Human telomerase activity: detectability and inducibility

Aubrey de Grey ag24 at mole.bio.cam.ac.uk
Tue Dec 21 11:39:36 EST 1999

I would welcome opinions, especially Sydney's, on what seems to me
to be a very remarkable oversight by both sides in the long-running
debate in Nature Biotechnology which I mentioned yesterday, begun
in 16(5):396 and continuing in 16(8):701, 17(1):4 and 17(4):313.

Briefly: Harry Rubin began by chastising the telomere community for
uncritically interpreting the immortalising effects of constitutive
telomerase expression as proof that telomere shortening is the sole
trigger of replicative senescence in human cells.  He preferred the
explanation that immortalisation occurs in occasional cells as the
result of chromosomal damage, which he argued is substantial in any
cell culture due to the trauma of explantation, and which might be
somehow exacerbated by telomerase.  In reply, Faragher et al. drew
attention to crucial differences between the increase of replicative
capacity induced by telomerase versus by v-myc, and made the strong
case that cells without telomere-maintenance capacity are doomed to
eventual senescence or apoptosis; they went on to suggest that the
presence of senescent cells in vivo and their delayed appearance in
calorically restricted mice, among other things, supports the role
of replicative senescence in organismal aging.  They also suggested
that the long telomeres of mice can explain the un-reduced tumour
formation of the telomerase knockouts.  Rubin retorted that telomere
length of human dermal fibroblasts does not fall after 20 years of
age, and he criticised the relevance of mouse telomere length.  21
authors later wrote (in one letter - fear not!) that the measurement
by Cristofalo et al. (PNAS 95:10614) of zero telomere loss with age
was explained by their use of donors and biopsy sites in which there
genuinely is no detectable turnover of those cells, implying (though
not explicitly stating) that such sites are atypical -- though they
then went on to say that "telomere erosion is ... not the way human
bodies count years", in contrast to the suggestion by three of the
same authors in a previous round that cell senescence does indeed
have something to do with aging.

Here is my question.  While I am not remotely persuaded by Rubin's
arguments that telomere loss does not limit replicative capacity, I
think that one of his points deserves extensive scrutiny, which it
did not receive in that exchange of letters.  He mentioned in Nature
Biotechnology 16(5):396-7 that several cell types have been shown,
both in vivo and in vitro, to induce telomerase activity when they
are stimulated into rapid cell division.  These were all cell types
which divide pretty often anyway.  Now: if I'm not mistaken, it took
until 1995 for anyone to discover telomerase activity in any human
somatic cell type (Counter et al, Blood 85:2315 is the first ref I
know of).  Am I right in saying that this was because the levels of
telomerase activity are below the pre-1995 level of detectability?
If so, it seems very plausible to me that relatively slowly-dividing
cell types may make even tinier amounts of the telomerase catalytic
subunit (perhaps only around cell division, perhaps not), but still
enough to maintain telomere length during aging.  When such cells
are cultured and required to divide unphysiologically often, they
might up-regulate telomerase, but they might not, or not enough to
maintain telomere length.  This would seem to be a reasonable way to
reconcile the Cristofalo result with the classical finding cited by
Rubin that fibroblasts in experimental animals turn over every two
months (Cameron and Thrasher, Topics Gerontol 10:108).

Thoughts and info welcome.

Aubrey de Grey

More information about the Ageing mailing list

Send comments to us at biosci-help [At] net.bio.net