It is difficult to use radiation to make a cell stop metabolizing. The amount of radiation necessary to actually disrupt ordinary metabolic processes (glycolysis, electron transport, ion mobility, etc.) is two or more orders of magnitude higher than the amount required to prevent the cell from producing viable daughter cells. Therefore the most commonly measured endpoint for quantitating the effect of radiation on cells is clonogenic or reproductive death, i.e. the inability of the cell to reproduce itself with fidelity.
To characterize a cell as capable of reproduction with fidelity we require that it be capable of producing viable offspring for several generations. The practical definition given by Alpen is that cell must be able to produce 50 offspring; this corresponds to just under six cell divisions, since 26 = 64, fairly close to 50.
Why is reproductive death so significant? Certainly we have some interest in an organism's ability to reproduce itself, so the effects on production of viable offspring are important in that regard. But even within the lifespan of a single individual organism, the ability of cells to divide correctly is significant. Many cells in an organism turn over, i.e. are replaced by newly-matured cells, a few weeks after they themselves mature, so the inability to replace those cells when the time comes will have a direct effect on the health of the individual. Also, note that we are concerned with the transmission of correct genetic information, not just any genetic information. Thus if a cell is capable of division but its phenotype has been altered due to mutations in its genes, then the cell will not perform its assigned role. Cancer is an instance of an improper or incomplete maturation of a group of cells; cancer cells are in general capable of dividing, but their genetic makeup and their phenotype differs from those of non-malignant cells.
In discussing the effects of radiation on cells, we will look at mechanisms by which cells fail to reproduce. We will also look at mathematical models for survival or its complement, cell death. In the early days of radiation research (up through the 1950's, say) these two areas of research were seen as distinct: biochemists and cell biologists studied how cellular reproduction was interfered with, and biophysicists looked at dose-response curves, viz.
| *
| *
| *
| *
ln( | *
surviv-| *
ing | *
frac- | *
tion) | *
| *
| *
| *
| *
| *
|_____________________*_
Dose
in an effort to find mathematically plausible models that fit the shape of curves like this one. In the last three decades the modelers and the biologists have begun to talk to one another. The modelers recognized that their mathematical models were likely to have more predictive power and fewer arbitrary constants in them if they could be connected to a molecular mechanism. Meanwhile, the mechanistic biologists recognized that one way to demonstrate the plaubility (although not the correctness) of proposed cellular and biochemical mechanisms is to show that the dose-response data are consistent with the mechanism.
It would also be useful to articulate why we're concerned with effects at the cellular level. It is a truism to say that biological organisms are composed of cells, so we are obliged to understand them from the cellular level if we're ever going to understand them at higher levels of organization. This is true, as far as it goes, but there is another reason. It is time-consuming, costly, and difficult to develop quantitative methods for studying dose-response relationships in whole animals. Apart from the difficulty of amassing large enough sample groups to get statistically meaningful results, animal researchers have to spend large amounts of time simply raising the animals before the experiment and analyzing their tissues after the experiment. In most places it is necessary to justify all animal experimentation in an Institutional Review Board (IRB); any experiment whose results could be obtained without use of whole animals is likely to be turned down by an IRB. It is comparatively easy to study dose-response relationships with cell cultures. Therefore, we tend to characterize the biological effects of radiation at the cellular level because cell culture is the most effective place to measure biological effects.
Bacterial cells were used for radiation studies in the 1930's and 1940's, but the applicability of bacterial studies to humans was always subject to dispute because the biochemistry of bacteria differs in some critical ways from that of higher organisms. DNA repair enzymes are different in bacteria from those in eukaryotes; different lists of metabolites are required by the cells to thrive; and the cells have different kinds of membranes around them, keeping the outside out and the inside in. The development in the early 1950's of ways of culturing mammalian cells enabled radiation researchers to look at radiation insult to mammalian tissue without having to deal with whole animals. Most of the modeling described in this unit on cell survival is based on experiments in mammalian cell culture.
The assumptions in target theory are:
Using these assumptions we can derive a variety of models for radiation's effects on cells. Alpen goes through these models in substantial detail, and in class we will discuss his derivations. There are several errors in his derivations that I will correct at the beginning of class.
The next level of sophistication beyond models of the kind that are based on Lee's assumptions are those that explicitly take into account the role of double-strand breaks (DSB's) in the target molecule, DNA. DSB's can be produced in two different ways: by the action of a single charged particle that disrupts covalent bonds in both DNA strands at the same time; and by the action of two different particles that disrupt the two chains separately, such that the disruptions are sufficiently close together in both time and space that the chains can separate and / or DNA repair enzymes cannot effectively repair the damage before replication. These models will be discussed in class as well.