A few months ago Scientific American published an article about new theories of cancer. In the article the authors mentioned that the development of cancer looks more and more like the evolutionary processes that have shaped organisms in the world around us for eons.
I came across a more detailed paper along that line recently. “Dynamics of Cancer Progression” was authored by Franziska Michor, Yoh Iwasa and Martin Nowak and was published this month in Nature Reviews. Fortunately, it’s also available as a PDF online without having a subscription to the journal.
Cancer is a genetic disease1. Although environmental and other non-genetic factors have roles in many stages of tumorigenesis, it is widely accepted that cancer arises because of mutations in cancer-susceptibility genes. These genes belong to one of three classes: gatekeepers, caretakers and landscapers. Gatekeepers directly regulate growth and differentiation pathways of the cell and comprise oncogenes and tumour-suppressor genes (TSGs). Caretakers, by contrast, promote tumorigenesis indirectly. They function in maintaining the genomic integrity of the cell. Mutation of caretakers can lead to genetic instability, and the cell rapidly accumulates changes in other genes that directly control cell birth and death. Landscaper defects do not directly affect cellular growth, but generate an abnormal stromal environment that contributes to the neoplastic transformation of cells.
While it’s a familiar idea that cancer arises from a series of genetic mutations—one is not enough—in a lineage of cells, the model in this paper incorporates the idea that the progression of changes involves the “selection” of cells that have become advantaged for reproduction in comparison to other cells in the same tissue, and that this “somatic selection”—not unlike natural selection among organisms in a population—is an important part of tumor formation. The other interesting thing about this model is that the variables that make a difference in whether or not a mutation in a cell’s DNA will lead to cell proliferation and to a tumor are quantifiable and can be related to each other mathematically. The correctness of the model is, therefore, testable in fairly straightforward way.
The first category of genes that is of concern are the “oncogenes”—genes that regulate the growth and differentiation of cells. An unaltered gene that promotes cell growth performs a normal and necessary function in the cell under certain circumstance. An analogy might be that these genes accelerate growth at certain times much as we use the gas pedal to make a car go faster under certain conditions. Trouble arises, however, when a mutation causes the gene to continue stimulating growth and to not respond to internal signals to stop. That’s akin to the problem you’d have if your car accelerator stuck when you wanted to back off.
Conversely, another category of genes has the specific normal function of suppressing cell growth. These tumor suppressor genes (TSGs) work to turn off the growth processes of cells so they do not over-proliferate and result in abnormal tissue growth. Mutations in these genes can cause them to fail to exercise their growth-cessation function. To continue the analogy, they are like the brakes on a car, and there’s trouble in a cell—or in a car—when the braking function fails. Cells normally maintain an internal balance between growth and retardation of growth (just as a driver both accelerates and brakes when driving on a freeway). The failure of the balance between these two functions can contribute to tumors.
Finally, a third important contributor to cell abnormality is chromosomal instability (CIN). DNA is aggregated into large chunks we know as chromosomes. Each chromosome contains millions of DNA molecules and hundreds of genes. Some of those genes normally produce products that enable chromosomes to maintain structure during their duplication and cell division. Mutations in those genes can produce unstable chromosomes, and when the chromosomes become unstable the rate of mutations in genes increases significantly—a situation usually called “genetic instability.” CIN is a common characteristic of cancerous cells. Whole chromosomes may be lost or gained or large pieces of chromosomes may be lost or gained. The exact role of CIN in carcinogenesis is, however, controversial.
But there’s more to the story than just what happens within a cell and whether or not its genes are intact and maintain the balance of growth start and stop. The mutations in genes mentioned above can have a variety of effects on the cell. A genetic mutation may be advantageous relative to other cells or it may be a big disadvantage. As with mutations in organisms, most cell mutations are disadvantageous, and a mutant cell may self-destruct (apoptosis) or it may fail to thrive among its neighboring cells. A genetically modified cell only begins to be a problem when it has characteristics that enable it to reproduce more rapidly or to die more slowly than normal cells in the tissue. Those advantages may enable the cell to dominate cell reproduction in a “compartment” of cells, and it may eventually create a population of descendant cells that displace the normal cells. This is “somatic evolution” as distinguished from organism population evolution.
There are a number of variables that affect whether or not a specific mutation will become a step toward on the path to some sort of tumor. The “fitness” of the change (how advantageous it is relative to the function of “normal cells”); how probable it is that a population of cells can “fixate” (become dominant in a tissue sub-section or “compartment”); whether the change is in a potential oncogene, TSG, or CIN gene, and what the sequence of genetic alterations is (oncogene before CIN or TSG or some other combination of events). It’s a long road from an initial mutation and a tumor and from an initial tumor and a life-endangering malignancy. The authors of “Dynamic Cancer Progression” bring these factors together and propose several sequences of events that model events in tumor formation and mathematical formulas that may quantify relationships among factors.
Personal impressions: To me a lot of the suggestions in this paper and others are remarkable. One way to look at organisms—including humans—is that they are huge colonies of living entities: cells. An organism is a vast ecosystem of sorts in which differentiated populations of cells constitute tissues and a complex system of signals between tissues keeps the whole organism in balance and able to function as a unified living entity. Problems start when internal genetic dysfunction or dysfunction is induced extrinsically by a carcinogen kicks off a population explosion of abnormal cells.
I think it would come as a great surprise to people to learn that, with each cell division, there is the possibility that some form of abnormal DNA reproduction will take place. The chance that one DNA base pair (one of the A, T, C or Gs we’ve all heard about) will make a copying mistake is .0000000001 per cell division. Obviously the molecular mechanisms of DNA replication are extremely reliable, but they’re not perfect. After all, the human DNA strand has three billion base pair molecules in it. Even high reliability in such a long series is bound to lead to some errors. Cells have mechanisms for identifying and fixing many DNA abnormalities, and that lowers the net error rate a lot. Still, with trillions of cells reproducing over decades of life, some cumulative degree of genetic abnormality is almost inevitable. Probably even more startling is the idea that, if a mutation in our DNA occurs, a process of evolutionary selection begins among cells that determines whether or not the event—in you or me—is a one-time thing or the first of a series of cycles which leads to a hearty renegade population with the power to endanger our lives.
I’m not sure Darwin would have guessed this.