Why might long telomeres be selected for in laboratory mice?

Why might long telomeres be selected for in laboratory mice?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

In a recent episode of The Portal, Eric Weinstein sits down with his brother Bret Weinstein to discuss Bret's Reserve-Capacity Hypothesis. It's an incredible story of scientific discovery and and academic injustice.

One prediction of the hypothesis was that laboratory mice, known at the time to have abnormally long telomeres for their lifespan, were not a representative sample of the wild mouse population because the environment the lab mice were being bred in was selecting for longer telomeres. This had serious implications for drug safety testing.

My surface level understanding as a non-biologist from listening to the Podcast goes like this. Please point out any misunderstandings.

Antagonistic Pleiotropy Hypothesis for Senescence

George Williams predicted that pleiotropic (multi-purpose) genes which are advantageous early in life but detrimental later in life will be selected for in nature since organisms often die or reproduce long before the disadvantageous effects manifest themselves. This is an evolutionary theory for aging, which argues that trade-offs exist between early life and later life fitness. This effect was observed, but no gene had been known to be the cause. This was the "missing pleiotropy".

The Hayflick Limit

Telomeres are repeating, non-coding regions of DNA at the end of each chromosome which aid in cell division. Each division, the telomeres shorten. Telomeres are counters of sorts, which provide a hard limit (the Hayflick limit) to the number of divisions a cell can undergo. Any cell that started dividing uncontrollably (cancer) would eventually reach its Hayflick limit and die. Aging (senescence) is the manifestation of cells in an organism reaching their Hayflick limit.

The Reserve-Capacity Hypothesis

This is Bret Weinstein's work. The way I understand it is that Weinstein pointed out that the missing plyotropy was not a gene at all, but telomeres. Weinstein observed that there existed a trade-off which was reflected in telomere length. Long telomeres allow cells to repair themselves many more times, slowing the effects of aging, but simultaneously leave the organism vulnerable to cancer, as cells which begin to divide uncontrollably will never reach their Hayflick limit. The length of telomeres is the result of a trade-off between dying of cancer, or dying of old age.

The JAX Lab Scandal

One critical prediction of the Reserve-Capacity Hypothesis was that mice (which were at the time thought to have abnormally long telomeres for their lifespan) don't naturally have long telomeres. Instead, it was specifically the mice coming from the Jackson (JAX) Laboratory (which supplied a significant percentage of North America with mice) which had abnormally long telomeres. Since JAX Lab had economic incentive to breed as many mice as possible as quickly as possible, they were artificially introducing selective pressures on the mice. Namely, the pressure of breeding the mice early in their lives. This meant that any late-life disadvantages, such as susceptibility to cancer as a result of longer telomeres, were not playing an important role since the mice were bred well-before they were likely to develop cancer. Since telomeres are a non-coding region of the chromosome, variation can happen very rapidly (no need for the traditional timescales of Darwinian selection). Over time, long telomeres became dominant in the mice population, and so research being done on mice purchased from JAX lab concluded that mice had very long telomeres.

The primary gap in my understanding is the following. Why was it that longer telomeres were selected for? It is unclear to me how discounting the late-life disadvantages of long telomeres would result in long telomeres becoming dominant in population. If anything, telomere length should become irrelevant. If so, we'd see a wide distribution of telomere length. That is, unless longer telomeres provide benefits to organisms early in life. Is that the case? If not, how can we explain long telomeres becoming not just common, but dominant?

Their arguments for increased telomere length in lab mice are:

  1. Rate of tumour formation is dependent on the number of cells within an organism. Additionally, tumours take time to form. Therefore, small animals that only need to reproduce early in life, like lab mice, are less prone to tumours and there is then less selective pressure favouring telomere shortening.
  2. Senescence has a negative effect on reproductive success, even early in life, and there is therefore selective pressure favouring increased telomere length that reduces senescence.
  3. While the lack of selection on senescence due to the short window for reproduction could have the opposite effect of reducing telomere length, point 2 is the dominant factor.

To be frank, this all sounds extremely speculative to me and the supporting evidence they did provide was not entirely convincing (though I am far from an expert in the field). Keep in mind that it's a hypothesis only.

In case you don't have access to their article, I quoted the relevant section below:

The unusual telomere system of lab mice may be an unintended consequence of captive breeding. Retirement of breeders after 8 months eliminates selection on late-life effects. Tumor-forming mutations take time to occur, tumors take time to become lethal, and the likelihood of tumor initiation is presumably a function of the number of cells in the body, so in small bodied animals like mice, tumors may be rare and inflict minimal cost in the first eight months of life, even absent a telomeric fail-safe. Further, selection for sustained high reproductive output (beginning early and maintained for 8 months) should strongly favor a reduction in senescent effects occurring in that window. Selection acting to eliminate senescent effects and increase early reproductive output may tend to elongate telomeres. Because of the inextricable connection between tumor suppression and somatic maintenance, telomere elongation should dramatically increase the risk of eventual tumor formation, but any effects manifesting after the breeding cut-off will be selectively irrelevant. By our model, selection for early high rates of reproduction in the absence of selection for longevity or tumor suppression should produce long telomeres and a strong propensity for eventual tumor incidence. Despite diminished senescence, we expect these mice to have reduced maximum longevity compared to wild conspecifics. At all ages, lab mice (with elongated telomeres) should be more likely to die of tumors than wild mice. These mice should also be unusually resilient to somatic damage and show few signs of aging other than tumor formation.

The hypothesis that an 8 month breeding cut-off should select for non-senescent, tumor prone mice seems paradoxical. One might expect the elimination of selection on late life effects to accelerate senescence, not retard it. But in lab mice, selection for high, sustained rates of breeding appears to be the dominant factor. The tumor fail-safe has effectively been turned off, condemning these animals to form tumors, but leaving an early-life window of reproduction within which there is minimal senescent decline.

Age-associated telomere shortening in mouse oocytes

Oocytes may undergo two types of aging. The first is induced by exposure to an aged ovarian microenvironment before being ovulated, known as ‘reproductive or maternal aging’, and the second by either a prolonged stay in the oviduct before fertilization or in vitro aging prior to insemination, known as ‘postovulatory aging’. However, the molecular mechanisms underlying these aging processes remain to be elucidated. As telomere shortening in cultured somatic cells triggers replicative senescence, telomere shortening in oocytes during reproductive and postovulatory aging may predict developmental competence. This study aimed to ascertain the mechanisms underlying altered telomere biology in mouse oocytes during reproductive and postovulatory aging.


We studied Tert expression patterns, telomerase activity, cytosolic reactive oxygen species (ROS) production, and telomere length in fresh oocytes from young versus reproductively-aged female mice retrieved from oviducts at 14 h post-human chorionic gonadotropin (hCG), in vivo or in vitro postovulatory-aged mouse oocytes at 23 h post-hCG. Oocytes were collected from super-ovulated C57BL/6 J mice of 6–8 weeks or 42–48 weeks of age. mRNA and protein expressions of the Tert gene were quantified using real-time quantitative reverse transcriptase polymerase chain reaction (Q-PCR) and immunochemistry. Telomerase activity was measured by a telomeric repeat amplification protocol assay, while telomere length was measured by Q-PCR and quantitative fluorescence in situ hybridization analyses.


The abundance of Tert expression in oocytes significantly decreased during reproductive and postovulatory aging. Immunofluorescent staining clearly demonstrated an altered pattern and intensity of TERT protein expression in oocytes during reproductive aging. Furthermore, relative telomerase activity (RTA) in oocytes from reproductively-aged females was significantly lower than that in oocytes from young females. In contrast, RTA in postovulatory-aged oocytes was similar to that in fresh oocytes. Oocytes from reproductively-aged females and postovulatory-aged oocytes showed higher ROS levels than oocytes from young females. Relative telomere length (RTL) was remarkably shorter in oocytes from reproductively-aged females compared to oocytes from young females. However, postovulatory aging had no significant effect on RTL of oocytes.


Long-term adverse effects of low telomerase activity and increased ROS exposure are likely associated with telomere shortening in oocytes from reproductively-aged female mice.

1. The Hayflick limit

Almost exactly 50 years ago, Leonard Hayflick and his colleague Paul Moorhead discovered that cultured normal human cells have a limited capacity to divide, after which they stop growing, become enlarged, engaging a new pathway in what has been termed replicative senescence [1]. In 1961 this was totally unexpected since the research community firmly believed that cells explanted into cell culture were immortal. To support the idea that normal human cells had a limited number of divisions, Hayflick and Moorhead cultured separate populations of male and female human fibroblasts simultaneously. One was derived from a male that had been passaged in culture longer than the one derived from a female. The older male cells and younger female cells were plated at equal densities and subculture as required. When the “older” unmixed male-derived cell population stopped dividing, they investigated the mixed population and discovered that only female cells were present [1]. Besides providing additional evidence that cells have a limited replicative lifespan, this experiment demonstrated that the older cells had a molecular counting mechanism regardless of being surrounded by younger cells. Ultimately, these experiments demonstrated that a counting mechanism was somehow programmed into each cell and once this biological clock (as opposed to a chronological clock) had expired, the cell would stop dividing [2]. One additional observation that Hayflick reported is that cryogenically preserved cells remembered the number of times that they had divided at the time they were frozen [3,4]. Today, this withdrawal from the cell cycle after a certain number of cellular divisions (replicative senescence) is known to be triggered as a result of shortened telomeres [5]. Studies on replicative senescence have begun to provide valuable information towards our understanding of certain aspects of tissue and organismal aging and, additionally, have created new opportunities in the area of regenerative medicine for aging tissues and telomeropathies (genetic diseases due to premature telomere shortening). Equally important, cancer cells have evolved the ability to overcome senescence [6,7] by using mechanisms capable of maintaining telomere lengths (such as expressing telomerase), which enables cancer cells to divide indefinitely [7], a biomarker of almost all advanced human cancers ( Fig. 1 ).

Certain male reproductive cells and embryonic stem cells retain full or almost full telomere length due to expression of telomerase activity. Pluripotent stem cells have regulated telomerase activity and thus they lose telomeres throughout life but at a reduced rate. Most somatic cell do not express telomerase activity and thus lose telomere length with each division at a faster rate until the cells uncap a few of their telomeres and undergo a growth arrest called replicative senescence. In the absence of cell cycle checkpoints (e.g. p53/pRB pathway), cells bypass senescence until they reach crisis. In crisis telomeres are so short that chromosome end fusions occur and there is increased genomic instability (probably due to chromosomal, breakage, fusion, bridge cycles). A rare cell that escapes crisis almost universally does so by reactivating telomerase and this cell can now become a cancer cell with limitless potential to divide. Almost all cancer cells have short telomeres and thus inhibitors of telomerase should drive such cancer cells into apoptotic cell death.


Although a number of previous studies measured telomere length in different species (30 ⇓ ⇓ ⇓ ⇓ –35), few of them determined the telomere shortening rates (4, 6 ⇓ ⇓ ⇓ –10, 15, 16). In this regard, some studies found a correlation between telomere shortening rates and species’ life spans, including previous work from our group in mice and humans (1, 4, 6 ⇓ ⇓ ⇓ –10) however, these studies did not compare side-by-side telomere shortening rates in phylogenetically distant species by using a single technique to measure telomeres.

In our current study, telomere length and the rate of telomere shortening from multiple species with very different life spans, including birds and mammals, was acquired in the same laboratory by using the sensitive HT Q-FISH technique which allows to determine absolute telomere length values in units of base pairs as well as individual telomere signals. A limitation of the current study is, however, the few available individuals for some species.

The results shown here indicate that the telomere shortening rate of a species can be used to predict the life span of that species, at least with the current dataset (Fig. 3E). We observed that mean telomere length at birth does not correlate with species life span since many short-lived species had very long telomeres, and long-lived species had very short telomeres. Future studies warrant determination of telomere shortening rate in species such as the naked-mole rat or the bat, which do not match their predicted life span well according to their body size (26, 36).

Finally, the fact that the rate of telomere shortening can be used to predict life span suggests that the cellular effects induced by short telomeres, such as cellular senescence, may be the critical factor determining species longevity. In this regard, some studies correlate DNA repair ability to species longevity (37 ⇓ –39). In particular, the ability to repair UV-induced damage positively correlates with life span in different species, including primates (37, 38). Also, DNA repair rates are higher in longer living rodent species compared with rodent species with a shorter life span (39). It is interesting to note that short telomeres induce DNA damage, and in turn certain types of DNA damage, such as UV irradiation or oxidative stress, can also lead to telomere shortening (40 ⇓ –42).

Telomeres, telomerase, cellular aging, and immortality

Because the lost sequences of the telomeres at each replication cycle can be resynthesized by telomerase, the enzyme has essential functions for cell immortalization ( Fig. 2 ). More than 20 years ago, Olovnikov showed that the loss of telomere sequences because of the end replication problem might explicate a possible role in regulating cellular lifespan ( 16). Telomere shortening was ascribed a function, i.e., to act like a mitotic clock that counts the number of cell divisions and limits endless cell proliferation by signaling entry into senescence ( 16)( 17). According to this telomere theory, a small amount of telomeric DNA is lost with each round of cell division. When the telomere length is reduced to a critical point, a signal is given to stop further cell division, the hallmark of cellular senescence. Replicating somatic cells in vitro lose 30–200 bp of telomeric sequences per population doubling, or ∼105–50 bp per year in vivo ( 18).

Telomere length and stability in adult somatic cells differ from those in fetal and germline cells. In the somatic tissues of an individual, the telomeres are noticeably shorter than in sperm cells, and lengths decrease with increasing age of the individual. On the other hand, telomeres in sperm maintain their length independently of increasing individual age ( 19).

In cultured cells, the loss of telomeric DNA depends on the number of cell divisions, and Allsopp et al. have noted that telomere length is a useful predictor of the residual proliferative capacity of cells ( 19). In this context, the question arises as to whether one could achieve unlimited replication capacity and immortality in somatic cells if telomere length could be maintained. Counter et al. ( 20) transfected normal human embryonic kidney cells with Simian Virus 40 tumor antigen (SV 40 T), forming a tumor virus protein that extended the lifespan of cultured cells. The transfected cells divided and entered a point of crisis, in which most of the cells died only some cells became immortal. During the period of cell division, the telomeres shortened continually and no telomerase activity could be detected. Those cells that survived the crisis point and became immortal had reactivated telomerase and stabilized their telomeres. This means that even somatic cells can gain the ability of endless replication if telomere length is maintained and (or) the enzyme telomerase is activated.

Cells in culture are thought to stop dividing because of activation of an antiproliferative mechanism termed “mortality stage 1” (M1). The stimulus for the induction of M1 may be DNA-damage signals from the altered expression of subtelomeric regulatory genes or from a critical shortened telomere. P53 and the retinoblastoma gene product pRb are involved in the execution of M1. One hypothesis for the induction of M1 postulates the following: (a) a single chromosome denuded of telomeric repeats produces a DNA-damage signal, which (b) induces p53 and p21 (c) p21 inhibits the cyclin-dependent kinases, which then (d) are prevented from phosphorylating pRb (e) the presence of unphosphorylated pRb coupled with other actions of p53 and p21 results in the M1 arrest ( 21). If these cell cycle regulators are mutated or blocked, the cells continue to divide and thus the telomeres continue to shorten. Cells divide until a second independent block in proliferation is reached, termed “mortality stage 2” (M2). The M2 mechanism is probably induced when so few telomere repeats remain that the unprotected chromosomal ends block further proliferation. The M2 block might be overcome in some cells by reactivation of telomerase, the repair of chromosome ends, the stabilization of telomere length, and the generation of an immortal cell clone (( 21), and Fig. 3 ).

The findings concerning telomerase activity in various human tissues are in accord with the differences in the telomere length described above: The enzyme is detectable in germline cells, but not in most postnatal somatic tissues ( Table 1 ).

Telomerase-independent mechanisms may also exist that stabilize telomere length. Murnane et al. ( 22) and Bryan et al. ( 23) reported telomere elongation in immortal human cells without detectable telomerase activity. So far, it is not clear whether telomere stabilization in these cells is achieved by the above-mentioned recombination or transposition events, or whether transformed cells may develop alternative mechanisms to circumvent any possible telomerase inhibition.

The present knowledge about the role of telomeres and telomerase in cellular senescence carries scientists a step forward toward understanding the phenomenon of human aging—a process accompanied by an accumulation of various cytogenetic changes and an increasing deficiency in DNA repair mechanisms. Lindsey et al. ( 24) found that in patients with syndromes of accelerated aging [progeria (i.e., Hutchinson–Gilford syndrome) and Werner syndrome], the mean telomere lengths in cell cultures were considerably shorter than in normal individuals. These premature aging syndromes are characterized in progeria by growth retardation and accelerated degenerative changes of the cutaneous, musculoskeletal, and cardiovascular systems in young patients ( 25), and in Werner syndrome, for which recently the a candidate gene has been identified ( 26), by an early-onset and accelerated rate of development of major geriatric disorders such as atherosclerosis, diabetes mellitus, osteoporosis, and various neoplasms ( 27). Recently, Kruk et al. ( 28) demonstrated that repair of DNA damage in telomeric regions decreases with age. Possibly this deficiency in telomeric repair is correlated with an age-related increase in genetic instability.

Why Do Medical Researchers Use Mice?

From formulating new cancer drugs to testing dietary supplements, mice and rats play a critical role in developing new medical wonders. In fact, 95 percent of all lab animals are mice and rats, according to the Foundation for Biomedical Research (FBR).

Scientists and researchers rely on mice and rats for several reasons. One is convenience: rodents are small, easily housed and maintained, and adapt well to new surroundings. They also reproduce quickly and have a short lifespan of two to three years, so several generations of mice can be observed in a relatively short period of time.

Mice and rats are also relatively inexpensive and can be bought in large quantities from commercial producers that breed rodents specifically for research. The rodents are also generally mild-tempered and docile, making them easy for researchers to handle, although some types of mice and rats can be more difficult to restrain than others. [Why Do Mice Poop So Much?]

Most of the mice and rats used in medical trials are inbred so that, other than sex differences, they are almost identical genetically. This helps make the results of medical trials more uniform, according to the National Human Genome Research Institute. As a minimum requirement, mice used in experiments must be of the same purebred species.

Another reason rodents are used as models in medical testing is that their genetic, biological and behavior characteristics closely resemble those of humans, and many symptoms of human conditions can be replicated in mice and rats. "Rats and mice are mammals that share many processes with humans and are appropriate for use to answer many research questions," said Jenny Haliski, a representative for the National Institutes of Health (NIH) Office of Laboratory Animal Welfare.

Over the last two decades, those similarities have become even stronger. Scientists can now breed genetically-altered mice called "transgenic mice" that carry genes that are similar to those that cause human diseases. Likewise, select genes can be turned off or made inactive, creating "knockout mice," which can be used to evaluate the effects of cancer-causing chemicals (carcinogens) and assess drug safety, according to the FBR.

Rodents also make efficient research animals because their anatomy, physiology and genetics are well-understood by researchers, making it easier to tell what changes in the mice's behaviors or characteristics are caused by.

Some rodents, called SCID (severe combined immune deficiency) mice, are naturally born without immune systems and can therefore serve as models for normal and malignant human tissue research , according to the FBR.

Some examples of human disorders and diseases for which mice and rats are used as models include:

Mice are also used in behavioral, sensory, aging, nutrition and genetic studies, as well as testing anti-craving medication that could potentially end drug addiction .

"Using animals in research is critical to scientific understanding of biomedical systems leading to useful drugs, therapies and cures," Haliski told Life's Little Mysteries.

Laboratory Life

Imagine you’re a wild fruit fly, of the species Drosophila melanogaster. You’re happily feasting on some yeast that’s growing on rotting fruit when, whoomf, you get sucked into a bottle and taken to a laboratory. From now on, this is your home.

Life in a bottle — or cage — is different from life in the wild. In nature, for example, fruit flies reproduce throughout their adult lives. Often, in the laboratory, they do not: flies grown in bottles may only be allowed to reproduce for the first five or six days after emerging from the pupa. (Wild flies can live for more than 80 days.) In nature, flies choose their mates. Often, in the laboratory, they do not: they are often assigned to one, and that one may be a close relative. On top of that, the food is different infectious diseases are rare predators are absent.

In short, the pressures of daily life have been transformed — and traits that were an advantage Out There may no longer be so Inside. Similarly, traits that would have killed you in the wild may help you get along inside a bottle.

If, for example, older flies are never allowed to reproduce, the ability to lay eggs later in life becomes irrelevant, so there’s nothing to prevent the appearance of mutations that interfere with that ability. Indeed, if those mutations increase early fertility, they may even be favored: the most fecund young flies are likely to leave the most descendants.

Thus, the switch from the wild to the laboratory immediately alters the evolutionary trajectory of a population — and sure enough, within a few generations, laboratory-bred life-forms become noticeably different from their wild cousins.

Exactly what happens depends on how the organisms are kept — different rearing methods create different evolutionary forces. But in general, laboratory Drosophila melanogaster evolve shorter lifespans than wild flies they become less able to cope with stresses like starvation or desiccation and their pattern of fertility changes. As you’d expect, females reared in bottles evolve to be hugely fecund as young flies but much less so when they are older.

Also as you’d expect, laboratory evolution is not unique to Drosophila melanogaster. In the wasp Nasonia vitripennis, females descended from a long line of laboratory wasps evolve to be more prone to promiscuous sexual behavior than wild wasps. In the Mediterranean fruit fly, Ceratitis capitata, laboratory-reared females evolve to be less fussy about who they mate with, and male sexiness changes. Wild female medflies don’t find laboratory-reared males as attractive as they find wild males. Mexican fruit flies, Anastrepha ludens, have the same problem: laboratory males have evolved in such a way that they are less popular with wild females.

Mice show a host of changes, too. Compared to their wild relations, laboratory mice are typically bigger, more docile, reach sexual maturity earlier and die younger. Some of these changes can appear quickly: one study found that the ability to reproduce later in life declined within 10 generations of the mice being bred in the laboratory.

Intriguingly, laboratory mice also have longer telomeres than wild mice. (Telomeres are the segments of DNA at the ends of chromosomes they are thought to play a role in aging and cancer.) Since no one is deliberately breeding mice for extra-long telomeres, this must arise as some consequence of laboratory life. But what?

That’s not clear. One possibility is that it’s due to inbreeding — for lab mice are often highly inbred. Consistent with this, one study of white-footed mice, Peromyscus leucopus, found that, when animals were forced to inbreed, telomeres lengthened substantially in fewer than 30 generations — although why this should be so is entirely mysterious.

All of which is fascinating. But does it matter?

That depends. For some scientific problems, the fact that laboratory life-forms evolve substantial differences from their wild relatives is irrelevant. For others, however, it matters a lot.

Let me give you two examples. Adaptation to the laboratory — or to captivity more generally — can make it much more difficult for organisms to thrive if they are later released to the wild. This has important implications for the conservation of endangered animals and for the control of pests. Captive breeding programs have been important tools for re-establishing wild populations of species such as the California condor but not all programs are successful. Genetic changes in captivity may be one reason. Similarly, many pest control programs depend on the “sterile male technique,” whereby males are bred in the laboratory, sterilized, then released into nature to mate with wild females. For this to work, the wild females must find the laboratory males attractive. Changes in mating behavior like the ones I mentioned earlier can, therefore, quickly reduce the effectiveness of the approach.

A second area where laboratory evolution can be a serious problem is in the study of subjects like the evolution of aging, and the diseases associated with it. For example, the study of laboratory populations may give a misleading impression of how easy it is to extend lifespans: since laboratory organisms tend to have unnaturally short lifespans, discovering ways to make them live longer may not be especially informative. We may simply be reversing the unnatural shortening that we created in the first place, a view supported by the fact that selection to increase lifespan in laboratory populations often simply restores it to levels seen in the wild.

Such realizations have led an increasing number of scientists to argue that long-established laboratory populations are “suspect starting material” for understanding aging, and that comparisons with wild populations “support the pessimistic interpretation that laboratory-adapted stocks of rodents may be particularly inappropriate for the analysis of the genetic and physiological factors that regulate aging in mammals.”

For some subjects, it’s better to go wild.

For an interesting overview of evolution in the laboratory, see Artamonova, V. S. and Makhrov, A. A. 2006. “Unintentional genetic processes in artificially maintained populations: proving the leading role of selection in evolution.” Russian Journal of Genetics 42: 234-246.

A large number of studies have found evidence of evolution to laboratory conditions. For Drosophila melanogaster, I drew, in part, on Sgrò, C. M. and Partridge, L. 2000. “Evolutionary responses of the life history of wild-caught Drosophila melanogaster to two standard methods of laboratory culture.” American Naturalist 156: 341-353. This paper shows how differences in laboratory rearing methods can affect evolutionary trajectories, and also shows how truncating the reproductive life of adult flies rapidly leads to flies evolving to reproduce more earlier compared to wild flies, laboratory flies had shorter lives. For laboratory populations being “suspect starting material” for aging studies, see page 351 of this paper.

For the lifespan of wild flies compared to laboratory flies, see Linnen, C., Tatar, M. and Promislow, D. 2001. “Cultural artifacts: a comparison of senescence in natural, laboratory-adapted and artificially selected lines of Drosophila melanogaster.” Evolutionary Ecology Research 3: 877-888. These authors show that wild flies live longer than standard laboratory flies, and that lines of flies that have been bred specifically to have long lifespans do not live longer than wild flies.

For laboratory rearing leading to loss of resistance to desiccation and starvation, see Hoffmann, A. A. et al. 2001. “Rapid loss of stress resistance in Drosophila melanogaster under adaptation to laboratory culture.” Evolution 55: 436-438. For promiscuous laboratory wasps, see Burton-Chellew, M. N. et al. 2007. “Laboratory evolution of polyandry in the parasitoid wasp Nasonia vitripennis.” Animal Behaviour 74: 1147-1154.

For evolution in the mating behavior of laboratory populations of medflies, see Rodriguero, M. S. et al. 2002. “Sexual selection on multivariate phenotype in wild and mass-reared Ceratitis capitata (Diptera: Tephritidae).” Heredity 89: 480-487. For the same phenomenon in Mexican fruit flies, see Rull, J., Brunel, O. and Mendez, M. E. 2005. “Mass rearing history negatively affects mating success of male Anastrepha ludens (Diptera: Tephritidae) reared for sterile insect technique programs.” Journal of Economic Entomology 98: 1510-1516. These papers also discuss the problems that laboratory evolution pose for pest control. An additional analysis of this is provided by Hendrichs, J. et al. 2002. “Medfly areawide sterile insect technique programmes for prevention, suppression, or eradication: the importance of mating behavior studies.” Florida Entomologist 85: 1-13.

For an overview of evolutionary changes in laboratory mice, see Miller, R. A. et al. 2002. “Longer life spans and delayed maturation in wild-derived mice.” Experimental Biology and Medicine 227: 500-508. This paper shows that wild-caught mice live much longer than most laboratory mice, and reach sexual maturity later. These authors are also responsible for the “pessimistic interpretation” quotation see page 507.

For the study showing that the ability to reproduce later in life can decline within 10 generations of laboratory residence, see Flurkey, K. et al. 2007. “PohnB6F1: a cross of wild and domestic mice that is a new model of extended female reproductive life span.” Journal of Gerontology, Biological Sciences 62A: 1187-1198.

For laboratory mice having weirdly long telomeres, see Hemann, M. T. and Greider, C. W. 2000. “Wild-derived inbred mouse strains have short telomeres.” Nucleic Acids Research 28: 4474-4478. For inbreeding producing long telomeres in white-footed mice, see Manning, E. L. et al. 2002. “Influences of inbreeding and genetics on telomere length in mice.” Mammalian Genome 13: 234-238.

For the possibility that evolution in captivity may pose a potential problem for captive breeding programs, see Woodworth, L. M. et al. 2002. “Rapid genetic deterioration in captive populations: causes and consequences.” Conservation Genetics 3: 277-288 and Williams, S. E. and Hoffman, E. A. 2009. “Minimizing genetic adaptation in captive breeding programs: a review.” Biological Conservation 142: 2388-2400.

The problem of laboratory mice in aging research has been discussed extensively by some authors. In addition to the papers I have already mentioned, see Harper, J. M. 2008. “Wild-derived mouse stocks: an underappreciated tool for aging research.” Age 30: 135-145 and Miller, R. A. et al. 1999. “Exotic mice as models for aging research: polemic and prospectus.” Neurobiology of Aging 20: 217-231.

Many thanks to Bret Weinstein for drawing my attention to the fact of long telomeres in laboratory mice, and for discussions about some of the implications this may have. Many thanks also to Nicholas Judson and Jonathan Swire for insights, comments and suggestions.


3.1 Replication of analyses in Gomes et al., ( 2011 )

Presented in Table 1 are the results reported by Gomes et al., ( 2011 ) and our re-analysis—both using two phylogenetic multiple regression models. Consistent with Gomes et al., ( 2011 ), we found that both mass and TL significantly predicted lifespan. However, the remaining results are all substantially discordant. While Gomes et al., ( 2011 ) found no association between TA and lifespan, we observed a significantly positive association. In predicting body mass, Gomes et al., ( 2011 ) found that TA was a significant predictor, but TL was not. We found the opposite: TA did not predict mass, but longer TL predicted decreased body mass. Using current estimates for mass and lifespan from the AnAge database (Table S11) did not qualitatively alter the findings described above (Table S2).

Gomes et al. Our re-analysis
(p-Value) β (p-Value)
Response: log10(mass)
Telomerase activity (.0082)** ** p < .01.
−0.0014 (.890)
log10(telomere length) (.71) −1.8435 (.007)** ** p < .01.
Response: log10(lifespan)
Telomerase activity (.34) 0.0039 (.029)* * p < .05.
log10(telomere length) (.0032)** ** p < .01.
−0.4118 (.002)** ** p < .01.
log10(mass) (<.0001)*** *** p < .001.
0.1401 (<.001)*** *** p < .001.
  • a Gomes et al. do not explicitly describe their model parameters, but these are our best guess at a reconstruction. More complete results from our re-analyses are given in Table S1.
  • * p < .05.
  • ** p < .01.
  • *** p < .001.

Gomes et al., ( 2011 ) reported a series of bivariate plots (their figure 4) for key variables in their analyses along with p-values. These p-values are identical to those presented in their figure 2 (our Table 1), strongly suggesting that these significance values were from phylogenetic multiple regressions, not phylogenetic bivariate regressions. A casual reader is likely to interpret these p-values as bivariate associations. Here, we present phylogenetic linear bivariate regression analyses of TL, body mass, maximum lifespan and TA to examine the sensitivity of the multiple regression results and more clearly define these bivariate associations.

The bivariate results (Figure 1) are qualitatively consistent with our multiple regression results, except lifespan did not show any association with TA (Figure 1d). Based on these bivariate results, a 1% increase in body mass predicts a 0.08% decrease in TL (Figure 1a), while a 1% increase in lifespan predicts a 0.43% reduction in TL (Figure 1b). The co-evolution of TL and body mass is illustrated in the phylogeny in Figure 2.

In an effort to further reconstruct the reasons for our discrepant results with those of Gomes et al. and explore the sensitivity of our analyses to modelling strategies, we experimented with alternative transformations of the TA measurements (see Supporting Information). Briefly, we find that in multiple regression models, as TA is first log-transformed (Table S3) and then binary transformed (Table S4), TL becomes less and less predictive of either mass or lifespan and the correlation between TL and TA increases (Table S5). When the bivariate association of mass predicting binary TA (0/1) is modelled in a phylogenetic logistic regression (see Supporting Information), complete telomerase repression, which was found in 39 out of 57 species, is significantly associated with higher body mass with an inflection point at 0.8 kg (Figure S1). Among the 18 species with nonzero TA (see Table S11), there were no significant associations between (untransformed) TA and mass or lifespan (Table S6). Inspection of the model residuals and quantile plots of the associations between TA, mass and lifespan (Figure 1c,d) indicated significant outliers violating the multivariate normal distribution assumptions of the models. To identify these outliers within the phylogenetic regression, we ran a post hoc leave-one-out deletion analysis that detects influential species on parameter estimates while accounting for phylogeny using the function “influ_phylm” in the r package “sensiPhy” (Paterno et al., 2018 ). After removing the most influential species (see Supporting Information), we reran the main analyses, which revealed a significant negative bivariate association between TA and mass (53 species, βlog(mass) = −0.5190 ± 0.2056 SE, p = .0148) that was, however, not robust to controlling for the effect of TL (Table S7).

3.2 Associations between neoplasia incidence and telomere biology

We found that the prevalence of neoplasia is a good predictor of malignancy prevalence (R 2 = .90, β = 0.7735 ± 0.0489 SE, p < .001, Figure S2, see also Figure S3) across 29 mammal species included in Boddy et al., ( 2020 ). Because neoplasia prevalence was available for more species, analyses proceeded using this variable.

As expected, neoplasia rate was positively associated with TL (Figure 3a) and binary TA (Figure 3d, marginally significant trend) across the 22 mammal species shown in Table S11. Thus, a 1% increase in TL predicts an absolute increase in neoplasia incidence of 0.21% (Figure 3a). Neoplasia rate was negatively associated with lifespan (Figure 3b). A 1% increase in maximum lifespan predicts an absolute decrease in neoplasia incidence of 0.14% (Figure 3b). Species with some TA (1) are predicted to show an absolute increase in neoplasia incidence of 20.6% (Figure 3d). There was no association between neoplasia rate and mass (Figure 3c). There was only a marginally significant positive association between TL and neoplasia rate, when accounting for binary TA in a phylogenetic multiple regression (Table 2), probably because TA and TL are correlated and some of the variation in TL is explained by TA (Table S5).

Response: Neoplasia rate Estimate SE t-Value p-Value
Intercept −31.0668 20.4645 −1.5181 .1455
log10(telomere length) 36.2935 17.6865 2.0520 .0542* * p < .10.
Binary telomerase activity (1) 14.0411 9.8713 1.4221 .1711
λ = 0.000 (95% CI: 0.000, 0.963), adjusted R 2 = .3071

Neoplasia rates could be biased towards higher estimates in domesticated species due to the easier observation of large numbers of individuals (Boddy et al., 2020 Ewald & Swain Ewald, 2015 Nagy et al., 2007 ), which could be associated with increased lifespan and senescence in captivity (Tidière et al., 2016 ). We therefore tested, in a post hoc analysis, whether the effect of TL on neoplasia rate was robust to controlling for potential domestication effects on neoplasia rates (Table S8) or if the effect of TL varied with domestication status (Table S9). TL remained significantly positively associated with neoplasia rate when controlling for domestication status and there was no effect of domestication status on neoplasia rate (Table S8). The effect of TL on neoplasia rate did not vary with domestication status (Table S9).

3.3 Effect of domestication on telomere length

The arithmetic mean (±SD) TL was 24.8 ± 14.3 kb for domesticated species and 20.3 ± 13.7 kb for nondomesticated species in the Gomes et al., ( 2011 ) data set. Controlling for phylogenetic nonindependence, there was a marginally significant positive trend, towards longer telomeres in domesticated species (bivariate regression βdomestication =0.1467 ± 0.0774 SE, p = .0633). Telomeres were significantly longer in domesticated species (multiple regression βdomestication =0.1416 ± 0.0615 SE, p = .0254, Table 3) when controlling for TA, body mass and lifespan. Thus, domestication is predicted to increase TL by 38.5% compared to nondomesticated species. Excluding the nine domesticated species from the replication analyses presented in Table 1 did not qualitatively alter our findings (Table S10).

Response: log10(TL) Estimate SE t-Value p-Value
Intercept 1.7168 0.1960 8.7594 <.001*** *** p < .001.
Domestication (1) 0.1416 0.0615 2.3020 .0254* * p < .05.
Telomerase activity 0.0060 0.0015 3.9197 .0003*** *** p < .001.
log10(lifespan) −0.3611 0.1203 −3.0023 .0041** ** p < .01.
log10(mass) −0.0099 0.0298 −0.3339 .7398
λ = 1.000 (95% CI: 0.000, 1.000), adjusted R 2 = .4121

3.4 Sensitivity analyses

The statistically significant bivariate associations between TL and mass (Figure 1a) and TL and lifespan (Figure 1b) were generally robust to sample size effects (mean change in β with 49% of the species included was 12%–17%) and the associations remained significant within most of the reduced data sets (85%–100%, Figures S4 and S5). However, the phylogenetic signal, λ, was unstable for smaller sample sizes and varied considerably between the bounds 0 and 1. The nonsignificant associations between TA and mass (Figure 1c) or lifespan (Figure 1d) were relatively unstable for smaller sample sizes with respect to changes in β (as expected for small β-values), but λ remained high and the associations remained nonsignificant under most simulations (83%–86%, Figures S6 and S7). The phylogenetic multiple regressions of lifespan and mass predicted by TL and TA were generally not robust to sample size effects (Figures S8 and S9), which may be attributed to the inclusion of TA. However, the phylogenetic signal remained high across simulations.

The bivariate associations between neoplasia rate and TL, lifespan or TA (22 species, Figure 3a,b,d) were relatively robust to sample size effects, but the significance of the associations decreased considerably as sample size was reduced, with ≥32% corresponding to a simulated sample size of just 15 species (Figures S10–S12). The association between neoplasia rate and mass remained nonsignificant under most simulations (Figure S13). The multiple regression of neoplasia rate predicted by TL and binary TA was relatively robust to sample size effects (Figure S14). The phylogenetic signal was not robust in the sensitivity analyses of neoplasia rate (Figures S10–S14).

Harvard scientists reverse the ageing process in mice – now for humans

Scientists claim to be a step closer to reversing the ageing process after rejuvenating worn out organs in elderly mice. The experimental treatment developed by researchers at the Dana-Farber Cancer Institute, Harvard Medical School, turned weak and feeble old mice into healthy animals by regenerating their aged bodies.

The surprise recovery of the animals has raised hopes among scientists that it may be possible to achieve a similar feat in humans – or at least to slow down the ageing process.

An anti-ageing therapy could have a dramatic impact on public health by reducing the burden of age-related health problems, such as dementia, stroke and heart disease, and prolonging the quality of life for an increasingly aged population.

"What we saw in these animals was not a slowing down or stabilisation of the ageing process. We saw a dramatic reversal – and that was unexpected," said Ronald DePinho, who led the study, which was published in the journal Nature.

"This could lead to strategies that enhance the regenerative potential of organs as individuals age and so increase their quality of life. Whether it serves to increase longevity is a question we are not yet in a position to answer."

The ageing process is poorly understood, but scientists know it is caused by many factors. Highly reactive particles called free radicals are made naturally in the body and cause damage to cells, while smoking, ultraviolet light and other environmental factors contribute to ageing.

The Harvard group focused on a process called telomere shortening. Most cells in the body contain 23 pairs of chromosomes, which carry our DNA. At the ends of each chromosome is a protective cap called a telomere. Each time a cell divides, the telomeres are snipped shorter, until eventually they stop working and the cell dies or goes into a suspended state called "senescence". The process is behind much of the wear and tear associated with ageing.

At Harvard, they bred genetically manipulated mice that lacked an enzyme called telomerase that stops telomeres getting shorter. Without the enzyme, the mice aged prematurely and suffered ailments, including a poor sense of smell, smaller brain size, infertility and damaged intestines and spleens. But when DePinho gave the mice injections to reactivate the enzyme, it repaired the damaged tissues and reversed the signs of ageing.

"These were severely aged animals, but after a month of treatment they showed a substantial restoration, including the growth of new neurons in their brains," said DePinho.

Repeating the trick in humans will be more difficult. Mice make telomerase throughout their lives, but the enzyme is switched off in adult humans, an evolutionary compromise that stops cells growing out of control and turning into cancer. Raising levels of telomerase in people might slow the ageing process, but it makes the risk of cancer soar.

DePinho said the treatment might be safe in humans if it were given periodically and only to younger people who do not have tiny clumps of cancer cells already living, unnoticed, in their bodies.

David Kipling, who studies ageing at Cardiff University, said: "The goal for human tissue 'rejuvenation' would be to remove senescent cells, or else compensate for the deleterious effects they have on tissues and organs. Although this is a fascinating study, it must be remembered that mice are not little men, particularly with regard to their telomeres, and it remains unclear whether a similar telomerase reactivation in adult humans would lead to the removal of senescent cells."

Lynne Cox, a biochemist at Oxford University, said the study was "extremely important" and "provides proof of principle that short-term treatment to restore telomerase in adults already showing age-related tissue degeneration can rejuvenate aged tissues and restore physiological function."

DePinho said none of Harvard's mice developed cancer after the treatment. The team is now investigating whether it extends the lifespan of mice or enables them to live healthier lives into old age.

Tom Kirkwood, director of the Institute for Ageing and Health at Newcastle University, said: "The key question is what might this mean for human therapies against age-related diseases? While there is some evidence that telomere erosion contributes to age-associated human pathology, it is surely not the only, or even dominant, cause, as it appears to be in mice engineered to lack telomerase. Furthermore, there is the ever-present anxiety that telomerase reactivation is a hallmark of most human cancers."

Although we attempted to credit and cite all relevant publications, we apologize if any important contributions may have been missed. We thank members of the Nandakumar lab for providing critical feedback at various stages of the writing process. During the writing of this review, the Nandakumar lab was supported by National Institutes of Health grants R01GM120094 and R01AG050509 and the American Cancer Society Research Scholar grant RSG-17-037-01-DMC (to J. N.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Author contributions

S. G. and J. N. developed the framework of the review. S. G. generated the full draft of the article text and figures with input from J. N.