Comet assay detects cold repair of UV-A damages in a human B-lymphoblast cell line

Mutation Research 408 Ž1998. 111–120

Comet assay detects cold repair of UV-A damages in a human B-lymphoblast cell line Claudia Bock ) , Heike Dittmar, Helgard Gemeinhardt, Eckhard Bauer, Karl-Otto Greulich Institut fur ¨ Molekulare Biotechnologie, Beutenbergstr. 11, 07745 Jena, Germany Received 23 February 1998; revised 27 April 1998; accepted 27 April 1998

Abstract During DNA repair studies, cells are occasionally kept on ice in order to suppress DNA repair. In the present studies cultivated human NC37 B-lymphoblasts were damaged by UV-A irradiation Ž365 nm. and DNA single strand breaks were detected at the single cell level with the alkaline comet assay in the temperature range from 48C to 448C. Single cell studies, in contrast to bulk experiments, allow to identify apoptotic or necrotic cells, which can be omitted for data analysis. Unexpectedly, similarly efficient single phase repair kinetics was found at all temperatures below 378C, i.e., particularly also in the cold. For recovery times below 20 min a linear decrease of DNA damage was detected. After 20 min, no additional repair was observed, i.e., complete repair of single strand breaks was not achieved. At 448C DNA damage increased with time, probably due to heat damage and cell death. Nucleotide excision repair inhibitors such as aphidicolin, 1-b-Darabinofuranosyl cytosine ŽaraC. and hydroxyurea, but not the base excision repair inhibitor methoxyamine caused a strong increase in DNA strand breaks. The use of repair inhibitors confirmed DNA repair at 48C. In conclusion, partial repair of UV-A damage is similar at 378C and 48C and is probably governed by nucleotide excision repair. Keeping samples on ice may not result in a total suppression of DNA repair. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Comet assay; UV-A radiation; Single strand break; DNA repair; Temperature dependence

1. Introduction It is a widely accepted view that DNA repair activity is slowed down at low temperatures. Therefore cells are commonly kept on ice when DNA Abbreviations: SCGE: single cell gel electrophoresis; araC: 1-b-D-arabinofuranosyl cytosine; Pol: polymerase; NER: nucleotide excision repair; BER: base excision repair ) Corresponding author. Tel.: q49-3641-656405r00; fax: q49-3641-656410.

repair has to be suppressed. Surprisingly, comparatively little is known about DNA repair at low temperatures. Only few articles are available about temperature dependence of DNA repair process or repair enzymes in the range from 188C to 408C w1,2x. The ability of cells to repair DNA lesions at temperatures below 188C has been reported in two papers on DNA repair after X-ray and g-irradiation w3,4x. To our knowledge no literature of low temperature dependence of DNA repair after UV irradiation is available.

0921-8777r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 8 7 7 7 Ž 9 8 . 0 0 0 2 3 - 8

112

C. Bock et al.r Mutation Research 408 (1998) 111–120

Among the studies of UV induced DNA damage and its repair, the range of UV-A Ž320–400 nm. has recently gained importance w5–7x. Interest has focused on UV-A radiation since its contribution to the total energy of sunlight with 5.1% is more than 10 times higher than that of UV-B Ž280 nm–320 nm, 0.3%. w8x. UV-A indirectly induces DNA lesions, such as single strand breaks, alkali-labile-sites, photomodification of bases and DNA-to-protein crosslinks w9x, probably via oxidative damage which is repaired either by nucleotide excision or base excision repair w10x. Detection of DNA damage and repair has often been studied in bulk experiments. The most common techniques to detect DNA single strand breaks are alkaline sucrose sedimentation w11x and alkaline filter elution w12x. Recently single cell techniques have gained interest since the individuality of a cell within a population could be considered w13,14x. Single cell gel electrophoresis ŽSCGE. or comet assay is a highly sensitive and rapid method to investigate the response of single cells to DNA damaging agents w15,16x Žfor reviews see Refs. w17–19x.. This technique allows us to detect damage levels as low as 0.1 DNA breaks per 10 9 Da w20x. SCGE has been applied to investigate UV radiation induced DNA strand breaks and DNA repair mostly in lymphocytes from peripheral blood w21,22x. The additional use of repair inhibitors, such as aphidicolin or enzymes which produce single strand breaks can be useful to complete investigations of repair mechanisms w23,24x. In the present experiments we investigated the temperature dependence of the DNA repair mechanism from 48C to 448C after UV-A irradiation in human B-lymphoblast cells with SCGE. The DNA polymerase inhibitors aphidicolin and araC were used to accumulate open repair sites and confirm that DNA repair takes places. Additionally methoxyamine was used in order to determine the involved DNA repair pathways.

2. Materials and methods 2.1. Cell lines and cell culture Human NC37 B-lymphoblast from caucasian is a commercially available cell line ŽICN Flow Labora-

tories, Germany.. Cells were grown in suspension in RPMI 1640 medium ŽSigma, Germany. with 10% fetal calf serum at 378C in a 5% CO 2 atmosphere. 2.2. UV source and irradiation procedures UV-A radiation was provided by the 365 nm line of a 100 W high pressure mercury lamp ŽZeiss, Germany., filtered by a bandpass filter ŽZeiss, Germany. with a full width at half maximum of 10 nm. The intensity on the sample was 280 Wrm2 and the time of exposure was varied from 1 to 30 min, corresponding to radiation fluences up to 500 kJrm2 . The intensity was measured with a power meter ŽPolytech, Germany.. For irradiation all cells were handled in 0.01 M PBS buffer at pH 7.4 ŽSigma, Germany., placed in a 300 m l cylindrical cuvette and stirred during UV-A exposure. The temperature of the cell suspension was kept at 48C during irradiation. 2.3. Alkaline single cell gel electrophoresis (comet assay) After UV-A treatment, cells were mixed with low melting point agarose ŽSigma, Germany. held at 378C, in order to obtain a final concentration of 0.8% agarose and about 1.5 Ø 10 5 cellsrml. The suspension Ž90 m l. was pipetted onto special microscope slides ŽLabcraft, UK., precoated with one layer of 0.5% and one layer of 1% normal melting point agarose Ždiluted in 0.005 M PBS pH 7.4. and covered with a coverslip. The agar was allowed to cool down for 5 min over ice. The coverslip was removed immediately and the microscope slides were immersed in cold lysis solution at pH 10 Ž2.5 M NaCl, 100 mM Na 2 EDTA, 10 mM Tris pH 10, 1% sodium sarcosinate, 1% Triton X-100, 10% DMSO. and kept at 48C for at least 1 h. Slides were then carefully dried, placed in alkaline electrophoresis buffer Ž333 mM NaOH, 1 mM Na 2 EDTA, pH 13. and left at 48C for 1 h to allow unwinding of the DNA. Slides were then transferred to an electrophoresis tank with fresh alkaline electrophoresis buffer. Electrophoresis was performed at a field strength of 1 Vrcm at 48C for 20 min. Subsequently, slides were carefully rinsed with 0.4 M Tris pH 7.5 for 5 min each, stained with 60 m l propidium iodide Ž30 m M.,

C. Bock et al.r Mutation Research 408 (1998) 111–120

covered with a coverslip and incubated for 5 min in the dark. 2.4. Repair studies After irradiation, cells were incubated in PBS at various temperatures in a water bath. Incubation of the cells in medium resulted in a strong increase of dead cells. The recovery time was varied up to 100 min. For the experiments with repair inhibitors 15 m M aphidicolin ŽFluka, Germany., 2 mM hydroxyurea ŽSigma, Germany . or 1-b -D -arabinofuranosyl cytosine ŽSigma, Germany. at a concentration of 5 m M was added to the cell suspension during recovery. In some experiments a combination of 2 mM hydroxyurea and 5 m M araC was used to study repair mechanisms. Methoxyamine at a concentration of 5 mM was already added 1 h before UV treatment and kept in the cell suspension during irradiation and recovery. Subsequently cells were washed in 0.01 M PBS buffer at pH 7.4.

113

mean value of tail moment is not an ideal parameter to characterise the distribution ŽFig. 1 and Ref. w25x.. The histograms were thus fitted by the x 2 function which is described by one single parameter, called ‘degree of freedom n’. With the curve fit software Sigma Plot ŽJandel Scientific, USA. the histograms were superimposed with the x 2-functions calculated by the best fit in a non linear regression. The x 2function is defined by: 2

x Ž x. s

n

1 2 n r2 P G

y1

n

P Ž x.

2

1

P ey 2 x

ž / 2

where G is the gamma-function and n is the degree of freedom. The parameter ‘degree of freedom n’ corresponds to the mean value of tail moments calculated by Gaussian statistics and is therefore a good parameter to describe DNA damage. In order to indicate that the ‘degree of freedom n’ replaces the mean value of

2.5. Detection Examination of slides was performed with an epifluorescence microscope ŽZeiss, Germany . equipped with an adequate fluorescence filter set Žexcitation: 510–560 nm, beam splitter: 580 nm, emission: 590 nm. at =25 magnification. Fluorescence was imaged with an intensified CCD video camera ŽComputer Optics, Germany. and analysed with appropriate software ŽKinetics Imaging, Great Britain.. 2.6. Statistical analysis The comets were evaluated by the tail moment, defined as the product of tail length and percentage of the fluorescence intensity in the tail. About 50 to 100 cells were measured for each exposure time and the distribution of tail moments within one sample was evaluated. To consider the individuality of each cell, histograms are chosen for the interpretation of results. The distribution of tail moments of individual cells within one sample is not represented by a Gaussian distribution. Therefore the widely used

Fig. 1. Curve fit of histograms with x 2-function: Normalised distribution of tail moments for 100 cells Žhistograms., curve fit with the x 2 -function Žstraight lines. and value of degree of freedom, corresponding to x 2-mean of tail moment after UV-A irradiation at 365 nm with Ža. 0 Jrm2 Žcontrol cells., Žb. 170 kJrm2 , Žc. 340 kJrm2 , Žd. 500 kJrm2 , Že. 680 kJrm2 .

114

C. Bock et al.r Mutation Research 408 (1998) 111–120

tail moment by Gaussian statistics we define this parameter as ‘ x 2-mean of tail moments’:

x 2 statistics: degree of freedom n '‘ x 2-mean of tail moments’ Gaussian statistics: ‘Gaussian mean of tail moments’ It has to be considered that small DNA damage cannot be fitted to n - 2 because the x 2-function has a singularity for n - 2 w25x. In our experiments n was evaluated to describe DNA damage. The curve fitting procedure includes the calculation of standard errors drawn in all the figures as error bars. Each standard error is a measure of the quality of the x 2-fit to the data points. All experiments were carried out at least twice to get an idea of the error inherent in the data.

3. Results For technical reasons the studies of the relationship between radiation fluence and ‘ x 2-mean of tail moments’ ŽFigs. 1 and 2., the repair kinetics ŽFig. 3a,b., the inhibitor experiments ŽFig. 4a,b. and the temperature dependence of DNA repair ŽFig. 5a,b. were not performed in the same experiment. Cells were taken after different cultivation times. This

Fig. 3. Repair studies of cultivated B-lymphoblasts NC37: ‘ x 2-mean of tail moments’ as a function of recovery time after UV-A treatment Ž365 nm. with 500 kJrm2 Ža. at a temperature of 378C and Žb. at 48C. Measured values of irradiated cells Žsquares. are fitted with two linear functions Žstraight lines.. Open triangles indicate unirradiated control cells. The inserts represent the distribution of tail moments after 100 min recovery time for both temperatures.

could result in a different distribution of cells in the cell cycle states G1 , G2 and S within the population. For this reason the absolute values of the ‘ x 2-mean of tail moments’ for different experiments are not identical. The relative values as well as the curve diagrams however were reproducible over all types of experiments reported below. Fig. 2. Relationship between radiation fluence and ‘ x 2-mean of tail moments’ of B-lymphoblasts NC37, corresponding to the degree of DNA single strand breaks plotted as a function of radiation fluence Ž365 nm. on the cell sample Žsquares. and linear regression Žstraight line.. An explanation for the variation in absolute values for the different experiments is given at the beginning of Section 3.

3.1. Linear dependence of UV-A radiation fluence and DNA damage UV-A induced DNA single strand breaks were produced immediately after UV-A treatment. Fig. 1

C. Bock et al.r Mutation Research 408 (1998) 111–120

115

relationship between the radiation fluence on the cell sample and the ‘ x 2-mean of tail moments’. 3.2. Incomplete DNA repair at 378C and 48C The diagram in Fig. 3 contains data for the repair studies of NC37 after UV-A irradiation Ž365 nm. with 500 kJrm2 at 378C and at 48C. The ‘ x 2-mean of tail moments’ is plotted as a function of recovery time. Repair of UV-A radiation induced DNA le-

Fig. 4. Repair inhibitor experiments: DNA repair studies after UV-A irradiation Ž365 nm. with 500 kJrm2 with use of various repair inhibitors Ža. at 378C and Žb. at 48C. Obviously a strong accumulation of strand breaks was achieved by the use of all repair inhibitors at 378C and 48C.

presents the histograms and the curve fit with the x 2-function for different radiation fluences with the investigated cell line NC37. The data were fitted well by this procedure. For untreated cells ŽFig. 1a. no better fit is possible because the x 2-function is infinite for values of the ‘ x 2-mean of tail moments’ smaller than 2 and small x Žfor n - 2 and x s 0, y s `.. After a radiation fluence of 170 kJrm2 , the distribution of tail moments within the cell population becomes asymmetric ŽFig. 1b. and is well described by the x 2-function. With increasing radiation intensity the distribution becomes symmetric and similar to the Gaussian distribution as described by Bauer et al. w25x. Fig. 2 reveals a linear

Fig. 5. Ža. Temperature dependence of DNA repair of cultivated B-lymphoblasts: ‘ x 2-mean of tail moments’ versus recovery time after UV-A irradiation Ž365 nm. with 500 kJrm2 at different temperatures. Obviously, at 448C degradation of DNA governs the process. For other temperatures DNA repair was observed Žfor details see Fig. 3 and Fig. 4b.. Žb. DNA strand breaks during recovery as a function of temperature: ‘ x 2-mean of tail moments’, indicating DNA single strand breaks as a function of temperature for 40 and 80 min recovery. Although the data appear to be fitted by curve Žconvex after 40 min, concave after 80 min.. The large error bars and consideration of possible biological background do not allow such a differentiated view.

C. Bock et al.r Mutation Research 408 (1998) 111–120

116

sions was observed at 378C and surprisingly also at 48C. The data were fitted by two linear functions. From 0 to 20 min recovery a linear decrease in the ‘ x 2-mean of tail moments’ could be observed. From 20 to 100 min recovery, the ‘ x 2-mean of tail moments’ remains constant, but did not reach the value of unirradiated cells. Similar parameters of the curve fit could be obtained for both recovery temperatures. The slope of the first function is m s y0.05. The constant obtained with the second curve fit was 2.8. Our results indicate that DNA damage is repaired with the same time constant at both temperatures. In total, these data show that the ‘ x 2-mean of tail moments’ is a suitable parameter to describe DNA repair process. 3.3. Accumulation of open repair sites by Õarious inhibitors Table 1 shows the different DNA repair inhibitors, used in the present studies. The inhibited enzymes and the efficiency in blocking either nucleotide excision repair ŽNER. or base excision repair ŽBER. are also indicated. For all experiments unirradiated cells were incubated for 80 min with the appropriate repair inhibitor in order to determine additional DNA damage. None of the inhibitors, used in the present studies cause DNA single strand breaks. The results of the repair studies with the above mentioned inhibitors at 378C and 48C are shown in Fig. 4a and b. The use of the DNA repair inhibitors aphidicolin, araC, hydroxyurea as well as araCrhydroxyurea resulted in a strong increase of the ‘ x 2-

mean of tail moments’ for both temperatures. This indicates that DNA repair took place and could be blocked by these substances. DNA repair of UV-A induced lesions is probably mediated by nucleotide excision repair, because araC as well as aphidicolin blocks DNA polymerases involved in nucleotide excision repair. Base excision repair may be not involved, because the use of methoxyamine revealed no change in the ‘ x 2-mean of tail moments’, compared to experiments without inhibitor Ždata not shown.. 3.4. DNA repair at 48C, 108C, 238C, 378C but not at 448C Fig. 5a shows the results on temperature dependence of DNA repair mechanism. At all investigated temperatures, except 448C repair of UV-A induced DNA lesions was observed. The differences in the ‘ x 2-mean of tail moments’ for temperatures between 48C and 378C are not significant as they are within the errors. At the recovery temperature of 448C the ‘ x 2-mean of tail moments’ is already significantly increased after recovery for 40 min. For 80 min recovery the ‘ x 2-mean of tail moments’ increases by a factor 6 compared to other temperatures. This indicates that cells are heat damaged and viability is decreased, resulting in cell death. DNA is then highly fragmented and large comets, corresponding to high values of the ‘ x 2-mean of tail moments’ are observed during SCGE. Fig. 5b shows DNA damage, described by the ‘ x 2-mean of tail moments’ after 40 and 80 min recovery for all investigated temperatures. There is

Table 1 Inhibitors of DNA repair which were used in the present studies: the inhibited enzymes as well as the efficiency in blocking either nucleotide excision repair ŽNER. or base excision repair ŽBER. are also indicated Inhibitor

Blocked enzymes

Inhibition indicates NER

Inhibition indicates BER

araC

Pol1 a Žstrongly., Pol b Žvery weakly. w23,24x low concentration of dCTP in the cellular precursor pool increases the efficiency of araC ribonucleotide reductase, enzyme responsible for production of deoxyribonucleotides ŽdNTPs. w24x ´ results in an depletion of DNA precursor pool HU increases the effectiveness of araC w24x Pol a , Pol d , Pol ´ w26x reacts with abasic sites in DNA, protecting them from enzymatic incision w27x ´ insensitive to AP-endodeoxyribonucleases

q

y

q

y

qq qq y

y y q

hydroxyurea ŽHU. araCrHU aphidicolin methoxyamine

C. Bock et al.r Mutation Research 408 (1998) 111–120

no temperature dependence of single strand break frequency between 48C and 378C. This indicates that DNA repair is processed with comparable activity all temperatures in this range.

4. Discussion 4.1. Good x 2-fit and linear relationship between radiation fluence and DNA damage Generally Gaussian statistics is used to describe the distribution of tail moments w6,21x. But for small DNA damage, represented by an asymmetric distribution of tail moments the symmetric Gaussian function is not adequate for statistical analysis. For this reason the standard deviations for the mean values of tail moment are very large, resulting in unrealistically large error bars w28x. Occasionally two different Gaussian functions are used to describe this asymmetry w29x. The x 2-function is suitable for asymmetric as well as for symmetric distributions w25x. While the Gaussian distribution is characterised by the mean value and the standard deviation, the x 2-function is described by one single parameter called ‘degree of freedom n’. As this parameter is equivalent to the mean value from Gaussian statistics we termed this the ‘ x 2-mean of tail moments’, as mentioned in Section 2. Because the x 2-function is very good in fitting asymmetric functions the standard error for the ‘ x 2-mean of tail moments’ are comparably small ŽFig. 2.. The standard error is calculated during curve fitting and is a measure for the quality of the fit. Only for the distribution of tail moments of undamaged cells ŽFig. 1a. the x 2-function is less suitable, because for a ‘ x 2-mean of tail moments’ smaller than 2, the x 2-function is infinite for small x values. This results in larger error bars for small ‘ x 2-mean of tail moments’. We obtained a linear relationship between the ‘ x 2-mean of tail moments’ and the radiation fluences ŽFig. 2.. This is in accord with data fitted by the x 2-function by Bauer et al. w25x. The linear dependence between radiation fluence and UV-A radiation induced DNA damage is in accordance with bulk experiments as well as single cell studies of many research groups Žfor example see Refs.

117

w7,11,25x.. This could be explained by the fact that the distribution of DNA damage among individual cells irradiated with UV-A was not more heterogeneous than for unirradiated cells and was independent of the radiation fluence ŽFig. 1.. In this case the experiments with the comet assay revealed similar results as with bulk experiments. In contrast, the repair studies showed a heterogeneous distribution of tail moments, which could lead to different results for single cell and for bulk experiments Žsee Section 4.2.. It has been reported in literature that after treatment with chemical mutagens, the distribution of tail moments is very heterogeneous and dose dependent w15x. 4.2. Single phase DNA repair and non-repairable residual damages Several articles about UV-A radiation induced DNA damage and cellular repair have been published w6,30x, but still little is known about the repair mechanisms of UV-A induced DNA lesions. The present studies revealed a single phase DNA repair within 20 min. The cells have been irradiated for 30 min and incision also occurs at 48C. Thus the breaks induced immediately after UV treatment could be composed by single strand breaks indirectly produced by UV-A light and nucleotide excision repair intermediates. No further decrease in DNA repair between 20 and 100 min was observed and complete rejoining of single strand breaks was not detected ŽFig. 3.. The residual breaks could be alkali-labilesites, which are known to be repaired more slowly than strand breaks w30x. The single cell experiments revealed that during DNA recovery the distribution of the tail moments is very heterogeneous. A distinction between cells, that have started and completed DNA repair and cells that show still DNA damage is possible. Furthermore cells with an immense tail moment that could not be evaluated by the software were neglected in the data analysis. These cells probably indicate dead cells, where strong DNA fragmentation was processed, probably by apoptosis. The percentage of cells with immense tail moment within one sample was equivalent to the percentage of dead cells determined by trypan blue staining preceding the comet assay. These advantages of the single cell technique could lead to more realistic

118

C. Bock et al.r Mutation Research 408 (1998) 111–120

results than with bulk experiments. This could be the reason for the differences in DNA repair kinetics in the present studies and the results reported by Henriksen et al. w6x. They observed exponential single phase kinetics with complete DNA repair after 365 nm irradiation within 120 min by alkaline elution assay w7x. Other authors reported about biphasic and complete DNA repair w30,31x. They suggested the involvement of different types of repair mechanisms. It could be possible that under the present experimental conditions only one type of DNA repair mechanism is activated, represented by the linear decrease in DNA damage. The remaining DNA damage after 20 min recovery, represented by a constant value of DNA damage may indicate that a second repair pathway could not be activated. The cell type could be crucial for this observation. Henriksen et al. used human lymphocytes from peripheral blood and obtained single phase kinetics w7x. Churchill et al. used chinese hamster ovary cells and observed biphasic kinetics w30x. Peak et al. w31x used human epithelioid P3 cells for their studies and also obtained biphasic repair kinetics. 4.3. Repair inhibitors increase DNA single strand breaks Inhibitors, such as aphidicolin and araC have widely been used to study DNA repair after UV-B and UV-C treatment w24,32,33x. Our results showed an increase in DNA strand breaks in the presence of the following repair inhibitors: aphidicolin, araC, hydroxyurea, as well as araCrhydroxyurea ŽFig. 4.. This was already observed after 40 min recovery at 378C as well as at 48C. The increase of single strand breaks could be explained by accumulation of open repair sites after the incision step during the repair process. This is due to the inhibition of DNA polymerases. The present data are in agreement with the data published by Henriksen et al. w7x. Since araC and aphidicolin block nucleotide excision repair, our data indicate that this process is also relevant for repair of UV-A damages Žsee also Ref. w31x.. The present studies with methoxyamine revealed no change in single strand breaks Ždata not shown., indicating that base excision repair is probably not activated in repair of UV-A induced DNA damage. In contrast DNA lesions after exposure to alkylating

agents have been shown by the use of methoxyamine to be repaired via base excision repair w34x. 4.4. Similar DNA repair between 48C and 378C, no repair at 448C Only a few groups have investigated the temperature dependence of DNA repair w4,5x. Our studies revealed similar repair of DNA lesions in the temperature range between 378C and 48C ŽFig. 5b.. The observed repair kinetics at 48C was equal with the one at 378C ŽFig. 3.. Probably enzymes involved in repair of UV-A induced DNA damage show similar activity in the investigated temperature range. It is also possible that cultivated cells have developed the ability to repair DNA damage under extreme temperature conditions. The observation that cultivated cells repair DNA damage at 48C was also reported by Nolan et al. w4x and Dube et al. Žunpublished observations.. Dube et al. Žunpublished observations. obtained similar results for repair studies at 48C with the same cell line. Nolan et al. reported on DNA repair after g-irradiation. In their studies repair activity was slightly reduced in the range of 378C to 208C and remained stable between 208C and 48C w4x. g-ray induced DNA lesions are repaired via other pathways than UV-A induced DNA damage. Therefore the temperature dependence of DNA repair may be different. The present studies with repair inhibitors at 48C showed that inhibition of polymerases revealed a strong increase in DNA strand breaks ŽFig. 4b.. Without repair inhibitors strand break repair was achieved. At 448C no DNA repair was observed. In contrast we obtained a strong increase in DNA single strand breaks. This could be explained by heat damage of cells, resulting in cell death and DNA fragmentation. In conclusion we observed that DNA repair of UV-A induced DNA damage is not temperature dependent in the range from 48C to 378C. Our studies revealed similar repair kinetics at 378C and at 48C. The use of repair inhibitors confirmed DNA repair at 48C. Acknowledgements We are grateful to Dr. Leo Wollweber, Hiltrud Munster, Helga Fritzke, Sabine Hoffmann and Kathe ¨ ¨

C. Bock et al.r Mutation Research 408 (1998) 111–120

Siller for cell culture. This work was supported by the Fond der Chemischen Industrie grant 161971. References w1x J.D. Chien, F.H.J. Yew, DNA replication and repair of tilapia cells: effects of temperature on DNA replication and ultraviolet repair in tilapia ovary cells, Cell Sci. 89 Ž1988. 263–272. w2x C. Delfini, E.V. de Venezia, G. Oberholtzer, C. Tomasello, T. Eremenko, P. Volpe, Cell-cycle dependence and properties of the HeLa cell DNA polymerase system, Proc. Natl. Acad. Sci. U.S.A. 82 Ž1985. 2220–2224. w3x R.P. Virsik-Peukert, D. Harder, Temperature and the formation of radiation-induced chromosome aberrations: II. The temperature dependence of lesion repair and lesion interaction, Int. J. Radiat. Biol. 49 Ž4. Ž1986. 673–681. w4x W.T. Nolan, J.E. Thompson, J.R. Lepock, J. Kruuv, Effect of membrane lipid perturbers on the temperature dependence of repair of sublethal and potentially lethal radiation damage, Int. J. Radiat. Biol. 39 Ž2. Ž1981. 195–205. w5x C. Alapetite, T. Wachter, E. Sage, E. Moustacchi, Fibroblasts exposed to UVC, UVB, UVA and g-rays, Int. J. Radiat. Biol. 69 Ž1996. 359–369. w6x E.K. Henriksen, J. Moan, O. Kaalhus, G. Brunborg, Induction of repair of DNA damage in UV-irradiated human lymphocytes. Spectral differences and repair kinetics, J. Photochem. Photobiol. B 32 Ž1996. 39–48. w7x A. de With, K.O. Greulich, UV-B-laser-induced DNA damage in lymphocytes observed by single-cell gel electrophoresis, J. Photochem. Photobiol. B 24 Ž1994. 47–53. w8x E. Sage, Distribution and repair of photolesions in DNA: genetic consequences and the role of sequence context, Photochem. Photobiol. 57 Ž1993. 163–174. w9x F. Urbach, R.W. Ganga, The Biological Effects of UV-A Radiation, Praeger Publishers, CBS Educational Publishing, 1996. w10x D.L. Croteau, V.A. Bohr, Repair of oxidative damage to nuclear and mitochondrial DNA in mammalian cells, J. Biol. Chem. 272 Ž41. Ž1997. 25409–25412. w11x M.J. Peak, J.G. Peak, B.A. Carnes, Induction of direct and indirect single-strand breaks in human cells DNA by far- and near-ultraviolet radiations: Action spectrum and mechanisms, Photochem. Photobiol. 45 Ž1987. 381–387. w12x J.G. Peak, M.J. Peak, Comparison of initial yields of DNAto-protein crosslinks and single strand breaks induced in cultured human cells by far- and near-ultraviolet light, blue light and X-ray, Mutat. Res. 246 Ž1991. 187–191. ¨ w13x O. Ostling, K.J. Johanson, Microelectrophoretic study of radiation-induced DNA damages in individual mammalian cells, Biochem. Biophys. Res. Commun. 123 Ž1. Ž1984. 291–298. w14x N.P. Singh, D.B. Banner, R.R. Tice, M.T. McCoy, G.D. Collins, E.L. Schneider, Abundant alkali-sensitive sites in DNA of human and mouse sperm, Exp. Cell Res. 184 Ž1989. 461–470. w15x N.P. Singh, T. McCoy, R.R. Tice, E.L. Schneider, A simple

w16x

w17x

w18x

w19x w20x

w21x

w22x

w23x

w24x

w25x

w26x

w27x

w28x

w29x

w30x

119

technique for quantitation of low levels of DNA damage in individual cells, Exp. Cell Res. 175 Ž1988. 184–191. P.L. Olive, D. Wlodek, R.E. Durand, P. Banath, ´ Factors influencing DNA migration from individual cells subjected to gel electrophoresis, Exp. Cell Res. 198 Ž1992. 259–267. V.J. McKelvey-Martin, M.H.L. Green, P. Schmezer, B.L. Pool-Zobel, M.P. De Meo, ´ A. Collins, The single cell gel electrophoresis assay Žcomet assay.: a European review, Mutat. Res. 288 Ž1993. 47–63. R.R. Tice, The single cell gelrcomet assay: a microgel electrophoretic technique for the detection of DNA damage and repair in individual cells, in: D.H. Philips, Svenitt ŽEds.., Environmental Mutagenesis, Bios Scientific Publishers, Oxford, 1994. D.W. Fairbain, P.L. Olive, K. O’Neill, The comet assay: a comprehensive review, Mutat. Res. 339 Ž1995. 37–59. C.M. Gedik, S.W.B. Ewen, A.R. Collins, Single-cell gel electrophoresis applied to the analysis of UV-C damage and its repair in human cells, Int. J. Radiat. Biol. 62 Ž3. Ž1992. 313–320. G. Speit, A. Hartmann, The contribution of excision repair to the DNA effects seen in the alkaline single cell gel test Žcomet assay., Mutagenesis 10 Ž6. Ž1995. 555–559. A.R. Collins, M. Ai-guo, S.J. Duthie, The kinetics of repair of oxidative DNA damage Žstrand breaks and oxidised pyrimidines. in human cells, Mutat. Res. 336 Ž1995. 69–77. C.M. Gedik, A.R. Collins, The mode of action of 1-b-Darabinofuranosyl-cytosine in inhibiting DNA repair: new evidence using a sensitive assay for repair DNA synthesis and ligation in permeable cells, Mutat. Res. DNA Repair 254 Ž1991. 231–237. A. Collins, D.J. Oates, Hydroxyurea: effects on hydroxyribonucleotide pool sizes correlated with effects on DNA repair in mammalian cells, Eur. J. Biochem. 169 Ž1987. 299–305. E. Bauer, R.-D. Recknagel, U. Fiedler, L. Wollweber, C. Bock, K.O. Greulich, The distribution of the tail moments in single cell electrophoresis Žcomet assay. obeys a chi-square Ž x 2 . not a Gaussian distribution, Mutat. Res. Ž1998. 398r12, 101–110. R.D. Wood, M.K.K. Shivji, Which polymerases are used for DNA-repair in eukaryotes, Carcinogenesis 18 Ž4. Ž1997. 605–610. M. Luizzi, M. Talpaert-Borle, ´ A new approach to the study of the base-excision repair pathway using methoxyamine, J. Biol. Chem. 260 Ž9. Ž1985. 5252–5258. A. Dube, C. Bock, E. Bauer, R. Kohli, P.K. Gupta, K.O. Greulich, Low intensity He–Ne Laser Ž632.8 nm. pre-irradiation decreases UVA-induced DNA damage in human Blymphoblast cells, J. Photochem. Photobiol. B Ž1998. submitted. P.L. Olive, M.R. Horsman, C. Grau, J. Overgaared, Detection of hypoxic cells in a C3H mouse mammary carcinoma using the comet assay, Br. J. Cancer 76 Ž6. Ž1997. 694–699. M.E. Churchill, J.G. Peak, M.J. Peak, Correlation between cell survival and DNA single strand break repair proficiency in the chinese hamster ovary cell lines AA8 and EM9

120

C. Bock et al.r Mutation Research 408 (1998) 111–120

irradiated with 365 nm ultraviolet-A radiation, Photochem. Photobiol. 53 Ž1991. 229–236. w31x J.G. Peak, B. Pilas, E.J. Dudek, M.J. Peak, DNA breaks caused by monochromatic 365 nm ultraviolet-A radiation or hydrogen peroxide and their repair in human epithelioid and xeroderma pigmentosum cells, Photochem. Photobiol. 54 Ž1991. 197–203. w32x A.R.S. Collins, S. Squires, R.T. Johnson, Inhibition of repair DNA synthesis, Nucleic Acid Res. 10 Ž4. Ž1982. 1203–1213.

w33x A. de With, K.O. Greulich, Wavelength dependence of laser-induced DNA damage in lymphocytes observed by single-cell gel electrophoresis, J. Photochem. Photobiol. B 99 Ž1995. 1–6. w34x C. Buschfort, M.R. Muller, S. Seeber, M.F. Rajewsky, J. ¨ Thomale, DNA excision repair profiles of normal and leukemic human lymphocytes: functional analysis at the single cell level, Cancer Res. 57 Ž1987. 651–658.