Genetic evolution is a clumsy way to adapt to a changing environment. Our genes were presumably shaped by very different conditions than we face now. Why wouldn’t natural selection favor organisms who can adapt to current conditions and pass on these adaptations to their children? Wouldn’t we be more fit if Lamarck was right and if so, why was he so wrong?
Turns out he wasn’t so wrong after all.
This was the first evidence, now confirmed multiple times, that an experience of the mother (what she eats) can reach into the DNA in her eggs and alter the genes her pups inherit. “There can be a molecular memory of the parent’s experience, in this case diet,” says Emma Whitelaw of Queensland Institute of Medical Research, who did the first of these mouse studies. “It fits with Lamarck because it’s the inheritance of a trait the parent acquired. There is even some evidence that the diet of a pregnant mouse can affect not only her offspring’s coat color, but that of later generations.”
That is from an article in Newsweek on epigenetics. Here is more. And here is a blog about epigenetics.
This raises the theoretical question: if you were to design the system of inheritance, where would it be optimal to draw the line between those characteristics that should be hard-wired in genes and those that can adapt at higher frequencies? And wouldn’t that depend on the environment? So would the line be hard-wired or epigenetic? And which side of the line is that trait on?
Cheap Talk 
4 comments
Comments feed for this article
March 30, 2009 at 5:59 pm
tam
Genes don’t, by and large, hard-wire characteristics. That, in fact, is what epigenetics is all about–modifying the actions of genes. Sometimes that’s a Good Thing and you end up with organisms from Lake Woebegon: above average. Other times, as in birth defects, not so much. Epigenetics is why identical twins are not really identical. It’s why chimps and people are so different even though they have pretty much the same genes.
March 30, 2009 at 9:05 pm
jeff
I see your point. But something is hard-wired right? I mean genetics is a formal system so it has to start with axioms.
March 31, 2009 at 8:47 am
tam
Hmmmm. Well, yours may be something of a philosophical point. So far as I know, in principle the action of any gene can be altered. Many different forces can potentially accomplish such alterations, for example an organism’s other genes and even what it eats. (Epigenetics.) The consequences of a particular alteration may of course be catastrophic for the organism, but that’s a different issue.
It’s important for people to understand that genes are not hard-wired in the sense I take it you mean. It’s certainly true that some genes are more buffered against outside manipulation than others. But even there, a range of “normal” actions may be available to a particular gene, depending on what else is going on, and any one of those actions will lead to a normal organism–or at least one that can function.
April 6, 2009 at 1:38 pm
Bob
I think the article is misleading in a significant way. She writes:
“This was the first evidence, now confirmed multiple times, that an experience of the mother (what she eats) can reach into the DNA in her eggs and alter the genes her pups inherit.”
But this is untrue. The experience does not ‘alter the genes’. In fact, she earlier states the opposite, that the fleas have ‘identical DNA sequences’. What appears to be changing is not the genes, but the expression of the genes. The author is sloppy with her words regarding this distinction.
I don’t see this as a repudiation of Darwin, so much as a reason to change our view of what genes really mean. Maybe there isn’t a gene for ‘bony helmet’ or ‘soft head’, as much as there is a gene for ‘bony head, if chemical A is present, soft head if chemical A is absent’. Or something even more complicated. The problem with our current thinking is the simple-minded approach to what a stretch of DNA ‘means’. Maybe its more complex than we thought, because we haven’t had enough time to uncover all of the complexity yet.