Balancing selection as the natural outcome of adaptation

Evolution relies on the presence of genetic variation in a population for natural selection to work, yet it is not entirely understood how such variation is maintained. Adaptation has been typically thought to reduce the amount of variation by favoring particular variants and removing others. However, HFSP Long-Term Fellow Philipp Wolfgang Messer and colleagues show that adaptation should often drive beneficial variants only to intermediate population frequencies where they can persist in a balanced state with other variants, thus promoting, rather than exhausting, genetic variation.

HFSP Long-Term Fellow Philipp Messer and colleagues
authored on Mon, 23 January 2012

When biologists first discovered the extraordinary wealth of genetic diversity present in most natural organisms, they were puzzled by the potential mechanisms that could maintain such high levels of variation. This is because natural selection should effectively favor particular beneficial genetic variants at the expense of others -- thereby lessening the genetic diversity of populations. Only a very specific form of selection, so-called balancing selection, keeps genetic variants at intermediate population frequencies, and thus actually preserves genetic variation. The textbook example of balancing selection is a specific variant of the human hemoglobin gene found in Africa, which confers resistance to malaria in those individuals who inherit the variant from only one of their parents (called heterozygotes), whereas the individuals who inherit the variant from both parents (called homozygotes) have sickle-cell anemia. Since heterozygotes with only one copy of the variant can survive much better than those with two copies, as well as better than those without any copy who can still get malaria, the variant is maintained at an intermediate population frequency. Yet despite a number of such famous examples, the prevailing view is that balancing selection is rare in evolution.

Figure: Plot of variant frequencies changing over time during simulated natural selection. Different colors represent the different variants of a particular gene that were observed in the population. Most beneficial mutations do not reach 100% frequency but are maintained at intermediate, balanced frequencies -- until they are displaced by new beneficial mutations that themselves often are balanced.

In their study, HFSP Long-Term Fellow Philipp Messer and colleagues show that balancing selection should actually be a natural consequence of adaptation in organisms that have two genome copies in their cells (so-called diploids, which make up a large fraction of species, including humans). One way to see this result is to consider that when a new beneficial variant first arises in a diploid population, individuals with this variant will typically have only one copy (i.e., be heterozygotes).  In order for the variant to be become more common in the population, it thus needs to be beneficial in those heterozygotes. However, the variant does not actually need to be beneficial in individuals with two copies of the variant in order for it to rise in frequency. Using Fisher’s famous model of adaptation, the authors showed that many beneficial variants should then behave as in the sickle-cell anemia example – where heterozygotes do better than the rest.

As a result, many adaptive DNA mutations should not reach 100% frequency in a population, but will instead persist at intermediate, balanced frequencies (see figure). Interestingly, recent experiments indicate that this pattern might indeed be common. In one example, fruit flies were evolved in the laboratory over many generations, having to adapt to new environmental pressures. Although many DNA variants were found to have changed their frequencies in response to natural selection, none of these variants ever reach 100% frequency in the population. It might thus be necessary to rethink the prevailing view of adaptation not as a process that predominantly removes genetic variation, but as one that could actually play a substantial role in promoting such variation.


Heterozygote advantage as a natural consequence of adaptation in diploids. Sellis D, Callahan BJ, Petrov DA, Messer PW (2011) Proc Natl Acad Sci U S A. 108(51):20666-71

 Pubmed link