Human malaria parasites are closely related to dozens of other Plasmodium species that infect non-human primates, rodents, and other mammals. Although several examples of recent adaptation to malaria in human populations have been discovered, little is known about the deeper evolutionary impact of the parasite on the mammalian lineage. From a set of 9,338 proteins conserved in 24 mammal species, we manually identified 412 proteins linked to malaria phenotypes in the literature. Models of codon evolution demonstrate that these 'malaria-interacting-proteins', or MIPs, have been exceptional targets of positive selection throughout mammalian evolution.
We find that MIPs with immune functions have been the primary targets of adaptation. Interestingly, about half of MIPs are also known to interact with viruses and bacteria, and adaptive patterns do not appear limited to lineages known to contract Plasmodium. In many cases, the pervasiveness of pleiotropy can make it difficult to attribute genetic adaptation to a single selective pressure. However, we use comparative analyses to show that overlap with viruses and bacteria cannot explain the excess of adaptation in MIPs. Instead, we suggest that Plasmodium, along with other insect-vectored Apicomplexan parasites, have driven pervasive adaptation in a large set of proteins throughout mammalian evolution.