Clearance of Senescent Cells is Fast in Youth, Slow in Aging, Tipping the Balance Towards Accumulation

The accumulation of senescent cells is a cause of aging, which is why a great deal of effort is presently going towards the development of senolytic therapies capable of selectively destroying these unwanted cells. Very little is known about the dynamics of senescent cells in old age, however. We know that older individuals have more senescent cells at any given moment in time, but is this because a small fraction of the many senescent cells created every day manage to linger persistently for years, resistant to the efforts of the immune system to remove them, or because clearance processes, while they will eventually destroy all senescent cells, are slowed to the point at which they cannot keep up? This open access paper suggests the second option to be more plausible. This has implications for therapies, such as how often a senolytic treatment would need to be applied.

In this study, we propose a framework for the dynamics of senescent cell (SnCs) based on rapid turnover that slows with age. Bleomycin-induced SnC half-life is days in young mice and weeks in old mice, causing critical slowing down, which greatly amplifies the differences between individual SnC levels at old age. We theoretically explore the implications of this slowdown in a model in which SnCs cause death when they exceed a threshold. The widening variation in SnC levels with age causes a mortality distribution that follows the Gompertz law of exponentially increasing risk of death.

The rapid removal of SnCs that we observe following bleomycin-induced DNA damage is in line with studies that showed efficient removal of SnCs in vivo following liver fibrosis or induction by senescence by mutant Ras. On the other hand, when senescence was induced in the skin by directly activating the cell-cycle inhibitor p14ARF, which was not associated with an increase in tissue cytokine expression or inflammation, the induced SnCs persisted in the tissue for several weeks. Clearance may thus depend on the tissue, on the method of senescence induction, and on the presence of the senescence-associated secretory phenotype (SASP).

The present analysis of longitudinal p16 trajectories suggests that SnC slow down their own removal rate. This effect may be due to several mechanisms, including SASP, disruption of tissue architecture, or SnC abundance exceeding immune capacity. For the latter effect, SnC abundance at old age needs to be comparable to the abundance of the immune cells that remove them, which make up on the order of 0.1% of the body's cells. Further research is needed to characterize these effects.

Our results suggest that treatments that remove SnCs can therefore have a double benefit: an immediate benefit from a reduced SnC load, and a longer-term benefit from increased SnC removal. Similarly, interventions that increase removal capacity, for example by augmenting the immune surveillance of SnC, are predicted to be an effective approach to reduce SnC levels. More generally, the present combination of experiment and theory can be extended to explore further stochastic processes in aging, in order to bridge between the population-level and molecular-level understanding of aging.

Link: https://doi.org/10.1038/s41467-019-13192-4