Michael Rose's book The Long Tomorrow discusses aging in the context of evolution. The book has something of the feel of a memoir— Rose writes about how he happened into this field, who helped and encouraged him, who was right and wrong, and what he was working on along the way.
Rose has spent most of his career single-mindedly breeding long-lived fruit flies (Drosophilia Melanogaster). This approach to achieving longevity has important implications for understanding the mechanisms of evolution and how they impact lifespan. Because the mechanism itself (selective culling and breeding) isn't applicable to humans, the implications for human longevity are indirect, even though they may turn out to be very important for us.
Rose begins his tale at a symposium he spoke at for the Templeton Fund, where eminent scientists and ethicists spoke against any attempt to pursue human longevity as respectively impossible and immoral. Christian theologians joined that chorus, but were opposed by a Jewish scholar, and by Rose himself, already pursuing longer-lived flies. This serves as a nice backdrop and introduction to the issues which allows him to refer back to the controversy and the parties later when he talks about his own views.
In the mid-70s, when Rose was trying to get started in biology, Crick & Watson's theories had been accepted, but the implications weren't yet clear. Scientists were starting to investigate the molecular mechanisms for all kinds of biological effects, both to see how they interacted with evolution and to find the cellular mechanisms that made life function. Hayflick's proposal on the impact of cell division problems on lifespan was getting publicity, but there were few posited mechanisms based on evolutionary reasoning. The Hayflick limit was an observed and accepted fact, but so far without much theoretical context. Thirty years earlier, J. B. S. Haldane had suggested that Huntington's disease (which waits until carriers of the causative genetic defect are in their 30's or 40's to attack) was a consequence of selection's pressure: genes that debilitate before the carriers reproduce will be weeded out, but diseases that crop up later can survive more easily in the population. Peter Medowar expanded on these ideas, but got the evolutionary causality wrong. George C. Williams straightened out a few clues and pointed out that processes that are helpful in the young, and therefore reinforced by evolutionary pressures, might be costly for mature animals but they wouldn't necessarily be corrected because reaching maturity is so much more important evolutionarily than surviving it. Finally, in 1966, William Hamilton put the pieces together and provided an argument based on evolutionary principles that explained why aging would arise. Once Rose studied the history, it was clear that there was some evolution-based theory, but no significant experimental results validating the evolutionary basis of aging. He thought he'd be able to make a mark on this new field fairly quickly.
Brian Charlesworth was Rose's mentor and supervisor when he started the research. Charlesworth had worked out the math for how much selective pressures should fall as a creature aged. Since selection pressures are strongest on the young, you'd expect to see less weeding out of deleterious mutations that affect the mature and the aged than of those that affect the young. Charlesworth proposed that Rose look for an experimental demonstration of that difference in drosophilia, by showing that the fall in fecundity of female flies followed his formulas. Rose spent more than a year in the laboratory, and counted more than a million eggs before analyzing his results. The results showed that Charlesworth's predictions were wrong.
But Rose had started a parallel experiment about half-way through the first one, based on a paper by J. M. Wattiaux (published in 1968) which showed that offspring of older parents lived longer than those born to younger parents. Wattiaux had been unable to demonstrate what he expected to be environmental causes of the difference, but when Rose read the paper, he realized that age of reproduction was the crucial variable that evolution could select on. If only eggs produced from older parents were allowed to reproduce, then the selective pressure (acting on potential parents) would be towards flies that stayed healthy long enough to meet the delayed date. A year after sharing the results of the first experiment, Rose showed that the new procedure (delaying breeding) produced flies that were already living about 10% longer than under standard fly care.
From that point, Rose (and later his students) studied the interactions between longevity, stress resistance (starvation and desiccation), diet and diet restriction, and body fat and other energy storage. They established a positive correlation between long life and all kinds of stress resistance, and showed that body fat and other energy stored in the flies' bodies increased stress resistance. Diet restriction seems to work because it encourages retention of more energy reserves. But all of these mechanisms (except diet restriction) work over evolutionary time scales. Rose wanted to find treatments that can lengthen the lives of those already alive. So he started to investigate molecular mechanisms to see if we can tell what's different within the cells of those predisposed to longer life when compared to their counterparts.
A later series of experiments demonstrated that the death rate, which increases with age, only actually increases from the start of reproduction to its cessation. What this means is that the death rate is strongly controlled by evolutionary pressures. For any species, evolution acts to reduce the death rate as much as possible before the onset of sexual maturity. After that point, it allows the death rate (from natural causes, primarily) to increase relatively smoothly. Once sexual reproduction ends (or possibly a bit later if individuals are still contributing significantly to the life success of their offspring) evolution stops being able to apply selective pressure. This means that the rate at which individuals die (in deaths per capita per annum) stops increasing. For most populations the rate is pretty high by this mature age, and the number of individuals reaching each later age is small, so it's hard to see that the rate isn't changing. But separate experiments done by Rose, Charlesworth, and Larry Mueller show that manipulating the age of last reproduction in populations of drosophilia directly effects the death rates of the resulting populations in only a few generations.
In humans, Rose says that 95 is the age at which mortality rates stop increasing. From 15 to 90, there is exponentially increasing mortality, but after 95, the rates stop increasing. This has two interesting implications for humans. First, many more people are going to make it to their 90s in coming decades, just because we're living so much more healthily and robustly than we used to. That means that there will be much larger cohorts which will see their mortality rates stop increasing, so we should see gradually rising maximum ages, even if nothing else changes.
The other implication is that mortality rates don't increase to 100%. There is always a chance of surviving another year if you aren't hit by a bus. And if Aubrey de Grey is right, our first step is just to clean up our constitutions so that we keep the mortality profile of a 40-something, and we'll vastly improve life-spans. As Rose says:
Aging is not an infinitely high wall of mortality, rising faster and faster as we get older, until everybody is dead. It is a ramp that takes us from a phase of low childhood mortality to a much later phase of high, but relatively stable, mortality. Postponing, retarding, or otherwise mitigating aging does not require pushing back a wall of death of infinite height. It requires smoothing out a ramp of mortality, and possibly lowering the height of the top of the ramp.
At this point, Rose seems to get serious about the implications of his research for human longevity. At the prodding of New Scientist, in 1984, he wrote a lead article proposing that a major, government-backed program to produce long-lived "methuselah" mice would be valuable. He pushed the idea with potential funders from government and the private sector, but never found anyone willing to actually underwrite the proposal. Rose never mentions Aubrey de Grey's Methuselah Mouse project, which I've talked about before. de Grey found private funders to endow a prize, rather than attempting to organize a single Manhattan project-style (Rose's own description) effort.
Having failed at his own efforts to start a project to find ways to apply his research to humans, Rose continues to be upbeat at the prospects that someone will succeed. He finishes the book by offering lots of advice and suggestions on which scientific paths are likely and unlikely to bear fruit, and what modes of organization are worth trying, particularly by people with different skills than his.
My own postscript involves pointing out that Rose didn't talk about what happened to his fruit flies in the end. He was a co-founder of the company Genescient, which uses modern gene assay techniques to look for nutraceuticals that can reproduce the cellular effects present in the Methuselah flies. Rose's name doesn't appear on Genescient's site, so there appears to have been some sort of split. Genescient has spun out Life Code, which markets Stem Cell 100, a nutraceutical based on this research. I haven't started taking it yet, but I do continue to investigate.