Mate choice in sticklebacks reveals that immunogenes can drive ecological speciation

Abstract

Adaptation to ecologically contrasting niches can lead to the formation of new species. Theoretically, this process of ecological speciation can be driven by pleiotropic "magic traits" that genetically link natural and sexual selection. To qualify as a true magic trait, the pleiotropic function of a gene must be reflected in biologically relevant mechanisms underlying both local adaptation and mate choice. The immune genes of the major histocompatibility complex (MHC) contribute to parasite resistance and also play a major role in sexual selection. Hence, the MHC may encode a candidate magic trait. Using diverging 3-spined stickleback populations from a connected lake-river habitat, we show with mate choice experiments in a flow channel that polymorphic MHC genes probably underlie assortative mating with respect to particular habitat-adapted ecotypes, potentially resulting in reproductive isolation. By manipulating olfactory cues in controlled experiments, we show that female sticklebacks employ MHC-dependent male olfactory signals to select mates with which they can achieve a habitat-specific MHC gene structure that optimally protects their offspring against local parasites. By using MHC-based olfactory signals, females thus select individuals of their own population as mates. Our results demonstrate how mate choice and parasite resistance may be functionally linked. These findings suggest that MHC genes are pleiotropic and encode a true magic trait of biologically significant effect.

Keywords: ecological speciation; magic trait; major histocompatibility complex; mate choice; parasite resistance.

Figure 1

Flow channel design. A gravid female stickleback was placed in the flow chamber that was fed by 2 water columns, to each of which stimulus water was continuously added to the constant water current. Fish were able to freely investigate the composition of water for 2 periods of 300 s each, with spatial reversal of water sources at halftime to control for side effects. Their choice between the front quarters of the chamber was video-recorded from above (Drawing by M.M.).

Figure 2

Control experiment—proving that pre-nesting males do not send the MHC signal yet: Lake (pink) and river (blue) females were each presented with the choice between water from a sympatric male (which had not build a nest) and had its water supplemented with MHC peptides and solvent (with peptides, darker color) and only solvent (without peptides, lighter color).Both lake (n = 12) and river females (n = 12) significantly preferred water from males which was supplemented with synthetic MHC peptides, confirming that the males of either population released only the validation factor and no MHC signal. Although statistics were performed on proportions, for better visualization we the time spent by the females in the front quarters of the flow channel. Means and SE are shown. *P < 0.05.

Figure 3

Experiment 1. Females prefer sympatric male odor occurs only in the presence of MHC signal. Each bar represents the time spent in each of the 2 front quarters of the flow channel (in seconds) during olfactory mate choice experiments. Females from either lake or river chose between odor from a lake (pink) and a river male (blue), tested in: (a) The presence of only the validation factor (light colors) (n = 24); (b) The presence of both the validation factor and MHC signal (dark colors) (n = 17). Although statistics were performed on proportions, for better visualization we depicted the actual time the females spent in the front quarters of the flow channel. Means and SE are shown. *P < 0.05.

Figure 4

MHC allele pools differ between lake and river populations. As there is locus duplication, abundances are used as proxy for frequencies. The abundances are calculated as the number of times an allele is found in the population divided by the number of fish genotyped (GPS lake, N = 42, MA-River, N = 40).

Figure 5

Predicted outcome of mate choice by females from lake and river between water from the tank of a lake and a river males that sent both the validation factor and the MHC odor signal depicted in Figure 3b. Predicted mean ± SE number of individual MHC class II B diversity in potential offspring with respect to MHC diversity of chosen (red) and rejected males (black), as a function of female origin from lake or river (N = 11). Individual MHC diversity in the offspring was predicted by calculating the potential combination of the female’s MHC alleles with each of the males’ sets (lake and river) of MHC alleles to be compared with the mean number of MHC alleles found in the female’s habitat of origin, lake or river. Preferred males were those with which, in combination with her own MHC genotype, the female would produce offspring with an individual diversity closer to the mean diversity (horizontal bars, pink for lake, blue for river) of the female’s population of origin (red points for chosen, black points for rejected males). See text for statistics.