Rapid Evolution of Assortative Fertilization between Recently Allopatric Species of Drosophila

Abstract

The virilis group of Drosophila represents a relatively unexplored but potentially useful model to investigate the genetics of speciation. Good resolution of phylogenetic relationships and the ability to obtain fertile hybrid offspring make the group especially promising for analysis of genetic changes underlying reproductive isolation separate from hybrid sterility and inviability. Phylogenetic analyses reveal a close relationship between the sister species, Drosophila americana and D. novamexicana, yet excepting their contemporary allopatric distributions, factors that contribute to reproductive isolation between this species pair remain uncharacterized. A previous report has shown reduced progeny numbers in laboratory crosses between the two species, especially when female D. novamexicana are crossed with male D. americana. We show that the hatch rate of eggs produced from heterospecific matings is reduced relative to conspecific matings. Failure of eggs to hatch, and consequent reduction in hybrid progeny number, is caused by low fertilization success of heterospecific sperm, thus representing a postmating, prezygotic incompatibility. Following insemination, storage and motility of heterospecific sperm is visibly compromised in female D. novamexicana. Our results provide evidence for a mechanism of reproductive isolation that is seldom reported for Drosophila species, and indicate the rapid evolution of postmating, prezygotic reproductive barriers in allopatry.

Figure 1

Egg hatch rate within and between D. novamexicana and six lines of D. americana(see Section 2). Egg hatch estimates “within N line” and “within A lines” are averages of within-line hatch rates from all experiments. Asterisks indicate egg hatch estimates that are significantly different between lines compared to within lines (P < 0.001, Tukey-Kramer HSD). A = D. americana, N = D. novamexicana. N4 is the D. novamexicana line used throughout the study (1031.4). All remaining two-letter abbreviations are the first two letters of each D. americana iso-female line i.d. (see Section 2).

Figure 2

(a) Proportion of fertilized and hatched eggs (error bars represent standard error). (b) Egg containing fluorescently labeled sperm tail (left) and eggs containing no sperm tail (right).

Figure 3

Proportion of inseminated females reared in 2% agar that contained (a) stored sperm and (b) had motile sperm. The left-most column in both (a) and (b) shows the proportion of conspecifically and heterospecifically mated females that laid eggs in 2% agar.

Figure 4

Average number of progeny produced by (a) D. novamexicana and (b) D. americana females when singly and doubly mated to conspecific and heterospecific males (error bars represent standard error).