Why is Alzheimer’s disease (AD) so prevalent? If it were merely maladaptive, we would expect it to be just as tragic but far more rare, perhaps even fixed out of some populations.

Over at Mind Hacks, a hint that the explanation might be selectable antagonistic pleiotropy: a situation in which a gene with a deleterious effect late in life can be positively selected because of a beneficial effect during the reproductive or child-rearing years. Apparently the Apolipoprotein-E epsilon-4 allele, one of the major genetic risk factors for AD, makes you smarter, fitter, and (for all we know) better-looking in early life:

A fascinating study published in this month’s Cerebral Cortex reports that a gene known to massively increase the risk of Alzheimer’s disease in later life is associated, in young people, with better memory performance and more efficient use of the brain’s memory structures. …

A key question is why people who carry the Epsilon 4 allele would have a more efficient brain system for memory in early life but are more likely to have these same memory systems degrade in later life, as happens in Alzheimer’s disease.

As Alzheimer’s typically strikes after the time most people have children, the researchers suggest that the Epsilon 4 allele could confer an evolutionary advantage without adversely affecting chances of reproduction.

Some evidence that supports this idea has been found in previous studies where the ApoE Epsilon 4 allele has been associated with higher IQ scores, reduced heart activity under stress, and reduced chance of difficulties during pregnancy and post-birth problems.

Regarding the question in paragraph 2 of the excerpt (“Why might something good for your memory early be bad for your memory late?”), it might be fruitful to visualize “memory” as an emergent property of a lot of neurons working together, rather than as the setting of a single imaginary dial (in which case one is forced to ponder why one gene would swing the dial in opposite directions in youth and old age).

ApoE-epsilon-4 might be comparatively better than other alleles at doing something good for neurons — for example, delivering the lipids that neurons need to make their membranes, which in turn are required for signaling and connectivity; the happy neurons then make a more efficient brain, which in turn is for some combination of reasons better at making babies — but the very same feature (or perhaps even the specific proportions of distinct types of lipid molecules delivered by this allele) might, slowly over time, create an environment more conducive to the growth of amyloid plaques.

That’s a total swing in the dark, of course; I’m merely presenting an example of the type of mechanism by which antagonistic pleiotropy might act, if that’s really what’s going on. It’s possible that the early-life IQ and memory benefits are epiphenomena of the true selectable benefit (e.g., the cardiovascular phenotype, which it’s not hard to imagine resulting in improved mental function) — or, for that matter, some as-yet-unheralded effect on sperm count.

It does seem particularly pernicious that the same trait might give some people better brains early in life and worse brains (and terrible suffering, for themselves and their families) later in life. It’s almost like evolution isn’t fair.