in lemon sharks because all of the mutations were silent with the exception of a single mutation found in two sharks. Very weak genetic structure was also detected in nuclear microsatellites (F_ST =0.014; p<0.088), but only between the Brazilian population and all of the others. A Bayesian analysis of the microsatellite data failed to reject a null hypothesis that there is one population in the region. Both classes of genetic marker indicated that geographic distance between sampling areas was correlated with genetic distance between them. All of these findings are consistent with my predictions and support the hypotheses of natal philopatry in females and high male-mediated gene flow. Fine-scale local genetic structure driven by behavior makes this species and others like it much more vulnerable to local fishing or habitat destruction than resource managers currently appreciate. I suggest future work should sequence larger regions of the mitochondrial genome to fully resolve population structure in this species and to investigate the possibility of local genetic adaptation in the coding regions.