TOR: The ancient link between cancer and ageing?

Could ageing and cancer be unified by TOR via the ‘compromise’ struck between soma and germline at the dawn of multicellular life?


original image

A highlight of this issue is the paper by Daniel Ackerman and David Gems 1, which discusses the role of fat metabolism in ageing. But there is something else that is interesting about that paper: it notes the role of germline cells in promoting ageing of the soma of Caenorhabditis elegans – specifically via TOR signalling. Indeed, one of the more recent and remarkable findings in gerontology research is that antagonising TOR signalling – for example with rapamycin (TOR = Target Of Rapamycin) – can prolong life 2 in yeast, nematodes and flies 3 (and reviewed in 4).

As I noted in an editorial comment recently 5, evolutionarily speaking, there is expected to be a ‘tension’ between the germline and soma of an organism. My synthesis drew together articles that directly and indirectly support the concept of cancer as an ancient proliferative survival program that overrides the good of the whole and the evolutionarily defined role of the germline as the organism's seat of proliferative survival. There might well be another twist in this tale: TOR signalling not only contributes to ageing in C. elegans, but its mammalian counterpart, mTOR signalling, is one of the principal signalling pathways activated in cancer. Could the observation of TOR-signalling-mediated organismal ageing be indirect support for the hypothesis of cancer as the enactment of a program for survival of the individual cell at any cost?

Some readers will notice a potential contradiction here: mTOR signalling has long been identified as one of the drivers of human cancer, and inhibitors of mTOR signalling are a major focus of research efforts (rapamycin, itself, is a potentially promising anti-tumour agent). Hence, rather than enacting mTOR signalling as part of an ageing program, a variety of human cancer cells are exquisitely dependent on mTOR signalling for proliferation. If, evolutionarily speaking, the soma were – at some deep level – still ‘in conflict’ with the germline, one might instead expect it to have some antagonistic mechanism towards germline-induced TOR-signalling.

But perhaps the very signal to age acted, over evolutionary time, as a trigger to escape the ageing: could mammalian cells indeed have a program for ‘reinterpreting’ a historically inhibitory signal as a growth (and pro-tumour) signal – a kind of switch that takes advantage of a different facet of TOR signalling developed over its evolutionary history? Pleiotropy of TOR is already recognised. Indeed, TOR has been referred to as ‘the ultimate example of an antagonistic pleiotropy gene’ 6 (and see also references therein): TOR drives developmental programs early in life, which, later in an organism's life, run on as an ‘aimless quasi-program of aging and aging-related diseases’ 6. But I would argue that, for the germline, and the response of a ‘renegade’ somatic cell, that program is not so ‘aimless’ after all…

Seen in a larger context, pleiotropy between developmental and ageing-related effects of a pathway may not be particularly exceptional: for example, the variable effects of TGF-beta (Transforming Growth Factor beta), which have been dubbed the ‘TGF-beta paradox’ 7 (and reviewed in 8): TGF-beta normally inhibits the development of cancer via signalling that restrains cell proliferation and forms a microenviroment that impedes cell motility, invasion, and metastasis: a manifestation of its crucial role in development to halt proliferation. However it is increasingly recognised that TGF-beta can also play a positive role in tumourigenesis, in particular the very processes of tissue invasion and metastasis: a manifestation of its role in the developmentally relevant epithelial to mesenchymal transition (EMT). A similar pleiotropy is manifested by the oncoprotein Myc 9, 10 (and also Ras, incidentally) to stimulate growth and senescence. Could these, too, be evolutionarily ‘reinterpreted’ ancient programmatic functions, rather than consequences of aging to which selection is indifferent?

The observation of germline-induced TOR signalling might be an example of the evolutionarily developed tension between germline and soma that could lead to cancer as an ‘escape route’ from the sometimes delicate relationship between the two lineages. In C. elegans, and other short-lived organisms, however, the soma might simply not exist for long enough to manifest resistance to the germline – i.e. cancer. But there does seem to be a program behind certain aspects of ageing, and the behaviour of the germline towards the soma could be an example. One of the reasons why there is such a debate about whether ageing is programmed, or whether it is simply the inevitable and cumulative effect of stochastic damage, is that it is probably both. Ageing must be seen from two evolutionary perspectives: (i) selection at the level of the individual, which works in the interest of fitness up till peak reproductive age, but cares little for the soma after that point; (ii) survival and renewal of life per se in a changing environment. Because of the way in which life works – and this might be a physical principle that is manifest by life – organisms respond plastically within generations, but (largely) via genetic changes between generations.

Each organism has a different reproductive rate and reproductive timespan, dependent on the niche it occupies. In a given species, those variables have become optimised for survival in a changing environment across evolution; otherwise the species would no longer exist. The genetic constitution of a given species has a statistical ‘shelf-life’ because the variability of environment with which its offspring have to cope increases probabilistically with time since the progenitor. That probably contributes to defining the timespan over which it reproduces before ageing and dying. Could it be, therefore, that the germline – whose function it is, after all, to produce the next generation – somehow ‘knows’ this? Might it then respond by enacting a program of ageing on the soma, evolutionarily ‘optimising’ its lifespan? That would also respect resources in the interest of the germline's progeny – progeny that, through divorce with the soma, recombine and mix their genes, increasing the likelihood of the procession of life in a changing environment.

The soma might very well not always ‘succumb’ to this ‘additional ageing’, programmatically imposed by the germline: i.e. cancer might be seen as an ‘escape route’ from a doomed host, enacted by an evolutionarily ancient proliferation program inherited from unicellular organisms 11. If so, one might expect one feature of cancer in higher animals to be a proliferative response to the ‘ageing signalling’ that the soma received earlier in evolutionary time from the germline in lower animals.

One of the possible evolutionary responses of a somatic cell to such TOR signalling induced by the germline would be to switch its life-history program to that of proliferation and the cancer phenotype; TOR signalling would become a proliferation signal to a cell that had ‘sensed’ its fate: to die with the rest of its ageing, disposable soma. Could the message to age, sent from germline to soma, and mediated by TOR signalling in lower animals, be the ancient origin of the role of mTOR signalling in human cancers today? Further, assuming a program behind germline TOR signalling to promote ageing, could this be additional support for the theory of cancer as an ancient program for survival of the single cell?

original image

Andrew Moore