Yann Klimentidis points me to this very exciting new paper. Some of its authors also contributed to another recent paper which investigated the role of climate in shaping human cranial variation.
The authors compared autosomal, X chromosome and Y chromosome genetic variation with that of mtDNA. For example an impressive 85% of autosomal microsatellite diversity in worldwide humans can be explained by their distance from Africa, but no apparent effect is seen for climate. The effect of distance from Africa is only 18% for HVS-I of human mtDNA, and a strong effect of climate measures (such as temperature) is seen. On the contrary, human Y chromosomes show no relationship with climate.
The effect of climate is also resilient to accounting for distance from Africa (which is very hot, and thus distance from it might be expected to correlate with a decrease in temperature; it is not), and by removing populations living in extreme arctic conditions, for which thermoregulation is of the utmost importance.
I would say that this paper represents the best evidence so far of selection acting on the human Y chromosome.
The implications of this are manifold, and they relate:
(i) to the use of mtDNA to assess population relationships. For example, South Asians have a strong affinity with West Eurasians in their Y-chromosomes and with East Asians in their mtDNA. Nonetheless they do not occupy an intermediate position between the two in autosomal DNA, but are predominantly aligned to West Eurasians.
(ii) to the overall mitochondrial time depth of humanity, i.e., the age of mitochondrial Eve, and what that implies about the recentness of our species (it may be older than its mtDNA time depth), and its possible admixture with other human populations such as Neandertals (it may have occurred, but the non-modern mtDNA may have been selected against)
(iii) to the potential of solving, at least in part, discrepancies between present-day and archaeological mtDNA gene pools, which may not in fact reflect processes of migration, but climate shifts over time.
Note that this is not the first time that adaptive evolution of human mtDNA has been proposed. For example five years ago, Ruiz-Pesini et al. noticed the conservation of internal branch amino acid substitutions at high latitudes, and made the case for energy-related climate adaptation in human mtDNA. The current paper makes a different argument, but arrives at a similar conclusion.
Proceedings of the Royal Society B doi:10.1098/rspb.2009.0752
Climate shaped the worldwide distribution of human mitochondrial DNA sequence variation
François Balloux et al.
There is an ongoing discussion in the literature on whether human mitochondrial DNA (mtDNA) evolves neutrally. There have been previous claims for natural selection on human mtDNA based on an excess of non-synonymous mutations and higher evolutionary persistence of specific mitochondrial mutations in Arctic populations. However, these findings were not supported by the reanalysis of larger datasets. Using a geographical framework, we perform the first direct test of the relative extent to which climate and past demography have shaped the current spatial distribution of mtDNA sequences worldwide. We show that populations living in colder environments have lower mitochondrial diversity and that the genetic differentiation between pairs of populations correlates with difference in temperature. These associations were unique to mtDNA; we could not find a similar pattern in any other genetic marker. We were able to identify two correlated non-synonymous point mutations in the ND3 and ATP6 genes characterized by a clear association with temperature, which appear to be plausible targets of natural selection producing the association with climate. The same mutations have been previously shown to be associated with variation in mitochondrial pH and calcium dynamics. Our results indicate that natural selection mediated by climate has contributed to shape the current distribution of mtDNA sequences in humans.