Fine-scale genetic breaks driven by historical range dynamics and ongoing density-barrier effects in the estuarine seaweed Fucus ceranoides L

Abstract

Background: Factors promoting the emergence of sharp phylogeographic breaks include restricted dispersal, habitat discontinuity, physical barriers, disruptive selection, mating incompatibility, genetic surfing and secondary contact. Disentangling the role of each in any particular system can be difficult, especially when species are evenly distributed across transition zones and dispersal barriers are not evident. The estuarine seaweed Fucus ceranoides provides a good example of highly differentiated populations along its most persistent distributional range at the present rear edge of the species distribution, in NW Iberia. Intrinsic dispersal restrictions are obvious in this species, but have not prevented F. ceranoides from vastly expanding its range northwards following the last glaciation, implying that additional factors are responsible for the lack of connectivity between neighbouring southern populations. In this study we analyze 22 consecutive populations of F. ceranoides along NW Iberia to investigate the processes generating and maintaining the observed high levels of regional genetic divergence.

Results: Variation at seven microsatellite loci and at mtDNA spacer sequences was concordant in revealing that Iberian F. ceranoides is composed of three divergent genetic clusters displaying nearly disjunct geographical distributions. Structure and AFC analyses detected two populations with an admixed nuclear background. Haplotypic diversity was high in the W sector and very low in the N sector. Within each genetic cluster, population structure was also pervasive, although shallower.

Conclusions: The deep divergence between sectors coupled with the lack of support for a role of oceanographic barriers in defining the location of breaks suggested 1) that the parapatric genetic sectors result from the regional reassembly of formerly vicariant sub-populations, and 2) that the genetic discontinuities at secondary contact zones (and elsewhere) are maintained despite normal migration rates. We conclude that colonization and immigration, as sources of gene-flow, have very different genetic effects. Migration between established populations is effectively too low to prevent their differentiation by drift or to smooth historical differences inherited from the colonization process. F. ceranoides, but possibly low-dispersal species in general, appear to be unified to a large extent by historical, non-equilibrium processes of extinction and colonization, rather than by contemporary patterns of gene flow.

Figure 1

Genealogy and distribution of the mtIGS haplotypes of Fucus ceranoides from NW Iberia. (a) Location of the study area (in black) in relation to Europe and the Iberian Peninsula. The geographical location and mtIGS lineages present in four Iberian populations previously analysed in Neiva et al. (2010, 2012) are also shown. (b) MtIGS parsimony networks of NW Iberian haplotypes. Sampled haplotypes are represented by circles sized to their frequency and black dots represent inferred, unsampled haplotypes. Links represent a single nucleotide change and black squares represent small indels. Inferred phylogroups are labelled by colour and letter (A-Red; B-Purple; C-Green). Shared and private haplotypes are depicted in bright and pale colour intensity, respectively. (c) Location of sampling sites, delimitation of the phylogeographic sectors considered (Western- W; Northwestern- NW; Northern- N). Pie charts depict haplotype frequencies at each site (see Table 1 for haplotype ID’s, haplotypes are coloured as in b)

Figure 2

MtIGS phylogeny and genealogic congruence in Iberian Fucus ceranoides. (a) 50% majority-rule consensus tree of mtIGS haplotypes of F. ceranoides, rooted with F. vesiculosus. Numbers above the branches are Bayesian posterior probabilities (> 0.70). Inferred phylogroups are highlighted in grey. (b) FCA plot based on individual multilocus genotypes. Individuals are labelled according to their mtIGS lineage. Note the correspondence between the mtIGS phylogeny and the nuclear population structure. The individuals from VIA, VIL, SAN and BAY were included in both analyses.

Figure 3

Mismatch distributions of the mtIGS phylogroups A, B and C of Fucus ceranoides. The grey bars and the solid lines depict the observed and expected (under the spatial expansion model) values, respectively.

Figure 4

Genetic subdivision of Iberian Fucus ceranoides based on STRUCTURE. Shown are the proportions of individual multilocus genotypes assigned to each of K virtual clusters, as illustrated by the different colours. The individuals from VIA, VIL, SAN and BAY were also included. Population codes are given in Table 1.

Figure 5

Isolation by distance in Fucus ceranoides from NW Iberia. Estimates of pairwise differentiation (F _ST/1-F_ST) are plotted against geographic distance for (a) NW Iberia (b) W sector, (c) NW sector and (d) N sector. The regressions are: y = 0,0045x + 0,1134), y = 0,0056x + 0,0934, y = −0,0006x + 0,3652 and y = 0,0023x + 0,0543, respectively.