Monday, September 04, 2006

Thrifty genes a development dilemma

I’ve been reading Ellen Shell’s The Hungry Gene: The Inside Story of the Obesity Industry. It provides a potted history of discoveries that confirm the genetic basis for obesity and is chatty, good light reading. One interesting observation that was new to me was the notion that obesity and associated Type 2 diabetes problems were particularly likely to arise in societies previously subject to food stresses and where, as a result, 'thrifty genes', had emerged to ensure survival in difficult times. Problems then arise if people come into contact with an abundant, sustained rather than periodic food supply.

Thrifty genotypes were constellations of genetic factors that encourage the conversion of calories into body fat and which decreased the sensitivity of the body to insulin to ensure adequate blood glucose levels in the brain during famine. This mechanism was essential for people to survive 'bottleneck' periods of extreme stress and food scarcity. ‘Thrifty genes’ became essential for survival given the ebb and flow of food availability.

Westernisation of such societies and their transition to stable, reliable food supplies means that the genes which once protected now condemn people to early deaths through obesity and diabetes illnesses. It is not what such people eat – those among them who have white ancestry, because of interbreeding of locals with randy early European settlers, have much lower incidence of such diseases though they share a common diet.

This story has one dramatic implication. Western societies have severe obesity-diabetes problems but these problems are not nearly as bad as societies which have had a history of food stresses. Examples of such societies include indigenous populations (e.g. the Australian aborigines, Hawaiians, Samoans, Nauruans) – where up to half the population is obese with diabetes - and those from Middle East cultures and the Sub-Continent.

More generally, an improved food situation in Asia and Africa might have dramatically adverse health consequences if these societies have previously been food stressed. The emergence of obesity issues in China and India suggests this is already a problem. Thus, the replication of the Western global obesity epidemic might be expected to have much more dramatic health implications in newly-emerging and developing countries.


conrad said...


you want to stop believing so much of this genetic stuff. Everyone needs to "confirm" genetic involvement in things like this to get their next grants, get short reports published etc. . The data itself on almost complex things (like metabolism, IQ, ...), is almost never conclusive, and everyone wants to believe self confirming correlations. The thrifty gene idea is no different in this respect.

hc said...

Conrad, But I find it is consistent with many facts - high levels of problems among certain populations. I didnt say it was the whole story but an important insight.

The evidence linking obesity to certain geneotypes is compelling.

conrad said...


no doubt genes are linked to various metabolic differences amongst people, but actually why that is so is extremely hard to determine -- and people always want a simple all encompassing reason for everything, because that is what sells (like thrifty genes). Its self-confirmation bias. A similar case can be made for things like Asthma in Australia -- why is it high ? People are going to tell you that it is the environment -- yet there are many many dustier places than Australia where people don't have high rates of asthma.

You can also look at things where there is no obvious explanation. Why do southern Chinese have the worst eye-sight in the world ? Why do European Jews have high rates of early AD (for which we do know some of the genetic stuff) ? Why do Brits have high rates of excma etc.

You can also look at the flip side where once really obvious things turned out to be not so obvious. The classic example is skin colour. Once upon a time everyone thought the closer you are to the equator the darker you are (most people still believe this -- much more so than the thrifty gene). But the actual evidence for this is not exceptionally thrilling -- Jared Diamond has written a lot about this, and it turns out that even the data is pretty weak (i.e,. exposure to UV light is only a weak correlate of skin colour).

If we go back to the data on which the "thrifty gene" is based, how accuracte really is it? Is it that Europeans really had a good food supply compared to other groups ? I'm sure the answer to this is no, because Europeans lived in temperate and cold climates, where you have all the problems of seasonal rain etc. They also lived inland a lot, so they couldn't just go catch some fish. What about Africans ? A lot of these guys lived in tropical areas which do (or did) have abundant sustained food supplies -- but they suffer terrible rates of obesity etc. in the US. The same is true of the Pacific Islanders -- surely these guys have had a sustained food supply for most of history (i.e,. fish).

These are of course just single data points, but I think it is safe to conclude that metabolic differences are at most only going to be weakly correlated with food supply history.

hc said...

Conrad, again it isn't the whole3 story but does seem to have much explanatory power.

The Pacific Islanders did have variable foiod supplies as did nomadic peoples - the Arabs, aboriginees, Americamn Indians all of which have high diabetes problems.

But I agree a formal empirical test is useful. I'll scout around and have a look.

Anonymous said...

Hi, a lot of discussion over this issue has lead to the main assumption that Neels (the author of the original thrifty genotype model) is not right. The reasons are:
1. the hypothesis is based on the assumption that the core genes were inhered in the pleistocene, but now we know that many cases of obesity and metabolic syndrome (the association with diabetes, etc.) are related to new mutations, specially in MCR-4 alleles.
2. Genomic maps of migration and population studies deny the conclusion that differences in prevalence are related to precise migration patterns (they can have Some importance.
3. Last but not least: Starvation were rare within the hominid history (they became more common recently), not representing such a strong selective pressure within the Adaptive Environment (mostly paleolitic).
But the guess was impressive!