Earlier this year, we discussed a pair of papers that proposed a role for the Wnt signaling pathway in aging. One of those studies focused on the klotho-/- mouse, which shows signs of progeria and has been taken as a model of accelerated aging.
In a letter to Science last week, however, we’re given cause to reconsider these results: U-Michigan’s Richard Miller, one of the more erudite and thoughtful (and outspoken) eminences grises of biogerontology, reminds us of some recent findings might influence our interpretation of any experiment using the klotho-/- mutant:
Liu et al. (Report, “Augmented Wnt signaling in a mammalian model of accelerated aging,” 10 August, p. 803) have elegantly shown how alterations in Wnt signals contribute to the suffering of klotho-deficient mice, but not every sick little rodent is a suitable model for human aging. The pathological features and short life span of klotho mutant mice have been shown to reflect hypervitaminosis D, secondary to ablated responses to Fgf-23 (1-3). The same syndrome appears in Fgf-23 mutants and can be cured by deleting the 1–hydroxylase gene that increases the activity of the vitamin. In both mutants, the features represented as evidence of “premature aging” can be eliminated simply by putting the mice on a diet low in vitamin D. Perhaps vitamin D deprivation will turn out to be the long-sought cure for aging, but in the meantime, it would be wise to view with some skepticism the claims that klotho and similar developmental mishaps provide convenient shortcuts for learning about mechanisms of “real” aging.
In other words, if the klotho-/- phenotype can be cured by twiddling the dietary levels of a single essential vitamin, to what extent can it be considered a legitimate model of accelerated aging? By extension, to what extent does this force a re-evaluation of findings based on the idea that klotho-/- is a bona fide progeria model?
There are many areas in aging research in which there is some disagreement. One question in dispute is the degree to which observations in simple organisms, such as postmitotic worms, can inform our understanding of mammalian aging. Similarly, reasonable people disagree on the role, if any, of cellular senescence in organismal aging. We appreciate that there is also considerable disagreement regarding how much mammalian models of accelerated aging can teach us about the normal aging process.
Our study centered on a set of observations suggesting that the Wnt family of proteins could bind to klotho, a protein whose absence has been linked to an accelerated aging phenotype in mice. Genetic evidence suggests that alleles of klotho are also associated with variation in human longevity (1). Nonetheless, we agree with Miller that considerable care must be taken when using the existing accelerated aging models as an indication of the normal aging process. Our opinion is that studying models of rapid aging will be useful in teasing out the underlying mechanisms of how we age, although we understand that Miller does not share that opinion. Hopefully, we will all live long enough to find out who is right.
I think the answer is a bit thin, as it doesn’t address Miller’s fundamental objection to the model (that the klotho-/- phenotype has a fundamental difference from real aging, namely, its ability to be cured by vitamin D deprivation), but others may disagree. Obviously, as the old refrain goes, more research is necessary. In the meantime, however, I wouldn’t start taking any Wnt inhibitors just yet.