Hypothesis: Whale and elephant cells will exhibit significant resistance to oncogenic transformation

There is substantial evolutionary selection pressure acting on all multicellular species to prevent the formation and uncontrolled growth of tumours prior to reproductive age. Interestingly, the quantum nature of tumourigenesis (specifically the monoclonal origin of most tumours) implies differing consequences for species with large differences in body size and life-span.

Introduction

Multicellular organisms are communities of individual entities (cells) cooperating for mutual benefit. This community is vulnerable to destruction by overgrowth of subpopulations of cells. As a consequence of the potential for exponential clone outgrowth, a single tumourigenic clone can result in the death of the entire organism, regardless of the initial contribution of that clone to the total number of cells. As an initial approximation, consider a “single-hit” model of carcinogenesis, where the tumour initiation rate (Rc) is proportional to the rate of mutation (Rm):

Rc ∝ Rm [1]

While use of a single-hit model is an oversimplification, as tumourigenesis is generally considered to require multiple sequential events (1), it may be argued that multilayered defenses against cancer are a consequence of the evolutionary pressures discussed below on ancestral organisms. I will argue that the depth and complexity of these defenses will vary in a somewhat predictable manner across species. Please note that the equations derived are employed for order of magnitude calculations only; greater precision is not intended.

Modeling carcinogenesis

The majority of fixed, heritable genomic DNA mutations occur as a consequence of misrepair of DNA damage during genome replication (2). Therefore, all else being equal, we can utilize genome replication events to estimate mutation probabilities. Let Rm(N) be the total rate of mutation in N cells. To generate a clonal population of N cells from a single precursor, N-1 genome replication events are required, so where Rm(N) evaluated at N=1 is small, the chance of at least one cell exhibiting a fixed mutation in such a population can be approximated as:

Rm(N) = (N-1)Rm(1) [2]

Combining equations [1] and [2], for large N we obtain:

Rc ∝ N [3]

If carcinogenesis in organisms were to follow single-hit kinetics, the following equation could therefore be used to track vulnerability as a function of size and maturation time:

Gorganism ≈ Nc + Fm Nc VmT [4]

Where G is the number of genome replication events required for an organism to reach a point where it can successfully reproduce and replace itself (to which, under the above approximations, the risk of developing cancer during this period is proportional, as per equation [3]). Nc is the number of cells in an adult individual and, therefore, also the number of genome replication events required to generate them. Fm is the fraction of cells actively turning over and being replaced, Vm is the average number of mitotic cycles (genome replication events) per dividing cell per unit time, and T is the survival time required. The first term represents the minimum number of cells required to construct the adult organism, while the second accounts for ongoing cell turnover.

At this point, in order to simplify further calculations and avoid use of variables whose values are unknown, it is useful to define the Mouse Equivalence Unit (MEU), for which we set Gmouse = 1. To a first approximation, mouse and human cells are similar in size (A. Elia, personal communication), and while cell-size data is hard to come by for organisms such as whales and elephants, for order-of-magnitude calculations it should be sufficient to assume cell numbers are roughly proportional to body mass. While Fm is unknown, its value is between 0 and 1.

Key relative variables are thus body mass, and minimal generation time. For mice, body mass averages 25 – 40 g, and generation cycle time is roughly 10 weeks (3). For humans, the values are ~60 kg and 25 years, and for elephants ~5000 kg and 30 years (4). In other terms, one human has a mass roughly equal to 2×103 mice, and requires ~100 times as long to reproduce. Thus, in order of magnitude, Ghuman falls between 103 (Fm = 0) and 105 (Fm = 1) MEU. Similarly, Gelephant falls between 105 and 107 MEU. Generation cycle period information on the largest whales is less accessible; however the blue whale is over an additional order of magnitude heavier (well over 100,000 kg) than an elephant, suggesting that Gblue-whale is in the range of at least 106 to 108 MEU.

Now consider the consequences of the uncontrolled replication of a single tumourigenic cell prior to reproduction. In a single human, the consequences are evolutionarily disastrous: deletion from the gene pool. In contrast, the equivalent cellular population of mice contains 103 to 105 individuals, and the consequences of a single tumour are hence overall relatively insignificant. In other words, at the level of population genetics, the effects of tumourigenesis, and thus the evolutionary pressures to control it, are expected to be much more significant for humans than for mice, and even more so for larger organisms such as elephants and whales.

Discussion

From this calculation, on a per-cell basis, humans would be predicted to have evolved significantly higher resistance to oncogenic transformation than mice. This prediction is borne out by a number of common observations. Mouse cells grown in culture will spontaneously immortalize (6), while human cells require additional mutagenesis or other genetic alteration (6). Telomere length maintenance via the telomerase system, permissive of indefinite clonal population expansion (7, 8), is tightly regulated in humans (9), but much less so in mice (10). The model thus appears reasonable in light of existing data.

This model makes the testable prediction that tumourigenesis on a per-cell basis will be even more tightly restricted in larger animals such as whales and elephants. Extension of the multi-step tumourigenesis discussed above is one potential route. Experiments in which cells from such organisms are exposed to known carcinogenic stimuli (i.e. radiation, chemical mutagens), and the frequency of oncogenic transformation (i.e. immortalization, growth in soft agar) determined relative to that of cells from smaller and more rapidly maturing organisms will provide evidence to support or refute this hypothesis.

Should it prove valid, certain implications will invite further thought. Firstly, the advantages and limitations of specific model organisms in cancer research may become clearer. Secondly, the failure of some animal research to translate to human treatment options might be understandable in terms of the selection pressures that differential defense mechanisms have imposed on those tumours that progress to a stage where they threaten the life of the host. Finally, the hypothetical transformation resistance mechanisms predicted to have evolved in very large, long-lived organisms could be studied to provide insight and inspiration into novel therapeutic approaches to human cancer.

It might be hoped that the above-mentioned transformation experiments would prove relatively straightforward in a laboratory with experience in such techniques. When I started thinking about this a few years ago, I was unable to find a source of cells from whales or elephants, however more recently such cells are becoming available (National Marine Cell Line Library, Wise Lab, www.usm.maine.edu/toxicology/research/nmcl.php), or alternatively cell samples might be obtained in conjunction with research conducted at zoos (e.g., at the Toronto Zoo, www.torontozoo.com/conservation/mammals.asp?nav=5). The first person to perform these experiments and write up a meaningful data set gets a free beer.

Acknowledgments: I would like to thank David Sealey and three reviewers for their constructive criticism of this manuscript.

References

1. S.H. Moolgavkar, E.G. Luebeck, Anticancer Res. 38:302 (2003).

2. J.H. Bielas, J.A. Heddle, Proc Natl Acad Sci USA 97:11391 (2000).

3. L.M. Silver in Mouse Genetics – Concepts and Applications, (Oxford Univ. Press, New York, 1995); Mouse Genome Informatics at The Jackson Laboratory, http://www.informatics.jax.org/silver/1.3.shtml.

4. P. Armbruster, P. Fernando, P et al., Animal Conservation 2:69 (1999).

5. A. Macieiera-Coelho, B. Azzarone, Genes Chromosomes Cancer 38:302 (2003).

6. W.C. Hahn, C.M.Counter, et al., Nature 400:464 (1999).

7. A.G. Bodnar, M. Ouellette, et al., Science 279:349 (1998).

8. H. Vaziri, S. Benchimol, Curr. Biol. 8:279 (1998).

9. N.W. Kim, M.A. Piatyszek, et al., Science 266:2011 (1994).

10. R.A. Greenberg, Allsopp, R.C. et al., Oncogene 16:1723 (1998).

Share your thoughts



Leave a Reply

You must be logged in to post a comment.