Today I’ll point out an open-access critique of programmed aging theories by the originator of the disposable soma theory of aging, one of the modern views of aging as accumulated damage rather than programming. The question of how and why we age is wrapped in a lot of competing theory, but of great practical importance. Our biochemistry is enormously complex and incompletely mapped, and thus the processes of aging, which is to how exactly our biochemistry changes over time, and all of the relationships that drive that change, are also enormously complex and incompletely mapped. Nonetheless, there are shortcuts that can be taken in the face of ignorance: the fundamental differences between young and old tissue are in fact well cataloged, and thus we can attempt to reverse aging by treating these changes as damage and repairing them. If you’ve read through the SENS rejuvenation research proposals, well, that is the list. The research community may not yet be able to explain and model how exactly this damage progresses, interacts, and spreads from moment to moment, but that effort isn’t necessary to build repair therapies capable of rejuvenation. You don’t need to build a full model of the way in which paint cracks and peels in order to scrub down and repaint a wall, and building that model is a lot most costly than just forging ahead with the painting equipment.
The engineering point of view described above, simply getting on with the job when there is a good expectation of success, is somewhat antithetical to the ethos and culture of the sciences, which instead guides researchers to the primary goal of obtaining full understanding of the systems they study. In practice, of course, every practical application of the life sciences is created in a state of partial ignorance, but the majority of research groups are nonetheless oriented towards improving the grand map of the biochemistry of metabolism and aging rather than doing what can be done today to create rejuvenation therapies. Knowledge over action. If we had all the time in the world this would be a fine and golden ideal. Unfortunately we do not, which places somewhat more weight on making material progress towards the effective treatment of aging as a medical condition – ideally by repairing its causes.
But what are the causes of aging? The majority view in the research community is that aging is a process of damage accumulation. The normal operation of metabolism produces forms of molecular damage in cells and tissues, a sort of biological wear and tear – though of course the concept of wear and tear is somewhat more nuanced and complex in a self-repairing system. This damage includes such things as resilient cross-links that alter the structural properties of the extracellular matrix and toxic metabolic waste that clutters and harms long-lived cells. As damage accumulates, our cells respond in ways that are a mix of helpful and harmful, secondary and later changes that grow into a long chain of consequences and a dysfunctional metabolism that is a long way removed from the well-cataloged fundamental differences between old and young tissues. An old body is a complicated mess of interacting downstream problems. In recent years, however, a growing minority have suggested and theorized that aging is not caused by damage, but is rather a programmed phenomenon – that some portion of the what I just described as the chain of consequences, in particular epigenetic changes, are in fact the root cause of aging. In the programmed view of aging, epigenetic change causes dysfunction and damage, not the other way around. That these two entirely opposite views can exist is only possible because there is no good map of the detailed progression of aging – only disconnected snapshots and puzzle pieces. There is a lot of room to arrange the pieces in any way that can’t be immediately refuted on the basis of well-known past studies.
There are two ways to settle the debate of aging as damage versus aging as evolved program. The first is to produce that grand map of metabolism and aging, something that I suspect is at the least decades and major advances in life science automation removed from where we stand now. The other is to build therapies that produce large degrees of rejuvenation, enough of a difference to put it far beyond argument that the approach taken is the right one. That is not so far away, I believe, as the first SENS rejuvenation therapies are presently in the early stages of commercial development. I think that, even with the comparative lack of funding for this line of development, ten to twenty years from now the question will be settled beyond reasonable doubt. Meanwhile, the programmed-aging faction has become large enough and their positions coherent enough that the mainstream is beginning to respond substantially to their positions; I expect that this sort of debate will continue all the way up to and well past the advent of the first meaningful rejuvenation therapies, which at this point look to be some form of senescent cell clearance.
Many people, coming new to the question of why and how aging occurs, are attracted naturally to the idea of a genetic programme. Aging is necessary, it is suggested, either as a means to prevent overcrowding of the species’ environment or to promote evolutionary change by accelerating the turnover of generations. Instead of programmed aging, however, the explanation for why aging occurs is thought to be found among three ideas all based on the principle that within iteroparous species (those that reproduce repeatedly, as opposed to semelparous species, where reproduction occurs in a single bout soon followed by death), the force of natural selection declines throughout the adult lifespan. This decline occurs because at progressively older ages, the fraction of the total expected reproductive output that remains in future, on which selection can act to discriminate between fitter and less-fit genotypes, becomes progressively smaller. Natural selection generally favours the elimination of deleterious genes, but if its force is weakened by age, and because fresh mutations are continuously generated, a mutation-selection balance results. The antagonistic pleiotropy theory suggests that a gene that has a benefit early in life, but is detrimental at later stages of the lifespan, can overall have a net positive effect and will be actively selected. The disposable soma theory is concerned with optimizing the allocation of resources between maintenance on the one hand and other processes such as growth and reproduction on the other hand. An organism that invests a larger fraction of its energy budget in preventing accumulation of damage to its proteins, cells and organs will have a slower rate of aging, but it will also have fewer resources available for growth and reproduction, and vice versa. Mathematical models of this concept show that the optimal investment in maintenance (which maximizes fitness) is always below the fraction that is necessary to prevent aging.
In recent years, there have been a number of publications claiming that the aging process is a genetically programmed trait that has some form of benefit in its own right. If this view were correct, it would be possible experimentally to identify the responsible genes and inhibit or block their action. This idea is, however, diametrically opposed to the mainstream view that aging has no benefit by its own and is therefore not genetically programmed. Because experimental strategies to understand and manipulate the aging process are strongly influenced by which of the two opinions is correct, we have undertaken here a comprehensive analysis of the specific proposals of programmed aging. On the principle that any challenge to the current orthodoxy should be taken seriously, our intention has been to see just how far the various hypotheses could go in building a convincing case for programmed aging.
This debate is not only of theoretical interest but has practical implications for the types of experiments that are performed to examine the mechanistic basis of aging. If there is a genetic programme for aging, there would be genes with the specific function to impair the functioning of the organism, that is to make it old. Under those circumstances, experiments could be designed to identify and inhibit these genes, and hence to modify or even abolish the aging process. However, if aging is nonprogrammed, the situation would be different; the search for genes that actively cause aging would be a waste of effort and it would be too easy to misinterpret the changes in gene expression that occur with aging as primary drivers of the senescent phenotype rather than secondary responses (e.g. responses to molecular and cellular defects). It is evident, of course, that genes influence longevity, but the nature of the relevant genes will be very different according to whether aging is itself programmed or not.
For various programmed theories of aging, we re-implemented computational models, developed new computational models, and analysed mathematical equations. The results fall into three classes. Either the ideas did not work because they are mathematically or conceptually wrong, or programmed death did evolve in the models but only because it granted individuals the ability to move, or programmed death did evolve because it shortened the generation time and thus accelerated the spread of beneficial mutations. The last case is the most interesting, but it is, nevertheless, flawed. It only works if an unrealistically fast-changing environment or an unrealistically high number of beneficial mutations are assumed. Furthermore and most importantly, it only works for an asexual mode of reproduction. If sexual reproduction is introduced into the models, the idea that programmed aging speeds up the spread of advantageous mutations by shortening the generation time does not work at all. The reason is that sexual reproduction enables the generation of offspring that combine the nonaging genotype of one parent with the beneficial mutation(s) found in the other parent. The presence of such ‘cheater’ offspring does not allow the evolution of agents with programmed aging.
In summary, all of the studied proposals for the evolution of programmed aging are flawed. Indeed, an even stronger objection to the idea that aging is driven by a genetic programme is the empirical fact that among the many thousands of individual animals that have been subjected to mutational screens in the search for genes that confer increased lifespan, none has yet been found that abolishes aging altogether. If such aging genes existed as would be implied by programmed aging, they would be susceptible to inactivation by mutation. This strengthens the case to put the emphasis firmly on the logically valid explanations for the evolution of aging based on the declining force of natural selection with chronological age, as recognized more than 60 years ago. The three nonprogrammed theories that are based on this insight (mutation accumulation, antagonistic pleiotropy, and disposable soma) are not mutually exclusive. There is much yet to be understood about the details of why and how the diverse life histories of extant species have evolved, and there are plenty of theoretical and experimental challenges to be met. As we observed earlier, there is a natural attraction to the idea that aging is programmed, because developmental programming underpins so much else in life. Yet aging truly is different from development, even though developmental factors can influence the trajectory of events that play out during the aging process. To interpret the full complexity of the molecular regulation of aging via the nonprogrammed theories of its evolution may be difficult, but to do it using demonstrably flawed concepts of programmed aging will be impossible.
Given that the author here has in the past been among those who dismissed the SENS initiative as an approach to treating aging by repairing damage, it is perhaps a little amusing to see him putting forward points such as this one: “despite the cogent arguments that aging is not programmed, efforts continue to be made to establish the case for programmed aging, with apparent backing from quantitative models. It is important to take such claims seriously, because challenge to the existing orthodoxy is the path by which science often makes progress.” Where was this version of the fellow ten years ago?