One of the newer “expansions” of the modern synthetic theory of evolution is the idea that the genetic variation “used” by either natural selection or genetic drift can arise not just through mutations within a species, but also through hybridization with another species. Hybridization between different species usually yields maladaptive offspring, but occasionally a fertile hybrid can be the source of a new gene that can spread through a species that didn’t originally have it. (I’ve previously written about work showing that an adaptive color gene in aphids was acquired not by hybridization, but by ingestion.)
Such a case of “adaptive introgression” (“introgression is simply the acquisition of genes by one species or population from another by interbreeding) was just reported in Current Biology (reference below), and you can read a summary of the results in a piece by Kai Kupferschmidt in Science NOW (I’m quoted).
As reported by Song et al., the house mouse, Mus musculus domesticus, acquired a gene for rodenticide resistance by mating with a close relative, the wild mouse Mus spretus. The rodenticide is the famous poison Warfarin, which is used widely to kill mice and rats. It works by inhibiting the synthesis of blood clotting factors that themselves are dependent on vitamin K. The mice then bleed to death internally. Warfarin is also used to inhibit the formation of blood clots in humans, but is called “Coumadin” for medical uses. Since its introduction in the 1950s, many populations of mice have become resistant to the poison: this is a classic case of natural selection that isn’t as widely known as examples of bacterial resistance to antibiotics.
Song et. al. found through DNA sequencing that populations of the house mouse in Spain and Germany had one region of their genome, on chromosome 7, that actually came from Mus spretus. (The two species were formerly “allopatric”, i.e., lived in different places, until they become geographically contiguous when the house mouse moved with its human host.) This region includes genes for Warfarin resistance, and experiments showed that house mice containing the introgressed spretus region lowered mortality from Warfarin and a similar poison from 84-100% to 9-20%. The amino-acid sequence of the introgressed gene (vkorc1spr ) differed between the two species. The authors haven’t done the definitive experiment—actually putting the spretus version of the gene into a house mouse genome and seeing whether it alone confers resistance to poison—but the mortality rates of introgressed house mus versus those lacking the gene are pretty good evidence.
Finally, the authors showed by population-genetic analysis that the spretus gene entered the house mouse population between 61 and 71 mouse generations ago, which corresponds to about 13-22 years in the wild, so the adaptive introgression occurred well after the poision was introduced. This again supports the notion that the spread of the spretus gene in house mice was promoted by natural selection for resistance to poison.
There are two questions to ask about this situation:
1. How common is adaptive introgression? My guess (which you can see in the ScienceNOW link above), is probably that it isn’t very common. Why? Because if it were, we would see it using DNA-based phylogenies. Adaptive introgression would show up as a region of the genome that was much more similar to a region in a related species than could have occurred by simply genetic drift or natural selection in the first species. We don’t see that kind of similarity very often. (It also could have other causes, like a variant in the common ancestor of the two species that simply was inherited by its descendants.) So I suspect that while the capture of adaptive genes by hybridization occurs occasionally, it won’t be an important source of variation compared to mutation.
Further, most species (we are one) simply can’t form fertile hybrids with a related species, a condition that is necessary for adaptive introgression to occur. The genus Drosophila (the flies on which I work) contains about 1500 described species, but hybrids are known in only about a dozen cases, and most of these hybrids are sterile.
2. Can species be selected to hybridize with other species? Some biologists—especially botanists—have theorized that natural selection will foster those traits that facilitate one species mating with another one, so that it can capture alleles to facilitate its own evolution. That doesn’t wash, because in the vast majority of cases the hybrids between different species are less fit than the parental species: hybrids can be sterile or inviable (the spretus/domesticus hybrids, for example, are largely sterile). So, although hybridization may occasionally capture a “good” gene, most of the time it produces maladapted hybrid offspring. Natural selection, then, would act to prevent rather than facilitate hybridization, because the beneficial effects of mating between species are far outweighed by the bad ones.
As the famous evolutionist Ronald Fisher wrote in 1930:
The grossest blunder in sexual preference, which we can conceive of an animal making, would be to mate with a species different from its own and with which hybrids are either infertile or, through the mixture of instincts and other attributes appropriate to different courses of life, at so serious a disadvantage as to leave no descendants. … it is no conjecture that a discriminative mechanism exists, variations in which will be capable of giving rise to a similar discrimination within its own species, should such a discrimination become at any time advantageous.
Mus spretus. Wild mice are incredibly cute, and I could never bring myself to kill them in traps.
______
Song, Y. et al. 2011. Adaptive introgression of anticoagulant rodent poison resistance by hybridization between Old World mice. Current Biology: doi:10.1016/j.cub.2011.06.043


















