Sunday, May 6, 2007

Genes in Conflict

"The conflict between maternal and fetus genes is one of the weirdest ideas in the modern theory of evolution."* According to evolutionary logic, the fetus "wants" to milk the mother for all it can get; the mother "wants" to restrict the fetus to what it needs to survive and save something for future offspring. The reason is that the fetus benefits from every bit of help the mother gives, while the mother's return on her investment diminishes with increasing investment, (i.e. ever greater investment won't necessarily increase her fitness).

One example of this struggle is the regulation of the mother's blood sugar, which is much than normal higher during pregnancy. Well this makes sense, you might think, because the mother is now feeding a growing fetus. But look at the mother's insulin level, it also is much higher than normal, and insulin is used to down-regulate blood sugar levels. Huh, that's weird, she is secreting more insulin, yet her blood sugar level is ever higher. She must be responding less to insulin. Why should that be?

David Haig suggested that the fetus is trying to increase blood sugar levels by releasing human placental lactogen (hPL, a hormone that reduces the effects of insulin) into the mother's bloodstream and the mother is trying to reduce blood sugar levels by increasing more insulin. The amount of hPL in the mother's blood is astonishing, on the order of 1000-2000x the level of comparable hormones, and the amazing thing is that hPL is entirely unnecessary because babies with nonfunctional hPL genes are completely normal at birth. So the mother and fetus are battling over the allocation of resources by pumping out more and more hormones.

But wait that's not the weird part! The weird part is, strictly speaking, the conflict is not between mother and fetus, but rather between genes within the same individual. The fetus that contains a gene causing it to release more hPL will someday, provided it is a female, grow up to be a mother whose offspring may contain that very same gene and will be trying to suck every bit of nourishment out of her. The gene is then, within the very same individual, advantageous during fetus-hood and disadvantageous during mother-hood. The net result is that the fetus loses out by possessing the hPL gene, its lifetime reproductive output is reduced, but it doesn't matter. The gene spreads anyway because the gene increases its representation in the gene pool.

There is then an inherent inefficiency in life. The intragenomic conflict leads to reduced reproductive output, through less efficient use of resources and, more importantly, in the hPL case, through increases in the onset of diabetes. 10% of pregnancies result in gestational diabetes; and 50% of the cases of gestational diabetes results in full diabetes later on in life. Would that be object of an intelligent designer?

Photo by Street Cow.
*Ridley, M. 2001. The Cooperative Gene. The Free Press.


  1. Another way to look at it is as the conflict between maternal and paternal genes, only the former of which are shared with the mother who is carrying the fetus.

  2. Cool phenomenon - thanks for the post.

    I've been thinking about this, and evolutionarily, it may not be all that strange. Remember that back in the early days of humanity, mothers probably only lived to be an average of 30 years old, due in large part to the high rate of death during childbirth.

    >"50% of the cases of
    >gestational diabetes results
    >in full diabetes later on >life.

    So, the odds were good that the mother wouldn't live long enough to ever develop that subsequent diabetes. Thus, shunting some extra resources to the offspring seems like a reasonable tradeoff. Those selfish genes demand propagation, no matter what the cost to the parent.

  3. I have to agree when you quote Ridley, "The conflict between maternal and fetus genes is one of the weirdest ideas in the modern theory of evolution." It definitely is one really weird (but quite correct) idea.

    I can't remember if I read the Cooperative Gene, I don't think I did. Is it worth checking out?

  4. Chris... precisely. Check out my post on George Williams for more on this.

    Stupac2... some stuff is slightly out of date, but on the whole it is quite entertaining. I recommend it.

  5. Great post. I love to puzzle over situations like this, in which a gene is detrimental to one gender but advantageous to the other, or detrimental to the mother but not the fetus, etc. Unfortunately I don't have the math skills to model the selective pressures on the alleles, but it's fun to think about anyway.