Telomeres vs DNA damage - battle of the aging theories

even though everyone has their favorite pet theory of aging, we need to remember that aging is multifactorial

This is a brief commentary on one of the recent NUS aging webinars "AgeX Therapeutics: Targeting Biological Ageing | Dr Michael D. West".

I have never been a fan of the telomere hypothesis/theory. Not saying it is wrong, but it did get hyped up a lot. This should not be very surprising since it is very appealing to laypeople (and thus potentially investors) and then there was the Nobel prize putting it into the spotlight. It is interesting that the speaker would mention DNA damage as a potentially weaker theory. In fact, both theories suffer from the same issues (1).

Neither theory has received a particularly strong backing from interventional mouse studies, contrary to assertions in the video.

In addition, we should not forget that many other types of evidence are relevant to assess the strength of a theory. The types of evidence that are used in biogerontology include:

  • comparative biology – long-lived species should display conserved mechanisms of aging
  • long-lived mouse mutants – if your trait is age-associated it should be changed in these mice
  • mouse intervention studies
  • observational studies in humans (longitudinal change with chronologic aging; change associated with health outcomes)
  • interventional studies in humans

Let us compare the two theories in light of these principles. Strength of the DNA damage theory:
1/ Conceptually cleaner and safer: linear (or at least monotonous) relationship between DNA damage and aging can be expected (6). Telomeres are a beast that has to be tamed through temporal or spatial expression of telomerase. Do not want to have it active in cancer cells.
2/ Better human data since we have overwhelming evidence that there are age-related diseases caused by DNA damage and longer telomeres, i.e. cancer. Apart from that, overall I would say both theories have decent support from observational studies in humans.
3/ has actual support from long-lived models like dwarf mice (Page 2009, Dominick 2017) (3)
4/ I feel that the comparative data for the DNA damage theory is stronger, although this could be my own bias due to more familiarity with the data. 

Strength of the telomere theory:
1/ better support from mouse intervention studies, but not by much (5). The best study I could find was de Jesus et al. 2012 (7). While the control mice in this study were healthy, the treatment group consisted of only 17 animals and the study as such was never repeated by any other lab. For comparison, the seminal work on rapamycin was performed on 150 controls and 300 animals in the intervention group and was by now replicated in perhaps a thousand animals in different labs across the globe. Sure, the NIA ITP is run by the best mouse aging labs in the world, but others can do it too. See for example Zhang et al. 2012 for absolutely stunning lifespan data in FGF21-transgenic mice with good median lifespans for the controls and 77 treated animals vs 67 controls. While FGF21 is being studied by conventional pharma companies I have not seen much hype in the aging field about this finding and no startup has yet formed to harness this potential – as far as I can tell.

The DNA damage theory is hard to test specifically since we lack the tools to do so. If we included e.g. Nrf2 data as indirect evidence for the DNA damage theory the data would be roughly on par but still weak for both theories (2).

References and notes

epistemic status: quick draft, late to the party, half-baked

1/ Let us ignore the multistress resistance theory which seems like a better-supported superset of the DNA damage theory.

2/ weak is relative, I do not consider the DNA damage theory to be weak, just exceedingly hard to test. Also the Nrf2 experimental data is not particularly good, but substantially better than nothing I would say. We have nothing (in mice) that resembles direct evidence for the DNA damage theory. We do, however, have overwhelming evidence from progeroid models that DNA damage is important to aging. While I put less stock into these models than the average biogerontologist, it is a moot point in the context of this blog post. As far as I can tell, there are decent telomere attrition mouse models suggesting telomeres matter leading to a "draw". Although, again, I feel like the DNA damage progeroid models are a tad bit better.

3/ Page, Melissa M., et al. "Mechanisms of stress resistance in Snell dwarf mouse fibroblasts: enhanced antioxidant and DNA base excision repair capacity, but no differences in mitochondrial metabolism." Free Radical Biology and Medicine 46.8 (2009): 1109-1118.

Dominick, Graham, et al. "mTOR regulates the expression of DNA damage response enzymes in long‐lived Snell dwarf, GHRKO, and PAPPA‐KO mice." Aging Cell 16.1 (2017): 52-60.
Note: DSB repair seems to be the currently favored mechanism in comparative studies, so these dwarf studies are somewhat contradictory because they highlight different pathways.

4/ Zhang, Yuan, et al. "The starvation hormone, fibroblast growth factor-21, extends lifespan in mice." elife 1 (2012): e00065.

5/ To my great surprise the mouse data is actually much better than I remembered it, not that I was ever very familiar with the telomere field.

6/ Obviously "linear" is an over-simplification between hormesis and feedback mechanisms, which were also (likely) the death knell for antioxidant studies.

7/ Bernardes de Jesus, Bruno, et al. "Telomerase gene therapy in adult and old mice delays aging and increases longevity without increasing cancer." EMBO molecular medicine 4.8 (2012): 691-704.