What Hazards?

One thing we can all agree on is that radiation is bad for you, right? Well.... First we have to be careful to define what we mean by ``radiation.'' Your fireplace radiates in the infrared (heat) and visible (light) parts of the electromagnetic (EM) spectrum; these forms of radiation are certainly beneficial as long as they don't get out of control. On the other hand, visible light in the form of a high-power laser can inflict damage, as can excessive heat or even microwave EM radiation. On the shorter-wavelength side of the EM spectrum, ultraviolet light can cause sunburn to the skin, while X-rays penetrate deeper and can do the same sort of microscopic damage as the still shorter-wavelength gamma () rays emitted by <sup>60</sup>Co (cobalt) radioisotopes. Can we make general statements about all of these? Perhaps, ``A little is good, but a lot is bad!'' Sorry, nothing so simple. It is certainly true that we cannot maintain health without both heat and light, and a certain amount of ``near ultraviolet'' may be required for natural vitamin D production in the skin, but we probably have no biological need for microwave or radio frequency radiation; and all EM radiation from ``far ultraviolet'' upward in frequency (downward in wavelength) is exclusively and unambiguously bad for the individual. [Whether or not genetic mutations are beneficial for the human race as a whole is a difficult question both scientifically and ethically; I will avoid trying to answer it.]

Why the big qualitative difference? What do ultraviolet, X-rays and -rays do that visible and infrared light don't? At last, a question to which there is a simple answer! They cause ionization of atoms and molecules inside cells, leaving behind a variety of free radicals -- types of molecules that quickly react chemically with other nearby molecules. If the free radicals react with the DNA molecules in which are encoded all the instructions to our cells for how to act and how to reproduce, some of these instructions can get scrambled.

Surprisingly, this does not always happen. The simplest detectable damage to a DNA molecule is a ``single-strand break,'' in which one of the strands of the double helix is broken by a chemical reaction with a radical. It is a testimony to the robustness of DNA that it is usually able to repair its own single-strand breaks in a few hours!

Whether this is because of multiple redundancy or context programming I do not know, but it sure is an impressive feat.

I should add an extra caveat at this point: what I have said about single- and double-strand breaks and healing times is what I recall from sitting on the PhD committee of a student working on pion radiotherapy about ten years ago. I don't imagine it has been substantially revised since then, but I am not absolutely sure. If you want a more reliable witness I will be glad to direct you to local experts.

What sorts of bad things are liable to happen when a DNA molecule sustains irreparable damage, scrambling some part of the instruction manual for the operation of the cell it inhabits?

• Cell Reproductive Death [most common] -- The cell containing the defective DNA may be unable to reproduce itself, so that although it may be able to function normally for its remaining natural lifetime, when it dies a natural death it will not have a new cell to replace it. Whether this causes a problem or not depends upon whether many other nearby cells have the same malady (one by itself will never be missed!) and upon the natural lifetime of that type of cell -- which ranges from a few days for hair follicles, skin and mucous membrane cells to ``forever'' for brain cells. Obviously, the loss of reproductive capacity is meaningless for a cell that never reproduces!

• Genetic Mutation [most subtle] -- If the cell in question happens to be a gamete destined for fusion with a member of the opposite sex, the resulting individual will have some scrambled instructions in the construction manual and will probably not grow up normally. In almost every case this will be fatal to the foetus, and in almost all the remaining cases it will be detrimental to the survival of the individual, although such mutations have presumably played a rôle in evolution to date. Note however that it is strictly impossible for any individual's genetic makeup to be retroactively altered by radiation (like the Hulk or Spiderman or any number of cheap sci-fi horrors), as this would require the same accidental scrambling to take place independently in every DNA strand in the victim's body!

  For men, there are two types of genetic damage: the sperm cells themselves have an active lifetime of only a few days, after which a new generation takes over; but the sperm-producing cells are never replaced and so can never repair damage to themselves. The latter applies also to women: the female gametes (eggs) are all produced early in life and, once damaged, cannot be repaired.

  If the altered cell is ``just any old cell'' then usually the change is harmless -- either the cell merely fails to do its part in the body until it dies or else the affected part of the DNA is irrelevant to the functioning of that cell in the first place -- but occasionally the change is related to cell division itself, and then there can be real trouble.

• Cancer [most unpleasant] -- Sometimes (very rarely) a damaged DNA molecule instructs a cell to mobilize all its resources and the resources of all its neighbours to reproduce as many copies of itself as possible. The offspring preserve the mandate, and a chain reaction takes place that ``crashes the system.'' This runaway reproductive zeal of a misguided cell is what we know as CANCER, and it is the worst hazard of radiation exposure. As far as anyone knows, any exposure to ionizing radiation increases one's chances of developing cancer, and so we can unambiguously say that ionizing radiation is bad for you.

Before we go on, it is interesting to note that all of the most potent therapies for treating cancer involve either ionizing radiation or chemical reactions that cause similar DNA damage; the strategy for these ``interventions'' is always to cause such overwhelming DNA damage to the cancer cells that every single cancer cell suffers ``cell reproductive death'' as described above. Although there are various techniques for making the cancer cells more susceptible to the radiation or harsh chemicals than normal cells, there are inevitably many casualties among the latter. It is not unusual, for instance, to kill off (in the sense of ``reproductive death'') as many as 90% of the normal cells in the tissues surrounding a tumour, relying upon the fantastic healing capacity of normal tissue to bounce back from this insult. Remember, the idea is to kill 100% (!) of the cancer cells.

It provides an important perspective to realize that the radiation used to kill the cancer may deliver a ``dose'' to healthy tissues that is more than 10,000 times the maximum legal limit for environmental radiation exposure, and yet the increased likelihood of developing another cancer from the radiation therapy is regarded as a negligible risk relative to allowing the existing cancer to progress unchecked. Whether or not oncologists have optimized their treatment strategies is another charged issue which I will avoid, but it is clear that a large radiation dose does not necessarily ``give you cancer'' immediately; rather it increases your chances of developing cancer in the long run. By how much? And over how long a run? These are the quantitative statistical questions that must be answered if one is to develop a rational scheme for evaluating radiation hazards.