REPAIR OF NEAR (365 nm)- AND FAR (254 nm)-UV DAMAGE TO BACTERIOPHAGE OF ESCHERICHIA COLI

Abstract: Intact bacteriophage have been irradiated at 365 nm or at 254 nm and then analysed for DNA photoproducts or injected into their bacterial host to test susceptibility of the damage to both phage and host-cell mediated repair systems. Both thymine dimers and single-strand breaks are induced in the phage DNA by 365 nm radiation. The dimers appear to be the major lethal lesion (approximately 2 dimers per lethal event) in both repair deficient bacteriophage T4 and bacteriophage λ. after irradiation with either 254 nm or 365 nm radiation. Damage induced in T4 by either wavelength is equally susceptible to x -gene reactivation (repair sector approximately 0.5). v -gene reactivation acts on a larger fraction of the near-UV damage (repair sector of 0.82 at 365 nm as against 0.66 at 254 nm). The host-cell mediated photoreactivation system is only slightly less effective for near-UV damage but host-cell reactivation (as measured by comparing survival of phage λ. on a uvr⁺ and a uvr^- host) is effective against a far smaller sector of near-UV damage (0.35) than far-UV damage (0.85). Weigle-reactivation (far-UV induced) of near-UV damage to phage λ is not observed. The results suggest that unless the near-UV damaged phage DNA is repaired immediately after injection. the lesions rapidly lose their susceptibility to repair with a consequent loss of activity of the phage particles.     

MUTATION INDUCTION BY AND MUTATIONAL INTERACTION BETWEEN MONOCHROMATIC WAVELENGTH RADIATIONS IN THE NEAR-ULTRAVIOLET AND VISIBLE RANGES

Abstract— The induction of mutations (reversion to tryptophan independence) by various UV (254, 313, 334 and 365 nm) and visible (405 and 434 nm) wavelengths was measured in exponential phase populations of Escherichia coli B/r thy trp and B/r thy trp uvrA by assay of irradiated populations on semi-enriched media. No mutations were induced in the repair proficient strain at wavelengths longer than 313 nm. Mutations were induced in the excisionless strain at wavelengths as long as 405 nm but less than expected from the known amount of DNA damage induced. Irradiation at the longer wavelengths (434, 405, 365 and 334 nm) suppressed the appearance of 254- or 313-nm-induced mutations in the repair competent strain but not in the excision deficient strain. The relative dose-requirement for mutation suppression was related to the relative efficiency of these wavelengths in inducing growth delay. These results suggest that the growth delay induced by near-UV and visible wavelengths allows more time for the 'error-free" excision repair process to act on the potentially mutagenic lesions induced by 254- and 313-nm radiations, thereby reducing the mutation frequency observed in the repair-proficient strain. The level of near-UV mutation induced in the excision deficient strain is lower than expected from the DNA damage known to be induced. It is possible that near-UV radiation induces a class of lethal lesions that are not susceptible to error-prone repair.     

LETHAL ACTION OF ULTRAVIOLET AND VISIBLE (BLUE-VIOLET) RADIATIONS AT DEFINED WAVELENGTHS ON HUMAN LYMPHOBLASTOID CELLS: ACTION SPECTRA AND INTERACTION SITES

Abstract— The repair proficient human lymphoblastoid line (TK6) has been employed to construcr an action spectrum for the lethal action of ultraviolet (UV) radiation in the range254–434 nm and to examine possible interactions between longer (334, 365 and 405 nm) and shorter wavelength (254 and 313 nm) radiations. The action spectrum follows a DNA absorption spectrum fairly closely out to 360 nm. As in previously determined lethal action spectra for procaryotic and eucaryotic cell populations, there is a broad shoulder in the334–405 nm region which could reflect the existence of either (a) a non-DNA chromophore or (b) a unique photochemical reaction in the DNA over this region. Pre-treatment with radiation at 334 or 365 nm causes either a slight sensitivity to (low fluences) or protection from (higher fluences) subsequent exposure to radiation at a shorter wavelength (254 or 313 nm). Pre-irradiation at a visible wavelength (405 nm) at all fluence levels employed sensitizes the populations to treatment with 254 or 313 nm radiations. These interactions will influence the lethal outcome of cellular exposure to broad-band radiation sources.     

Action spectra for inactivation of normal and xeroderma pigmentosum human skin fibroblasts by ultraviolet radiations

—Action spectra for UV-induced lethality as measured by colony forming ability were determined both for a normal human skin fibroblast strain (lBR) and for an excision deficient xeroderma pigmentosum strain (XP4LO) assigned to complementation group A using 7 monochromatic wavelengths in the range 254-365 nm. The relative sensitivity of the XP strain compared to the normal skin fibroblasts shows a marked decrease at wavelengths longer than 313 nm. changing from a ratio of about 20 at the shorter wavelengths to just greater than 1.0 at the longer wavelengths. The action spectra thus indicate that the influence on cell inactivation of the DNA repair defect associated with XP cells is decreased and almost reaches zero at longer UV wavelengths. This would occur, for example, if the importance of pyrimidine dimers as the lethal lesion decreased with increasing wavelength. In common with other studies both in bacterial and mammalian cells, our results are consistent with pyrimidine dimers induced in DNA being the major lethal lesion in both cell strains over the wavelength range 254-313 nm. However, it is indicated that different mechanisms of inactivation operate at wavelengths longer than 313 nm.     

UV-INDUCED PROTEIN ALTERATIONS AND LIPID OXIDATION IN ERYTHROCYTE MEMBRANES

Certain ultraviolet radiation-induced effects in skin may result from primary photochemical alterations in cell membranes. We have studied isolated erythrocyte membranes in order to determine the UV-fluence and wavelength dependence for protein alterations and lipid oxidation. Protein crosslinking was detected as high molecular weight protein (greater than 200,000 DA) on polyacrylamide/agarose gel electrophoresis. Spectrin decreased more rapidly than the other membrane proteins upon exposure to lambda = 250-380 nm radiation. Nitrogen-purging inhibited the UV-induced decrease in spectrin by 60% and decreased crosslinking to an even greater degree. The decrease in spectrin was not inhibited by superoxide dismutase, catalase, or sodium azide. Radiation at 280 nm was most effective for spectrin loss, 265 and 297 nm were less effective and 254 and 313 nm were not effective. Prior irradiation at 280 nm did not sensitize the membranes to subsequent irradiation at 313 nm indicating that photodecomposition products of tryptophan are not involved. Lipid photooxidation was measured with the thiobarbituric acid assay and was induced at higher fluences of UV radiations than those required for loss of spectrin. These results indicate that the major effects of UV radiation on cell membranes are alterations of proteins and suggest that tryptophan is the major chromophore for these alterations.     

INACTIVATION ACTION SPECTRA OF BACILLUS SUBTILIS SPORES IN EXTENDED ULTRAVIOLET WAVELENGTHS (50–300nm) OBTAINED WITH SYNCHROTRON RADIATION

Five types of Bacillus subtilis spores (UVR, UVS, UVP, RCE, and RCF) differing in repair and/or recombinational capabilities were exposed to monochromatic radiations at 13 wavelengths from 50 to 300 nm in vacuum. An improved biological irradiation system connected to a synchrotron radiation source was used to produce monochromatic UV radiation in this extended wavelength range with sufficient fluence to inactivate bacterial spores. From the survival curves obtained, the action spectra for the inactivation of the spores were depicted. Recombination-deficient RCE (recE) and RCF (recF) spores were more sensitive than the wild-type UVR spores in the entire range of wavelengths. This was considered to mean that DNA was the major target for the inactivation of the spores. Vacuum-UV radiations of 125-175 nm were effective in killing the spores, and distinct peaks of the sensitivity were seen with all types of the spores. Insensitivities at 190 and 100 nm were common to all five types of spores, indicating that these wavelengths were particularly impenetrant and absorbed by the outer layer materials. The vacuum-UV peaks centering at 150 nm were prominent in the spores defective in recombinational repair, while the far-UV peaks at around 235 and 270 nm were prominent in the UVS (uvrA ssp) and UVP (uvrA ssp polA) spores deficient in removal mechanisms of spore photoproducts. Thus, the profiles of the action spectra were explained by three factors; the penetration depth of each radiation in a spore, the efficiency of producing DNA damage that could cause inactivation, and the repair capacity of each type of spore.     

AN ACTION SPECTRUM FOR ULTRAVIOLET RADIATION-INDUCED MEMBRANE DAMAGE IN Escherichia coli K-12

A wild-type Escherichia coli K-12 strain was irradiated using monochromatic radiation in the range 254 to 405 nm. A measure of the cell membrane damage induced at each wavelength was investigated by comparing cell viability after irradiation on nutrient agar and on minimal medium containing either a low or high inorganic salt concentration. An action spectrum for lethality and for cell membrane damage was then determined. From 254 to 310 nm lethality closely corresponded to the absorption spectrum of DNA, and there was no indication of membrane damage. However, above a wavelength of 310 nm, the direct absorption of radiation by DNA could not account for the sensitivity observed. Moreover, at wavelengths longer than 310 nm, cell membrane damage was induced and by an increasing factor up to a peak at 334 nm. At the longer wavelengths of 365 and 405 nm, there was a gradual decrease from the peak of damage to cell membranes induced by 334 nm radiation. These results indicate that cell membrane damage may contribute significantly to near-UV radiation-induced cell lethality in wild-type E. coli K-12.     

ALPHA POLYMERASE INVOLVEMENT IN EXCISION REPAIR OF DAMAGE INDUCED BY SOLAR RADIATION AT DEFINED WAVELENGTHS IN HUMAN FIBROBLASTS

Abstract Cultured fibroblasts derived from normal human skin have been irradiated at a series of monochromatic wavelengths throughout the ultraviolet region and exposed to the specific α polymerase inhibitor, aphidicolin (1 μg/m l , 2 days) prior to assay for colony forming ability. Repair of 75-80% of the lethal damage induced by UVC (254 nm) or UVB (302 nm, 313 nm) radiation is inhibited by aphidicolin suggesting that such damage is repaired by a common α polymerase dependent pathway. Exposure to aphidicolin after irradiation at longer UVA (334 nm, 365 nm) or a visible (405 nm) wavelength leads to slight protection from inactivation implying that the processing of damage induced in this wavelength region is quite distinct from that occurring at the shorter wavelengths and does not involve α polymerase.     

RESTRICTION ENZYME CLEAVAGE OF UV-IRRADIATED DNA: A COMPARISON OF FAR-UV AND MID-UV WAVELENGTHS

UV-irradiated DNA is less susceptible to restriction by Type II endonucleases than unirradiated DNA presumably due to photolesions formed in the recognition sites. Previous reported studies have used 254 nm radiation or 313 nm plus acetophenone, both treatments which introduce pyrimidine dimers in preference to other photolesions. To assess the effect of a longer wavelength, at which the ratio of pyrimidine dimer formation to the formation of other photolesions is reduced, two different DNAs were irradiated with UV of either 254 or 313 nm and restricted with suitable restriction endonucleases. Restriction patterns were analysed for novel fragments resulting from UV-induced alteration of enzyme recognition sites. EcoRI restriction of 254 nm irradiated lambda DNA produced six novel bands, only three of which were observed following restriction of 313 nm irradiated lambda. These three represented the largest fragments resulting from single site blocks. Novel fragments involving adjacent site blocks observed at 254 nm were not found with 313 nm radiation. Comparison of 254 nm irradiated pSV₂gpt to that irradiated at 313 nm, both restricted with Dral, revealed a more complex pattern. Although all sites were singly blocked by radiation of both wavelengths, multiple site blocks produced by 313 nm radiation did not occur in the order predicted by the 254 nm radiation dose response. These data suggest that certain sites in pSV₂gpt may be more refractory to multiple site blocks than others when irradiated at 313 nm.     

CORRELATION BETWEEN ENDOGENOUS GLUTATHIONE CONTENT AND SENSITIVITY OF CULTURED HUMAN SKIN CELLS TO RADIATION AT DEFINED WAVELENGTHS IN THE SOLAR ULTRAVIOLET RANGE

Abstract— Glutathione depletion of cultured human skin fibroblasts by treatment with buthionine-S,R-sulfoximine (BSO) sensitises them to radiation at a series of defined wavelengths throughout the solar UV range. We now show that there is a close quantitative correlation between cellular glutathione content (as depleted by BSO) and sensitivity to radiation at 365 nm. A weaker correlation is observed when cells are depleted of glutathione using diethylmaleimide. Both fibroblasts and epidermal keratinocytes derived from the same foreskin biopsy are sensitised to radiation at 313 nm by glutathione depletion. However, the keratinocytes are sensitised to a much lesser extent, an observation which agrees quantitatively with the higher residual levels of cellular glutathione remaining after maximum depletion by BSO (approximately 25% for the keratinocytes vs less than 5% for the fibroblasts). At low to intermediate fluence levels, 10 mM cysteamine present during irradiation at 302 nm is able to almost completely reverse the sensitising effects of glutathione depletion suggesting that the endogenous thiol protects against radiation at this wavelength by a free radical scavenging mechanism. At 313 nm, the sensitisation is not reversed by cysteamine suggesting that glutathione plays a more specific role in protection against radiation at longer wavelengths. Xeroderma pigmentosum group A fibroblasts (excision deficient) are also sensitised to radiation at 313 and 365 nm by depletion of glutathione but since the sensitization is less than that observed for the normal strain, we cannot conclude that glutathione protects against a sector of DNA damage susceptible to excision repair. The results provide further evidence that endogenous glutathione is involved in protecting human skin cells against a wide range of solar radiation damage and suggest that while free radical scavenging is involved at the shortest wavelength (302 nm) tested, a more specific role of glutathione is involved in protection against radiation at longer wavelengths.     

MEMBRANE DAMAGE CAN BE A SIGNIFICANT FACTOR IN THE INACTIVATION OF ESCHERICHIA COLI BY NEAR-ULTRAVIOLET RADIATION

Abstract A DNA repair competent strain of Escherichia coli K-12 showed sensitivity to inorganic salts (at concentrations routinely used in minimal media) after irradiation with broad spectrum near–UV radiation, at fluences that caused little inactivation when plated on complex growth medium. This effect was not observed with cells that had been exposed to 254 nm radiation. This sensitivity to minimal medium was increased by increasing the salt concentration of the medium and by increasing the pH of the medium. This sensitivity was greatly increased by adding to the medium a low concentration of commercial glassware cleaning detergent that had no effect on unirradiated cells or far-UV irradiated cells. These findings may explain the large variability often observed in near-UV radiation survival data, and demonstrate that, at least on minimal medium plates, membrane damage contributes significantly towards cell killing. This phenomenon is largely oxygen dependent.     

LEAKAGE OF 86Rb+ AFTER ULTRAVIOLET IRRADIATION OF Escherichia coli K-12

Abstract— Stationary phase cultures of a DNA repair proficient Escherichia coli K-12 strain showed a release of intracellular material as assessed by three different methods (260 nm absorption; [methyl-³H]thymidine leakage and ⁸⁶Rb⁺ leakage) after broad-band (Black-Light Blue) near-UV radiation but not after far-UV (254 nm) radiation. As a control response for membrane damage to cells, this leakage of intracellular material was also determined by each method after mild-heat (52°C) treatment of E. coli K-12. An action spectrum for the release of ⁸⁶Rb⁺ from E. coli K-12 after irradiation with monochromatic wavelengths, from 254 to 405 nm, is also presented. The action spectrum for lethality (F₃₇ values) obtained for this strain, shows that leakage of ⁸⁶Rb⁺ occurs at fluences equivalent to or slightly less than fluences causing inactivation at wavelengths above 305 nm. In contrast, at wavelengths below 305 nm, leakage of ⁸⁶Rb⁺ from irradiated cells can be induced but only at fluences significantly greater than was required to cause cell inactivation. These results indicate, therefore, that near-UV radiation can induce a damaging effect on the cell's permeability barrier which may be significant in causing the death of the cell, whereas the effect is not significant in causing the death of cells by far-UV radiation where DNA damage is known to be the main cause of lethality.     

Sunlight-induced killing of nondividing human cells in culture

Nondividing populations of human diploid fibroblasts that are DNA excision repair proficient (WS-1, KD. SSCW) and repair deficient (XP12BE) were exposed to mid-day summer sunlight for a determination of survival based on an ability of cells to remain attached to a culture vessel surface. Whereas mid- and far-UV wavelengths and radiation emitted from a sunlamp cause a gradual degeneraton and detachment of cells in a dose-dependent manner, sunlight does not promote cell killing that is evidenced by these criteria in repair proficient cells. Detachment of repair deficient cells is promoted to a limited extent but only at sunlight exposure times that are low with respect to the amount of DNA damage (pyrimidine dimers) induced. Repair proficient and deficient cells exposed to sunlight for longer times do not detach but are incapable of excuding a viable stain several days after exposure and appear as histologically fixed cells. Pyrimidine dimer levels in these sunlight irradiated cells were great enough to have promoted detachment had these levels been induced by UV (254 nm) alone. Other photodamage induced by these exposures evidently inhibits the dimer-induced cell degeneration that leads to cell detachment. We conclude that pyrimidine dimers are responsible for cell killing at short sunlight exposure times (< 40 min) but that at longer exposures (> 80 min) cells arc killed by a different mechanism that is independent of dimer-caused death.     

Induction of Lambda Prophage by 213 nm Laser Radiation: A Quantitative Comparison with 193 nm Excimer Radiation Using Image Analysis

Abstract— We compared the DNA damage produced by radiation from two UV laser wavelengths, 213 nm and 193 nm, with that produced by noncoherent 254 nm radiation. Following irradiation of Escherichia coli BR339, a bacteriophage lambda lysogen containing the lacZ gene, prophage induction was measured by assaying for β-galactosidase. Because of the limited penetration by UV laser wavelengths an agar overlay of the lysogen was used as the irradiation target. Irradiation of 254 nm was performed in buffer suspension followed by transfer of 5 μL spots onto assay plants. Computer image analysis was used to monitor the rate of product formation, observed as an increase in optical density of the irradiated zones on assay plates. We found that the rate of product formation was a more reproducible unit of comparison than the optical density present at the end of the reaction. Although the rate of product formation was not linearly related to enzyme concentration, the data could be fit to a simple logarithmic function. Using this method, we concluded that the DNA damaging ability of 213 nm radiation was 10 times more efficient than 193 nm radiation and about 100 times less efficient than 254 nm noncoherent radiation.     

INDUCTION OF DIRECT AND INDIRECT SINGLE-STRAND BREAKS IN HUMAN CELL DNA BY FAR- AND NEAR-ULTRAVIOLET RADIATIONS: ACTION SPECTRUM AND MECHANISMS

Abstract— An action spectrum for the immediate induction in DNA of single-strand breaks (SSBs, frank breaks plus alkali-labile sites) in human P3 teratoma cells in culture by monochromatic 254-, 270-, 290-, 313-, 334-, 365-, and 405-nm radiation is described. The cells were held at +0.5C during irradiation and were Iysed immediately for alkaline sedimentation analysis following the irradiation treatments. Linear fluence responses were observed over the fluence ranges studied for all energies. Irradiation of the cells in a D₂O environment (compared with the normal H₂O environment) did not alter the rate of induction of SSBs by 290-nm radiation, whereas the D₂O environment enhanced the induction of SSBs by 365- and 405-nm irradiation. Analysis of the relative efficiencies for the induction of SSBs, corrected for quantum efficiency and cellular shielding, revealed a spectrum that coincided closely with nucleic acid absorption below 313 nm. At longer wavelengths, the plot of relative efficiency vs . wavelength contained a minor shoulder in the same wavelength region as that observed in a previously obtained action spectrum for stationary phase Bacillus subtilis cells. Far-UV radiation induced few breaks relative to pyrimidine dimers, whereas in the near-UV region of radiation, SSBs account for a significant proportion of the lesions relative to dimers, with a maximum number of SSBs per lethal event occurring at 365-nm radiation.     

LETHAL EFFECTS OF NATURAL SOLAR ULTRAVIOLET RADIATION IN REPAIR PROFICIENT AND REPAIR DEFICIENT STRAINS OF ESCHERICHIA COLI: ACTIONS AND INTERACTIONS

Abstract— The inactivation of repair proficient ( Escherichia coli K12 AB 1157, E. coli B/r) and repair deficient ( E. coli K12 AB 1886 uvrA , AB 2463 recA and AB 2480 uvrA recA ) strains of bacteria by noon sunlight has been measured. The use of biological dosimetry based on an ultraviolet (UV) sensitive strain of Bacillus subtilis spores has allowed a quantitative comparison of bacterial inactivation by solar, 254 and 302 nm radiations. Our analysis indicates that: (1) uvrA and recA gene products are involved in repair of a substantial portion of the solar DNA damage, (2) 302 nm is a more appropriate wavelength than 254 nm to represent the DNA-damaging action of sunlight and that (3) repair proficient strains are inactivated by sunlight more rapidly than expected from the levels of DNA damage induced. When populations of repair proficient bacteria are exposed to noon sunlight for 20 min, they become sensitive to the lethal action of far-UV (254 nm), MMS (0.1 M ) and to a lesser extent, mild heat (52°C).     

INDUCTION OF DNA-PROTEIN CROSSLINKS IN HUMAN CELLS BY ULTRAVIOLET and VISIBLE RADIATIONS: ACTION SPECTRUM

Abstract— DNA-protein crosslinking was induced in cultured human P3 teratocarcinoma cells by irradiation with monochromatic radiation with wavelengths in the range254–434 nm (far-UV, near-UV, and blue light). Wavelength 545 nm green light did not induce these crosslinks, using the method of alkaline elution of the DNA from membrane filters. The action spectrum for the formation of DNA-protein crosslinks revealed two maxima, one in the far-UV spectrum that closely coincided with the relative spectrum of DNA at 254 and 290 nm, and one in the visible light spectrum at 405 nm, which has no counterpart in the DNA spectrum. The primary events for the formation of DNA-protein crosslinks by such long-wavelength radiation probably involve photosensitizers. This dual mechanism for DNA-protein crosslink formation is in strong contrast to the single mechanism for pyrimidine dimer formation in DNA, which apparently has no component in the visible light spectrum.     

SINGLE-STRAND BREAKS INDUCED IN BACILLUS SUBTILIS DNA BY ULTRAVIOLET LIGHT: ACTION SPECTRUM and PROPERTIES

Abstract— The induction of single-strand breaks (alkali-labile bonds plus frank breaks) in the DNA of Bacillus subtilis irradiated in vivo by monochromatic UV light at wavelengths from 254 to 434 nm was measured. The spectrum consists of a major far-UV (below 320 nm) component and a minor near-UV shoulder. A mutant deficient in DNA polymerase I accumulates breaks caused by near-UV (above 320 nm) wavelengths faster than the wild-type strain proficient in polymerase I. Measurable breaks in extracted DNA are induced at a higher frequency than those induced in vivo. Anoxia, glycerol, and diazobicyclo (2.2.2.) octane inhibit break formation in extracted DNA. Alkali-labile bonds induced by 365-nm UV radiation are largely (78%) covalent bond chain breaks, the remainder consists of true alkali-labile bonds, probably apurinic and apyrimidinic sites.     

ACTION SPECTRA IN ULTRAVIOLET WAVELENGTHS (150-250 nm) FOR INACTIVATION AND MUTAGENESIS OF Bacillus subtilis SPORES OBTAINED WITH SYNCHROTRON RADIATION

Bacillus subtilis spores were exposed in vacuo to monochromatic UV radiation from synchrotron radiation in the wavelength range of 150 nm to 250 nm. Survival and frequency of mutation to histidine-independent reversion were analysed for three types of spores differing in DNA-repair capabilities. UVR spores (wild-type DNA repair capability) exhibited nearly equal sensitivity to the lethal effects of far-UV (220 nm and 250 nm) and of vacuum-UV radiation (150 and 165 nm), but showed marked resistance to 190 nm radiation. UVS spores (excision-repair and spore-repair deficient) and UVP spores (a DNA polymerase I-defective derivative of UVS) exhibited similar action spectra; pronounced sensitivity at 250 and 220 nm, insensitivity at 190 nm and a gradual increase of the sensitivity as the wavelength decreased to 165 nm. In all strains, the action spectra for mutation induction paralleled those for the inactivation, indicating that vacuum-UV radiation induced lethal and mutagenic damages in the spore DNA. The insensitivity of the spores to wavelengths around 190 nm may be explicable by assuming that radiation is absorbed by materials surrounding the core in which DNA is situated.     

PHOTOREACTIVATION OF ICR 2A FROG CELLS AFTER EXPOSURE TO MONOCHROMATIC ULTRAVIOLET RADIATION IN THE 252–313 nm RANGE

Abstract— Exposure of ICR 2A frog cells to photoreactivating light after treatment with monochromatic ultraviolet (UV) radiation in the 252–313 nm range resulted in an increase in survival with similar photoreactivable sectors for each of the wavelengths tested. As photoreactivating enzyme is specific for the repair of pyrimidine dimers in DNA, these findings support the hypothesis that these are critical lesions responsible for killing of cells exposed to UV radiation in this wavelength range. The action spectra for cell killing and production of UV-endonuclease sensitive sites were similar to the DNA absorption spectrum though not identical. Because the number of endonuclease sensitive sites is a reflection of the yield of pyrimidine dimers, these data also suggest that the induction of dimers in DNA by UV radiation in the 252–313 nm range is the principal event leading to cell death.