Models for Mechanisms of Protection against Damaging Effects of Radiation 159
niques be obtained experimentally. It should hereby be stressed that the
purpose of any theoretical studies in natural science, be it in physics,
biology, cosmology or other fields, is to obtain new knowledge about
Nature, in the ideal case by narrowing down the possibilities to just those
which are actually realized in Nature. Consequently, only such models fulfill their purpose which allow both the deduction of quantitative predictions about all experimentally measurable quantities and the working out
of explicit recipes for quantitative determination of the numerical values of
the parameters of the model from actually measured quantities.
Now, in studies of biological damage due to ionizing radiations, as in
most other fields of research, models may be constructed in various ways
depending on which basic assumptions we start from (chap. 1 in [5]). As
discussed above, it is with present experimental techniques beyond our
powers to check these assumptions in direct experimentation. The best
procedure is, therefore, to investigate how the phenomenon might have
been formed from each set of assumptions. If one set of results looks more
like the real phenomenon than another set we shall probably have discovered not only how the phenomenon is actually formed, but also which
basic assumptions are correct.
At the Wisconsin symposium in 1963 I presented a review of quantitative models for the analysis of the mechanisms of carcinogenesis [1, 6]
satisfying the criteria just discussed. I shall at this symposium present a
short review of a quantitative model for the analysis of the mechanisms of
protection against the damaging effects of ionizing radiation. For details I
beg to refer you to the original paper [3] and to a short summary of it [2].
In quantitative studies of the protective, and also the sensitizing, effects
of added substances on the inactivation of enzymes in dilute aqueous
solution by ionizing radiation, a most fundamental problem is met withnamely, how to define a quantitative measure of radiosensitivity. If the
data fit a straight line when the remaining enzyme activity, E, is plotted
semilogarithmically versus the radiation dose, D, the quantity 1/D37 is
usually chosen, where D37 denotes that dose for which the enzyme activity,
E, is reduced to 1 Ie" 37 % of the initial activity, Eo. However, in several
experiments nonexponential dose-inactivation curves are met with [8], and
the quantity 1 /D37 then loses any deeper meaning as a quantitative measure
of radiosensitivity. In these cases the only sensible procedure is to construct
a hypothetical model of the reaction mechanisms behind the inactivation in
question, from this model to deduce a mathematical formula for the doseinactivation curves, and then to fit this formula to the experimental data
in question by numerical or graphical adjustment of the parameters of
the model. The radiosensitivity of the given enzyme under given experimental conditions is then expressed by these numerical values of the
parameters.
niques be obtained experimentally. It should hereby be stressed that the
purpose of any theoretical studies in natural science, be it in physics,
biology, cosmology or other fields, is to obtain new knowledge about
Nature, in the ideal case by narrowing down the possibilities to just those
which are actually realized in Nature. Consequently, only such models fulfill their purpose which allow both the deduction of quantitative predictions about all experimentally measurable quantities and the working out
of explicit recipes for quantitative determination of the numerical values of
the parameters of the model from actually measured quantities.
Now, in studies of biological damage due to ionizing radiations, as in
most other fields of research, models may be constructed in various ways
depending on which basic assumptions we start from (chap. 1 in [5]). As
discussed above, it is with present experimental techniques beyond our
powers to check these assumptions in direct experimentation. The best
procedure is, therefore, to investigate how the phenomenon might have
been formed from each set of assumptions. If one set of results looks more
like the real phenomenon than another set we shall probably have discovered not only how the phenomenon is actually formed, but also which
basic assumptions are correct.
At the Wisconsin symposium in 1963 I presented a review of quantitative models for the analysis of the mechanisms of carcinogenesis [1, 6]
satisfying the criteria just discussed. I shall at this symposium present a
short review of a quantitative model for the analysis of the mechanisms of
protection against the damaging effects of ionizing radiation. For details I
beg to refer you to the original paper [3] and to a short summary of it [2].
In quantitative studies of the protective, and also the sensitizing, effects
of added substances on the inactivation of enzymes in dilute aqueous
solution by ionizing radiation, a most fundamental problem is met withnamely, how to define a quantitative measure of radiosensitivity. If the
data fit a straight line when the remaining enzyme activity, E, is plotted
semilogarithmically versus the radiation dose, D, the quantity 1/D37 is
usually chosen, where D37 denotes that dose for which the enzyme activity,
E, is reduced to 1 Ie" 37 % of the initial activity, Eo. However, in several
experiments nonexponential dose-inactivation curves are met with [8], and
the quantity 1 /D37 then loses any deeper meaning as a quantitative measure
of radiosensitivity. In these cases the only sensible procedure is to construct
a hypothetical model of the reaction mechanisms behind the inactivation in
question, from this model to deduce a mathematical formula for the doseinactivation curves, and then to fit this formula to the experimental data
in question by numerical or graphical adjustment of the parameters of
the model. The radiosensitivity of the given enzyme under given experimental conditions is then expressed by these numerical values of the
parameters.