Phylogeographic history of flat periwinkles, Littorina fabalis and L. obtusata

Abstract

Background: The flat periwinkles, Littorina fabalis and L. obtusata, are two sister species widely distributed throughout the Northern Atlantic shores with high potential to inform us about the process of ecological speciation in the intertidal. However, whether gene flow has occurred during their divergence is still a matter of debate. A comprehensive assessment of the genetic diversity of these species is also lacking and their main glacial refugia and dispersal barriers remain largely unknown. In order to fill these gaps, we sequenced two mitochondrial genes and two nuclear fragments to perform a phylogeographic analysis of flat periwinkles across their distribution range. Results: We identified two main clades largely composed by species-specific haplotypes corresponding to L. obtusata and L. fabalis, with moderate to strong support, respectively. Importantly, a model of divergence with gene flow between the two species (from L. obtusata to L. fabalis) was better supported, both in Iberia and in northern-central Europe. Three mitochondrial clades were detected within L. fabalis and two within L. obtusata, with strong divergence between Iberia and the remaining populations. The largest component of the genetic variance within each species was explained by differences between geographic regions associated with these clades. Our data suggests that overall intraspecific genetic diversity is similar between the two flat periwinkle species and that populations from Iberia tend to be less diverse than populations from northern-central Europe.Conclusions: The phylogeographic analysis of this sister-species pair supports divergence with gene flow. This system thus provides us with the opportunity to study the contribution of gene flow and natural selection during diversification. The distribution of the different clades suggests the existence of glacial refugia in Iberia and northern-central Europe for both species, with a main phylogeographic break between these regions. Although the genetic diversity results are not fully conclusive, the lower diversity observed in Iberia could reflect marginal conditions at the southern limit of their distribution range during the current interglacial period.

Hybridization patterns between two marine snails, Littorina fabalis and L. obtusata

“…This suggests that hybridization could have been more frequent in the past, prior to the evolution of multiple reproductive barriers, including in currently allopatric sites. However, because mtDNA is unlinked to selected nuclear loci, we cannot exclude that mtDNA introgression can persist and spread over long into L. obtusata (Sotelo et al, 2020). Asymmetric mtDNA introgression into L. fabalis is in line with previous work showing that males of both species prefer larger females (Saltin, 2013), which suggests that most successful interspecific crosses in Northern…”

“…and "*" indicates allopatric locations. Results are consistent across replicates sorting, introgressive hybridization is supported by three main lines of evidence: (a) hybrids between the two species were detected, including F2s and backcrossed individuals (this study andCarvalho et al, 2016), suggesting not only that hybridization is possible but also that at least some F1s are viable and fertile; (b) a model of divergence with gene flow for this system has a better fit than a model without gene flow(Sotelo et al, 2020); and (c) in contrast with mtDNA, sequences from two nuclear fragments showed no shared haplotypes between the two species (except in Cabo do Mundo)(Sotelo et al, 2020), suggesting that incomplete lineage sorting is unlikely to explain the observed patterns in mtDNA versus nuclear genes.…”

“…Chromatograms were visually inspected and sequences aligned using Sequencher 4.1.4 (Gene Codes Corporation), with subsequent trimming for equal length (569 bp) and manual correction of artifacts. From the 819 samples analyzed in this study, 117 (14.3%) had previously been sequenced for this fragment by Sotelo et al, Table ).…”

“…Three lines of evidence suggest that mtDNA introgression is asymmetric in Iberia: (a) a higher proportion of L. fabalis populations with signatures of introgression when compared with L. obtusata (59% vs. 47%); (b) a slightly higher overall proportion of mtDNA introgression into L. fabalis than the other way around (18.5% vs. 16.2%); and (c) individuals from the different hybrid classes show a higher proportion of typical L. obtusata over L. fabalis haplotypes. However, this asymmetry in Iberia is much weaker than the one observed in Northern Europe, where the overall proportion of mtDNA introgression into L. fabalis was 35% versus 6% into L. obtusata (Sotelo et al, ). Asymmetric mtDNA introgression into L. fabalis is in line with previous work showing that males of both species prefer larger females (Saltin, ), which suggests that most successful interspecific crosses in Northern European sites preferentially involve a L. obtusata female and a L. fabalis male.…”

“…A fragment of the mitochondrial gene cytochrome b ( Cyt‐b ) was amplified using the cytbF‐cytbR primer pair (Panova et al, ). Amplification was carried out as in Sotelo et al, and Sanger sequencing was performed at Macrogen Europe (Amsterdam, The Netherlands) using the forward primer. Chromatograms were visually inspected and sequences aligned using Sequencher 4.1.4 (Gene Codes Corporation), with subsequent trimming for equal length (569 bp) and manual correction of artifacts.…”