An ancient gene inversion under selection in Europe

New York Times news story on this inversion

Stefansson and colleagues report on an inversion on chromosome 17 (at 17q21.31) that is interesting in terms of selection and human history. The inversion is polymorphic in humans with both a normal and an inverted allele. The inverted allele is more rare, especially in Asia and in Africa, but has a frequency of around 20 percent in Europeans. The inversion encompasses 900 kb of sequence and includes at least two coding genes. The effects of the allelic variation between the inverted and normal forms of these genes have not been fully characterized, but it appears that individuals with at least one copy of the inversion have a higher overall rate of recombination. The region around the inversion itself is strongly linked and appears to have undergone little or no recombination during the lifetime of the polymorphism.

By sequencing two regions within the inversion area, the researchers were able to determine that the inversion polymorphism was very ancient. Assuming a 6 million year divergence between humans and chimpanzees, the age of this polymorphism was determined to be around 3 million years. This is very old compared to most human nuclear genetic variation, but not exceptionally so, since a number of other genes have variation dating to before one-and-a-half million years ago (e.g. PDHA1, M1).

What is perhaps more exceptional is its pattern of geographic variability. The ancient variant is far more common in Europe than in other parts of the world. This might be taken to indicate that this allele once had a uniquely European distribution, but such a hypothesis has an obvious problem: some of the copies of the inverted allele present in Africans appear to be more closely related to the noninverted allele than any of the inverted copies found in Europeans. This observation argues for at least the ancient presence of this allele within Africans, and is consistent with the hypothesis that this allele originated in Africa and only then moved to Europe. Answering this question will require a larger sample of variation to be examined, since it could easily turn out that the rare alleles noted in Africa might also be present in Europeans that were not sampled.

So what explains the high frequency of the allele in living Europeans? Stefansson and colleagues use the Icelandic database to test whether this a little variation is associated with any phenotypic traits. They find that carriers of the inverted allele presently appear to have a selective advantage in Iceland, because they have more offspring on average. This advantage is fairly high, around 3.5 percent more offspring compared to the non-inverted allele.

Considering the strong level of selection on the inversion now, there are a number of questions that should be examined. Is the frequency of the inverted allele currently increasing? If so its current frequency may be an artifact of relatively recent cultural or technological changes. Is this a balancing polymorphism? Such a hypothesis might explain the retention of the allele over long periods of history at relatively low frequencies (similar to the ABO blood groups). What elements of the European Environment appeared to make this allele advantageous. If its functioning is indeed related to the level of genomic recombination, why wouldn't it be advantageous in other parts the world? The researchers couldn't examine all of these questions, partly because the low frequency of the allele makes it difficult to test its effects in homozygotes.

Does the inversion address the "out of Africa" hypothesis? Not really. The researchers raise the question of whether the gene may have been inherited from archaic human species, but there's really no evidence to suggest that it was. Its current presence in Africa and Asia, even at very low frequencies, is consistent with gene flow between these continental regions and Europe coupled with natural selection favoring the inversion in Europe itself. So the story is really the mystery of why this inversion would be favored in Europe and not elsewhere.


  author = {Hreinn Stefansson and Agnar Helgason and Gudmar 
    Thorliefssonand Valgerdur Steinthorsdottir and Gisli Masson and 
    John Bernard and Adam Baker and Aslaug Jonasdottir and Andres
    Ingason and Vala G. Gudnadottir and Natasa Desnica and Andrew
    Hicks and Arnaldur Gylfason and Daniel F. Gudbjartsson and 
    Gudrun M. Jonsdittir and Jesus Sainz and Kari Agnarsson and 
    Birgitta Birgisdottir and Shyamali Ghosh and Adalheldur 
    Olafsdottir and Jean-Baptiste Cazier and Kristleifur Kristjansson
    and Michael L. Frigge and Thorgeir E. Thorgeirsson and Jeffrey
    R. Gulcher and Augustine Kong and Kari Stefansson},
  year = {2005},
  title = {A common inversion under selection in Europeans},
  journal = natgenet,
  volume = {Advance Online Publication},
  pages = {}  }

Nature Genetics online