Daily Archives: April 1, 2008


A rambling introduction to chemical carcinogenesis

This is an archive from the old blog, originally written in 2008.

Regular readers may have noticed that I get rather annoyed by the casual use of the word “chemical” to mean “synthetic chemical”, and the use of the naturalistic fallacy (natural good, chemical bad) that is associated with this phenomenon. It is really quite inexcusable. I’ve known plenty of intelligent and educated people who are aware of why they are mistakes, and yet continue to make them. Such usage is particularly associated with concerns about cancer, so I thought it may be time for a basic (or intermediate?) concepts post on chemical carcinogenesis.

In healthy cells, the processes of cell growth, division, and death — the cell cycle — are carefully regulated. Carcinogenesis is the loss of that regulation, and carcinogens are the agents which cause that loss. Such a loss transforms the cell into a cancer cell, which is then able to proliferate unchecked. Though the terms “carcinogenesis” and “carcinogen” most commonly refer to the events and agents which cause that initial transformation, they are sometimes also used to refer to those events and agents which promote tumour growth later in the disease. In reality, those two categories heavily overlap, and in fact, it takes several of these events before a cell line can really be called “cancer”. Physical and biological agents can be carcinogens, but we’ll ignore their mechanisms for now (after all, it’s just too easy to point out that the sun and viruses are “natural”), and concentrate on the chemicals.

Some chemical carcinogens are able to directly influence the cell cycle. For example, Bisphenol A (which I mentioned a few weeks ago) worries some people because it is able to mimic the effect that estrogen has on regulating the cell cycle — it binds to and activates estrogen receptors, whose job it is to promote cell division. And then there’s tobacco (not to be confused with tobacco smoke, which is a cocktail of chemicals), which has all sorts of effects on cell signaling, which in turn feed back on gene expression and the cell cycle.

Most carcinogens, however, are mutagens: they damage the DNA. They punch holes in useful genes, switch on genes that should not be switched on, and subtly change what genes do. In this respect, cancer is a genetic disease. The mutations are, of course, acting largely at random, though it is easy to forget this (and a similar problem of overlooking the random nature of mutation exists in popular views of evolutionary biology). Billions of mutations have occurred within you, but the overwhelming majority will have had no effect whatsoever. They may have hit part of the 80+% of the genome which is junk, or had a neutral or insignificant effect on the gene that they landed in. Of those that actually hit something important, most will be dealt with by the cell’s DNA repair mechanisms, or will trigger cell suicide. Finally, to contribute to cancer, the mutagens must happen upon certain types of genes: those involved in regulating the cell cycle (you may remember the terms “oncogene” and “tumour suppressor gene” from previous posts), cell death, or the DNA repair mechanisms themselves.

Mutagens are typically divided into direct-acting carcinogens, and procarcinogens — the latter requiring modification before they become harmful. That modification comes in the form of your own metabolism. It’s not the job of your metabolism to turn procarcinogens into full (or “ultimate”) carcinogens. Rather, we have sets of enzymes which break down classes of complex molecules into less complex molecules, and one of the complex molecules in its class might just happen to be a procarcinogen. Such procarcinogens are found in everything from wood smoke to the fungus that grows on the out-of-date food in your fridge.

Pick any topic in biology and you will be able to discuss it in terms of evolution. I have discussed cancer in terms of mutation and selection events occurring within the tumour before (here, for example), but the chemical carcinogenesis story can also tell us about the evolution of our inherent defences against cancer, and about a couple of important evolutionary principles. At the same time, we can learn about whether artificial really does mean “bad”. First we have to flip the necker cube and look at carcinogens — and other chemicals — from a different angle. Rather than talking about those chemicals which are, in some way, poisonous to us, and those which are safe, we can recognise that no chemical is inherently good or bad, and call the categories “chemicals to which we are vulnerable” and “chemicals to which we are resistant”. It is only in the context of our evolved bodies that chemicals take on such properties as “carcinogenic”.

Evolution has equipped us with a number of strategies to cope with harmful chemicals. One is to find a physiological or biochemical context in which they are not harmful: if the harmful chemical is interfering with your vital molecular machine (i.e. enzymes), change the shape of the vital molecular machine. Or you could develop kidneys that are capable of filtering the chemical out, or an enzyme capable of breaking the chemical down. Another strategy is a change in behaviour: a fear or brightly coloured bugs, or a distaste for bitter plants. A third strategy is to make you have children, and then stop caring.

Humans have all of these defences and more. Alongside our DNA repair mechanisms, we have strong, stable and specific enzymes; we have a digestive system, liver and kidneys, which do their best to keep the bad guys out; and most importantly, we preciously guard a small set of genes to pass on to the next generation. The defence mechanisms that evolution has given us work wonders, but have failed to make any individuals immortal. Evolution cares little for those who have passed reproductive age: selfish genes have little to gain from maintaining a dead end. Our DNA repair mechanisms are therefore tuned to keep our bodies from cracking and crumbling just long enough to bring a new set of gene vehicles into the world.

The second evolutionary principle that the chemical carcinogenesis story highlights is that evolution has no foresight. Evolution works by trial and error, life and death, and if you put a product of evolution in a novel situation, it may find itself in trouble. Our throat and lung cells were therefore wholly unprepared for the chemicals in cigarette smoke (fortunately, evolution has equipped us with one last defence — a brain — and it looks like we won’t have to adapt to cigarette smoke the hard way). Similarly, enzymes don’t “know” not to break down procarcinogens into carcinogens, because our ancestors did not consume any of them in significant volumes. In this respect, we can consider the hypothesis that “natural” things are less likely to cause cancer: we’ve probably had longer to adapt to them.

The trouble arises when one tries to turn all of this into a set of simple rules. Carcinogenesis is not simple, and chemistry is not simple. When it comes to the safety of chemicals, the empirical approach is the best we’ve got. And that’s something that I ought to leave for another post.