Mutagenesis vol. 26 no. 6 pp.

689695, 2011 Advance Access Publication 21 July 2011

doi:10.1093/mutage/ger034

The effects on DNA migration of altering parameters in the comet assay protocol such as agarose density, electrophoresis conditions and durations of the enzyme or the alkaline treatments

ller* Clara Ersson and Lennart Mo

Department of Biosciences and Nutrition, Karolinska Institutet, SE-141 83 Huddinge, Sweden.
* To whom correspondence should be addressed. Tel: 46 8 524 810 75; Fax: 46 8 774 55 38; Email: lennart.moller@ki.se

Received on February 22, 2011; revised on May 13, 2011; accepted on May 17, 2011

The single cell gel electrophoresis (comet assay) is a popular method for measuring DNA migration as an estimate of DNA damage. No standardised comet assay protocol exists, which make comparisons between studies complicated. In a previous inter-laboratory validation study of the comet assay, we identied important parameters in the protocol that might affect DNA migration. The aim of this study was to assess how different comet assay protocols affect DNA migration. The results in this study suggest that (i) there is a signicant linear doseresponse relationship between the agarose gels density and DNA migration and that damaged cells are more sensitive to the agarose gels density; (ii) incubation with formamidopyrimidine DNA glycosylase for 10 min is inadequate, whereas 30 min is sufcient; (iii) the typically used 20 min of alkaline treatment might be to short when analysing samples that contain particular alkali-labile sites (ALS) and (iv) the duration of electrophoresis as well as the strength of the electric eld applied affects the DNA migration. By using protocol-specic calibration curves, it is possible to reduce the variation in DNA migration caused by differences in comet assay protocols. This does, however, not completely remove the impact of the durations of alkaline treatment and electrophoresis when analysing cells containing ALS that are relatively resistant to high alkaline treatment.

activity or by the subsequent alkaline treatment. By their ability to relax DNA supercoiling, DNA breaks allow the DNA to migrate out of the nucleoids under the inuence of an electric eld. A comet-like image is created with a tail consisting of relaxed and mainly single-stranded DNA. Several guidelines for the comet assay have been published (25), but no standardised protocol exists and there are considerable differences in the protocols used by different research groups. These differences negatively affect inter-laboratory comparisons of results. In a previous inter-laboratory validation study of the comet assay, we identied important parameters in the comet assay procedure that might affect the DNA migration, such as the agarose density, duration of the enzyme treatment, duration of the alkaline treatment and duration of the electrophoresis (6). The duration of the alkaline treatment as well as the duration of the electrophoresis have been shown to signicantly affect the level of DNA migration in c-irradiated cells (7,8). Forchhammer et al. (7) concluded that differences could be reduced to a large extent by using protocol-specic calibration curves. In this study, the impact on DNA migration by different comet assay protocols was investigated. Key parameters in the comet assay protocol including agarose density, durations of enzyme treatment, alkaline treatment and electrophoresis and strength of the electric eld were altered and the impact on DNA migration was assessed.

Downloaded from http://mutage.oxfordjournals.org/ by guest on December 17, 2013

Materials and methods

Study design Cells treated with Ro 19-8022 plus visible light were analysed with the FPG comet assay, whereas c-irradiated cells and cells treated with hydrogen peroxide (H2O2) were analysed with the alkaline comet assay. Only one parameter at a time in a standard comet assay protocol was altered: (i) the agarose density, (ii) the duration of the enzyme treatment, (iii) the duration of the alkaline treatment, (iv) the duration of the electrophoresis or (v) the strength of the electric eld during electrophoresis. Each experiment was repeated three to four times. This study did not take into account the effect of these parameters on the detection of cross links. The selected densities of the agarose gels, the durations of the enzyme treatment, the alkaline treatment and the electrophoresis as well as the strengths of the electric eld were to a large extent based on the protocols used by our collaborators in the European Comet Assay Validation group (ECVAG) (6). FPG was kindly provided by Prof. A.R. Collins (Department of Nutrition, University of Oslo, Norway). Preparation of cells containing DNA damage c-Irradiation. Human acute monocytic leukaemia THP1 cells in phosphatebuffered saline (PBS) were irradiated with c-rays at 0, 2.5, 5 and 7.5 Gy as previously described (7). The aliquots, predominately containing DNA breaks, were frozen slowly down to 80C in freezing medium [50% foetal bovine serum (FBS), 40% RPMI 1640 medium and 10% dimethyl sulfoxide (DMSO)]. H2O2 treatment. A549 human type II alveolar epithelial cells were cultured and harvested as previously described (9). The cells were exposed to 200 lM H2O2 in PBS for 5 min on ice. Following exposure, the cells were washed with PBS, isolated by trypsination (5 min) and re-suspended in cell medium. After centrifugation, the cell pellet was suspended in freezing medium (50%

Introduction The single cell gel electrophoresis (comet assay) is a frequently used method to measure DNA migration as an estimate of DNA damage. With the alkaline comet assay (pH . 13), DNA breaks, i.e. strand breaks and alkali-labile sites (ALS) are measured. The increasing use of the comet assay is attributed to the low number of cells required, its sensitivity in detecting low levels of DNA damage and the ability to measure different types of DNA damage by adding lesion-specic enzymes (1). In the formamidopyrimidine DNA glycosylase (FPG) comet assay, the level of 8-oxoguanine as well as other altered purines is measured by incorporating a digestion with the enzyme FPG (1). FPG recognises and removes specic oxidated purines creating apurinic (AP) sites. The AP sites are transformed to DNA strand breaks either by the enzymes associated lyase

The Author 2011. Published by Oxford University Press on behalf of the UK Environmental Mutagen Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

689

 ller C. Ersson and L. Mo Dulbeccos modied Eagles medium, 40% FBS and 10% DMSO) and aliquots were frozen slowly down to 80C in a polystyrene freezing box. Exposure to photosensitiser and light. A549 cells were treated with 0.6 lM photosensitiser Ro 19-8022 in PBS and visible light, which predominantly generates 8-oxoguanine (10). Cells were irradiated on ice for 5 min from a distance of 33 cm with a 500-W tungsten halogen lamp. Following irradiation, cells were immediately washed with PBS, isolated by trypsination (5 min) and re-suspended in cell medium. Cells were centrifuged and cell pellets were suspended in freezing medium and frozen slowly down to 80C in a polystyrene freezing box. Experimental set-up Standard comet assay protocol. Cells were embedded in 0.7% low melting point agarose (nal concentration) and then lysed (2.5 M NaCl, 0.1 M EDTA, 10 mM Tris, pH 10) for a minimum of 1 h at 4C. In the FPG comet assay, the samples were washed three times for 5 min in buffer (40 mM HEPES, 0.1 M KCl, 0.5 mM EDTA, 0.2 mg/ml bovine serum albumin, pH 7.9, 4C) and then incubated with only buffer or buffer containing FPG for 30 min at 37C. This enabled an estimation of both DNA breaks and calculation of net FPG-sensitive sites. After FPG treatment, the samples were placed in an alkaline solution (0.3 M NaOH, 1 mM EDTA, pH . 13) for 40 min at 4C. In the alkaline comet assay (where only DNA breaks are measured), the samples were placed in the alkaline solution directly after lysis. Alkaline treatment was followed by electrophoresis in the same solution for 30 min (1.15 V/cm, 300 mA). The DNA was stained with ethidium bromide and comets with tails of mainly single-stranded DNA could then be seen with a uorescence microscope. The comets were analysed using computerised image analysis (Komet 4.0; Kinetic Imaging Ltd, Bromborough, UK). The magnitude of the DNA tail (% DNA in the tail) provides information about the level of DNA lesions. One parameter at a time of the comet assay protocol was altered in each experimental set-up. Each experimental set-up was repeated three to four times, and each time, 100 cells were scored (50 cells from each of two elds on a microscope slide) and the mean value reported. Agarose density. To investigate whether the agarose gels density affects the level of DNA migration, cells irradiated with different doses of c-radiation (0, 2.5, 5 and 7.5 Gy) and cells exposed to H2O2 were analysed with the alkaline comet assay using different agarose densities (i.e. 0.4, 0.7, 0.9 and 1.3%, of nal agarose concentrations). Duration of enzyme treatment. To assess how the duration of the enzyme treatment affects DNA migration, cells treated with Ro 19-8022 plus visible light were analysed with the FPG comet assay with enzyme treatment for 10, 30 and 45 min. Duration of alkaline treatment. To investigate the impact of the alkaline treatment, DNA migration was analysed in cells treated with (i) 0 Gy of c-radiation, (ii) 5 Gy of c-radiation, (iii) H2O2 and (iv) Ro 19-8022 plus visible light, with either the alkaline comet assay or the FPG comet assay applying 20, 40 and 60 min of alkaline treatment. Duration of electrophoresis. To investigate how the duration of the electrophoresis affects the DNA migration, cells irradiated with c-radiation (0, 2.5, 5 and 7.5 Gy) and cells treated with H2O2 were analysed with the alkaline comet assay with 20 and 30 min electrophoresis. Strength of the electric eld during electrophoresis. To assess how the strength of the electric eld affects the DNA migration during electrophoresis, cells irradiated with c-radiation (0, 2.5, 5 and 7.5 Gy) and cells treated with H2O2 were analysed with the alkaline comet assay using different strengths of the electric eld (0.70, 1.15 and 1.6 V/cm). Statistics Statistical analyses were performed using PASW Statistics 18.0 (SPSS Inc., Chicago, IL, USA). Linear doseresponse relationships were analysed with analysis of variance (ANOVA) and Pearsons correlation coefcients (r) are reported. Differences between doseresponse curves were analysed with twoway ANOVA and whether the slopes of the curves were signicantly different was assessed. All other comparisons of the impact of different protocol details were carried out with Students t-test. Normal distribution was assessed with the ShapiroWilk test and homogeneity of variances was assessed with Levenes test and it was concluded that parametric tests were appropriate. Pvalues ,0.05 were considered statistically signicant. The unit of variability was the experiment. Fig. 1. DNA migration in untreated cells, i.e. cells exposed to 0 Gy (lled diamonds) and cells exposed to 5 Gy (lines) of c-radiation measured by the alkaline comet assay using different agarose densities. Each increase in agarose density caused a signicantly decreased DNA migration in cells irradiated with 5 Gy (Students t-test, where *, ** and *** means P , 0.05, P , 0.01 and P , 0.001, respectively). The linear doseresponse relationships were signicant both for untreated cells (r 5 0.71, P , 0.01, ANOVA) and cells exposed to 5 Gy of c-radiation (r 5 0.95, P , 0.001, ANOVA). The slopes of the two regression curves were signicantly different (P , 0.001) from each other. The experiment was repeated four times. Each lled diamond/line represents the mean value of 100 scored cells (50 cells from each of two separate gels).

Results Agarose density Each increase in agarose density caused a signicant decrease in DNA migration in cells irradiated with 5 Gy of c-radiation as shown in Figure 1. The linear doseresponse relationship (r 5 0.950) was highly signicant (P , 0.001). No signicant differences in DNA migration were observed in the untreated cells (0 Gy of c-irradiation) when comparing different agarose densities with Students t-test. However, the linear doseresponse relationship (r 5 0.709) for the untreated cells was signicant (P , 0.01, ANOVA). The slopes of the two regression curves in Figure 1 were signicantly different (P , 0.001) from each other. Duration of enzyme treatment The duration of the incubation with FPG signicantly affected the DNA migration (net FPG-sensitive sites) when analysing cells treated with Ro 19-8022 plus visible light. As shown in Figure 2, the DNA migration was signicantly increased when incubation with FPG was performed for both 30 and 45 min compared to for 10 min (P , 0.05). The DNA migration was not signicantly affected by prolonging the incubation time with FPG from 30 to 45 min. Duration of alkaline treatment The DNA migration was not signicantly affected by the duration of the alkaline treatment in untreated cells (0 Gy of c-irradiation) (Figure 3A).

Downloaded from http://mutage.oxfordjournals.org/ by guest on December 17, 2013

690

Comet assay protocol affects DNA migration

Downloaded from http://mutage.oxfordjournals.org/ by guest on December 17, 2013

Fig. 2. DNA migration (DNA breaks 5 lled diamonds and net FPG-sensitive sites 5 lines) in cells treated with Ro 19-8022 plus visible light measured by the FPG comet assay. The DNA migration was signicantly increased when incubation with FPG was performed for 30 or 45 min compared to for 10 min (P , 0.05, Students t-test). The DNA migration was not signicantly affected by prolonging the incubation with FPG from 30 to 45 min. The experiment was repeated three times. Each lled diamond/line represents the mean value of 100 scored cells (50 cells from each of two separate gels).

The DNA migration increased signicantly when the duration of alkaline treatment was increased from 20 to 40 min, from 40 to 60 min (P , 0.05) and from 20 to 60 min (P , 0.01) in cells irradiated with 5 Gy of c-radiation (Figure 3B). The linear doseresponse relationship (r 5 0.969) was highly signicant (P , 0.001). The slopes of the regression curves for untreated cells and cells irradiated with 5 Gy of c-radiation were signicantly different from each other (P , 0.001). In cells treated with H2O2, the DNA migration signicantly increased when the duration of alkaline treatment was prolonged from 20 to 40 min (P , 0.05) and from 20 to 60 min (P , 0.001) as shown in Figure 3C. When increasing the duration of alkaline treatment from 40 to 60 min, the DNA migration was not signicantly affected (P 5 0.056). The linear doseresponse relationship (r 5 0.954) was highly signicant (P , 0.001). The slopes of the regression curves for untreated cells and cells treated with H2O2 were signicantly different from each other (P , 0.001). Both DNA breaks and net FPG-sensitive sites were measured with the FPG comet assay in cells treated with Ro 19-8022 plus visible light using different durations of the alkaline treatment as shown in Figure 3D. The DNA migration (net FPG-sensitive sites) was not signicantly affected by increasing the alkaline treatment time from 20 to 40 min, whereas a further increase to 60 min caused a signicant increase in DNA migration (P , 0.05). The linear dose response relationship (r 5 0.784) was signicant (P , 0.01). When measuring the DNA migration without FPG (i.e. the DNA breaks) in cells treated with Ro 19-8022 plus visible light, no effect by the duration of alkaline treatment was observed when analysed with Students t-test. However, the

Fig. 3. DNA migration (DNA breaks 5 lled diamonds and net FPG-sensitive sites 5 lines) measured with the comet assay with 20, 40 and 60 min of alkaline treatment. (A) The DNA migration was not signicantly affected by the duration of the alkaline treatment in untreated cells, i.e. cells irradiated with 0 Gy of c-radiation (Students t-test). The experiment was repeated three times. (B) The DNA migration increased signicantly when the duration of the alkaline treatment was prolonged from 20 to 40 min, from 40 to 60 min (P , 0.05) and from 20 to 60 min (P , 0.01) in cells irradiated with 5 Gy of c-radiation (Students t-test). The linear dose response relationship (r 5 0.97) was highly signicant (P , 0.001, ANOVA). The slopes of the regression curves for untreated cells and cells irradiated with 5 Gy of c-radiation were signicantly different from each other (P , 0.001). The experiment was repeated three times. (C) In cells treated with H2O2, the DNA migration signicantly increased when the duration of alkaline treatment was prolonged from 20 to 40 min (P , 0.05) and from 20 to 60 min (P , 0.001, Students t-test). When increasing the duration of alkaline treatment from 40 to 60 min, the DNA migration was not signicantly affected (P 5 0.056, Students t-test). The linear doseresponse relationship (r 5 0.95) was highly signicant (P , 0.001, ANOVA). The slopes of the regression curves for untreated cells and cells treated with H2O2 were signicantly different from each other (P , 0.001). The experiment was repeated three times. (D) When analysing the DNA migration (net FPG-sensitive sites) in cells treated with Ro 19-8022 plus visible light with the FPG comet assay, the DNA migration was not signicantly affected by increasing the alkaline treatment from 20 to 40 min, whereas an increase to 60 min signicantly (P , 0.05, Students t-test) affected the DNA migration. The linear doseresponse relationship (r 5 0.78) was signicant (P , 0.01, ANOVA). The slopes of the regression curves for DNA breaks and net FPG-sensitive sites were not signicantly different from each other. The experiment was repeated four times. Each lled diamond/line represents the mean value of 100 scored cells (50 cells from each of two separate gels).

linear doseresponse relationship (r 5 0.651) was signicant (P , 0.05, ANOVA). The slopes of the regression curves for DNA breaks and net FPG-sensitive sites were not signicantly different from each other (Figure 3D). 691

 ller C. Ersson and L. Mo

Duration of electrophoresis The DNA migration in cells irradiated with 0, 2.5, 5 and 7.5 Gy of c-radiation measured with the alkaline comet assay with 20 or 30 min electrophoresis are shown in Figure 4. The level of DNA migration was signicantly increased when running electrophoresis for 30 min compared to 20 min (P , 0.001, two-way ANOVA) and the slopes of the two regression curves were signicantly different from each other (P , 0.001). The linear doseresponse relationships (r 5 0.989 for 20 min and r 5 0.992 for 30 min, respectively) were highly signicant for both curves (P , 0.001). In cells treated with H2O2, the DNA migration was signicantly increased when the samples had been subject to 30 min compared to 20 min electrophoresis (P , 0.01) as shown in Figure 5A. After transforming the DNA migration into DNA breaks per 106 base pairs by using the calibration curves in Figure 4 as previously described (6), the difference in the number of breaks detected was not as great, although still signicant (P , 0.05) (Figure 5B). The strength of the electric eld during electrophoresis The DNA migration was signicantly affected by the strength of the electric eld during electrophoresis in cells irradiated with different doses of c-radiation measured with the alkaline comet assay (Figure 6). The DNA migration was signicantly increased (P , 0.001, two-way ANOVA) when comparing 1.60 with 1.15 V/cm and when comparing 1.15 with 0.70 V/ cm. The linear doseresponse relationships (r 5 0.991, 0.985 and 0.993 for 0.70, 1.15 and 1.60 V/cm, respectively) were highly signicant for all curves (P , 0.001). The slopes of the

regression curves were all signicantly different from each other (P , 0.001). Untreated cells were not signicantly affected by the strength of the electric eld, whereas cells irradiated with 2.5, 5 and 7.5 Gy of c-radiation were

Downloaded from http://mutage.oxfordjournals.org/ by guest on December 17, 2013

Fig. 5. DNA migration in cells treated with H2O2 analysed with the alkaline comet assay. (A) There was a signicantly increased level of DNA migration in cells that had been subject to 30 min compared to 20 min electrophoresis (P , 0.01, Students t-test). (B) After transforming the DNA migration into DNA breaks (lesions per 106 base pairs) by using the calibration curves in Figure 4, the difference in the number of breaks detected was not as great, although still signicant (P , 0.05, Students t-test). The experiment was repeated four times. Each lled diamond represents the mean value of 100 scored cells (50 cells from each of two separate gels).

Fig. 4. DNA migration in cells irradiated with different doses of c-radiation measured with the alkaline comet assay with 20 or 30 min of electrophoresis. The DNA migration was signicantly increased (P , 0.001, two-way ANOVA) when running electrophoresis for 30 min (lines) compared to 20 min (lled diamonds), and the slopes of the two regression curves were signicantly different from each other. The linear doseresponse relationships for both curves were highly signicant (r 5 0.99, P , 0.001, ANOVA). The experiment was repeated four times. Each lled diamond/line represents the mean value of 100 scored cells (50 cells from each of two separate gels).

Fig. 6. DNA migration in cells irradiated with different doses of c-radiation measured with the alkaline comet assay using different strengths of the electric eld (0.70, 1.15 and 1.60 V/cm) during electrophoresis. The DNA migration was signicantly increased (P , 0.001, two-way ANOVA) when comparing 1.60 (closed circles) to 1.15 V/cm (lines) and when comparing 1.15 to 0.70 V/ cm (lled diamonds) and the slopes of the regression curves were signicantly different from each other (P , 0.001). The linear doseresponse relationships (r 5 0.99 for all strengths of the electric elds) were highly signicant for all curves (P , 0.001, ANOVA). The experiment was repeated three times. Each lled diamond/line/closed circle represents the mean value of 100 scored cells (50 cells from each of two separate gels).

692

Comet assay protocol affects DNA migration

Fig. 7. DNA damage in cells treated with H2O2 analysed with the alkaline comet assay using different strengths of the electric eld (0.70, 1.15 and 1.60 V/cm). (A) The DNA migration was signicantly increased when applying 1.60 V/cm when compared with 1.15 V/cm (P , 0.05, Students t-test) and 0.70 V/cm (P , 0.01, Students t-test). When comparing 1.15 with 0.70 V/cm, the DNA migration was not signicantly affected (P 5 0.050, Students t-test). (B) After transforming the DNA migration into DNA breaks (lesions per 106 base pairs) by using the calibration curves in Figure 6, there were no signicant differences in DNA breaks when comparing 0.70 and 1.15 V/cm or 1.15 and 1.60 V/cm (Students t-test). When comparing 0.70 with 1.60 V/cm, the amounts of DNA breaks were still signicantly different (P , 0.05, Students t-test). The experiment was repeated three times. Each lled diamond represents the mean value of 100 scored cells (50 cells from each of two separate gels).

signicantly affected by each change in the strength of the electric eld (Students t-test). The strength of the electric eld during electrophoresis signicantly affected the DNA migration in cells treated with H2O2 analysed with the alkaline comet assay as shown in Figure 7A. There was no statistically signicant difference in DNA migration when using 1.15 V/cm compared to 0.70 V/cm (P 5 0.050). The DNA migration was signicantly increased when using 1.60 V/cm compared with 1.15 V/cm (P , 0.05) and 0.70 V/cm (P , 0.01). The linear doseresponse relationship (r 5 0.964) was highly signicant (P , 0.001). After transforming the DNA migration into DNA breaks (lesions per 106 base pairs) by using the calibration curves in Figure 6, no signicant difference remained between samples that had been subject to 0.70 and 1.15 V/cm or 1.15 and 1.60 V/cm (Figure 7B). There was, however, a signicant impact of the strength on the electric eld when comparing 0.70 with 1.60 V/cm. The linear doseresponse relationship (r 5 0.720) was signicant (P , 0.05). Discussion Comet assay is an increasingly used popular method to measure DNA damage (1113). No standardised protocol exists and due to considerable differences in protocols, it is complicated to compare different studies. The inuence on DNA migration by applying different protocols has been discussed previously (14). In the present study, the impact on DNA migration by some key parameters in the comet assay protocol was investigated. The DNA migration was affected by the density in the agarose gel both in untreated cells and cells exposed to 5 Gy of c-radiation. The damaged cells were more sensitive to

alterations in agarose density, which indicated that the sensitivity of the comet assay could be modied by altering the agarose concentration (Figure 1). The data in the present study suggest that 10 min of incubation with FPG is inadequate but that 30 min is sufcient for the enzyme to nd all oxidative lesions (Figure 2). Unpublished data from our laboratory suggest that this also is the case for human 8-oxoguanine-DNA glycosylase 1 (hOGG1), the human parallel to FPG (J. Kain, personal communication). It is essential that investigators make sure that they use enzyme incubation conditions where the saturation level is achieved (by using an appropriate concentration of the enzyme and an appropriate duration of the enzyme treatment). ALS are transformed to DNA strand breaks during alkaline treatment (15). Typically, 20- or 40-min alkaline treatment is applied in the comet assay (6). Tice et al. (5) stated that 20-min alkaline unwinding is considered sufcient for most purposes. c-Irradiated cells have previously been shown to be sensitive to the duration of alkaline treatment (7,8,16), which is in accordance with the ndings in the present study (Figure 3B). Vijayalaxmi et al. (8) speculate that this is due to radiationinduced ALS that are relatively resistant to high alkaline treatment. In the present study, cells treated with H2O2 were also observed to be sensitive to alkaline treatment (Figure 3C), which suggests that these samples also contain ALS that are relatively resistant to high alkaline treatment. Untreated cells were not signicantly affected by the duration of the alkaline treatment (Figure 3A). In addition, the signicantly different slopes of the untreated and treated cells indicate that the sensitivity of the method could be improved by altering the duration of the alkaline treatment. When investigating cells treated with Ro 198022 plus visible light, no statistically signicant impact on DNA migration (net FPG-sensitive sites) was observed when increasing the alkaline treatment from 20 to 40 min (Figure 3D). Taken together, these results suggest that the effect on DNA migration by the duration of alkaline treatment depends on the type of DNA damage measured. Typically used durations of alkaline treatment might be inadequate, provided that one wants to measure ALS. In order to be able to adjust the comet assay protocol to the type of damage being investigated, a thorough investigation of how different DNA damages are affected by alkaline treatment is required. In addition, Speit and Hartmann (17) suggested that durations of alkaline treatment and electrophoresis should be adjusted to obtain valid and reproducible results for each cell type. The duration of the electrophoresis signicantly affected the DNA migration in c-irradiated cells (Figure 4), which supports previous ndings by Forchhammer et al. (7). Electrophoresis is performed in the same solution as alkaline treatment. It can be speculated that the additional 10 min that the samples are subject to alkaline solution contributes to an increased conversion of ALS with a relatively high resistance to alkaline treatment into DNA strand breaks. The signicantly steeper slope of the 30-min electrophoresis calibration curve in Figure 4 supports this hypothesis. If a sample irradiated with a higher dose of c-radiation contains more ALS, this sample will be more sensitive to the duration of alkaline treatment. Thirty minutes compared to 20-min electrophoresis also leads to a signicantly increased DNA migration in cells treated with H2O2 (Figure 5A). After transforming the DNA migration into DNA breaks per 106 base pairs, the impact of the duration of electrophoresis was not as great although the difference in the number of breaks detected was still signicant (Figure 5B). As 693

Downloaded from http://mutage.oxfordjournals.org/ by guest on December 17, 2013

 ller C. Ersson and L. Mo

shown in Figure 3, the duration of alkaline treatment affected both H2O2-treated cells and c-irradiated cells suggesting that both treatments cause formation of ALS that are relatively resistant to high alkaline treatment. Different proportions of such ALS in the c-irradiated cells and H2O2-treated cells could be an explanation to why the impact of the duration of electrophoresis was not completely removed by using protocolspecic calibration curves. The strength of the electric eld signicantly affected the level of DNA migration in both c-irradiated cells (Figure 6) and H2O2-treated cells (Figure 7A). When using the calibration curves in Figure 6 to convert the DNA migration into DNA breaks (lesions per 106 base pairs) in cells treated with H2O2, the impact of the strength of the electric elds was decreased (Figure 7B). Both high and low strengths of the electric eld can cause practical problems. Highly damaged cells could end up outside the methods range of linear detection (3) when a strong electric eld is applied together with a sensitive protocol. Another problem according to our experience is that when only a small fraction of the DNA is in the comet head, the computer programme might have difculties to dene the comet head. When using a weak electric eld, the DNA fragments might not migrate long enough, i.e. some DNA fragments might not have migrated out from the comet head leading to an underestimation of DNA damage. The significantly different slopes of the regression curves for the different strengths of the electric eld in Figure 6 indicate that the sensitivity of the assay can be improved by altering the strength of the electric eld. Forchhammer et al. (7) have shown previously that it is possible to reduce the variation in DNA migration caused by using different comet assay protocols by using protocolspecic calibration curves. Observations in previous studies performed by ECVAG (6,18) as well as in the present study support this nding. In addition to facilitating comparisons between studies that have used different protocols, a real unit such as lesions per 106 base pairs is easier to understand for individuals who are not familiar with the comet assay compared to primary comet assay end points such as for example %DNA in tail. In the present study, we observed that the transformation of DNA migration into DNA breaks (lesions per 106 base pairs) by using calibration curves consisting of c-irradiated cells is not without problems. ALS with a relatively high resistance to alkaline treatment in c-irradiated cells and/or the cells being investigated could complicate the transformation of DNA migration into DNA breaks. Conversion of primary comet assay end points into a real unit by calibration with ionising radiation is useful, but the method has some problems that need to be considered. Whether results in different studies could be directly compared when a standardised protocol is used needs to be elucidated. ECVAG is currently performing a study where we use a reference protocol to enable comparisons between laboratories without the impact of some critical protocol differences. One should, however, keep in mind that it might be appropriate to adjust the comet assay protocol to the type of DNA damage and cell type being investigated. In conclusion, key parameters in the comet assay protocol including (i) agarose density, (ii) duration of enzyme treatment, (iii) duration of alkaline treatment, (iv) duration of electrophoresis and (v) strength of the electric eld during electrophoresis affect DNA migration. By using protocol-specic calibration curves to transform the DNA migration into strand breaks per 694

106 base pairs, it is possible to reduce the variation in DNA damage caused by using different comet assay protocols. Transformation did not completely remove the variation in DNA migration, which could possibly be explained by the presence of ALS that are relatively resistant to high alkaline treatment in cells treated with c-radiation or H2O2. The results presented in this paper indicate that the sensitivity of the comet assay can be optimised in relation to the specic performed study by altering different parameters in the assay protocol such as the agarose concentration, the duration of the alkaline treatment, the duration of the electrophoresis and the strength of the electric eld. Funding Environmental Cancer Risk, Nutrition and Individual Susceptibility, a network of excellence operating within the European Union 6th Framework Program, Priority 5: Food Quality and Safety (Contract No. 513943); the Swedish Research Council (Vetenskapsra det); Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (FORMAS). Acknowledgements
L. Forchhammer (Institute of Public Health, University of Copenhagen, Denmark) is acknowledged for preparing the c-irradiated cells. J. Kain (Department of Biosciences and Nutrition, Karolinska Institutet, Sweden) is acknowledged for blinding the slides. The photosensitiser Ro 19-8022 was a generous gift from F. Hoffmann-La Roche, Basel, Switzerland. The authors are members of ECVAG which was created within the ECNIS network of excellence in order to validate the comet assay. Conict of interest statement: None declared.

Downloaded from http://mutage.oxfordjournals.org/ by guest on December 17, 2013

References
1. Collins, A. R. (2004) The comet assay for DNA damage and repair: principles, applications, and limitations. Mol. Biotechnol., 26, 249261. 2. Albertini, R. J., Anderson, D., Douglas, G. R. et al. (2000) IPCS guidelines for the monitoring of genotoxic effects of carcinogens in humans. Mutat. Res., 463, 111172. 3. Collins, A. R. (2011) The use of bacterial repair endonucleases in the comet assay. Methods Mol. Biol., 691, 137147. 4. Hartmann, A., Agurell, E., Beevers, C. et al. (2003) Recommendations for conducting the in vivo alkaline Comet assay. Mutagenesis, 18, 4551. 5. Tice, R. R., Agurell, E., Anderson, D. et al. (2000) Single cell gel/comet assay: guidelines for in vitro and in vivo genetic toxicology testing. Environ. Mol. Mutagen., 35, 206221. 6. Johansson, C., Mller, P., Forchhammer, L. et al. (2010) An ECVAG trial on assessment of oxidative damage to DNA measured by the comet assay. Mutagenesis, 25, 125132. 7. Forchhammer, L., Bra uner, E. V., Folkmann, J. K., Danielsen, P. H., Nielsen, C., Jensen, A., Loft, S., Friis, G. and Mller, P. (2008) Variation in assessment of oxidatively damaged DNA in mononuclear blood cells by the comet assay with visual scoring. Mutagenesis, 23, 223231. 8. Vijayalaxmi, Tice, R. R. and Strauss, G. H. (1992) Assessment of radiationinduced DNA damage in human blood lymphocytes using the single-cell gel electrophoresis technique. Mutat. Res., 271, 243252. 9. Johansson, C., Rytter, E., Nygren, J., Vessby, B., Basu, S. and Mo ller, L. (2009) Down-regulation of oxidative DNA lesions in human mononuclear cells after antioxidant supplementation correlates to increase of gammatocopherol. Int. J. Vitam. Nutr. Res., 78, 183194. 10. Will, O., Gocke, E., Eckert, I., Schulz, I., Paum, M., Mahler, H. C. and Epe, B. (1999) Oxidative DNA damage and mutations induced by a polar photosensitizer, Ro19-8022. Mutat. Res., 435, 89101. 11. Dusinska, M. and Collins, A. R. (2008) The comet assay in human biomonitoring: gene-environment interactions. Mutagenesis, 23, 191205. 12. Fairbairn, D. W., Olive, P. L. and ONeill, K. L. (1995) The comet assay: a comprehensive review. Mutat. Res., 339, 3759. 13. Karlsson, H. L. (2010) The comet assay in nanotoxicology research. Anal. Bioanal. Chem., 398, 651666.

Comet assay protocol affects DNA migration 14. Tice, R. (1995) The single cell gel/comet assay: a microgel electrophoretic technique for the detection of DNA damage and repair in individual cells. In Phillips, D. H. and Venitt, S. (eds), Environmental Mutagenesis. BIOS Scientic Publishers, Oxford, pp. 315339. 15. Kohn, K. W. (1991) Principles and practice of DNA lter elution. Pharmacol. Ther., 49, 5577. 16. Speit, G., Trenz, K., Schu tz, P., Rothfuss, A. and Merk, O. (1999) The inuence of temperature during alkaline treatment and electrophoresis on results obtained with the comet assay. Toxicol. Lett., 110, 7378. 17. Speit, G. and Hartmann, A. (2006) The comet assay: a sensitive genotoxicity test for the detection of DNA damage and repair. Methods Mol. Biol., 314, 275286. 18. Forchhammer, L., Johansson, C., Loft, S. et al. (2010) Variation in the measurement of DNA damage by comet assay measured by the ECVAG inter-laboratory validation trial. Mutagenesis, 25, 113123.

Downloaded from http://mutage.oxfordjournals.org/ by guest on December 17, 2013

695