Inbreeding uncovers fundamental differences in the genetic load affecting male and female fertility in a butterfly

Abstract

Inbreeding depression is most pronounced for traits closely associated with fitness. The traditional explanation is that natural selection eliminates deleterious mutations with additive or dominant effects more effectively than recessive mutations, leading to directional dominance for traits subject to strong directional selection. Here we report the unexpected finding that, in the butterfly Bicyclus anynana, male sterility contributes disproportionately to inbreeding depression for fitness (complete sterility in about half the sons from brother-sister matings), while female fertility is insensitive to inbreeding. The contrast between the sexes for functionally equivalent traits is inconsistent with standard selection arguments, and suggests that trait-specific developmental properties and cryptic selection play crucial roles in shaping genetic architecture. There is evidence that spermatogenesis is less developmentally stable than oogenesis, though the unusually high male fertility load in B. anynana additionally suggests the operation of complex selection maintaining male sterility recessives. Analysis of the precise causes of inbreeding depression will be needed to generate a model that reliably explains variation in directional dominance and reconciles the gap between observed and expected genetic loads carried by populations. This challenging evolutionary puzzle should stimulate work on the occurrence and causes of sex differences in fertility load.

Figure 1

Mating scheme for one pair (A/B) of inbred lines used to produce different combinations of inbred and hybrid crosses (not shown: C/D and E/F). Generation refers to parents rather than offspring. For each pair of inbred lines the parental cross (XP) was performed in one direction only. Letter and number codes in parentheses denote crosses used, together with equivalent crosses from the other pairs of lines, for figures 2 and 3.

Figure 2

Frequency distributions of the proportion of eggs hatching per mated female in the outbred base population and three classes of daughter lines (see figure 1): (a) outbred base population or P (n=35); (b) F₁ (n=265); (c) F₂ (n=692); and (d) F₂ cross or XP (n=74). Arrows indicate the mean hatch rate.

Figure 3

Frequency distributions of the proportion of eggs hatching per mated female in different classes of cross at the F₃ level (see figure 1): (a) inbred males×inbred females (n=175); (b) hybrid males × hybrid females (n=94); (c) inbred males×hybrid females (n=74); (d) hybrid males × inbred females (n=74). Arrows indicate the mean hatch rate.