Population genetics of speciation in two closely related wild tomatoes (Solanum section Lycopersicon)

Abstract

We present a multilocus sequencing study to assess patterns of polymorphism and divergence in the closely related wild tomato species, Solanum peruvianum and S. chilense (Solanum section Lycopersicon, Solanaceae). The data set comprises seven mapped nuclear loci (approximately 9.3 kb of analyzed sequence across loci) and four local population samples per species that cover much of the species' range (between 80 and 88 sequenced alleles across both species). We employ the analytical framework of divergence population genetics (DPG) in evaluating the utility of the "isolation" model of speciation to explain observed patterns of polymorphism and divergence. Whereas the isolation model is not rejected by goodness-of-fit criteria established via coalescent simulations, patterns of intragenic linkage disequilibrium provide evidence for postdivergence gene flow at two of the seven loci. These results suggest that speciation occurred under residual gene flow, implying that natural selection is one of the evolutionary forces driving the divergence of these tomato species. This inference is fully consistent with their recent divergence, conservatively estimated to be <or=0.55 million years. We discuss possible biases in the demographic parameter estimates due to the current restriction of DPG algorithms to panmictic species.

Figure 1.—

Geographic ranges of S. peruvianum (dark shading) and S. chilense (cross hatching) and approximate locations of the sampled populations. The S. peruvianum populations, from south to north (open diamonds), are Tarapaca (TAR, elevation 400 m), Arequipa (ARE, 2180 m), Nazca (NAZ, 2140 m), and Canta (CAN, 2050 m). The S. chilense populations, from south to north (solid dots), are Antofagasta (ANT, 2900 m), Tacna (TAC, 1250 m), Moquegua (MOQ, 2450 m), and Quicacha (QUI, 1830 m). ANT and TAR were previously studied by Städler et al. (2005; see their Figure 1). Note the region of sympatry in northernmost Chile and southern Peru. S. peruvianum sensu lato extends to northern Peru, but populations north of ∼9°–10°S latitude are now regarded as the separate species S. arcanum (see text).

Figure 2.—

Scatter plot of average locus-specific rates of recombination and median P-values for single-population contrasts (LD test of gene flow). Recombination is expressed as γ per site, and the locus-specific values are the means obtained for the S. peruvianum populations (i.e., both species are plotted with the S. peruvianum recombination estimates). The P-values are locus-specific medians obtained from up to nine interspecific population contrasts. The lower left area of the plot contains the data for loci CT066 and CT166 (cf. Table 6).