Our study shows a strong negative correlation between telomerase activity and body mass. The correlation was significant both for the composite telomerase activity coefficient for six tissues and for the total raw telomerase activity. Analysis of individual tissues showed that telomerase activity in spleen, liver, and kidney negatively correlates with body mass. We propose the following model to explain coevolution of telomerase activity and body mass (). Evolutionary increases in body mass increase cancer risk, as larger animals contain more cells in their bodies and malignant transformation may occur in any single cell. The increased cancer mortality rate drives the adaptive evolution of tumor-suppressor mechanisms (Nunney, 1999
; Leroi et al., 2003
). Our results provide evidence that such an adaptive tumor-suppressor mechanism — somatic repression of telomerase activity — has evolved with body mass in animals. Other tumor-suppressor mechanisms such as more efficient DNA repair may also evolve with body mass (Promislow, 1994
Coevolution of telomerase activity and body mass. Increase in body mass leads to increased cancer risk. To counteract this risk large species evolve additional tumor-suppressor mechanisms, such as repression of somatic telomerase activity.
Previous studies did not allow differentiation between the contributions of body mass and lifespan on evolution of somatic repression of telomerase activity, as the existing data on telomerase activity among mammals was derived from a limited number of species, primarily mice and rats that are small and short lived, and on large long-lived species such as humans and cattle. The importance of considering body mass when analyzing evolution of longevity has been previously emphasized. By controlling for the effect of body mass and phylogeny Promislow (1994)
concluded that DNA repair rates correlate with body mass rather than with longevity in mammals. Similarly, Lorenzini et al. (2005)
found that replicative capacity of fibroblasts positively correlates with body mass. The latter result may appear inconsistent with our finding of negative correlation between telomerase activity and body mass. This apparent contradiction can be easily explained by considering the definition of replicative capacity employed by Lorenzini et al. (2005)
. In that study fibroblasts were cultured in 20% oxygen and out of 59 cultures 21 spontaneously immortalized after a period of reduced growth rate. It has been demonstrated by Parrinello et al. (2003)
that senescence of mouse fibroblasts in 20% oxygen is caused by oxidative stress rather than by telomere shortening, while at physiological oxygen concentration mouse fibroblasts are immortal. Since half of the species used by Lorenzini et al
. were rodents, the observed positive correlation between replicative capacity and body mass is likely to reflect the positive correlation between oxidative stress resistance and body mass, but has no direct relation to telomerase activity. The largest species (cattle, gorilla, and human) examined by Lorenzini et al
. display telomere-mediated senescence, and do not express telomerase activity in the soma (Thomas et al., 2000
; Steinert et al., 2002
), which is in agreement with our finding of negative correlation between body mass and telomerase activity.
In contrast to telomerase activity, we did not find correlations between telomere length and body mass or lifespan. The majority of rodents in this study had telomeres longer than 30 kb. Notable exceptions are beaver and capybara that also had the lowest telomerase activity. Short telomeres in beaver and capybara suggest the presence of replicative senescence in these two species. It is likely that the species with high telomerase activity do not have replicative senescence, thus telomere length in these species does not affect cancer rates, and therefore does not coevolve with body mass or lifespan. Previous studies have shown telomere length to be highly variable even within closely related inbred mouse strains, and found no correlation of telomere length with lifespan (Hemann & Greider, 2000
). The importance of telomerase activity level but not telomere length in the species with high somatic telomerase activity can be explained as follows. Multiple studies suggest that telomerase has oncogenic and growth promoting functions independent of its role in telomere maintenance (reviewed in Chang & DePinho, 2002
; Gorbunova & Seluanov, 2003
). Thus it is possible that in the species at the high end of telomerase activity spectrum the level of telomerase activity affects cancer rates independently of telomere length.
We found that several long-lived rodent species such as grey squirrel and naked mole rat have high telomerase activity in somatic cells, and are therefore unlikely to use replicative senescence as an anticancer mechanism. The presence of telomerase activity in somatic cells may provide benefits such as better wound healing and stronger immune response. It is important to note, however, that the regulation of telomerase activity is one of many anticancer adaptations. Thus small long-lived rodents may maintain high telomerase activity in the soma and rely on other tumor suppressor mechanisms. Since human cancer is believed to originate from stem cells (Reya et al., 2001
), and stem cells have telomerase activity, understanding anticancer mechanisms employed by long-lived rodents with high telomerase activity may provide valuable information to fight cancer in humans.
Our inability to detect an effect of lifespan on the evolution of telomerase activity is puzzling, because any increase in lifetime cell divisions should increase the opportunity for tumorigenic somatic mutation. The lack of correlation between lifespan and a tumor-suppressor mechanism may suggest that large body mass presents greater risk for cancer development than long lifespan. Lifetime cancer risk is thought to depend on the number of cell divisions that occur during development and the number of cell divisions needed for maintenance during lifespan. Thus, a speculative explanation for why body mass but not lifespan coevolves with tumor-suppressor mechanisms could be that initial events leading to cancer occur during growth and development. It is important to note, however, that tumor-suppressor mechanisms other than telomerase repression may coevolve with lifespan. Future studies aimed at understanding why telomerase coevolves with body mass but not lifespan may shed new light on the mechanisms of cancer development and evolution of longevity.