Abstract
“Separa et impera”
On June 6th, a review appeared
1
concerning the state of aging research and the promising ways forward for the field. So far, so good. But this was not any old review. Here's why: • it appeared in Cell, one of the most influential journals in biology; • it is huge by Cell's standards— 24 pages, with well over 300 references; • all its five authors are exceptionally powerful opinion-formers—senior, hugely accomplished and respected scientists' • above all,
It will not escape those familiar with SENS that this last feature is not precisely original, and it may arouse some consternation that no reference is made in the paper to that prior work. But do I care? Well, maybe a little—but really, hardly at all. SENS is not about me, nor even about SENS as currently formulated (though a depressing number of commentators in the field persist in presuming that it is). Rather, it is about challenging a profound, entrenched, and insidious dogma that has consumed biogerontology for the past 20 years, and which this new review finally—finally!—challenges (albeit somewhat diplomatically) with far more authority than I could ever muster.
The above is a pretty dramatic characterization, I realize, so I will not cut too many corners in defending it. I appreciate that readers will vary in their familiarity with the relevant history of the field, so I shall start from the beginning.
Let's step back 30 years, to well before my time. In the early 1980s, biogerontology was a backwater in biology, disrespected for its “descriptive” nature, i.e., for its paucity of opportunity for hypothesis-driven work. This makes no sense to me—it stands to reason that any field must start out as descriptive, and that the scientists who piece together enough of the puzzle to enable a progressive transition to more hypothesis-driven work are every bit as important and worthy of praise as those who come after them—but it is a documented and undisputed fact. But also biogerontology was a conceptually fragmented discipline. Over the preceding decades, starting in the 1950s with Denham Harman's free radical theory and continuing right up to the early 1980s with the introduction of glycation and epigenetic noise as potential players, a positive zoo of “theories of aging” had emerged—and that's even if you count only the mechanistic ones, which at the end of the day are those that matter when considering strategies for intervention. This led to a view that aging is a chaotic, un-selected, un-programmed phenomenon—and to great despondency in the field with regard to any prospect for ever doing anything about it. (That despondency fueled an increasing tendency to claim, at least in the presence of anyone with deep pockets, that intervention was not even a main goal of the field—but I digress, and will do so no further, not least because my views on such matters have already been aired ad nauseam in this space and elsewhere.)
In parallel with all this, however, another strand of biogerontology was moving forward—the study of calorie restriction (CR). CR had been known since the 1930s to extend rodent longevity by 30%–50%, and had inevitably been the focus of intense study. But attempts to dovetail it with “theories of aging” had been thoroughly unsuccessful; in the words of the pioneering CR researcher Roy Walford, CR supported every theory of aging.
Into this morass, in 1983, came Michael Klass's discovery of a Caenorhabditis elegans mutant that greatly increased longevity. 2 Huh???? How could such a mutant exist? Since aging is chaotic, surely it was impossible to slow it down in toto (as was clearly required to extend life span much) by disrupting just one gene. Indeed, despite increasingly clear follow-up work, it was not for another decade (until Cynthia Kenyon's discovery of a different mutation with an even bigger effect 3 ) that this advance was given real credence and consideration.
Those new to this timeline may think that the next step is obvious: That these mutations are ways to get to the bottom of the tension between the chaotic nature of aging revealed by mechanistic work and the unitary nature revealed by CR. And indeed they are—and rather blatantly so, because both Klass's and Kenyon's mutants were in genes involved in the dauer pathway, the alternative developmental strategy adopted by C. elegans larvae when deprived of adequate nutrients. But that's not what happened.
What actually happened, once Kenyon's contribution made it unequivocal that aging could be modulated by genes, was that large numbers of ambitious (and talented, sure) biologists flocked to biogerontology, attracted by the seductive combination of public interest and new-found tractability to the fashionable cell and molecular biology techniques at which they were already expert. A positive torrent of publications in the highest-impact journals duly ensued and still shows no sign of diminishing. But these publications were, necessarily, focused on manipulations and dissections of precisely those aspects of aging that were the most amenable to such intervention. It was virtually forgotten, at least if the highest-impact journals were anything to go by, that that might not be the whole story.
I am not disputing that to some extent, a rising tide floats all boats. But only to some extent. As an example, consider the role of telomere shortening in aging—something on which those familiar with my work will know I harbor doubts, but that isn't the point here. It was back in 1997 that telomerase knockout mice were shown to have undiminished life span at first, but there were various problems after a few generations of inbreeding, problems that could be clearly linked to failed maintenance of stem cells in rapidly renewing tissues. 4 The single most obvious, but somewhat expensive, next experiment was to generate a mouse with a knockout of endogenous telomerase but a drug-inducible transgenic copy so that telomerase could be reactivated late in life, thereby determining which aspects of telomere-induced aging could be reversed by re-lengthening of telomeres. And this experiment was indeed done. But was it done at once? Not as such: It took FOURTEEN YEARS before such work saw the light of day. 5 I confess I have not consulted Ron DePinho to determine how hard he tried to obtain funding for such work at earlier times, but I think you can see my point.
It has been blindingly obvious for the past decade, and not only to me, that there is actually no data conflict here. CR is a simple intervention, and aging is complex and chaotic. That doesn't pose a paradox: It just says that CR triggers a programmed response that fans out into a bunch of processes with the combined effect of slowing aging. It's easy to come up with plausible evolutionary advantages for the possession of such a response, so its existence is not paradoxical. But what does it mean for biomedical intervention?
Two decades–long studies of CR in monkeys have reported results in recent years, 6,7 and although many commentators have over-simplistically favored one over the other, the studies actually tell a broadly similar story: Too much refined food is really bad for you, but severe restriction of natural food is not measurably better for you (at least not if you're at least averagely healthy for your age to begin with) than slight restriction. Because monkeys are so similar to humans, this is strong evidence that both CR itself (however performed, e.g., by intermittent fasting) and CR mimetics (drugs that seek to trick the body into thinking it's on CR even when it isn't) will be minimally if at all effective against aging in humans.
We may summarize the foregoing as follows. Aging has been shown, over several decades, to consist of a multiplicity of loosely linked processes, implying that robust postponement of age-related ill-health requires a divide-and-conquer approach consisting of a panel of interventions. Because such an approach is really difficult to implement, gerontologists initially adopted a position of such extreme pessimism that all talk of intervention became unfashionable. The discovery of genetic and pharmacological ways to mimic CR, after a brief period of confused disbelief, was so seductive as a way to raise the field's profile that it was uncritically embraced as the fulcrum of translational gerontology for 20 years, but finally that particular emperor has been decisively shown to have no biomedically relevant clothes. The publication of so authoritative a commentary adopting the “paleogerontological” position, that aging is indeed chaotic and complex and intervention will indeed require a panel of therapies, but now combined with evidence-based optimism as to the prospects for implementing such a panel, is a key step in the elevation of translational gerontology to a truly mature field.