Female heterogamety and speciation: reduced introgression of the Z chromosome between two species of nightingales

Abstract

Several lines of evidence suggest that the X chromosome plays a large role in intrinsic postzygotic isolation. The role of the Z chromosome in speciation is much less understood. To explore the role of the Z chromosome in reproductive isolation, we studied nucleotide variation in two closely related bird species, the Thrush Nightingale (Luscinia luscinia) and the Common Nightingale (L. megarhynchos). These species are isolated by incomplete prezygotic isolation and female hybrid sterility. We sequenced introns of four Z-linked and eight autosomal loci and analyzed patterns of polymorphism and divergence using a divergence-with-gene flow framework. Our results suggest that the nightingale species diverged approximately 1.8 Mya. We found strong evidence of gene flow after divergence in both directions, although more introgression occurred from L. megarhynchos into L. luscinia. Gene flow was significantly higher on the autosomes than on the Z chromosome. Our results support the idea that the Z chromosome plays an important role in intrinsic postzygotic isolation in birds, although it may also contribute to the evolution of prezygotic isolation through sexual selection. This highlights the similarities in the genetic basis of reproductive isolation between organisms with heterogametic males and organisms with heterogametic females during the early stages of speciation.

Figure 1

Geographic distribution of the Common Nightingale (dark gray) and the Thrush Nightingale (light gray) with indicated sampled populations. (1) Distal allopatry for the Common Nightingale (Spain), (2) close allopatry for the Common Nightingale (Czech Republic, SW Poland), (3) close allopatry for the Thrush Nightingale (NE Poland), and (4) distal allopatry for the Thrush Nightingale (Finland). The zone of sympatry is depicted only schematically owing to uncertainty about the exact position and width of the hybrid zone.

Figure 2

Neighbor-joining trees for four Z-linked (A) and eight autosomal (B) loci. The nodes that do not have majority bootstrap support have been compressed. Because recombination events were detected for most loci, the trees represent overall similarity between the haplotypes; portions of each gene may have slightly different genealogies.

Figure 3

The marginal posterior probability distributions for the demographic parameters of the IM model. Effective population size is shown for L. megarhynchos (N₁), L. luscinia (N₂), and ancestral population (N_A). Migration rates are from L. luscinia to L. megarhynchos (m₁), and from L. megarhynchos to L. luscinia (m₂). Distributions are shown for all genes (A), autosomal genes (B), and Z-linked genes (C).