FURTHER EVIDENCE AGAINST MEIOTIC-DRIVE MODELS OF HYBRID STERILITY.

Abstract

One of the most striking generalizations in evolutionary biology is the observation that if only one sex is sterile or inviable in the offspring of a cross between two species, it is nearly always the heterogametic sex (Haldane, 1922). This rule applies regardless of which sex is heterogametic: in birds and butterflies, for example, it is the heterogametic females that are sterile or inviable (Haldane, 1922; Coyne, 1992). The dependence of Haldane's rule on heterogamety and not on gender implies that the sex chromosomes play an important role in postzygotic reproductive isolation. This, in turn, is supported by genetic studies in several groups, which have repeatedly shown that the sex chromosomes have very large effects on hybrid sterility and inviability (Coyne and Orr, 1989). Coyne and Orr (1989) and Coyne (1992) review the many explanations of this rule. No explanation, however, is satisfactory, as they all require unproved assumptions or are contradicted by data (Coyne, 1992). One recently proposed evolutionary explanation posits that both Haldane's rule and large sex-chromosome effects results from genes that cause meiotic drive (Frank, 199 la; Hurst and Pomiankowski, 1991). According to this view, mutations causing preferential segregation of a chromosome may arise frequently, but are more likely to increase in frequency when on sex chromosomes than on autosomes. This is because the restricted recombination between heterogametic sex chromosomes may allow linkage disequilibrium to build up between loci causing meiotic-drive and insensitivity loci that prevent the driven chromosome from self-destructing. (Such systems may involve the X chromosome driving the Y, or vice versa.) When such drive produces distorted sex ratios within a species, natural selection will favor its suppression through the accumulation of modifier alleles that restore the sex ratio to normal. If different sex-linked drive-and-suppression systems evolve independently in two related species, and the suppressor alleles are not completely dominant, then meiotic drive will reappear in the hybrids. This could lead to the mutual annihilation of the sex chromosomes in heterogametic hybrids, causing sterility of only the heterogametic sex. This is the meiotic-drive explanation for Haldane's rule. Moreover, genetic analysis would show that the sterility is due largely to the sex chromosomes, explaining large X (and Y) effects. It should be noted that this theory explains only hybrid sterility and not inviability, because inviability is not caused by meiotic drive. Other assumptions, such as pleiotropic effects of drive on mitosis (Frank, 1991 a) or the mobilization of drive-inducing transposons (Hurst and Pomiankowski, 1991), must be invoked to produce a general explanation of postzygotic reproductive isolation. These meiotic-drive theories have been criticized by Coyne et al. (1991), Charlesworth et al. (1992), and Johnson and Wu (1992) on both theoretical and empirical grounds [see Frank (1991b) for counterarguments]. For example, Coyne et al. (1991) note that, in many cases, meiotic-drive alleles may actually accumulate faster on autosomes than on sex chromosomes. Here we provide a direct experimental test of these hypotheses. A natural prediction of meiotic-drive theories of hybrid sterility is the appearance of sex-chromosome drive in species hybrids (Hurst and Pomiankowski, 1991; Coyne et al., 1991; Johnson and Wu, 1992). One can, for example, produce interspecific hybrids between species obeying Haldane's rule. If the sterility of heterogametic hybrids is produced by meiotic-drive alleles suppressed within one species but reexpressed in hybrids (i.e., the suppressor alleles are not completely dominant), then meiotic drive may reappear in semisterile hybrids, who should then produce progeny with distorted sex ratios. Johnson and Wu (1992) report such a test in hybrids between D. simulans and D. sechellia, a pair of species that obey Haldane's rule and show strong X-effects on sterility (Coyne and Kreitman, 1986). In this experiment, semisterile backcross hybrids produced progeny with a normal sex ratio, militating against a meioticdrive theory. We investigated this possibility in three other pair of Drosophila species that are not closely related to D. simulans and D. sechellia. All three species obey Haldane's rule, and genetic analysis has shown that hybrid male sterility is due largely to the X chromosome (Orr, 1989; Orr and Coyne, 1989). All three hybridizations produced semisterile backcross males whose progeny were examined for sex-ratio distortion. D. virilisID. texana. -D. virilis females from a Pasadena, California, strain were crossed to males of a D. texana strain from Morrilton, Arkansas (the latter strain was used in the genetic analysis of Orr and Coyne, 1989). The semisterile F, males were backcrossed to D. virilis females, and the male and female offspring