The science news circuits have been buzzing about Jan van Deursen’s recent paper, in which mice stayed younger, longer when their senescent cells were removed. They’re right to call this technology a game changer for anti-aging medicine. They’re wrong to say this is new—in fact, the recent paper advances only incrementally over van Deursen’s stunning results from 2011. But what they don’t explain very well is that so far this technology only works for specially-prepared mice. The mice are genetically engineered (before birth) to make their senescent cells vulnerable to a trigger that can be administered later, when they are old. We don’t yet have a way to selectively kill senescent cells in a natural mouse, or a natural human.
(Background is in my blog post of last spring.)
As we get older, a tiny minority (~1 in 10,000) of cells becomes senescent, usually through telomere shortening that goes so far as to compromise the integrity of the chromosomes. The number of cells is small, but they do outsized damage, by secreting signals into the surrounding region and even into the body as a whole that turn on inflammaging, which is one of the primary modes by which the body destroys itself. Chronic, systemic inflammation is linked to all the diseases of old age, especially arthritis, cancer, arterial diseases and Alzheimer’s.
The original paper from 2011 reported on a novel idea to test the hypothesis that getting rid of this tiny number of cells could have a positive impact on the whole body. The experiment required genetically engineered mice. That means their genes were modified in the egg stage, when the incipient mouse is still a single cell, and there’s only one set of genes to modify. Mice could be prepared in such a way that a particular gene called p16 was associated with an added gene that made the cells extremely vulnerable to a drug that wouldn’t otherwise have damaged them. This was done because senescent cells express p16, while normal cells don’t. So administration of the drug would kill just the senescent cells, while leaving normal cells alone.
The results of the experiment were dramatic. Animals that had their senescent cells removed lived 20-25% longer, and were healthier and more active at an age when other mice were in steep decline. In the recent paper, life extension was bumped up marginally to 24-27%.
From my perspective as theorist, I take this as confirmation of the idea that aging is part of the developmental program, not an unavoidable side-effect or “accumulated damage” as standard thinking allows.
- The body is assassinated by signaling, not by damage.
- Much of the signaling comes from a tiny minority of cells that the body could eliminate, but doesn’t.
- And furthermore, there is no need for this minority to become senescent in the first place. They become senescent for want of telomerase, despite the fact that every cell in the body includes the telomerase (TERT) gene, and has the potential to produce telomerase, if it were instructed to do so. (There are many species that DO produce telomerase through their lifetimes, including mice, pigs, and cows.)
Most scientists have yet to assimilate this paradigm shift, and the popular press glosses over it with glib quotes.
This seems perverse, but there’s method to the body’s madness. Cells undergo senescence because they accumulate damage that could potentially lead to cancer, and the molecules they secrete prompt the immune system to come over and clear them. “It’s a very potent anti-cancer mechanism,” says Baker. But as we get older, the immune system falters, and senescent cells accumulate. Now, the molecules they secrete become problems rather than solutions.
Even then, senescent cells have benefits. Last year, Campisi showed that these cells help to heal wounds. And sure enough, Baker and van Deursen found that their mice heal more slowly after such cells were removed. [quote from TheAtlantic]
But (1) the cancer hypothesis has been abandonned even by its principal proponent, Judith Campisi. Senescent cells cause a net increase in cancer deaths. And (2) the idea that secretions from senescent cells may marginally increase wound healing efficiency cannot explain their evolutionary provenance if the small good is outweighed by a larger harm. The net result is that they kill us. (I wrote a related column last year.)
This technology holds up the possibility of a quick avenue toward life extension in humans that could be delivered in a treatment starting in middle age or even later. But promising as this idea is, it remains an idea and not a treatment that can be tested. Up until now, it only works in genetically engineered animals, and not in natural mice or you or me. What we need is a medication that will kill senescent cells while leaving normal cells undamaged. This is akin to the idea of chemotherapy, but perhaps somewhat easier because we have already identified a single genetic marker (p16) to identify the cells we want to kill, and because the cells are not proliferating and mutating as they are in a cancer patient. Nevertheless, there is a substantial challenge in finding the medication that can kill almost all senescent cells while leaving almost all other cells undamaged.
The word for such an agent is senolytic. Last year, two effective senolytic agents were reported: quercetin, a common botanical extract, and Dasanatib, a chemo drug [my blog post from last spring, including reference]. Though they prove the principle, they don’t distinguish senescent cells efficiently enough to offer an attractive therapy.
Some promising anti-aging technologies are being ignored by researchers and pharmaceutical companies, but this isn’t one of them. The good news is that there is a race on to test senolytic agents, with at least half a dozen labs competing to find powerful and non-toxic senolytic agents. Oisin Biotech is a start-up with a liposomal technology. Van Deursen and Campisi have their own for-profit spinoff, called Unity Biotechnology. This is now a problem of synthetic chemistry and testing, and we should know within a year or two if they are finding success.