Mutation Research 558 (2004) 75–80

Assessment of DNA damage by comet assay on frozen total blood:method and evaluation in smokers and non-smokers

Isabelle Hiningera,∗, Alexia Chollat-Namya, Sylvie Sauvaigob, Mireille Osmana,H. Faurea, Jean Cadetb, Alain Favierb, Anne-Marie Roussela

a L.B.S.O, Université Joseph Fourier, Grenoble, Franceb Laboratoire des Lésions des Acides Nucléiques, DRFMC/SCIB and UMR 5046 CEA Grenoble,

17 avenue des Martyrs, 38054 Grenoble, France

Received 22 July 2003; received in revised form 5 November 2003; accepted 17 November 2003

Abstract

This study was undertaken with the aim to develop an optimised protocol for the evaluation of DNA damage in frozenwhole blood. This was achieved by use of the single-cell gel electrophoresis (SCGE) or comet assay in its alkaline version.After collection of blood, the total blood sample was mixed with dimethyl sulfoxide (DMSO), a cryoprotectant commonlyused for prevention of freezing-induced damage to living cells, and then stored at−80◦C. We observed no statisticallysignificant differences in the level of DNA damage between fresh blood samples and frozen blood samples, as assessed bythe comet assay. Considering the absence of effects of the freezing step, a frozen blood sample was included as a controlsample in subsequent experiments. Thus the protocol was applied to blood samples of twenty healthy subjects includingsmokers and non-smokers. The comparative analysis indicated that the level of DNA damage was 56% higher in smokersthan in non-smokers (P = 0.01). Altogether, this study strongly suggests that frozen whole blood could be utilised inassociation with the comet assay in human epidemiological bio-monitoring for the assessment of genetic damage in populationsat risk.© 2003 Elsevier B.V. All rights reserved.

Keywords:Comet assay; Whole blood; Frozen samples

1. Introduction

There is an increasing interest in reliable biologicalmarkers of exposure to reactive oxygen species (ROS).The damaging effects of ROS on the DNA structuremay lead to genomic instabilities that are implicatedin the aetiology of cancer[1,2]. The assessment of amoderate increase in the level of DNA damage in hu-

∗ Corresponding author. Tel.:+33-4-7663-7131;fax: +33-4-7663-7180.E-mail address:isabelle.hininger@ujf-grenoble.fr (I. Hininger).

man cells, as an early indicator of genetic diseases orcancer, is of growing interest. The single-cell gel elec-trophoresis or comet assay (SCGE/comet assay), as agenotoxicity assessment method, is currently used forthe estimation of DNA damage at the level of the in-dividual cell. The method allows the detection of sev-eral classes of DNA alterations such as double-strandbreaks, single-strand breaks, alkali-labile sites, in-complete repair of a-basic sites and cross-links[3].In human studies, the comet assay is widely usedwith isolated lymphocytes[4]. However, the isola-tion of lymphocytes represents a possible source of

1383-5718/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.mrgentox.2003.11.004

76 I. Hininger et al. / Mutation Research 558 (2004) 75–80

artefacts and involves time-consuming experiments.Therefore, the objective of the present work was toevaluate the possibility to apply the comet assay toassess the steady-state level of DNA damage in frozenwhole blood. This appears to be a valuable approachto detect small amounts of DNA damage in humantotal blood, with possible applications in clinicalstudies. Indeed, using the comet assay we observedan increased amount of DNA damage in frozen totalblood samples from a group of smokers exposed toan oxidative risk[5,6].

2. Material and methods

2.1. Subjects

Twenty subjects, apparently healthy, including 10smokers (5 women, 5 men), and 10 non-smokers (5women, 5 men), aged 24–51 years, were recruitedamong the technical and student personnel of the Sci-entific and Medical University of Grenoble (France).None of the subjects received any medication or nutri-tional supplements. In the smoker group, the subjectssmoked for 18± 9 years and were still active smok-ers. In the non-smoker group, none of the subjects hadever been a smoker.

2.2. Blood sample collection

Blood was collected between 8 and 10 a.m. inLi-heparin tubes. The SCGE/comet assay was per-formed in triplicate on 10�l of fresh blood fromnon-smokers as control samples in an experiment totest the effect of freezing. In parallel, 500�l of bloodfrom all the volunteers was immediately stabilisedwith 500�l of a 20:80 (v/v) mixture of dimethylsulfoxide (DMSO) and RPMI 1640 cell culturemedium. Aliquots of these samples were progres-sively frozen to−80◦C by use of a cryopreservationvessel (Bicell®, Fischer Bioblock Scientific, France)placed in a−80◦C freezer. After one night, thesamples were transferred from the cryopreservationvessel to storage at−80◦C until analysis, 4 monthslater. The effect of freezing was assessed by compar-ing the results obtained on fresh total blood againstthose obtained on frozen total blood. The effects ofsmoking were assessed on frozen total blood from

non-smokers and smokers. A frozen sample of totalblood from a non-smoker was included in each seriesof experiments as a control.

2.3. Chemicals

Low-melting temperature agarose was purchasedfrom FMC Bioproducts (Rockland, ME, USA);sodium sarcosinate, ethylenediamine tetra-acetic aciddisodium salt (Na2-EDTA), Tris base, Triton 100 Xand ethidium bromide from Sigma Chemical Com-pany (St. Louis, MO); phosphate-buffered saline(PBS) without calcium and magnesium, and RPMI-1640 cell culture medium from Gibco (Grand Island,NY).

2.4. Preparation of cells

After quick thawing in a water bath at 37◦C, bloodsamples were checked for cell viability by use of thetrypan blue exclusion technique. Blood samples werecentrifuged at room temperature at 300× g for 5 min.The cell pellet was washed twice and re-suspendedin Ca++- and Mg++-free phosphate-buffered saline(PBS) to obtain about 20,000 cells in 10�l.

2.5. SGCE/comet assay

The comet assay was performed according to Singh[7], with the following modifications. Thus, 10�l offresh or thawed blood (around 20,000 cells) weremixed with 110�l of 0.6% low-melting agarose inRPMI-1640 at 37◦C. Subsequently 110�l of the mix-ture were layered onto a coded slide pre-coated withthin layers of 1% agarose, and immediately coveredwith a cover-glass. Slides were left for 10 min on icein order to allow the agarose to solidify. After gentlyremoving the cover-glass, the slides were immedi-ately immersed in an ice-cold freshly prepared lysissolution (2.5 M NaCl, 10 mM Na2-EDTA, 10 mM Trishydroxymethyl-aminomethane, 1% sodium sarcosi-nate, 1% Triton 100 X and 10% DMSO, pH 10) to lysethe cells and to allow DNA unfolding. After 1 h in thedark at 4◦C, the slides were immersed in freshly pre-pared alkaline electrophoresis buffer (300 mM NaOH,200 mM Na2-EDTA, pH 13) for unwinding (25 min)and then electrophoresed (25 V/300 mA, 25 min). Allthe steps were carried out under minimal illumination.

I. Hininger et al. / Mutation Research 558 (2004) 75–80 77

The electrophoresis tank was covered with a blackpaper in order to avoid any additional light-inducedDNA damage. Once electrophoresis was completed,slides were neutralized (3×5 min; 0.4 M Tris pH 7.5).

2.6. SCGE/comet assay analysis

The dried microscope slides were stained withethidium bromide (20�g/ml in distilled H2O;50�l/slide), covered with a cover-glass prior toanalysis with a Zeiss fluorescence microscope un-der green light. The microscope was connected to acharge-coupled device (CCD) camera and a personalcomputer-based analysis system (comet AnalysisSoftware, version 4.0, Kinetic Imaging Ltd., Liver-pool, UK) to determine the extent of DNA damageafter electrophoretic migration of the DNA fragmentsin the agarose gel. Results were expressed as percent-age of DNA in the tail (% Tail) or in Tail Moment(product of tail DNA/total DNA by the tail centerof gravity, in arbitrary units), or tail length (distancefrom the head center to the end of the tail (�m)).Three samples per subject were assayed implicating50 cells per sample. The means of the three sampleswere calculated for each subject.

2.7. Statistics

The results are reported as means± standard de-viation. All data are processed statistically using theStudent’st-test paired (to measure the effect of freez-ing) and unpaired (to measure the effect of smoking).

Table 1Level of DNA damage in frozen vs. fresh total blood in individual non-smokers (n = 10) given as various parameters in the comet assay

Subject serialnumber

Tail DNA (%) Tail moment (arbitrary units) Tail length (�m)

Fresh totalblood

Frozentotal blood

Fresh totalblood

Frozen totalblood

Fresh totalblood

Frozen totalblood

1 5.54 4.26 9.80 9.28 22.43 28.062 4.94 3.51 8.82 7.53 22.44 27.183 3.76 6.33 7.64 11.34 26.14 23.154 4.79 4.98 8.45 8.72 22.14 23.395 4.68 3.44 8.15 6.34 22.48 23.216 3.62 4.66 7.29 10.20 28.10 27.527 2.48 3.11 5.53 7.18 28.13 29.08 2.83 3.18 6.28 6.96 28.16 27.929 3.86 3.19 12.02 6.71 38.07 26.84

10 4.29 4.24 11.22 8.82 35.70 26.72

Table 2Comparative effect of frozen vs. fresh total blood from 10non-smokers on various parameters of the comet assaya

Parameters Fresh total blood(n = 10)

Frozen totalblood (n = 10)

Tail DNA (%) 4.08 ± 0.96 4.09± 1.02Tail moment

(arbitrary units)1.25 ± 0.30 1.22± 0.24

Tail length (�m) 27.38± 5.65 26.66± 2.20

a All values are mean± S.D.

Normality was assessed by the Shapiro–Wilk’sW-test(P < 0.05). Statistical significance was set atP <

0.05.

3. Results

3.1. Influence of freezing

After freezing and thawing of the blood, cell via-bility measured by trypan blue was greater than 85%in all subjects. We compared the DNA damage infresh total blood and after storage at−80◦C, for thesame donor and according to the protocol describedin the methodology section. The frozen samples havebeen processed on different days from the fresh sam-ples. We present here the values for each individual(Table 1) and the means (Table 2) for 10 non-smokers,comparing their fresh and frozen blood samples. In-terestingly, the freezing step did not lead to any sig-nificant increase in the level of DNA damage assessed

78 I. Hininger et al. / Mutation Research 558 (2004) 75–80

Table 3Characteristics of the population of smokers and non-smokers

Subject serialnumber

Non-smokers Smokers

Gender Age (years) Gender Age (years) Number of cigarettesper day

Yearssmoked

1 F 24 F 26 11 102 M 51 M 50 17 323 F 46 M 30 13 94 F 32 M 32 10 155 M 31 F 38 13 146 M 35 F 36 15 207 F 33 M 40 11 258 M 27 F 45 20 289 F 42 M 26 9 610 M 33 F 35 12 20

Mean± S.D. 35± 8 36 ± 8 13 ± 3 18 ± 9

by various DNA-damage parameters, when comparedwith fresh whole blood (Table 2). Samples stored at−80◦C were stable for 4 months.

Additionally, a mean value from a sample of totalfrozen blood was first evaluated within a single exper-iment (n = 10; tail moment= 1.35± 0.11) and thenin a series of 10 experiments (n = 10; tail moment=1.38±0.19). No significant differences were observedamong the respective mean values in each type of ex-periment. An aliquot of this sample was then includedas a control in the following experiment on the effectof smoking.

3.2. Effect of smoking

Characteristics for individual non-smokers andsmokers are presented inTable 3. The groups arenot different in age (35± 8 versus 36± 8 years fornon-smokers and smokers, respectively) and gen-der (5 women and 5 men in each group). In thesmokers group, the subjects had smoked for 18± 9years and were still active smokers. The estimationof smoking-induced DNA damage was made forfrozen total blood samples according to the proce-dure described in the methods section. The level ofDNA damage for each individual in the non-smokers(n = 10) and smokers (n = 10) group is givenas tail moment inTable 4. Smokers present signif-icantly higher values in the level of DNA damagethan do non-smokers (tail moment in arbitrary units1.92 ± 0.70 versus 1.23 ± 0.24, respectively;P =

Table 4Level of DNA damage in frozen total blood in each individualnon-smoker (n = 10) and smoker (n = 10), given as tail moment(arbitrary units)

Subject serial number Non-smokers Smokers

1 1.38 3.072 1.07 2.893 1.40 2.394 1.27 1.655 1.25 2.066 1.16 1.097 1.38 1.618 1.60 1.949 1.04 0.8810 0.74 1.63

Mean± S.D. 1.23± 0.24 1.92± 0.70

0.01). The level of DNA damage is 56% higher insmokers than in non-smokers. (Fig. 1).

4. Discussion

The aim of this study was to establish an accurateassay for DNA damage in frozen whole blood, in orderto determine this parameter in epidemiological inves-tigations, since DNA alterations may cause cytotoxic,mutagenic, and carcinogenic effects. The analysis wasperformed with whole blood, reducing the samplingtime and avoiding the risk of increased damage dur-ing cell isolation. The experiments were carried out

I. Hininger et al. / Mutation Research 558 (2004) 75–80 79

Fig. 1. Effect of smoking on total blood DNA damage expressed as % increase in arbitrary units. (Frozen samples were used in this assay.)∗P = 0.01.

with fresh blood and compared with the results ob-tained with frozen samples. DMSO is a cryoprotectantwidely used in cell culture methodology to store cells[8]. Since no significant differences in DNA damagewere observed before and after freezing, the resultssuggest that frozen blood could be used for delayedanalysis. Furthermore, this possibility permits the useof an aliquot of a frozen total blood sample as a con-trol in subsequent assays.

To check the feasibility of our methodology usingthe comet assay with frozen total blood in epidemi-ological studies, a clinical study was performed onsmokers and non-smokers. Using the modified comettechnique, we revealed a significantly higher level ofDNA damage in smokers compared to non-smokers.Previous studies on human lymphocytes have shownthat cigarette smoking increased DNA damage[9–11]and these results are in agreement with an earlier studyfrom our group[12] and others[13], which reportedan increased 8-oxo-deoxyguanosine level in lympho-cytes of smokers. Some authors did not find differ-ences in the level of DNA damage between smokersand non-smokers, probably because the population ofsmokers was different by the number of cigarettes,the duration of smoking, or the history of the sub-jects (ex-smokers referenced as non-smokers)[14,15].Smoking leads to the release of large quantities of freeradical species and many other carcinogenic products[5]. DNA damage could be due to the cumulative orindividual effects of carcinogenic agents contained in

cigarette smoke[4,16,17]and to the reduction in an-tioxidant status in smokers[18]. The ease to work onfrozen total blood, avoiding the step of lymphocyteisolation, could facilitate the use of the comet assay inepidemiological studies to screen populations at riskfor oxidative damage.

5. Conclusion

Our data demonstrate the possibility to apply thecomet assay analysis to whole blood. The method alsoworks on frozen whole blood samples. The advantageof this modified technique lies in its ease to collect andstore samples, and to analyse them simultaneously.The protocol of storage is not technically demand-ing and is, therefore, suitable for use in studies withlarge populations and could permit the use of standardsamples in the assays. Estimation of oxidative DNAdamage by use of the comet assay applied to frozenwhole blood could therefore prove successful for thebiomonitoring of human populations in epidemiolog-ical studies.

References

[1] C.L. Limoli, M.L. Kaplan, J.W. Philips, G.M. Adair, W.F.Morgan, Differential induction of chromosomal instability byDNA strand-breaking agents, Cancer Res. 57 (1997) 4048–4056.

80 I. Hininger et al. / Mutation Research 558 (2004) 75–80

[2] D. Anderson, T.W. Yu, D.B. McGregor, Comet assayresponses as indicators of carcinogen exposure, Mutagenesis13 (1998) 539–555.

[3] F. Kassie, W. Parzefall, S. Knasmüller, Single cell gel elec-trophoresis assay: a new technique for human biomonitoringstudies, Mutat. Res. 463 (2000) 13–31.

[4] C. Betti, T. Davini, L. Giannessi, N. Loprieno, R. Barale,Microgel electrophoresis assay (comet test) and SCE analysisin human lymphocytes from 100 healthy subjects, Mutat. Res.307 (1994) 323–333.

[5] Ly. Zang, K. Stone, W.A. Pryor, Detection of free radicals inaqueous extracts of cigarette tar by electron spin resonance,Free Rad. Biol. Med. 19 (1995) 161–167.

[6] D.M. Parkin, P. Pisani, A. Lopez, E. Masuyer, At least one inseven cases of cancer is caused by smoking. Global estimatesfor 1985, Int. J. Cancer 59 (1994) 494–504.

[7] N.P. Singh, M.T. Mc Coy, R.R. Tice, E.L. Schneider, A simpletechnique for quantification of low levels of DNA damage inindividual cells, Exp. Cell. Res. 175 (1988) 184–191.

[8] J. Paul. Preservation, storage and transportation of livingtissues and cells, in: J. Paul (Ed.), Cell and TissueCulture, Churchill Livingstone, Edingburgh and London,1973, pp. 308–315.

[9] G. Frenzilli, C. Betti, T. Davini, M. Desideri, E. Fornai, L.Giannessi, F. Maggiorelli, P. Paoletti, R. Barale, Evaluationof DNA damage in leukocytes of ex-smokers by single cellgel electrophoresis, Mutat. Res. 375 (1997) 117–123.

[10] C. Betti, T. Davini, L. Giannessi, N. Loprieno, R. Barale,Comparative studies by comet test and SCE analysis in humanlymphocytes from 200 healthy subjects, Mutat. Res. 343(1995) 201–207.

[11] A. Dhawan, N. Mathur, S. K Seth, The effect of smokingand eating habits on DNA damage in Indian population asmeasured in the comet assay, Mutat. Res. 474 (2001) 121–128.

[12] S. Pourcelot, H. Faure, F. Firoozi, V. Ducros, M. Tripier,J. Hee, J. Cadet, A. Favier, Urinary 8-oxo-7,8-dihydro-2′-deoxyguanosine and 5-(hydroxymethyl) uracil in smokers,Free Rad. Res. 30 (1999) 173–180.

[13] S.M. Piperakis, E.E. Visvardis, M. Sagnau, A.M. Tassiou,Effects of smoking and aging on oxidative DNA damageof human lymphocytes, Carcinogenesis 198 (1998) 695–698.

[14] M. Wojewodzka, M. Kruszewski, T. Iwanndo, A.R. Collins,I. Szumeil, Lack of adverse effect on smoking habit on DNAstrand breakage and base damage, as revealed by the alkalinecomet Assay, Mutat. Res. 440 (1999) 19–25.

[15] A. Gonjito, F. Elias, D. Salvadori, M.-L. de Olivera, L.Correa, J. Goldberg, J. trindade, J. de Camargo, Single-cell gel(comet) Assay, detects primary DNA damage in nonseoplasticurothelial cells of smokers and ex-smokers, Cancer Epidemiol.Biomarkers Prevent. 10 (2001) 987–993.

[16] H.L. Falk. Chemical agents in cigarette smoke, in: D.H. Lee(Ed.), Handbook of Physiology: Reaction to EnvironmentalAgent, Waverly, Baltimore MD, 1977, pp. 199–211.

[17] B.S. Cohen, M. Eseinbud, N.H. Harley, Alpha radioactivityin cigarette smoke, Radiat. Res. 83 (1980) 190–196.

[18] I. Hininger, M. Chopra, D. Thurnam, F. Laporte, M.J.Richard, A. Favier, A.M. Roussel, Effect of increased fruitand vegetable intake on the susceptibility of lipoprotein tooxidation in smokers, Eur. J. Clin. Nutr. 51 (1997) 601–606.