INTRODUCTION
Telomeres are short repetitive DNA sequences that cap the ends of linear
chromosomes and are highly conserved across most eukaryotic organisms
(Meyne et al., 1989; Blackburn, 1991). Telomeres shorten during cell
replication and may be further degraded due to oxidative stress (von
Zglinicki, 2002). When telomeres get too short, the cell typically
ceases to replicate. The enzyme telomerase can rebuild telomeres
(Greider, 1990) but is normally repressed in somatic cells of many
mammals, presumably as a tumor suppression mechanism (Gomes et al.,
2010; Tian et al., 2018). In the absence of telomerase, telomere
shortening triggers cell senescence, which can prevent proliferation of
precancerous cells (Campisi, 2001). Thus, replicative senescence has
been interpreted as an adaptive tumor-suppression mechanism (e.g.
Campisi, 2001).
Cross-species analyses have furthered understandings of evolutionary
trade-offs shaping telomere length (TL) and telomerase activity.
Haussmann et al. (2003) showed that telomere loss rate, but not TL, was
inversely correlated with maximum lifespan in birds, which has since
been corroborated across tetrapod species (Dantzer & Fletcher, 2015;
Sudyka et al., 2016; Tricola et al., 2018, Whittemore et al., 2019;
Pepke & Eisenberg, 2020). Seluanov et al. (2007) found that telomerase
activity was negatively correlated with body mass, but not lifespan in
rodents. Since, all else equal, increased telomerase activity is
expected to lead to slower telomere erosion, Seluanov et al.’s results
apparently contrast with the results on telomere shortening rates
reported above. Seluanov et al. also found no significant associations
of TL with body mass or lifespan. In an analysis of 57 mammalian
species, Gomes et al. (2011) reported a negative relationship between
telomerase activity and mass, but not lifespan across mammals.
Furthermore, Gomes et al. found that lifespan was inversely correlated
with TL while accounting for body mass, but there was no independent
association between TL and mass. Here we build on these past findings,
first trying to replicate the analyses of Gomes et al., then proceeding
to further examine how telomere biology may influence cross-species
variation in cancer risk and finally how telomere biology may have been
shaped by selection related to domestication.
Telomere biology is widely assumed to influence cancer risk and thus for
different species to evolve different telomere dynamics in response to
differing fitness costs of cancer. Cancer occurs in almost all
vertebrates, but among wild animals, cancer incidence rates are little
studied, but seem to vary considerably (Effron et al. 1977; Pesavento et
al. 2018; Albuquerque et al. 2018; Boddy et al. 2020). For instance,
cancer is rarely reported in long-lived whales, bats or naked mole-rats,
while rats and mice are known to be much more prone to cancer
(Albuquerque et al. 2018). Different species have evolved different
anti-cancer strategies and mechanisms; e.g. long-lived species that
reproduce late in life are expected to be under stronger selection
pressure to evolve efficient cancer defenses compared to short-lived
species (Weinstein and Ciszek, 2002; Seluanov et al. 2008; 2018).
Despite larger and longer-lived species having more cells that exist for
a longer duration, they do not show higher cancer rates than smaller,
short-lived species (Peto et al. 1975). This phenomenon, known as
‘Peto’s Paradox’, is generally assumed to be explained by species with
larger body sizes and longer lives having increased fitness costs from
cancer development. Correspondingly larger and longer-lived species are
thought to tend to evolve shorter telomeres and decreased telomerase
activity (Caulin & Maley, 2011; Gomes et al., 2011; Tian et al., 2018;
Risques & Promislow, 2018). Thus, longer telomeres are predicted to
increase the risk of cancer (e.g. Aviv et al., 2017). However, there are
reasons to question the associations between telomere length and cancer
risk (reviewed in Eisenberg & Kuzawa, 2018). Shorter telomeres can
cause chromosomal instability that promotes cancer. Longer telomeres
improve immune function, which is important in combatting many
infectious diseases which cause cancers and in combatting the
development of malignancies. To better characterize the association
between telomere biology and cancer, using cross-species datasets we
examine whether longer telomeres and telomerase activity predict
increased cancer incidence as is often assumed.
In both mice (Mus musculus and Peromyscus leucopus ) and
pearl millet (Pennisetum glaucum ), domesticated strains have been
observed to have longer TLs than their wild counterparts (Bickle et al.,
1998; Hemann & Greider, 2000; Manning et al., 2002; Weinstein and
Ciszek, 2002; Sridevi et al., 2002; Kotrschal et al., 2007; Seluanov et
al., 2008). Multiple explanations for this putative association between
domestication and longer TL have been suggested (Manning et al., 2002;
Weinstein and Ciszek, 2002; Eisenberg, 2011). To better characterize how
general the association between domestication and TL is, we examine this
association across a cross-species dataset including 9 domesticated and
48 non-domesticated mammalian species.
Our analyses primarily build on a cross-species dataset generated by
Gomes et al. (2011). Gomes et al. measured mean telomere length (TL) and
telomerase activity among 43 mammal species and combined with comparable
published measurements on 14 species. Gomes et al. then examined
relationships between TL, telomerase activity, body mass and maximum
lifespan. As a first step to test our hypotheses regarding domestication
and cancer risk, we aimed to replicate the findings from Gomes et al.’s
analysis. We found considerably discrepant results. To establish the
robustness of the findings we conducted additional sensitivity analyses.
Thus, in this manuscript we first present our re-analysis of Gomes et
al. and try to explain these discrepant results. Then we proceed to test
whether TL and TA predict cancer risk and that TL was elongated by
domestication.