Telomerase overexpression slows aging

Somatic cells may divide only a limited number of times before undergoing replicative exhaustion. Originally mysterious, the Hayflick limit (named after the scientist who first characterized it) is now understood at a mechanistic level: Each time a cell divides, the telomeres (repetitive DNA sequences found at the ends of each chromosome) become shorter; critically short telomeres trigger a permanent growth arrest known as cellular senescence. (Telomere shortening is in itself a consequence of the physical limitations of template-directed DNA replication; using the inevitable shortening of chromosome ends as a kind of physiological “clock” is a wonderful example of nature making a virtue out of necessity.)

Not all cells are mortal, of course: Germ line cells and the pluripotent somatic cells known as stem cells express telomerase, the enzyme the lengthens telomeres, and thereby sidestep the Hayflick limit. Stem-ness and germ-ness of a given cell aside, introducing the gene for TERT (the catalytic subunit of telomerase) appears to be sufficient to confer clonal immortality. Within a tissue, telomerase-expressing cells provide a theoretically infinite reserve of replacements for other cells that die due to tissue injury, wear and tear, or even the clonal death resulting from hitting the Hayflick limit.

Expressing telomerase in a wide variety of somatic cells would therefore seem a tempting strategy for lifespan extension. Specifically, telomerase expression could prevent any age-related decline in tissue function can be attributed to decreased regenerative potential.

We know why this is a non-starter, of course. Telomerase is tightly repressed in most somatic cells, and for a very good reason: What do you call a cell with an unlimited division potential that’s not a stem cell or germ cell? Usually “cancer.” Even for a tumor cell that has overcome the senescence checkpoints, the physical rules of DNA replication still apply, and telomeres will shorten every division until the cell is eating into its own coding DNA. Therefore, it’s essential for an ambitious young cancer cell to find a way to lengthen its own telomeres; indeed, this problem is significant enough that it’s considered one of the major steps in tumor progression. In any case, an organism with widespread telomerase expression in its somatic cells would very likely find itself dealing with multiple neoplasias — hardly the right animal in which to ask questions about division potential and lifespan.

But what if cancer couldn’t form for other reasons? In such a case, we could test the hypothesis that increased regenerative capacity confers increased lifespan. That’s precisely what a multi-lab collaboration from Spain has done; they find that mice that express TERT in most of their cells live significantly longer than the wildtype. From Tomás-Loba et al.:

Telomerase Reverse Transcriptase Delays Aging in Cancer-Resistant Mice

Telomerase confers limitless proliferative potential tomost human cells through its ability to elongate telomeres, the natural ends of chromosomes, which otherwise would undergo progressive attrition and eventually compromise cell viability. However, the role of telomerase in organismal aging has remained unaddressed, in part because of the cancer-promoting activity of telomerase. To circumvent this problem, we have constitutively expressed telomerase reverse transcriptase (TERT), one of the components of telomerase, in mice engineered to be cancer resistant by means of enhanced expression of the tumor suppressors p53, p16, and p19ARF. In this context, TERT overexpression improves the fitness of epithelial barriers, particularly the skin and the intestine, and produces a systemic delay in aging accompanied by extension of the median life span. These results demonstrate that constitutive expression of Tert provides antiaging activity in the context of a mammalian organism.

In mice expressing higher levels of three different tumor suppressors, cancer is essentially unable to form (one way to think about it is that tumors are delayed longer than the lifespan of the animal). In these animals, TERT indeed confers increased regenerative capacity and a significant increase in median lifespan.

Two questions, of many that one might raise:

First, why is the effect only on median lifespan? Inspection of the figures in the paper reveals that the cancer-resistant/TERT-expressing mice have a 50% survival time that is 20-30% longer than the wildtype — but by the time all of the oldest wildtype animals have died, so have all of the painstakingly engineered mutants. The clear implication is that exhaustion of regenerative potential is more relevant to early-life mortality than late-life mortality — counterintuitive, because one would expect regenerative failure to get progressively worse as a function of time, and to make an increasingly important contribution to mortality later in life.

Second: Mouse cells have really long telomeres, and telomerase expression is widespread in mouse tissues (though not usually at high enough levels to prevent some telomere shortening at every cell division). It takes mouse TERT knockouts around four generations of homozygosity to even begin to see a phenotype. Granted, mouse generations are far shorter than mouse lifespans, so this is not the same as saying that it takes four lifetimes for TERT to make a difference, or for replicatively senescent cells to begin to appear within a given mouse. But still, it makes me wonder what’s going on. Could telomerase be doing something else — i.e., something other than lengthening telomeres — that is particularly important in determining median lifespan?


  1. I think it would be telling to know the causes of death among those mice.

    For instance, if they underwent cellular senescence, it could be because despite the enhaced telomerase, there was a lack of growth signals (after all, telomerase by itself does not increase cell turnover…), or because the overexpression of tumor-suppressors caught up with the increased telomerase, or a combination of both.

    If, on the other hand, they didnt undergo senescence but all died of cancer, well, it would be telling as well…

    BTW: supposedly there was an experiment a while ago with telomerase-enhaced mice having increased maximum lifespan. This isn´t that one, right?

  2. There is evidence in the literature that overexpressing telomerase in human fibroblasts can protect mitochondria from sustaining OS-induced damage: the paper demonstrated that OS resulted in active export of TERT and co-localisation with mitochondria. This “extra-telomeric” effect resulted in improved mitochondrial function, but telomere attrition continued. It is interesting that in the Blasco paper above, telomere attrition was significantly delayed.
    The work certainly needs to be repeated in a model that has a shorter mean telomere length…

  3. I think it is interesting that those commenting on telomerase think that just because cancer cells turn on telomerase, turning on telomerase in non-cancerous somatic cells will inevitably result in cancer cells. This simply makes no sense. Cancer cells are going to turn on telomerase without our help, and normal somatic cells are not going to become cancerous because telomerase is turned on. In fact, there is some evidence that cells that reach senescence (without telomerase) turn cancerous as an attempt to prevent cell death.

    Also, if some cells do become cancerous, there doesn’t seem to be any evidence that *all* cells do. This means that cancer cures which target only cancer cells (see will solve the problem of any cancer cells being created.

    If telomerase turns out to be a universal cell regeration mechanism and any cancer risk is removed, this may be quite a medical miracle.


  4. For evidence of the fact that telomerase does not cause cancer (but is instead only permissive of cancer that is otherwise happening), go to and search using:

    telomerase does not cause cancer

  5. Telomerase per se doesn’t cause cancer; but neither is it merely a permissive factor.

    That term (“permissive factor”) is a little bit misleading in this context anyway. There are several different alterations that have to take place in a cell before it develops into a tumor. All of these changes are necessary but none of them is sufficient on its own. For example, cancer cells have to overcome DNA damage checkpoints — if they don’t, no tumor will form — but by itself, abolishing a single checkpoint won’t create a tumor. So rather than talking about what does (or doesn’t) cause cancer, we should instead refer to what is (or is not) a necessary step in the multi-stage process of tumorigenesis.

    Reactivation of telomerase (or activation of the ALT pathway) is an essential step in tumorigenesis. It’s a significant enough barrier to tumor formation that animals who overexpress telomerase in the soma will basically all get premature cancer. No one is arguing that all TERT-overexpressing cells form tumors, but the rate of tumorigenesis among these cells is extremely high, and it is therefore the case that all TERT-overexpressing animals will.

    This is why the authors of the paper described above had to overexpress three separate tumor suppressors — otherwise, the TERT-overexpressing mice simply died of early cancer. The cancers may have been “caused” by something other than TERT expression, e.g. mutation (though even this gets into the issue of proximate vs. ultimate causes) — but the population studies demonstrate that TERT was necessary for those mutations to result in lethal tumors.

  6. I recall reading, in this blog and elsewhere, that telomerase, aside lenghtening telomeres, also had some role at fixing DNA. Particularily, one of it´s sub-units. (in particular, I recall reading here that someone used gene engineering to breed a mouse which had a gene for a “portion” of telomerase, and he saw some effects that pointed towards that conclusion) We also know that oncogene knockouts tend to have increased cell proliferation (and increased rate of cancer, as a tradeoff), and vice versa, that tumor suppressors activation severely limits cell proliferation (such as in the aged bone marrow, with Pl6). So I wonder: could it be that in these telomerase-overexpressing mice we are seeing increased DNA damage repair (or DNA protection. Whichever applies), which in turn results in a lower rate of tumor supressor genes firing off, at least for a while?

  7. What would likely occur if telomerase expression was turned on for a sufficient period of time to lengthen telomeres and (perhaps) assist in the repair of DNA – per Mstudent’s comment – and then turned off? Would the cell’s “clock” be effectively reset with the associated risk of tumor development returned to normal?

  8. We’re pretty sure that p16 activation results in chromatin reorganization and the formation of permanent heterochromatic regions where all or most gene expression is silenced.

  9. This whole mouse study is tainted because the mice they are studying are inbred countless times which causes telomeres to be unnatually extended. Naturally found field mice do NOT have telomeres this long so the experimentation of these mice is flawed from the start. It takes several generations to wear down the telomeres in lab mice.

  10. Hey sorry about this stupid question but what does it mean by median lifespan? Thanks 🙂

Comments are closed.