Monday, June 02, 2008

Selection speculation: CLOCK and reward-dependence in Africans   posted by agnostic @ 6/02/2008 02:39:00 AM

Since so many comments lead off with some variant of "I would guess," why not try to corrall them all into one post where they could serve a purpose? Each week I'll find some area of the human genome that shows signs of recent selection, see what phenotypes the gene affects, and although I'll likely provide the most convincing story, readers can conjecture to their heart's content about what might have driven selection. It may, for once, improve the discussion to comment while still deranged from last night's drink.

Let's start with the data: using Haplotter, we see that for the gene CLOCK, there is a signal of recent selection in Africans but not in Europeans or Asians. The CLOCK gene is involved in maintaining our circadian rhythm, and I started this search looking for between-group differences in being "day people" vs. "night people." However, other genes related to circadian rhythm -- PER1, PER2, PER3, Tim, CSNK1E, Cry1, and Cry2 -- show no signals of recent selection anywhere. So perhaps there is something else that CLOCK does that these others do not.

It turns out that CLOCK is also involved in the dopaminergic system, which regulates mood and behavior. None of the OMIM entries for the other circadian rhythm genes mentions dopamine, mood, the reward system, etc. So it is more likely that CLOCK has undergone selection for its effects on mood and on the brain's reward system, rather than on circadian rhythm per se. In mice, and so potentially in humans as well, mutations in CLOCK make individuals lower in anxiety and higher in risk-taking (Roybal et al. 2007; free full text).

Turning to data from Lynn & Martin (1995), which I turned into a convenient graph here, we see that Nigerians -- the African group that HapMap data come from -- are indeed the lowest in the world in the personality trait Neuroticism (the Israeli data-point must be a mis-coding by Lynn, who is known for doing so). They are also the highest in the world in Extraversion. Neuroticism measures a tendency toward anxiety and related states, while Extraversion measures, among other things, pace of living (manics living a faster-paced life) and excitement-seeking. So Nigerians show a psychological and behavioral phenotype that's fairly comparable to the mice with mutations in CLOCK, strengthening the hypothesis that selection at CLOCK has acted on personality rather than circadian rhythm per se.

Now comes the fun part -- telling a story about why this phenotype was more advantageous in Western Africa than in Central Europe or Northeast Asia over the past 10,000 or so years. The low-anxiety and high-excitement-seeking phenotype is suspiciously like that associated with derived alleles at the DRD4 locus, and the novelty-seeking 7R allele at that locus does show up at intermediate frequencies in Africans (Chen et al. 1999; see a map here). I concur with Harpending & Cochran (2002) (free full text), who argued in the DRD4 case that the phenotype is more advantageous in "cad societies" than in "dad societies."

More concretely, I suggest, based on the teachings of the pickup artist community, that low anxiety aids in approaching more females in a shorter time -- you are less paralyzed during the approach, and you get over rejection more easily -- and that higher risk-taking motivates you to approach more females. Contingent upon having other traits that are appealing to females (a muscular body, artistic skill, dancing or singing ability, a knack for flirting, and so on), the phenotype here would have increased male success in courtship of (multiple) females, a stronger determinant of reproductive success in a "cad society" than in a "dad society."

Lastly, here are two testable predictions: 1) lowland South American and highland New Guinean tribes should also show signs of selection at the CLOCK locus, since they are even more prototypically "cad societies," and they have the highest frequencies of the 7R allele at DRD4. And 2) African Bushmen should show no signs of selection at CLOCK since they are more "dad societies" and have lower frequencies of the 7R allele at DRD4. (ALFRED does not have data on CLOCK.)

The fact that Northeast Asians do not show signs of selection at CLOCK already supports the hypothesis, since they are more "dad societies" and have a very low frequency of 7R at DRD4 (and even the 2R allele, which shows up somewhat there, has a much more muted effect than does 7R). The lack of selection at CLOCK in Central Europeans is neither here nor there: they are more "dad societies" than Western Africa but not so much as East Asia; Europeans also have intermediate frequencies of 7R and DRD4. So we could have an example of "Rushton's Rule," where East Asians show no selection for low-anxiety / high-novelty-seeking alleles at either DRD4 or CLOCK, Europeans show selection at just one locus, and West Africans show selection at both.

To the best of my knowledge, as they say, this is the first argument for why Africans show a signal of selection at CLOCK. Consider it part of the graduate student with a PC trend in studying human evolution. You heard it here first.


Chen, C., M. Burton, E. Greenberger, & J. Dmitrieva (1999). Population migration and the variation of Dopamine D4 receptor (DRD4) allele frequencies around the globe. Evolution and Human Behavior, 20(5): 309-324.

Harpending, H. & G. Cochran (2002). In our genes. PNAS, 99(1): 10-12.

Lynn, R., & T. Martin (1995). National differences for thirty-seven nations in extraversion, neuroticism, psychoticism and economic, demographic and other correlates. Personality and Individual Differences, 19(3): 403-406.

Roybal, K., D. Theobold, A. Graham, J.A. DiNieri, S.J. Russo, V. Krishnan, S. Chakravarty, J. Peevey, N. Oehrlein, S. Birnbaum, M.H. Vitaterna, P. Orsulak, J.S. Takahashi, E.J. Nestler, W.A. Carlezon, Jr., & C.A. McClung (2007). Mania-like behavior induced by disruption of CLOCK. PNAS, 104(15): 6406-11.

Labels: , , , , , ,