disease

Will the swine flu lead to the next big evolutionary change for humans? No. But it has already begun to affect the way people interact with each other. I wandered onto campus a couple of weeks ago and saw people wearing face masks! We've been asked to plan for our "essential" classes in the event of a pandemic.

It's a good time to be on leave. But my kids' school has been closed the rest of the week. It's the flu -- more than a third of the whole school was out sick yesterday.

In the August Smithsonian magazine, writer Rob Dunn discussed a hypothesis that tries to relate cultural diversity and xenophobia (fear of the other) to the rate of infectious disease ("The culture of being rude"):

In a series of high-profile papers, Corey Fincher and Randy Thornhill, both at the University of New Mexico, and Mark Schaller and Damian Murray of the University of British Columbia argue that one factor, disease, ultimately determines much of who we are and how we behave.

Their theory is simple. Where diseases are common, individuals are mean to strangers. Strangers may carry new diseases and so one would do best to avoid them. When people avoid strangers—those outside the tribe—communication among tribes breaks down. That breakdown allows peoples, through time, to become more different.

Differences accumulate until in places with more diseases, for example Nigeria or Brazil, there are more cultures and languages. Sweden, for example, has few diseases and only 15 languages; Ghana, which is a similar size, has many diseases and 89 languages. Cultural diversity is, in this view, a consequence of disease.

On the surface, this seems a poor example -- the population has been thin in Sweden relative to Ghana for most of the last 6000 years, until the rise of the Swedish state. It's no surprise that a recent population expansion coupled with political centralization would result in a relatively uniform language and culture area.

But as the article goes on to explain, some of the theorists think that the rise of states is itself a dependent variable. They would propose that the growth of polities was limited in Ghana because of a high disease load, retaining and fostering a cultural diversity that would have been wiped out by natural political consolidation in a less-disease-prone region of the world.

That's the logic, at least.

This kind of topic is interesting but endlessly frustrating. The frustration -- at least for me -- comes from the ready confusion of biological and cultural processes of change. Dunn's article says as much:

As a rule, it is good to be skeptical of biologists who, like Fincher and Thornhill, propose to explain a whole bunch of things with one simple theory. More so when those biologists are dabbling in questions long reserved for cultural anthropologists, who devote their careers to documenting and understanding differences among cultures and their great richness of particulars. Biologists, and I am no exception, seem to have a willingness–or even need—to see generalities in particulars. Fincher’s new theory would offer an example of these desires (and a little hubris) run amok, of biologists seeing the entire history of human culture through one narrow lens. It would offer such an example, if it didn’t also seem, quite possibly, right.

As a rule, I'm skeptical of everything. "Wary of strangers" wouldn't keep our school open -- those kids are all catching the flu from their friends. It doesn't take very many contacts between groups of people to spread an epidemic far and wide.

In that respect, it's a problem for percolation theory. You've got a network of people through which the flu can be transmitted. If the people tend to be highly interconnected, with each other, the flu spreads widely. But even if links between groups are rare, the strong interconnections within a group can keep the epidemic alive long enough to make it to the next hop. At some critical level, the links are no longer enough, and the pathogen can't propagate.

So does it do you personally much good to be wary of strangers, if you have a few cosmopolitan friends? Everything can make a little difference to the percolation network, but it seems to me that people carrying the "xenophobia" gene would still maintain large social networks including a few people who interacted far and wide with strangers. It's not that the direction of the effect is wrong; it's that a wholesale elimination of stranger contact by an individual may have little effect on her probability of infection, since friends may have gotten the pathogen from strangers.

It's awful hard for a "xenophobia" gene to get going in that scenario.

Jenny Tung of Duke University and colleagues report in Nature (online early) that yellow baboons have evolved a Duffy antigen-related defense against a baboon relative of malaria.

Most Africans carry a null allele for the Duffy antigen, coded by the DARC gene, which functions to protect them from vivax malaria. It's not the worst kind of malaria (that would be falciparum), but it is a major cause of disease outside those regions where the Fy*O allele is near fixation.

The baboon version of DARC is not convergent with the human null allele; the paper reports that it actually increases the gene activity, whereas the baboon variant allele actually increases the activity.

Presumably, the different defense in baboons is because they're fighting a different parasite:

Baboons are not generally infected by Plasmodium in the wild, but are vulnerable to infection by several closely related haematoprotozoans (4, 5) including Hepatocystis kochi, a blood parasite nested within the paraphyletic Plasmodium genus (15). Hepatocystis parasites do not produce the cyclical fever spikes typical of malaria in humans, but do produce anaemia and visible merocyst formation, followed by scarring on the liver (4).

They were able to show that the infection rate with Hepatocystis is significantly lower in baboons that cary the protective allele.

Based on their comparisons, the locus looks like a case of balancing selection:

We detected an increased level of population differentiation among East African baboon populations around FY, by comparing a FY-linked microsatellite with 35 neutral microsatellites (F_st = 0.31, P F_st for the neutral markers was 0.008–0.346; F_st is a metric describing genetic divergence between populations based on allele frequency differences at variable sites; Supplementary Fig. 3). We also detected a higher value for the Tajima's D statistic (D = 1.26) in this region relative to nine of nine other resequenced putative cis-regulatory regions in the Amboseli population and 11 of 12 resequenced transcribed regions (range of D for all other loci was -1.60 to 2.12). The only locus with a higher value of D, a transcribed portion of the gene MSR1, exhibited an even more extreme value than that identified for the MHC DQA1 promoter in baboons (22), which is known to evolve under strong trans-specific balancing selection (20).

A balance really would be necessary for them to be likely to have any evidence of selection. There's no reason to think that baboons are in the kind of demographic and disease transient that humans are in, so if a protective allele were always beneficial, it would likely be fixed. Still, it's not obvious what the disadvantage of the protective allele would be, although in humans it has been suggested that altering Duffy expression may impair immune response by reducing white blood cell count (Reich et al. 2009). Considering the high stress and cortisol levels of wild baboons, it may be that changes in immune activity have even more disadvantages.

One important aspect of the study is that the allele affects the cis-regulatory region of the gene; that's the general research topic covered by Gregory Wray's research group. I think that it's important because it raises the prospect that targeted sequencing of cis-regulatory elements in primate genomes might lead to the discovery of more adaptive variations within primate species. In their concluding paragraph, the authors emphasize the strengths of a combined field and in vitro approach to characterizing functional variants:

In vivo gene expression measurements are complicated by variation in genetic background and in the environment, both of which can modify functional cis-regulatory effects (25, 26). Indeed, our results show that even baboons that are homozygotes at the C/T site sometimes exhibit allelic imbalance in FY expression, suggesting that other, unidentified functional cis-regulatory variants are also segregating in the population. In contrast, in the in vitro comparisons, only a single cis-regulatory site differed between the experimental constructs, thus controlling for both environment and genetic backgrounds. Using both approaches in tandem can be synergistic: while in vitro experiments can help pin down specific functional sites, in vivo results demonstrate that these effects are relevant to the biology of individuals in the wild.

When it comes to primate genetics, looking for defenses to infectious diseases should be low-hanging fruit. Just take human genes that have alleles that defend against diseases, and sequence them for variations. Hopefully we'll find many others -- and in a few cases, those variations may prove useful in human health contexts, as they may reveal new pathways to deter or defeat pathogen infections.

References:

Tung J, Primus A, Bouley AJ, Severson TF, Alberts SC, Wray GA. 2009. Evolution of a malaria resistance gene in wild primates. Nature (advance online publication) doi:10.1038/nature08149

Reich D, Nalls MA, Kao WHL, Akylbekova EL, Tandon A, et al. 2009. Reduced Neutrophil Count in People of African Descent Is Due To a Regulatory Variant in the Duffy Antigen Receptor for Chemokines Gene. PLoS Genet 5(1): e1000360. doi:10.1371/journal.pgen.1000360