Why do we age? Part III

As the years go by, I have become more interested in ageing, Perhaps it is more accurate to say that my interest has been reignited since, as mentioned in an earlier article, my Ph.D. advisor was interested in this subject – this was in the 1960s – and, since he was an articulate explainer, his interest sparked mine.

Nevertheless, I suspect that an increased interest in the phenomenon of ageing, as one gets older, is not uncommon amongst biologists. The process forces itself upon your attention. You begin to notice its signs in yourself, even if not dramatic. At the same time, you become aware of the obituaries of recently deceased famous people, many of whom you had not thought of as particularly old. Then you read of their ages and you think “well, they were not exactly spring chickens”. Often, there will be a mention of their cause of death and frequently it is an age-related one. Intrigued by the inevitability of ageing, you do more reading and begin to think, “hey, this is a really interesting biological problem” especially because it is still largely a mysterious one, even if a lot better understood in its particulars than in the 20^th century.

In this newsletter, so far, I have devoted one article to Alzheimer’s disease, one of the main consequences of ageing and a common contributor to the deaths of elderly people (see The terrible tangle of Alzheimer’s disease) and two to the more general issue of ageing and the deterioration (“senescence”) that accompanies it (see Why do we age? Parts I and II). My initial plan had been to write a third on the two alternative strategies that biologists concentrate on to counter the effects of ageing. These can be denoted as, respectively, “healthy ageing” and “life extension”. The first has to do with amelioration strategies for ageing to reduce its effects and make it less of a burden. The second deals with ideas to substantially expand the human life-time beyond its generally accepted upper limit of 120 years.

Yet, I began to sense a two-fold problem in devoting an article to this subject. First, I did not feel that I had a particularly interesting or worthwhile take on “healthy ageing” though it certainly seems like a good idea. Beyond recommending agreed-upon things like a) don’t smoke, b) reduce your alcohol consumption, c) have a healthy diet, d) keep physically active, e) get enough sleep, f) have meaningful social relationships in your life, I was not sure I could go much beyond such common sense advice to anything thought-provoking. As to “life extension”, though I have heard it discussed for at least 20 years, I have seen nothing that convinces me that there is really something substantive there in the science. (I would be glad to hear from any readers if I have missed something important in this respect.) My current belief is that when we understand senescence better, that may well suggest some worthwhile ideas on life extension. However, for the moment, I think I will leave the topic of life-extension alone.

Yet, I have recently become aware of some work, over the last 10-15 years, on the biology of ageing that strikes me as fascinating and worth some attention. This article will focus on this material. It concerns evidence that there are stages of ageing and short time periods when the process speeds up somewhat. This phenomenon was not really predicted in any of the earlier theorizing, which makes it even more interesting. ¹

The conventional idea about the rate of ageing in humans is that, from the early 20s onwards, it increases steadily, at least until the early 90s, when it seems to level off. (This last fact, namely that one can get past further ageing, if you live long enough, is itself very interesting.) This is depicted in the famous Gompertz Law, a sigmoidal curve of mortality vs. age, named after its discoverer, Benjamin Gompertz (1779-1865). It dates back to 1825 and is based on human mortality data, by now a lot of it. W.M. Makeham made a later, significant contribution, publishing it in 1860 and it often is referred to now as the Gompertz-Makeham Law.

It has not been overthrown, falsified, by the new findings, but it was always a statistical approximation and it may need modification. Its underlying assumption, made explicit in the 20^th century, was that the rate of normal mortality (mortality not involving accidents or murders) was a reflection of the process of senescence.

Of course, every death is an individual case but over-all mortality from natural causes should reflect the generalities of the ageing process, hence mortality curves should provide significant clues to that process. Nevertheless, mortality is a downstream consequence of senescence and the longer the interval between the damage done by the process of senescence, whatever it is, and the death of the individual, the poorer the temporal correlation between the two events will be. Another way of saying this is that the greater the contribution of factors that are not a product of ageing per se, the harder it will is to dissect out those due to senescence. This was Makeham’s contribution. ²

The new findings suggest that there are upward blips in ageing rates in humans, at certain specific points, and that these reflect processes in physiological ageing that contribute to mortality. For humans, these periods map to, approximately, every 20 years, with the points of acceleration being roughly at ages 20, 40, 60 and 80. Furthermore, several kinds of data suggest that this is not just a human phenomenon or restricted to animals like ourselves, that is large mammals. There is evidence that this is a general fact of ageing in animals, being seen in fruitflies and nematodes, though of course, at different ages in them, given their much shorter lifetimes.

In fact, the first clue that ageing may exhibit discrete stages was found in fruit flies, coming from a laboratory in France and published in 2011. The investigators had fed a specific non-toxic blue dye to fruit flies of different ages; the aim of the study was to measure the amount of food intake in the flies, with the dye, which was not digested, thus being a reliable marker of the quantity of food.

In young fruit flies, the molecule, a medium-sized one was primarily confined to the gut. However, as the flies began to enter old age, around 40 days, there was an abrupt transition: the flies began to turn blue, and the transition was often seen within the space of a day. The change was due to a much greater permeability in the gut, as a result of the ageing process. Indeed, this marked the near end-of-life stage. These “Smurfs”, named for the blue cartoon character, soon died. This was not due to the dye, which, recall, was non-toxic and non-digestible but probably largely a consequence of the increased gut permeability, which is probably harmful in itself, as well as a marker of other changes due to senescence. Fruit flies tend to die within two days of acquiring Smurfness. ³

This first finding was consistent with a two-stage model of ageing, the first and longest involving flies of good health and low mortality and the second that of Smurfness, which was a harbinger of early death. Yet, work from other labs had indicated that there was a succession of physiological “hallmarks of ageing” in fruitflies, not just markers of passing chronological age. These included reduced energy stores, reduced mobility, and reduced fertility (for both males and females). The laboratory that had discovered the Smurf phenomenon soon published data showing there were about six discernible transcriptional stages associated with physiological ageing, not chronological age, and that while there was quite a spread of times for the earlier stages in middle-aged flies, there was a convergence toward showing all as flies became much older. ⁴

In other words, ageing is a process that is linked with time, hence with becoming chronologically older, but not strictly determined by time; it apparently is programmed but does not take place according to any biological chronometer. And at a certain point, in the ageing process, a tipping point or threshold is reached, and the end of life is not far away. For fruit flies, Smurfness is a visible indicator of passing that threshold. An increasing body of data on human ageing points in the same direction: that senescence is not a gradual, linear process of the body becoming less and less functional, but something that happens in reasonably discrete stages, which can be tracked by specific physiological and molecular hallmarks; this material is complicated and not fully understood. ⁵

There is much more (as with all my topics) that could be said about this subject but I will bring this piece to a close with two general thoughts.

In the first article in this mini-series (Why do we age? Part I) I suggested that there were two broad ideas about the nature of ageing and that most of the specific hypotheses are variants of one or the other. The first is that ageing in some sense is “programmed”, that its timing is built into each species’ biology. The second is that senescence is basically a case of cumulative wear-and-tear on our tissues and organs and that when someone dies of “old age”, they are actually dying from the loss of functionality brought about by long-time wear-and-tear. (It might seem hard to reconcile this latter view with the nature of cancer, a leading cause of death, but this can be accommodated by viewing cancer as the loss through some form of wear-and-tear of the restraining forces that keep it at bay in most non-elderly people.)

In that piece, I said that ageing-as-a-programmed process seemed unlikely to me. My reason for skepticism about programming as the explanation was that, in general, complex properties of biological systems are products of evolution and it is difficult to see how a mechanism for death, especially late in life could be selected. In general, selected properties arise because they give some advantage in early life for reproductive capacity. In normal evolutionary thinking, it is hard to imagine something that is selected that comes in only after reproductive activity has ceased, which is generally the case for humans.

That still seems like sound reasoning to me but the findings reviewed in this article actually strongly argue for an important element of programming in senescence. Of course, senescence might be an accidental by-product of something that has been selected for early-in-life advantages. Indeed, there is an important line of thinking about senescence that runs along these lines. However, the findings summarized here look like evidence for something rather intricate and complicated and it is hard to imagine something like that arising as an incidental side-effect. Not least, the late-stage plateau aging rate in humans is consistent with programming of senescence. Altogether, I think that the debate about programming vs cumulative wear-and-tear as the basis of senescence is still unresolved but I am more open to the former idea than I was.

The second general point concerns the practical implications of these findings about ageing-as-a-process-in-stages, which in turn brings us back to the matter of “healthy ageing” vs. “life extension”. In principle, if senescence occurs in stages, then there might be interventions that delay the onset of these stages. Indeed, there are some preliminary indications from the Drosophila work that this is possible. If so, this has implications for new ideas to promote healthy ageing. Those, in turn, might improve chances for life-extension, They might not be dramatic, e.g. to life spans of 200 or 300 years but still significant. We do not know but, in general, if someone asks you “Is X possible?”, the answer is almost always “yes” (though the probability might be very low.) The one certainty is that while the field of senescence research has had a long and venerable history (going back 200 years to the original Gompertz paper), it still has a lot of life in it.

1. A good popular account of this work is in Lawton, G. (2025). Rapid bursts of ageing are causing a total rethink of how we grow old. The New Scientist, July 12, 2025, www.newscientist.com/article72485338

2. A good review, with the underlying mathematics of mortality curves explained, see Golubev, A. (2023). An underappreciated peculiarity of late life human mortality kinetics assessed through the lens of a generalization of the Gompertz-Makeham law. Biogerontology 25: 479-490. Doi/org/10. 1007/s10522-023-10079-2

3. For a later and more detailed study, see M. Rera, C. Vallot, C. Lefrançois (2018). The Smurf transition: new insights on ageing from end-of-life studies in animal models. Curr. Op. in Oncology 30(1). 38-44 doi.org/10.1097.0000000000000419.

4. See F. Zane et al. (2023). Smurfness-based two-phase model of ageing helps deconvolve the ageing transcriptional signature. Aging Cell: doi/org/10.1111/acel.13946

5. A good review of human ageing seen from the stages-perspective is that of M. Olecka et al. (2025). BioEssays: 47: e202400222