Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Chapter 2 INTRODUCTION It is well known that ultraviolet radiation (W) can be harmful to plants and animals including humans. The effects of UV on living cells and organisms depend on the wavelength of the radiation. The ultraviolet portion of the electromagnetic spectrum is conventionally divided into three parts--UV-A, UV-B, and UV-C--in order of decreasing wavelength (Figure 2.1). The divisions are somewhat arbitrary, based largely on our understanding of how UV affects humans. For the purposes of this report, UV-A is the wavelength region from 320 nanometers (nary) to 400 nm (near- W); W -B. wavelengths from 290 nm to 320 nm (mid-UV); and WV-C, wavelengths from 190 nm to 290 nm (far-UV). m e known harmful effects per unit dose of the shorter wavelengths, UV-C and UV-B, are greater than those of the longer wavelengths, UV-A (Blum 1959; Harm 1980b; NRC 1975, 1976a, 1979a; Parrish et al. 1978). A familiar effect of UV on humans is sunburn (Figure 2.1). UV also affects the metabolism of, kills, and mutates cells in culture, and is carcinogenic for animals, including humans. The ozone layer provides protection from W by absorbing the most harmful wavelengths. The spectrum of solar radiation reaching the surface of the earth for the current atmospheric distribution of ozone is shown schematically in Figure 2.1. Radiation in the UV-C band is essentially completely absorbed by stratospheric ozone and does not reach the surface of the earth; even with large reductions (tens of percents) in the concentration of stratospheric ozone almost no UV-C would be trans- mitted to the earth. Most of the solar W -B also does not reach the surface of the earth. Absorption in the W -B band is a sensitive function of the amount of ozone, 37
38 1.0 in z ~ 0.6 - o J O 0.4 m > '5 0.2 UJ Or: \ UV-C I UV-B I UV-A | Visible ~ Human Sunburn (Erythema) _ ~ 1 \ .1/' \ ol I I I 260 280 300 320 340 / Current | / Solar Radiations at Earth's / | it/ 1 1 WAVELENGTH (nary) 1 .0 0.8 `,, z UJ A 0.6 I J 0.4 u' > - 0.2 LL 360 400 420 O FIGURE 2.1 Schematic plots of the W portion of the solar electromagnetic radiation currently reaching the surface of the earth and the biological sensitivity curve for human sunburn (erythema) are shown as functions of wavelength. however, and so if ozone concentrations decrease, either as a result of natural causes or as a result of human activity, the amount of W -B reaching the surface of the earth will increase and the harmful effects of W will also increase. The amount of UV-A reaching the surface of the earth is not sensitive to changes in ozone concentration. Changes in ozone abundance resulting from the release of chlorofluorocarbons, and other human activities, would take place only over a long period of time, probably decades. It is conceivable that many living creatures with relatively fast reproductive cycles could adapt biologically to a slow increase in the average intensity of W. because they would go through many generations in the time it takes for the intensity to reach some new steady state value. Humans, on the other hand, could not adapt biologically nearly as rapidly. Furthermore, if an increase in W gives rise to an increased incidence of skin cancer, the increased cancer incidence is not likely to be detected for many years after the increase in W. Thus, the continued release of chlorofluorocarbons may lead to reductions in stratospheric ozone some time in the future, and that may lead to increases in the
39 incidence of cancer in humans even farther in the future. The effects of human activities on stratospheric ozone are of concern for the long term, but the effects of current events on one rucuLe won. ~ '=~, ~= reversible. The second part of this report addresses the biological effects to be expected from changes in solar W . The uncertainties in understanding are large in spite of substantial advances in basic knowledge. These advances have not answered all of the important questions. A long-term commitment to research designed to answer the remaining critical questions is needed to facilitate predictions about the effects of enhanced W on biological systems. With new knowledge comes the possibility of reduced or increased concerns about ozone reduction, either from changes in understanding of the effects currently recognized or from previously unknown effects. Continuous surveillance of the problem by knowledgeable photo- biologists is highly desirable, not only directly but also indirectly via basic research. For example, a number of years ago the fact that visible light can ameliorate the damaging effects of UV on human cells was not suspected. Now, as a result of experiments of a hectic nature on cells in culture (Harm 1980a, Sutherland et al. 1974), this amelioration is recognized as an important factor (D'Ambrosio et al. 1981b, Sutherland et al. 1980b). _ . ~ ~ ~ _ _ A ~ ~ - THE PROBLEM At the surface of the earth the intensity of sunlight is a strong function of wavelength, decreasing rapidly for wavelengths below 320 nm (Figure 2.1). Intensities at wavelengths below 320 nm are affected most by changes in stratospheric ozone. Figure 2.2 shows the effect of large reductions in ozone on the spectrum of light reaching the earth. The net effect is a shift in the entire spectrum of UV at the surface of the earth toward shorter wavelengths; that is, the intensity of the short-wavelength W increases. While the reductions of ozone illustrated in Figure 2.2 are much larger than is generally anticipated, the figure illustrates the point. For example, an approximate 50 percent decrease in stratospheric ozone gives rise to a change in intensity that increases from a factor of about 2 at 305 nm to a factor of about 50 at 295 nm. In general, for any change
40 1 .0 0.1 z `~, 0.01 o o oh 0.001 > - LD ?~ It 0.0001 1.0 >G'` Sunlight /\.? Through Ozone If. \ \ \ \(b) (a)\ \ \ \ \ \ \ \ \ \ \ - UV-B ~ ~UV-A 290 300 310 320 330 WAVELENGTH (nary) 0.1 en in UJ 0.01 Z J z en 0.001 > J LL 0.0001 FIGURE 2.2 The relative intensity of sunlight (solar elevation of 60°) reaching the surface of the earth for different amounts of stratospheric ozone (the normal amount is close to 3.4 atmosphere mm). The shapes of two biological sensitivity curves are also shown: (a) damage to DNA multiplied by the transmission of human epidermis, and (b) human erythema or sunburn. Curve (c) is the response of the Robertson-Berger meter (discussed in Chapter 5~. (Source: The three curves of sunlight intensity are from U.S. Congress, Senate (1975~; the two biological sensitivity curves are from Setlow (1974) and Scott and Straf (1977~; the Robertson-Berger meter curve is from Berger et al. (l975~.)
41 in ozone concentration, one can compute with reasonable confidence the change in the W spectrum striking the surface of the earth. Hence a Predicted decrease in stratospheric ozone will give rise to predicted increases in intensity as a function of wavelength of solar UV (Johnson et al. 1976). The extent of the known deleterious effects of UV also depends strongly on wavelength and, as a rule, increases rapidly for wavelengths below 320 nm. Figure 2.2 shows two curves of biological sensitivity (Scott and Straf 1977, Setlow 1974). The figure illustrates the findings that UV-A wavelengths are much less biologically effective for damaging DNA or causing sunburn than W-B, and that in the W-B region the biological sensitivity per unit dose is an extremely sensitive function of wavelength. Thus, even if the increase in the absolute amount of UV penetrating the ozone layer is small, the changes will occur in a region of the spectrum that is very effective biologically. Plots of biological sensitivity as a function of wavelength--so-called action spectra--are obtained experimentally. These experiments are difficult to do on simple biological systems and even more difficult to do on animals, plants, and ecosystems. Thus there are uncertainties in our understanding of the dependence of effects on wavelength. Furthermore, ethical considera- tions prevent the controlled investigation of some action spectra, specifically those for various types of skin cancer in humans. Most of the available data do not derive from direct experiments on the biological systems of interest. For example, the basic data on human skin cancer are epidemio- logical: incidence, prevalence, and mortality at a relatively small number of locations in the United States. The locations differ in many ways, for example, in the average UV-B exposure during the year, the maximum UV-B exposure at any time during the year, the amount of visible light, and the ethnic and occupational backgrounds and life styles of the populations. Without data from many more locations that differ widely in the variables that might affect skin cancer incidence, it is not possible to use epidemiological data alone to determine the important variables or the action spectrum responsible for skin cancer. Thus we must draw inferences about the action spectrum for human skin cancer from animal experi- ments and molecular theories. Without knowing the action spectrum for a particular effect, that is, without knowing
42 the biological sensitivity curve, it is not possible make even rough predictions. For example, if curve (a) in Figure 2.2 were not the proper one to use because the major effect arose from wavelengths in the UV-A region, there would be no real consequences of ozone depletion on the biological system of interest. But if the sensitivity curve were as given in curve (a) of Figure 2.2, the depletion of ozone would have a large effect. There are two general approaches to measuring and predicting the effects of increased UV on biological systems. to 1. A straightforward approach is to irradiate a system with solar simulators, which mimic the spectrum of the sun, as a function of time and for various concentra tions of ozone. _: ~~~~ This approach is useru' ror scuay~ng ertects on crop plants and small animals, but, even if there were large numbers of such simulators available, it is impractical for studying effects on ecosystems because they are too large. In addition, the experimental irradiation of people is not ethical, even though large segments of the U.S. population willingly participate in a natural experiment of a sort through their propensity for sunbathing. 2. A second approach is to expand arid apply photobio- logical theories of effects on molecules, microorganisms, cells in culture, plants, and animals in order to improve the data base and our understanding. Predictions about the effects of ozone depletion on complex biological systems, such as humans and ecosystems, can then be made from fundamental principles. THE UNDERLYING BIOLOGICAL QUESTIONS Part II of this report builds on the large amount of photobiological data accumulated in the U.S. Department of Energy's Climatic Impact Assessment Program and in two extensive National Research Council reports: Environ- mental Impact of Stratospheric Flight (NRC 1975) and Protection Against Depletion of Stratospheric Ozone by Chlorofluorocarbons (NRC 1979a). Since those reports were written, there have been important additions to the basic knowledge of photobiological processes and some modest increases in basic epidemiological data. The changes and refinements in knowledge are summarized in the chapters that follow.
43 Because straightforward solar-simulation experiments cannot be used to estimate most of the~biological effects likely to result from a change in stratospheric ozone, the problem must be approached by determining directly, or indirectly by extrapolation from simpler biological systems, the answers to four key questions. With the answers to these questions, models and theories can be constructed from which reasonable predictions of photo- biologi al responses can be made. 1. What is the shape of the dose-response curve? An . increasing dose of UV produces an increasing biological effect, but the effect is usually not linearly propor- tional to the dose. The quantitative relationship between dose and response may be described by a dose-response curve. Figure 2.3 shows the general shapes of three possible dose-response curves. If the dose-response curve were a straight line, a 10 percent change in dose would give a 10 percent increase in effect. If, on the other hand, it were curved sharply upward, as in curve _ (c) in Figure 2.3, a 10 percent increase in dose would give rise to different increases in the biological effect, depending on the initial dose. For an initial dose of 2.5 arbitrary units per year, a 10 percent increase in dose would give a 70 percent increase in the biological effect. If the dose-response curve were curved downward, as in curve (a) in Figure 2.3, a 10 percent increase in the same initial dose would give rise to only about a 4 percent increase in the biological effect. Hence it is necessary to know the actual form of this relationship for the shortest wavelengths of W that penetrate the ozone layer, say 290 nm, to the longest that have an important biological effect on the system being investigated. For the induction of cancer in mice, for example, this longest wavelength is near 320 nm. If the dose-response curves have similar shapes for all wavelengths investigated, one can have confidence that the fundamental photobiological processes are the same at all wavelengths. On the other hand, if the curves do not have the same shape at all wavelengths, different types of photochemical or photobiological mechanisms must operate at different wavelengths. 2. IS there a reciprocal relationship between intensity and duration of exposure in responses? In a number of biological systems, low intensities delivered for a long time give the same result as high intensities delivered for a short time, as long as the same total
4 UJ 11 UJ ° 2 o cr. LO - ~: J U] CC o S ~' / 44 4 Percent / 10 / 70 Percent l(c) I 10 Percent Dose Increase 1 1 , 1 0 1 2 RELATIVE DOSE PER YEAR 3 4 FIGURE 2.3 Hypothetical dose-response curves (a), (b), and (c), illustrating the effect on the changes In anticipated biological effects resulting from a 10 percent dose increase. (In the W-B spectral region, a 5 percent change In ozone concentration will probably produce an approximate 10 percent change in dose.) dose is given--the response is simply the product of intensity and time, or the total, time-integrated dose. Exposure time and dose rate are then said to be related reciprocally, and the reciprocity law holds. If the reciprocity law does not hold, one must know not only the dose-response relationship, but also the dependence of the response on the exposure time and dose rate. For example, in simple cellular systems, a given effect usually requires a higher dose at low intensities than at high intensities, presumably because during low-intensity irradiation repair processes take place and little damage accumulates (Harm 1980b). In rats a single dose is more tumorigenic than an equal dose fractionated over 12 weeks (Strickland et al. 1979). On the other hand, to produce tumors in 50 percent of mice by W irradiation, a higher dose is required at high intensities than at low inten- sities (see Chapter 5). It is not known how intermittent exposures, as might actually be experienced by humans at it.
45 work or during recreation, affect the dose-response relationship. A difficulty in extrapolating the effects of laboratory-type experiments to the outside world is the fact that most laboratory experiments involve acute exposures usually taking only a fraction of a cell cycle time. In sunlight, however, many biological systems are exposed to low intensities for long times (chronic exposure). Since exposures to sunlight, and in particular to W -B. may often be weak, early or late in the day, or during the winter, many chronic exposures may be at dose rates well below those used in the laboratory to determine whether reciprocity holds. 3. How does biological sensitivity depend on wavelength? It is clear from descriptions given above that the specific biological effects resulting from a change in amount of ozone depends critically on the action spectra. If these curves are not known from direct experiment or cogent theory, there is no theoretical basis for making a prediction of the effects of ozone depletion. The answers to questions (1) and (2) above must be known before the shape and the wavelength dependence of each action spectrum, that is, the relative effectiveness of different monochromatic wavelengths in producing the observed effect, can be determined. The product, wavelength by wavelength, of the action spectrum and the spectrum of sunlight at the surface of the earth gives the relative effectiveness of sunlight in producing the specific biological effect (Caldwell 1971, NRC 1979a, Setlow 1974), provided that interactive effects (see question (4) below) are small. Any ozone depletion will change the spectrum of sunlight at the surface of the earth. This change, when multiplied by the action spectrum, will give the radiation amplification factor, i.e., the percentage increase in biologically damaging W per percentage decrease in ozone. The radiation amplification factor depends on the action spectrum. ~ Ar" there effects at different wavelengths that interact? The studies that have been conauccea In one three areas discussed above have used single wavelengths of W . An extrapolation to the effects of sunlight on crops, ecosystems, and humans from experiments in which the effects of single wavelengths are studied can be made only if the effects of the isolated wavelengths are purely additive, and not synergistic or antagonistic. Hence it is crucial to determine whether biological systems irradiated with a range or band of wavelengths
46 act as one would predict from the sum of the effects at discrete wavelengths. In addition, there may be other synergistic or antagonistic agents in the environment to consider, such as visible light, temperature, and chemicals. As will be discussed in Chapter 3, there are large synergistic effects between W -B and longer wavelengths in many simple photobiological systems. Despite the present uncertainties in understanding, there has been impressive progress in the extent of knowledge and in the delineation of the types of questions that can be answered easily. Certain questions may take several years and much data accumulation to answer, and some appear at present to be unanswerable but perhaps could be answered in the future with the help of a strong program of basic research.