Update to Holden and Mace's "Phylogenetic Analysis of the Evolution of Lactose Digestion in Adults" (1997): Revisiting the Coevolution of Human Cultural and Biological Diversity

In 1997, Clare Holden and I used a formal phylogenetic comparative method to show that high levels of the ability to digest lactose in adulthood were coevolving with a history of dairying at the population level (Holden and Mace 1997). This not only confirmed what was already suspected—that lactase persistence was an adaptation to pastoralism—but also allowed us to show statistically the direction of the change: Dairying preceded the evolution of lactose tolerance, providing stronger evidence in favor of the route of causation. Any model in which the evolution of lactose tolerance either was independent of pastoralism or preceded the cultural evolution of pastoralism was far less likely than models in which pastoralism evolved first, which then caused lactase persistence to be selected for. Since 1997, when we wrote our paper, I am happy to say that we still agree with the conclusions of our study, and in fact, had we to do it over again, we would not do it in any substantively different way. However, there have been a number of exciting developments in the field, and I am delighted to take the opportunity to report them here.

First, when we examined the coevolution of lactose tolerance and milk drinking, although the association was known (Simoons 1978), little was known about the genetic basis of the persistence of the ability to digest lactose into adulthood. This is no longer the case, and the story is intriguing (Ingram et al. 2009). Several candidate genes have been identified that appear to be highly correlated with lactase persistence. One of the tightest correlations is with the -13910*T allele, which is a single nucleotide polymorphism (13.9 kb) located close to the lactase gene (C-13910 > T). The allele is widespread in Europe, and its high frequency combined with a long region of haplotype conservation (500 kb) suggests recent origin combined with strong selection (Bersaglieri et al. 2004). This story continues to provide not only one of the best examples of gene culture coevolution but also one of the strongest cases of recent selection on a gene being documented. Why the selection in favor of lactase persistence in Europe is so strong is not entirely clear, but it clearly enabled enhanced survival and growth and provides further support for the notion that it was the Indo-Europeans' agricultural innovations that underpinned their dominance in Europe.

What is intriguing, however, is that the same genes do not appear to predict lactase persistence in different regions of the world. This allele does not predict lactose tolerance in Africa, where it is actually rare and where at least three other alleles (also close to the lactase gene) have been implicated (Tishkoff et al. 2007). This suggests, as predicted by the culture historical hypothesis, that lactase persistence evolved recently after the spread of modern humans around the world, and the problem of how to digest milk has been "solved" by natural selection in a number of different ways. The independent evolution of different gene-regulating mechanisms is the best explanation, and lactase persistence may be one of the first examples of a polymorphism in a regulatory element. In addition, cultural traditions, especially curdling milk before drinking, may have contributed to relatively low levels of lactose tolerance in some populations, especially in Africa. Some have suggested that high levels of lactase persistence might have hidden costs in malarial areas, because flavins generated by milk drinking may enhance the risk of catching malaria. There is also some evidence that, among Somali nomads, a particular gut flora is enabling lactose digestion in genetically "lactose intolerant" individuals (Ingram et al. 2009). There is much more to run in this story as all these diverse mechanisms are further investigated.

The reason we were able to make firm conclusions about the direction of gene-culture evolution in our 1997 paper was because the methods we were using were a considerable methodological advance on previous work in...