Paradoxes can be intellectually challenging and illuminating.TheeponymousPetoparadoxoriginatedapproximately40 years agowhenPeto, alongwithhis colleaguesDoll andCairns, highlighted the observation that cancer risk does not appear to scale with size in the animalkingdom.1,2Theunderlying premise was thatmore cell division (tomake and sustain a larger animal) along with longer life span might be expected tocarryaproportionallygreatermutationalandmalignancyrisk. Canceroccurs throughout themulticellular animalphyla,3 and the observation made by Peto and colleagues appears to be generally correct. Cancer rates, insofar as reliable data are available, do not increasewith size,2,4 which is extraordinary given that the compoundof size (or cell numbers) and life span imply a billion-fold range of intrinsic risk.1 However, the expectation of a straightforward correlation between animal size, cell numbers, and risk of cancer is somewhat simplistic.Anaveragehumanadulthas around 1014 cells. But turnover of cells, for example, by proliferation in the adult intestinalepitheliaorbloodoccursatapproximately1011perday. Therefore, in 50 years, the number of cells generated by silent cellcycles inthehumanbodyoutnumbersthesteady-statenumberbyseveral fold.Little isknownabouttheburdensofcell turnover in very large mammals. In addition, evolution has prudentlyarranged formostproliferation tooccurnot in stemcells but in transit or amplifying cells in which potentially hazardous new mutations are filtered out by cell differentiation and death.What should reallymatter are total numbers of proliferating stem cells and progenitor cells over a lifetime. Despite these caveats, the “invention” of multicellularity around600millionyearsagowillhave incurredan intrinsic risk ofmutant “cheater” cells3 that selfishly replicate, evading normal social restraints, and disseminate between tissues as cancers.Becausecontinuedcellproliferationwasnecessary for the integrity of tissue function, resilience, and reproductive fitness, this conflict needed resolution.5 It is reasonably clear that thischallengewasmetearly inevolutionwiththeaccrualofmultiple adaptations.5 With the progressive increase in the complexity, size, and longevityofanimalsovermillionsofyears, the risk ofmutations empowering cancerous cell growthwill have escalated. So the expectation is that this would have provided selectivepressure for furtheradaptations torestraincancer risk. In that context, it is not unreasonable to assume that very large animals, such as some of the large dinosaurs and present-day elephants and whales, might have benefited from some extra cancer-suppressingmechanisms.Thecrucialassumption is that lethalcancer inancestral individualsprovidedtheselectivepressure for adaptations that currently restrain risk. In this issueof JAMA, Abegglen and colleagues6useddata from the zoo in San Diego, California, to illuminate this intriguingquestion.Theauthors firstverified fromextensivenecropsy data that variations in the incidence of cancer in captive mammals have no relationship to size. They specifically focused on elephants and found that aging elephants seem to have low cancer rates. Next, the authors turned their attention to thep53gene (TP53). The rationale is theprominent role the p53 protein has in cancer suppression, and the clinical experience of those who study families with Li Fraumeni syndrome, inwhich loss of 1 copy of the p53 gene in the germline results in substantially elevated cancer risk. TheDNA from the blood of anAfrican elephantwas investigated using cloning and sequencing, with confirmation of a previoussuggestionthatelephantshavemultiplecopiesofTP53. Of at least20copiesof theTP53gene identified, 19were intronlesspseudogenesor retrogenes.The resultwas replicatedwith DNAfromanAsianelephantbutnot in thedistantly related,but comparatively midget species hyrax (Procavia capensis). This provides a broad time window (approximately 6-60 million years ago) for the origins of elephant TP53 retrogenes. The key question thenwas: Are any of these pseudogenes producing functional protein, and, if so, what difference does it make to a cell’s response to DNA damage? The authors provideevidencethatthesepseudogenesweretranscribedintoRNA and appear to have functional protein in so far as x-irradiation results in the anticipated changes in downstream effectors of p53function,andelephantcellsaresignificantlymorelikelythan human cells to undergo apoptosis. One of theTP53 retrogenes (No. 9) was expressed at the protein level and binds toMDM2 when transfected into fibroblasts and the cells are exposed to x-irradiation. The conclusion that elephants owe their relatively cancer-free longevity to the acquisition, in an ancestor, of extra copies of TP53 seems plausible. Even if this provocative result and deduction are correct, therearesomecaveats.First, acquisitionofextracopiesofTP53 is likely to incur a significant cost or trade-off, notably in attrition of stem cells, accelerated aging, and decreased longevity.7 Howdoold elephants survive? Attribution of ancestral adaptations toparticular functionsor selectivepressurescanbesomewhat presumptuous. Stephen Jay Gould referred to this tendency inbiologists asa“just-so”explanation, aproposKipling’s “just-so” stories that included the extravagant explanation of how the elephant got its long trunk.Accrual of extra functional copiesofTP53couldhaveservedsomeotherfitnessbenefitother Related article page 1850 Opinion
Read full abstract