TY - JOUR
AU - Collins, Andrew R.
AB - Abstract The comet assay is the method of choice for measuring DNA damage, of various sorts, in human cells such as lymphocytes obtained in the course of population-based studies of environmental and occupational exposure to different genotoxic agents, including radiation, chemicals and oxidative stress. It is noted for its versatility and the breadth of its possible applications. In terms of simplicity, cost, small amount of material required, sensitivity and reliability, the comet assay in its various modifications has few serious competitors. When standardized and validated, the comet assay can provide invaluable information in the areas of hazard identification and risk assessment of environmental and occupational exposure, diseases linked with oxidative stress (e.g. diabetes and cardiovascular disease), nutrition, monitoring the effectiveness of medical treatment and investigating individual variation in response to DNA damage that may reflect genetic or environmental influences. The information obtained could lead to individual advice on lifestyle changes to promote health and especially on relative risks of genotoxic exposure to environmental pollution. Biomonitoring and biomarkers—an introduction Environmental disease is seen as the result of exposure to environmental stressors (including nutrition) modulated by individual susceptibility factors. The molecular epidemiological approach using molecular markers has been in use for almost two decades in the fields of environmental medicine, human biology, pathology and especially in monitoring environmental and occupational exposure. Biomarkers may be defined as indicators of molecular and cellular events in biological systems that may illuminate relationships between environmental hazards and human health effects and disease processes. Biomarkers are measured in biological material (generally blood or urine) collected from patients or volunteer subjects in observational or intervention studies. For environmentally induced diseases, molecular biomarkers play a key role in understanding the relationships between exposure to toxic environmental chemicals and the development of chronic diseases and in identifying individuals at increased risk. They can be used to monitor levels of exposure to some disease-causing agent or to protective factors, or they may inform about inter-individual variation in response to these factors. Finally, they can be studied in relation to genotype, which is increasingly seen as a crucial factor influencing individual responses to exposure and susceptibility to diseases of various kinds. Biomarkers thus cover a wide field, but this review will focus on biomarkers of genotoxic exposure and cancer risk, and in particular on biomarkers of DNA damage and DNA repair assessed using the comet assay. The molecular epidemiological approach, measuring molecular or cellular indicators of disease risk or exposure to causative or preventive factors, is valuable as an adjunct to conventional epidemiology (case–control and cohort or intervention studies) and in its own right as a means of experimenting on humans. Its main advantages are that it requires far smaller numbers of subjects and much less time (and is therefore more economical) than conventional epidemiology; in addition, the biomarkers, if carefully chosen, can give useful information about molecular mechanisms involved in disease aetiology, for example if they reflect an early stage in the progression of the disease. Characterization, validation and optimization of biomarker assays should be seen as crucially important to their use in biomonitoring studies. Surprisingly, many commonly used biomarkers and methods for their measurement have not yet been rigorously examined for their accuracy, reproducibility, specificity and sensitivity or assessed for the feasibility and cost-effectiveness of their application (1,2). Relatively speaking, though, considerable effort has gone into validating the comet assay for use in biomonitoring studies (3–9). The correct evaluation of results and consequently their interpretation are dependent on appropriate statistical analysis. It should be emphasized that false negative or positive results can result from a poor design (selection criteria, poor matching of groups, inappropriate biomarkers, too few subjects, etc.). If—as is customary—a P value of 0.05 is set to define statistical significance, then by definition 1 in 20 apparently significant results will have arisen by chance—an important consideration when numerous biomarkers are measured and many comparisons and correlations are studied. The reader should refer to the article in this issue by Lovell and Omori (10) for elucidation of these and other statistical issues. The comet assay The comet assay is described in detail in other articles in this special issue. Briefly, it is a simple, rapid and sensitive method for measuring DNA breaks in small numbers of cells (typically lymphocytes, in human biomonitoring studies). Cells are embedded in a thin layer of agarose on a glass slide or plastic film, and then lysed in a solution containing detergent and 2.5 M NaCl. Thus membranes and soluble cell constituents, as well as histones, are removed, leaving the DNA still supercoiled and attached to the nuclear matrix. Subsequent alkaline incubation and electrophoresis causes DNA loops containing breaks to move towards the anode, forming a ‘comet tail’ when visualized by fluorescence microscopy with a suitable stain. The images resemble comets, and the relative content of DNA in the tail indicates the frequency of breaks. Strictly speaking, ‘breaks’ in the context of the alkaline comet assay include apurinic/apyrimidinic (AP) sites, which are alkali labile. An early modification of the assay, incorporating an extra step of digestion with a lesion-specific endonuclease following lysis, allowed quantitative detection of damaged bases; most commonly, oxidized bases have been measured (see section on the use of lesion-specific enzymes, below). The comet assay has applications in areas of biomedical and environmental health science such as biomonitoring of human populations for occupational/environmental exposure to genotoxic agents, assessment of DNA damage and oxidative stress in connection with various diseases and assessment of the antioxidant protection afforded by micronutrients in the human diet (Figure 1). Fig. 1 View largeDownload slide The range of applications of the comet assay, measuring various experimental end points (below the arrow) and reflecting different levels of risk determination from exposure to disease (above the arrow). Fig. 1 View largeDownload slide The range of applications of the comet assay, measuring various experimental end points (below the arrow) and reflecting different levels of risk determination from exposure to disease (above the arrow). Practical considerations to ensure reliability of the comet assay as a biomonitoring tool While the comet assay is eminently suitable for use in biomonitoring, special consideration should be given to certain practical and theoretical questions. Every step from initial sampling to final evaluation of data plays an important role and can influence the reliability of results. Therefore, each laboratory should set up and implement standard procedures for all experimental procedures, manipulations of samples and analyses. While it is unrealistic to expect all laboratories performing human biomonitoring studies to adopt identical protocols, inter-laboratory validation studies continue to be extremely valuable exercises. Meanwhile, we have collected together scattered observations and recommendations to ensure the reliability and robustness of the assay. Sampling time and transport Sampling both exposed (treated) and control (referent) individuals on the same day reduces the likelihood of day-to-day experimental variation influencing results. This may not be feasible; but what should definitely be avoided is collecting samples from all exposed subjects and then from all controls (or vice versa), over different time frames. Betti et al. (11) found that the time of year when sampling takes place plays a greater role in the comet assay than in other cytogenetic assays, namely chromosome aberrations and micronuclei. This variation with time is loosely referred to as ‘seasonal variation’. The quality of biological material used for measurements of markers is dependent on sampling conditions and operational aspects including freezing, storage, transport and retrieval (i.e. thawing prior to analysis). This is particularly important for the comet assay. It is recommended to collect samples of early morning fasted blood and to transport them in cool conditions (but not directly in contact with ice). In the course of a biomonitoring study in Slovakia, blood samples were collected from three different towns always at the same time in the morning; we found that the variability of estimates of DNA strand breaks and formamidopyrimidine DNA glycosylase (FPG)-sensitive sites was not affected by time of transport/storage up to 4 h (Figure 2). (The apparent variation in endonuclease III-sensitive sites is possibly attributable to the place, rather than the time of collection.) A joint study in two independent laboratories showed that samples can be stored for up to 4 days at 4°C or at room temperature without any untoward effects on DNA breaks (12). However, the long-term stability of oxidized bases—as opposed to frank breaks—has not been investigated. Fig. 2 View largeDownload slide Effect of time of transportation of blood sample (0, 2 or 4 h) on level of damage detected with the comet assay. Results shown here are from several population studies combined (data from control subjects only). SEMs are shown. The number of subjects was >400 for 0 h, ∼100 for 2 h and ∼40 for 4 h (M. Dusinska, K. Volkovova, M. Staruchova, A. Kazimirova, M. Barancokova, A. Horska, Z. Dzupinkova, unpublished data). Fig. 2 View largeDownload slide Effect of time of transportation of blood sample (0, 2 or 4 h) on level of damage detected with the comet assay. Results shown here are from several population studies combined (data from control subjects only). SEMs are shown. The number of subjects was >400 for 0 h, ∼100 for 2 h and ∼40 for 4 h (M. Dusinska, K. Volkovova, M. Staruchova, A. Kazimirova, M. Barancokova, A. Horska, Z. Dzupinkova, unpublished data). Reference standards and standard validated protocol Despite efforts in many laboratories within Europe and further afield (6,13–16), a standard comet assay protocol and quality control programme have not been adopted. Existing protocols for the most commonly used alkaline version of the comet assay are generally comparable, but many variations exist in different steps such as lysis, alkaline unwinding and electrophoresis. The use of reference standards (positive/negative controls with known levels of damage, ideally included in each electrophoresis run) and a standard protocol with quality assurance during the whole study helps to minimize variation within one laboratory and allows the comparison of results with those of other population studies. Freshly isolated or cryopreserved human lymphocytes or mammalian cell lines have been used as a negative control and cells exposed to genotoxic agents as a positive control (6,13–20) (sometimes incorrectly called an internal standard). The effort to develop a real internal standard by using molecular probes or specific cells which can be clearly distinguished from the cells under investigation in the same gel is still continuing. Currently, the best approach is to use as an external reference standard frozen aliquots from a single or pooled collection of lymphocytes or HeLa or other cells, either untreated or treated with an appropriate agent to induce a significant but not saturating level of damage. If the reference standards are exchanged between laboratories, results from those laboratories can be directly compared. Otherwise, calibration against X- or γ-irradiated cells (with defined amounts of damage) (see section below on calibration) can control for inter-laboratory variation. If more specific DNA lesions are measured by the comet assay using repair enzymes (see below), enzyme specificity should be assessed before use. It is recommended that enzyme from the same batch is used and no changes are made during the same study. Inter- and intra-individual variations among the human subjects can be related to both individual factors and laboratory conditions. Surrogate and target cells The choice of cells to be sampled in biomonitoring studies is often limited to lymphocytes (or leukocytes). However, it must be remembered that white blood cells are not representative of all cells in the body, and in particular they are not target cells for cancer. More appropriate cells from various tissues and organs may sometimes be available. Exfoliated bladder cells, nasal and buccal epithelial cells (21–25), tear duct epithelial cells (26) and cells from biopsies (27) have all been used, but obtaining ‘good’ comets (with reasonably low levels of damage) can be problematic, especially when tissue must be disaggregated or when many dead or senescent cells are present. There are several studies using sperm (28–30) but the methods have not yet been fully optimized; the DNA in sperm is packaged very differently from DNA in somatic cells, and this affects comet production. In this review, we focus mainly on leukocytes and lymphocytes. Human blood; leukocytes and lymphocytes, fresh or frozen Human blood cells are particularly useful for biomonitoring purposes as they are easily acquired. Most studies use fresh whole-blood cells or isolated leukocytes or lymphocytes. They cannot be regarded as typical somatic cells but because they circulate in the body, their cellular, nuclear and metabolic state (including DNA) reflects overall body exposure. Several protocols for freezing and storage of cells have been developed (13,31). Duthie et al. (31) evaluated the influence of cryopreservation on endogenous and induced DNA strand breakage, oxidized purines, oxidized pyrimidines and misincorporated uracil, antioxidant capacity and DNA repair capability in human peripheral blood lymphocytes. Freezing did not increase endogenous levels of DNA damage in freshly isolated human lymphocytes. We have been using a constant protocol, freezing lymphocytes slowly in cell culture medium (minimal essential medium (MEM) or Roswell Park Memorial Institute (RPMI) medium) with 10% foetal bovine serum and 10% dimethylsulphoxide (DMSO, a common cryoprotectant) and keeping them at −80°C (13). Hininger et al. (32) developed a protocol for the evaluation of DNA damage in frozen whole blood. The total blood sample was mixed with an equal volume of medium containing 20% DMSO, and then stored at −80°C. There were no differerences in DNA strand breaks between fresh and frozen blood. Though differences in the level of DNA damage between fresh and frozen samples of blood cells may be observed, they are generally slight and consistent, and the use of frozen white blood cells (or whole blood) has logistic advantages when numerous samples are collected in a short time, as in many biomonitoring studies. Evaluation of comet assay results: how to express data Since the comparison of independent studies is an essential part of evaluation of the impact of environmental and occupational exposure, it is crucially important that results are expressed in a statistically valid way, using understandable and comparable units. In the case of the comet assay, the important parameter is the mean or median comet score for each sample; the distribution of comets within a gel is irrelevant (indeed, if individual comet scores are used for statistical analysis of population groups, a misleadingly low standard error will be obtained). At least two replicate gels are needed to obtain the mean value for each sample. These mean values are taken forward into the statistical analysis. Using appropriate statistical tests is crucial for evaluating results and identifying differences between exposed and control (or treated and untreated) groups [see article by Lovell and Omori in this issue (10)]. The choice of test depends on whether or not results are normally distributed. Results obtained using computer-based image analysis are usually expressed as % DNA in tail, tail moment or tail length. Meta-analysis of biomonitoring data has shown that tail moment and % tail DNA give good correlations with the dose of genotoxic agents used (16). However, since tail moment is measured in non-standard units and different image analysis systems give different values, % tail DNA is more meaningful and easier to conceptualize. Tail length is dependent on electrophoresis conditions and therefore is also difficult to use for comparison. % DNA in tail was confirmed as the most appropriate parameter for comparison of data from different studies though tail moment may be given as well (3,13,16,33–35). Visual scoring is often used, assigning comets to classes 0–4 depending on relative tail intensity and giving an overall score (for 100 comets per sample) of between 0 and 400 arbitrary units. In some instances, samples have been analysed by both image analysis and visual scoring (3,6), and results expressed as either % tail DNA or arbitrary units correlate extremely well (Figure 3). Fig. 3 View largeDownload slide Equivalence of visual scoring and image analysis. A set of slides showing a range of levels of DNA damage were analysed by both visual scoring (results expressed as arbitrary units on a scale of 0–400) and by computerized image analysis giving results in terms of the % of DNA in the comet tail. One hundred comets were analysed per slide (A.R. Collins and C.M. Gedik, unpublished data). Fig. 3 View largeDownload slide Equivalence of visual scoring and image analysis. A set of slides showing a range of levels of DNA damage were analysed by both visual scoring (results expressed as arbitrary units on a scale of 0–400) and by computerized image analysis giving results in terms of the % of DNA in the comet tail. One hundred comets were analysed per slide (A.R. Collins and C.M. Gedik, unpublished data). Calibration It is useful and informative if comet assay results, whether obtained as % tail DNA or as arbitrary units, can be expressed in ‘real’ units, such as DNA breaks per 109 Da. This depends on (indirect) calibration, normally using ionizing radiation to induce DNA breaks, and relying on the equivalence of 0.31 breaks per 109 Da/Gy that was established many years ago using very different irradiation conditions. Problems associated with calibration are discussed elsewhere in this issue (36). What can we measure? We should keep in mind that DNA damage consists mainly of transient lesions. The damage we measure represents a dynamic steady state, with a balance between input and removal (repair). The alkaline comet assay detects a mixture of lesions (single- and double-strand breaks, AP sites, repair intermediates as well as—in proliferating cells—breaks associated with replication). Ideally, we should know what is the normal background level and what is the biological half-life of the damage measured. What is the relationship between DNA damage and repair rate, at the level of individuals? A low repair capacity might suggest that the steady-state level of damage will be relatively high, or it may imply a lack of induction because the level of damage is low. These and other questions remain to be definitively answered. The use of lesion-specific enzymes The original alkaline method measures only DNA breaks (and alkali-labile sites that are converted to strand breaks). Much DNA damage is not direct strand breakage, but rather modification of bases. There are several enzymes involved in DNA repair that have specific activity against certain lesions, introducing a break at sites of base damage. We include these enzymes in an extra step in the comet assay, following lysis, to increase its sensitivity as well as its specificity. Endonuclease III is used to detect oxidized pyrimidines (37,38), FPG recognizes the oxidized purine 8-oxoGua (39,40) and AlkA removes alkylated bases (41). Using repair enzymes is particularly useful in estimating oxidative damage, with the proviso that the specificity of the enzymes is limited. FPG recognizes 8-oxoGua—but also the purine breakdown products FaPyAde and FaPyGua. FPG also detects alkylation damage (N7 methylGua) (42–44). Generally, digestion with FPG increases the number of DNA breaks up to 2-fold. Though the presence of FPG-sensitive sites is not absolute evidence for the presence of oxidative damage, it is generally considered that 8-oxoGua is the major substrate for FPG in vivo, and so it is a reasonable approximation to equate FPG-sensitive sites with oxidized guanine. The mammalian analogue of FPG—8-oxoGua DNA glycosylase (OGG1)—was recently applied in the comet assay (44). OGG1 has a higher substrate specificity than FPG, recognizing only 8-oxoGua and Me-FaPyGua (45). The varied specificities of repair enzymes make them potentially very useful tools for analysing the spectrum of damage induced in cells by a particular exposure to genotoxic agent, but conversely care must be taken when interpreting the results obtained with these repair enzymes as evidence for the presence of particular lesions. It is also worth noting that damage may be underestimated if lesions are inaccessible in the DNA or occur in clusters in such proximity that they register as a single break or if digestion conditions are not optimal. However, there is evidence that these are not important considerations or that errors cancel each other out: in experiments with cells treated with Ro 19-8022 and light to induce different amounts of 8-oxoGua, high-performance liquid chromatography (HPLC) and the comet assay gave dose–response curves with identical slopes indicating equal efficiency of detection (though the background level of damage was several times higher with HPLC than with the comet assay) (46). What affects the background level of DNA damage? Reports from biomonitoring studies show that the basal level of DNA damage in leukocytes is influenced by a variety of lifestyle and environmental exposures, including exercise, air pollution, sunlight and diet. The detailed review of Møller examines these aspects. What is the baseline or background level of DNA damage measured by the comet assay against which individual levels should be compared? Møller pooled results from 125 studies, recalculating visual score on a scale of 0–100, which he found allowed direct comparison with % tail DNA. Median values of % tail DNA/visual score (with range, ±25% from median, in brackets) were as follows: DNA strand breaks 8.6 (4.4–14.5), endonuclease III-sensitive sites 11.0 (4.2–19.5) and FPG-sensitive sites 7.6 (3.2–14.2). Seasonal and geographical differences in comet scores have been detected in several studies (6,16,47–50). Fluctuations in strand breaks through the year in a group of subjects who were repeatedly sampled were attributed to differences in sun exposure (16,49,50). Møller in his review of a large number of studies found a significant (but weak) negative correlation between latitude and basal damage; relatively low levels of damage were reported in northern Europe (16). DNA repair as a biomarker DNA repair is a crucial element in protection against cancer, as it removes potentially mutagenic changes in DNA. Individual differences in capacity for repair are bound to influence susceptibility to cancer and therefore it is an important marker in biomonitoring studies. Also worth considering is the possibility that genotoxic chemicals can act by interfering with repair. Several methods for measuring repair based on the comet assay have been applied as biomonitoring assays (51,52). The most straightforward approach to measuring DNA repair in human cells is to measure the kinetics of repair of DNA breaks or enzyme-sensitive sites during incubation of cells after treatment with specific damage-inducing agent. The comet assay is particularly useful for monitoring repair after low doses of damage. This approach has been used in a study of subjects exposed to traffic pollution (53,54), in a rubber tyre factory (55), among styrene-exposed workers (56), in a study of nuclear power plant workers (57) and in several other studies (51). However, rejoining of strand breaks induced by hydrogen peroxide (H2O2) by freshly isolated lymphocytes seems to be much slower than in most cultured cells (and slower than the repair of radiation-induced breaks) and there is some doubt over the validity of this approach (36, 58). As an alternative to the cellular repair assay, an in vitro assay has been developed, based on the incubation of cell extract with a damage-containing DNA substrate (59). This in vitro approach was applied to measure repair in samples of cells collected in the EC FIBRETOX project (60–62). Substrate HeLa cells were treated with photosensitizer Ro 19-8022 plus light, and OGG activity was measured in both exposed and control individuals. The in vitro comet repair assay was modified by Langie et al. (63) to measure nucleotide excision repair (NER) on a substrate containing bulky adducts induced by treatment with benzo(a)pyrene diolepoxide. Gaivão et al. (52) applied in vitro assays for both NER (with UV-damaged substrate) and base excision repair (BER, using Ro 19-8022 + light) and estimated inter- and intra-individual variability in a group of >30 healthy subjects. Individual repair rates were consistent across time. Between individuals, variation was high; the range was ∼4-fold for BER and 10-fold for NER. Guidelines for biomonitoring studies Earlier reviews have dealt with different aspects of the use of the comet assay in human biomonitoring studies (3–7,64,65), but without providing any specific, practical guidance for using the comet assay in human biomonitoring. Several general articles on biomonitoring are available (1,2,67–69) which can be recommended to help in the design of biomonitoring studies using the comet assay. Certain basic principles should be followed in planning and performing the study, in order to avoid obtaining false-positive and -negative results. Study design is critically important: exposed groups should be matched with respect to sex, age, smoking habit, alcohol consumption, nutrition and lifestyle with control (referent) groups. The size of investigated groups should be large enough to be able to obtain valuable and meaningful results and so should be estimated in advance by carrying out a power calculation. This depends on knowing (or estimating) the precision of the assay method, the coefficient of variation (or standard deviation) of the biomarker and the minimum difference or change that would be expected to have a biologically significant effect. The coefficient of variation for separate analyses of 8-oxoGua in identical cell samples using the comet assay is typically 10–20% (70). As a rule of thumb, it is usually advisable to have at least 50 subjects in each category. Inclusion and exclusion criteria have to be clearly defined and confounding factors (such as age, sex and smoking), which influence the background level of DNA damage and may bias the study, should be taken into consideration. Environmental and occupational monitoring relies on data from exposure measurement and personal monitoring and information on dose–response relationships is valuable, if available. Critics often refer to the subjectivity of the comet assay. To avoid this, samples should be coded, and analyses should be randomized and blinded wherever possible. Time of sampling, seasonal and geographical details and operational aspects (e.g. retrieval conditions, transport and storage conditions) should be recorded as all these might increase the variability. There follows a checklist to help in planning a biomonitoring study with the comet assay (e.g. monitoring populations for effects of exposure to a genotoxic agent or effects of differences in diet or lifestyle or age; also intervention studies, e.g. with dietary constituents). We assume that the samples consist of lymphocytes (the usual biomonitoring material for the comet assay). Make sure you have sufficient subjects in the study to obtain statistically meaningful results; carry out a ‘power calculation’. Include a control group of subjects, i.e. unexposed or untreated or taking a placebo (according to the type of study). Always obtain ethical approval. Sampling of subjects should be performed in the same way throughout the study. Be aware of the possibility of ‘seasonal effects’; collect samples from controls and exposed/treated at the same time, rather than in consecutive phases. Carry out a pilot study for every critical aspect of the study—from sample collection to comet scoring—to check for unforeseen problems and to assess (and if possible control for) experimental variation. Even if it is possible to carry out the comet assay on the day the samples are collected, it is advisable to store aliquots of lymphocytes at −80°C or in liquid nitrogen. Freeze them slowly in freezing medium (MEM or RPMI medium with foetal calf serum and 10% DMSO). When using frozen stored samples, consider whether to select them completely randomly for comet assay analysis or in batches (e.g. all samples from one subject together in the same comet experiment). Consider using lesion-specific endonucleases to increase the sensitivity and selectivity of the comet assay. Make at least two parallels from each sample (i.e. two gels or slides, for each end point—strand breaks, FPG-sensitive sites, etc.). Include a negative control (e.g. untreated lymphocytes) in each comet assay experiment. Preferably, these cells should be frozen aliquots from a single collection of lymphocytes. Include a positive control (e.g. lymphocytes treated with H2O2) in each comet assay experiment. Preferably, these cells should be from a single collection of lymphocytes, treated with the damaging agent and then frozen. Ensure that comet scoring is done ‘blind’. When analysing results in terms of the overall effect of exposure or treatment, it is the overall comet score for each subject/sample that counts—not the values for all the comets scored (which would give a misleading view of variation). Read the article by Lovell and Omori (10) on statistical issues in the use of the comet assay (in this special issue) for help in planning and executing your study. Archive your slides (leave on the bench for a day or two to dry and then store in boxes at room temperature). While carrying out a particular study, always use the same protocols and chemicals from the same company, and avoid making any change in procedure, however slight it may seem. As far as possible follow the principles of Good Laboratory Practice. (Note: The lesion-specific enzymes are mostly available commercially, or several well-known comet researchers produce their own and may give some away. It is advisable to work out for yourself the optimal enzyme incubation conditions since these may vary from batch to batch.) The comet assay as a biomarker for oxidative stress Oxidative stress is a factor (as cause or effect) in many diseases. When selecting a biomarker to examine effects of oxidative stress—or antioxidant protection—in humans, DNA oxidation products are an obvious choice, in view of the link between DNA damage and cancer, and also because they are relatively easy to measure. As an indicator of the overall redox state of the organism, oxidized DNA bases can provide a useful independent marker in studies of other chronic diseases, for example, vascular disease or diabetes. A substantial increase in base oxidation was found in DNA of lymphocytes from diabetics, subjects at risk of cardiovascular disease as well as sufferers from ankylosing spondylitis (71,72). For cardiovascular disease, peripheral mononuclear cells, especially monocytes, are important target cells, and so the assessment of oxidative stress via DNA damage measurement can have much wider implications than simply a concern for genotoxicity (73). What is the best way to measure DNA oxidation? Historically, before the use of FPG facilitated the measurement of 8-oxoGua with the comet assay, the method of choice for characterizing products of free radical attack on DNA was gas chromatography/mass spectrometry (GC-MS) since it allows the unambiguous identification of these products as well as quantitative measurement. Naturally, when interest shifted towards the background level of oxidation in DNA from cultured cells, peripheral white blood cells, tumour or normal tissue, GC-MS was applied. HPLC with electrochemical detection (HPLC-ECD) can also identify certain oxidation products; nucleosides are detected with higher sensitivity than are bases. 8-Oxo-7,8-dihydro-2′-deoxyguanosine (8-oxodGuo) is most commonly measured. The combination of FPG-digestion with a method for measuring DNA strand breaks—the comet assay or alkaline unwinding or alkaline elution—provided an alternative approach. Adherents of each of these approaches have their own favourite ways of expressing results—whether in femtomole of 8-oxoGua per microgram of DNA or 8-oxodGuo per 106 bp or FPG-sensitive sites per 109 Da. Results with the different methods were seldom compared. However, by the mid-1990s, it became clear that there were serious discrepancies in estimates of background levels of oxidized bases in human cells. Measuring the same cell types in one laboratory by HPLC-ECD and with the comet assay, for example, gave results that differed by a factor of at least 10 (39). A far greater range of values is seen in the literature overall (74). Typically, GC-MS has given figures in the hundreds of guanines oxidized per million, HPLC estimates are around 5–30 and the enzymic approach finds <1. Though few recognized it at that time, it is now apparent that further oxidation of DNA occurs during preparation of samples for chromatographic analysis (75). A small amount of oxidation of Gua in terms of parts per million could completely mask a low background concentration. The enzymic methods—using FPG to convert 8-oxoGua to DNA breaks—are less prone to spurious oxidation, as only very mild and brief sample processing is involved. Establishing the background level of oxidized bases was a major aim of the ESCODD project (76). 8-OxoGua was measured in the DNA of lymphocytes collected in different ESCODD member countries, using HPLC-ECD or using FPG to convert the oxidized bases to breaks, followed by the comet assay. The X-ray calibration curve was used to convert comet assay results to breaks per 109 Da and thence to 8-oxoGua per 106 Gua. The likely true background level of 8-oxoGua (in ‘real’ units) was estimated as between 0.3 and 4 per 106 Gua—the range between the median values (across countries) for the comet assay with FPG and for HPLC-ECD, respectively. The ESCODD study provides a benchmark for assessing the credibility of published reports on background DNA oxidation; values of more than ∼5 8-oxoGua per 106 Gua should be regarded as dubious. It seems clear that while the comet assay with FPG may not have the precision of chromatographic methods, it is generally much more accurate in determining the level of background DNA damage in cells. Whether or not oxidative damage to DNA in human cells, at the low level we find it, represents an increase in cancer risk remains a matter for speculation (77). DNA damage as a marker of environmental exposure to genotoxins DNA damage can be seen as a biomarker of exposure to genotoxic agents and as an index of biologically effective dose. In risk assessment, the comet assay can help in hazard identification. DNA damage is also commonly regarded as a marker of cancer risk. It is clearly relevant to cancer since DNA damage is the initiating event in carcinogenesis, but there is as yet no evidence that a high level of DNA damage measured in white blood cells (the usual human sample) reflects an elevated risk of cancer. In contrast, chromosome aberrations and micronuclei have been shown to have predictive value (78,79). Several reviews summarize the use of the alkaline comet assay for environmental and occupational monitoring (4–6,17,64,65,80). These reviews reported a few hundred studies which have employed the comet assay to measure, in most cases, strand breaks and AP sites. An overview of biomonitoring studies performed between 2000 and 2003 using the alkaline comet assay on lymphocytes is given by Faust et al. (5,65). The most recent review of Møller (16) includes 125 biomonitoring studies. The use of repair endonucleases increases the specificity of the assay and thus makes it particularly suitable for biomonitoring studies. A few studies have measured also these more specific DNA lesions, especially oxidized bases. Levels of DNA breaks and base oxidation in lymphocytes of subjects occupationally exposed to ionizing radiation were measured by Wojewodska et al. (81) and Kruszewski et al. (82). Oxidative DNA damage was measured in workers exposed to antimony trioxide (83), in workers in a rubber tyre factory (55,84) and in styrene-exposed workers (56,85). Significant increases in DNA damage as well as in micronuclei and chromosomal aberrations were found in lymphocytes of workers in a rubber tyre factory, compared with administrative workers in the factory or research institute employees. DNA damage correlated with chromosome aberrations, micronuclei, number of polymorphonuclear leukocytes, lymphocyte proliferation and also with years of exposure (84). Higher levels of DNA damage were found in styrene-exposed subjects compared with controls (56,85). The FIBRETOX project was set up to investigate the possible health risks of occupational exposure to mineral fibres used as substitutes for asbestos. The design of the study included both exposed and controls (reference group). The approach involved measuring biomarkers of exposure, effect and individual susceptibility. In this project, we asked several questions: What is the role of oxidative DNA damage in exposure to asbestos and mineral fibres? How do factors such as nutrition and polymorphisms in metabolic and DNA repair genes influence genetic stability? We looked for differences in specific biomarkers between exposed and control subjects; correlations between various biomarkers and age, exposure, smoking, nutrition, sex etc. and associations between various biomarkers and genetic polymorphisms. Different groups (exposed, controls, men, women, smokers and non-smokers) and subgroups (exposed men, exposed women, control men, control women, exposed smokers, exposed non-smokers, control smokers and control non-smokers) were analysed; the appearance of a significant effect in more than one subgroup was taken as confirmation of the significance of these findings. We monitored exposure to asbestos (70 controls and 61 exposed), mineral wool (43 controls and 97 exposed) and glass fibres (37 controls and 81 exposed). Altogether, 389 subjects were investigated (150 controls and 239 exposed). Exposed subjects worked for at least 5 years in a factory. Controls were matched for sex, age, smoking habit and alcohol use. All subjects were clinically examined, with biochemical tests, X-ray, spirometry, skin test for allergies and lifestyle and nutritional questionnaires. External exposure to mineral fibres and polycyclic aromatic hydrocarbons (PAHs) was measured in the workplace four times a year including sampling time and using personal dosimeters. Various biomarkers of exposure, effect and individual susceptibility were measured (60–62, 86) (Figure 4). In asbestos-exposed subjects, we found higher levels of oxidized bases, strand breaks as well as chromosomal aberrations though we did not find any correlation between chromosomal aberrations and DNA damage (60). However, there was a positive correlation between base damage (endonuclease III-sensitive sites) and micronuclei in the subjects of the mineral wool monitoring study (61). Fig. 4 View largeDownload slide Upper panel: effect of asbestos exposure on endonuclease III-sensitive sites. Significantly higher levels were seen in exposed men (P = 0.04). Lower panel: effect of mineral wool exposure on DNA strand breaks. Significantly higher levels were seen in exposed subjects overall (P = 0.05) and in exposed non-smokers (N-S) compared with control non-smokers (P = 0.004) (drawn from data in refs 60,61). Fig. 4 View largeDownload slide Upper panel: effect of asbestos exposure on endonuclease III-sensitive sites. Significantly higher levels were seen in exposed men (P = 0.04). Lower panel: effect of mineral wool exposure on DNA strand breaks. Significantly higher levels were seen in exposed subjects overall (P = 0.05) and in exposed non-smokers (N-S) compared with control non-smokers (P = 0.004) (drawn from data in refs 60,61). Recently, Kopjar et al. (87) with 170 healthy randomly selected volunteers from the general population reported a positive correlation between DNA migration in the comet assay and the total number of chromosome aberrations. Are certain population groups more susceptible to DNA damage than others? In view of its sensitivity, the comet assay may be useful to identify possible differences in levels of damage between men and women, adults and children, smokers and non-smokers or as a function of age. NewGeneris project ‘Newborns and Genotoxic exposure risks: Development and application of biomarkers of dietary exposure to genotoxic and immunotoxic chemicals and of biomarkers of early effects, using mother-child birth cohorts and biobanks’ has as its overall goal the development and application of biomarkers in relation to dietary exposures and childhood disease, to investigate whether maternal exposure results in in utero exposure and subsequently induces carcinogenic and immunotoxic events in the unborn child, thereby leading to increased risk of cancer and immune disorders in later childhood. This project explores the currently undefined contribution of dietary genotoxins consumed by the mother to disease risk in the unborn child. The comet assay is being further developed for the measurement of DNA damage and repair in white blood cells isolated from maternal and cord blood (52,88,89) as well as for the detection of DNA damage in sperm. Mothers and newborn infants did not differ in their susceptibility to DNA breaks (or in their ability to repair them) after an in vitro challenge of lymphocytes with H2O2 (88). There are several studies of children exposed to different environmental and industrial pollutants. Higher levels of DNA damage were found in children exposed to arsenic and lead compared to controls in the most polluted area in Mexico (90), and exposure to DDT metabolites increased DNA strand breaks but not oxidative DNA damage in children (91). The background level of DNA damage in children was the same as in adults. A Mexican study of children suffering infectious disease and malnutrition showed that children with diseases had higher levels of DNA damage both before and after treatment with antibiotics (92). Ortiz-Pérez et al. (93) investigated effects of the insecticide deltametrin in soil and its potential genotoxicity in children but no difference between exposed and non-exposed in the comet assay was found. In the German PAH study, exposure to PAH was investigated in workers and in mother–child pairs living close to a coke-oven plant. Increased levels of DNA strand breaks were detected in the mothers and their children (aged ∼6 years) compared to mothers and children from a rural area (94). There were only weak indications of an association between exposure and DNA damage measured with the comet assay, but there was a strong correlation between DNA strand breaks in mothers and their children (94). Synergism with other potential genotoxic chemicals or other DNA stressing factors may account for these observations: shared genetic predisposition may also play an important role. It is still not clear if the background levels of DNA damage differ between men and women. In our biomonitoring studies, we did not find any differences between men and women in reference groups, in any of the DNA lesions studied—strand breaks, FPG-, endonuclease III- or Alk A-sensitive sites (60,61) (Figure 5), though men tend to show a greater resistance of lymphocyte DNA towards oxidation by H2O2. Gender-related differences have been found in an Indian population: healthy men had higher levels of basal DNA damage in lymphocytes compared with women (95). Fig. 5 View largeDownload slide DNA damage of different types analysed with the comet assay, classified according to sex. Data from control subjects in several population studies are here combined; numbers insider the bars indicate the number of subjects providing data. SEMs are shown with P values where these are significant. FPG- and endonuclease III-sensitive sites and H2O2-induced breaks are shown net, i.e. after subtraction of strand breaks (M. Dusinska, K. Volkovova, M. Staruchova, A. Kazimirova, M. Barancokova, A. Horska, Z. Dzupinkova, unpublished data). Fig. 5 View largeDownload slide DNA damage of different types analysed with the comet assay, classified according to sex. Data from control subjects in several population studies are here combined; numbers insider the bars indicate the number of subjects providing data. SEMs are shown with P values where these are significant. FPG- and endonuclease III-sensitive sites and H2O2-induced breaks are shown net, i.e. after subtraction of strand breaks (M. Dusinska, K. Volkovova, M. Staruchova, A. Kazimirova, M. Barancokova, A. Horska, Z. Dzupinkova, unpublished data). The effect of smoking on the level of DNA damage measured by the comet assay has not been clarified as conflicting results have been obtained (4,65). In our study on 150 middle-aged men, oxidative damage in lymphocyte DNA was significantly higher in smokers compared with non-smokers (96). However, other biomonitoring studies with smaller groups of smokers and non-smokers, did not find any difference between smokers and non-smokers (60,61). Hoffmann et al. (80) evaluated 38 studies and found higher levels of DNA damage in smokers than in non-smokers in studies investigating the effect of smoking as a genotoxic exposure. From the pooled analysis including 125 monitoring studies, Møller (16) showed that sex and smoking did not influence genotoxicity detected by the comet assay; however, there was a positive association between age and level of DNA damage. Several studies have reported on DNA damage in white blood cells in the elderly. Results vary, but often a slight increase in strand breaks and/or oxidized bases is seen with age (97). Combining results from unexposed subjects from various population studies, we find no correlation between various kinds of damage and age (Table I), although associations were found in mineral fibre-exposed subjects in the FIBRETOX study, possibly reflecting years of exposure rather than age itself (62). Table I DNA damage as a function of age in human control subjects from several population studies combined   Number of subjects  Pearson correlation coefficient  P  Strand breaks  468  0.06  0.21  Net FPG-sensitive sites  462  −0.07  0.16  Net endonuclease III-sensitive sites  458  0.015  0.75  Net H2O2-induced breaks  463  −0.003  0.96    Number of subjects  Pearson correlation coefficient  P  Strand breaks  468  0.06  0.21  Net FPG-sensitive sites  462  −0.07  0.16  Net endonuclease III-sensitive sites  458  0.015  0.75  Net H2O2-induced breaks  463  −0.003  0.96  View Large DNA repair assays have only recently begun to be applied in studies of exposure to environmental and occupational agents. Aka et al. (57) used a cellular assay to follow repair of strand breaks induced by ionizing radiation in lymphocytes from 31 control and 32 exposed male workers in a nuclear power plant. There was no significant difference in the rate of rejoining between exposed workers and controls or between smokers and non-smokers. In contrast, Cebulska-Wasilewska et al. (98) conducted a population study to investigate whether occupational exposure to mercury can cause genotoxicity or affect DNA repair efficiency after irradiation with UV or X-rays. Blood samples from 25 exposed workers and 50 matched controls were investigated. The data indicate that occupational exposure to mercury did not cause direct genotoxicity but did cause a significant decline in DNA repair. Mohankumar et al. (99) studied UV-induced DNA repair in lymphocytes of smokers and non-smokers. A significant difference in DNA migration 10 min after UV exposure implies that cigarette smoking perhaps interferes with the incision steps of NER. The effect of air pollution on DNA damage and repair after irradiation with UV or exposure to H2O2 was investigated in 80 healthy subjects living in urban or rural areas (100). The extent of damage and subsequent repair appear to be influenced by the residential environment. Marcon et al. (101) studied variability in DNA repair in 31 healthy subjects. Fresh blood samples were irradiated with γ-rays (2 Gy) and the kinetics of DNA repair in leukocytes assessed by the comet assay 0, 15 and 30 min after irradiation. Palyvoda et al. (102) irradiated lymphocytes of cancer patients with 2 Gy and found high DNA damage, low repair rate and high levels of residual non-repaired DNA damage compared with healthy controls. Levels of induced DNA damage and repair kinetics in isolated human blood lymphocytes of individuals from two cities (Kosice and Sofia) were studied after irradiation with X-rays (3 Gy). A significant decrease in repair efficiency associated with exposure to PAHs was observed in the exposed individuals from both cities, when analysed separately or together. A negative influence of tobacco smoking on the efficiency of DNA repair was also observed (54). Piperakis et al. (103) studied DNA damage and repair after H2O2 in human treatment of lymphocytes taken from 116 healthy individuals, both controls and exposed to pesticides. There was no significant difference in basal DNA damage between the study groups nor did the H2O2-induced damage or ability to rejoin the strand breaks vary between control and exposed groups. In vitro repair of 8-oxoGua was one of the biomarker assays applied in the FIBRETOX project, investigating workers from factories producing asbestos and man-made mineral fibres. Sixty-one asbestos workers, with exposure of between 5 and 40 years, were compared with 21 unexposed factory workers (60). Overall, there was no difference in repair rates between the groups, although extracts from exposed female workers showed a lower repair activity than those from female controls. Smokers and non-smokers had equal repair activities. Mineral wool factory workers (98 exposed workers and 43 unexposed workers from the same factory) were also studied (61); there was no effect of exposure or smoking, but in this factory men had significantly higher repair rates than women. Vodicka et al. (104) used both the cellular assay for repair of radiation-induced strand breaks and the in vitro comet assay for repair of 8-oxoGua to examine effects of exposure in a rubber tyre factory (55) and among lamination workers (56). In the tyre plant, 94 exposed workers were compared with 16 controls. Smokers had higher strand break rejoining rates than non-smokers. Strand break rejoining rates were higher in exposed subjects and showed significant positive correlations with styrene concentrations measured in air and in blood. OGG activity showed a similar pattern, increasing with increasing styrene concentration in air and in blood, and also correlating with urinary styrene metabolites. This study suggested that both strand break repair and removal of 8-oxoGua are induced by the styrene exposure. Overall, it is clear that DNA repair, both NER and BER, may be relevant biomarkers for assessing cancer risk in environmental and occupational exposure, but larger studies with more subjects are needed. Assessing nutritional effects on endogenous DNA damage and antioxidant status There is strong evidence that antioxidants present in fruits and vegetables can increase the resistance of cellular DNA to oxidative attack, both in vivo and ex vivo. Challenging lymphocytes ex vivo with oxidative-damaging agent (generally H2O2) and measuring induced DNA breaks with the comet assay allow indirect measurement of antioxidant status. The increased resistance of lymphocytes to oxidative attack was used as evidence of improved antioxidant status in various nutritional intervention studies (105–108). In a recent study, 168 healthy volunteers consumed a drink of blueberry/apple juice providing 97 mg quercetin and 16 mg ascorbic acid a day; after 4 weeks, 20% greater protection (P < 0.01) against ex vivo H2O2-induced oxidative DNA damage was seen (109). Many intervention trials have used the comet assay to monitor endogenous DNA damage. For example, endogenous base oxidation was significantly lower after daily supplementation with vitamins C and E and β-carotene for 20 weeks (110), a daily dose of carrot juice for 2 weeks (111), several weeks of consuming 1 litre of soya milk each day (112), one to three kiwifruits per day over a 3-week period (113) (Figure 6) or 480 ml of grape juice daily for 8 weeks (114). On the other hand, Giovannelli et al. (50) in 71 healthy subjects found a positive association of DNA base damage with tomato consumption (P = 0.05). Fig. 6 View largeDownload slide Effects on DNA damage/repair biomarkers of 3-week supplementation of diet with two kiwifruits per day. Data shown are net FPG- and endonuclease III-sensitive sites, net H2O2-induced breaks and in vitro repair activity on a DNA substrate containing 8-oxoGua. Fourteen subjects took part in the trial. (Data are redrawn from ref. 113.) Fig. 6 View largeDownload slide Effects on DNA damage/repair biomarkers of 3-week supplementation of diet with two kiwifruits per day. Data shown are net FPG- and endonuclease III-sensitive sites, net H2O2-induced breaks and in vitro repair activity on a DNA substrate containing 8-oxoGua. Fourteen subjects took part in the trial. (Data are redrawn from ref. 113.) The application of the comet assay in nutritional studies of this kind has been thoroughly reviewed, and tables are presented summarizing results (115,116). Overall, endogenous strand breaks are seen (not surprisingly) as a poor indicator of the effects of antioxidants. Assessing resistance of lymphocyte DNA to H2O2-induced damage, or measuring endogenous base oxidation, is more likely to reflect antioxidant intake. While about half the studies examined showed apparently protective effects of antioxidants, others showed no effect—whether because of poor study design, inadequate numbers of subjects or inappropriate source of micronutrients. In addition to intervention studies, there have also been some observational studies. Oxidative DNA damage was investigated in a study of vegetarians (117). The group with traditional dietary habits had significantly higher levels of oxidative DNA damage (strand breaks and oxidized purines, P = 0.005) compared with vegetarians. A significant positive correlation between age and oxidative DNA damage (net FPG-sensitive sites) was found in non-vegetarians, while the opposite trend (negative association) was seen in vegetarians. Traditionally, in central/eastern Europe, the antioxidant intake is substantially higher in summer than winter. We set out to discover whether the seasonal difference still applies and whether there are consequences in terms of biomarkers of cancer and of cardiovascular disease. Plasma antioxidant levels, lipid peroxidation and lymphocyte DNA damage were measured in 150 middle-aged men, and dietary and lifestyle information obtained from food frequency questionnaires. Differences in both antioxidant intake and plasma antioxidant levels between winter and summer were less than expected, suggesting that the availability of fruits and vegetables throughout the year has altered the traditional diet. Variations in DNA damage were not clearly related to antioxidant intake or plasma levels (48,118,119). Correlations between DNA damage and consumption of fruits and vegetables were investigated in the FIBRETOX study. Intake of fruits (P = 0.05), vegetables (P = 0.01) and cereals (P = 0.05) inversely correlated with FPG-sensitive sites in all 389 investigated subjects as well as in all 239 exposed subjects (86,120). Data from a cross-sectional study of 164 generally healthy non-smoking African Americans and Whites in North Carolina, aged 20–45 years, equally distributed by race and sex, show that levels of oxidative DNA damage, measured using the alkaline comet assay, were lower in African Americans than Whites though they had lower levels of antioxidants (121). An inverse association between lycopene and oxidative DNA damage was found in the combined study population after adjusting for sex, age, body mass index, passive smoke exposure, physical activity, education, income and alcohol intake. There was, however, a ‘positive’ association of vitamin E with oxidative DNA damage in the total population and in African American men after adjusting for covariates. We found a similar association between DNA damage and vitamin E (120). Gene–environment interactions: the comet assay and genetic polymorphisms How do genetic polymorphisms in metabolic and DNA repair genes influence the effect of exposure on genetic stability? Many biomonitoring studies include genetic markers (single nucleotide polymorphisms) as biomarkers of individual susceptibility to assess modulating effects of exposure on DNA damage and repair (55,62,96,122–128). Unfortunately, many studies measure DNA damage only as strand breaks and not specific DNA lesions. Moreover, the size of the study population is often quite small, especially in studies of occupational exposure, which increases the likelihood of false-positive or -negative results. Therefore, a good strategy for assessing significance is needed. We normally split the population under investigation into subgroups (such as men/women, smokers/non-smokers and exposed/control) and analyse the subgroups separately. We consider correlations convincing only if significance is found in the whole group and/or in several subgroups. Several of our studies show that DNA damage as well as repair appear to be modulated by interactions between environmental and genetic factors. Response to environmental factors often depends on specific genetic polymorphisms. It is of course impossible to analyse all genetic polymorphisms but at least the most relevant depending on exposure, population studied and confounding factors should be taken into consideration. DNA damage and repair and glutathione-S-transferase polymorphisms DNA response to exposure can be modulated by genetic polymorphisms in metabolic genes. Here, we focus on glutathione-S-transferase (GST) as a representative of the phase II metabolism pathway. Several biomonitoring studies have shown an association between DNA damage and GST polymorphism but there are also studies which show no effect (124,125). Pavanello and Clonfero (126) reviewed international scientific publications until 2000 on the influence of metabolic genotypes on biological indicators of genotoxic risk in environmental or occupational exposure. Biomarkers of exposure and of effects have been evaluated according to different genotypes (or phenotypes) of several activating/detoxifying metabolic activities. In fewer than half the studies (43 of 95) was an influence of genotype on the examined biomarkers (including the comet assay) found. No association of DNA damage with GST polymorphism was found in 32 exposed workers and 32 controls from a Portuguese rubber tyre factory (127). No significant effects of the genotypes of GSTM1, GSTP1 and GSTT1 on DNA damage were found in an Estonian shale-oil mine study of 50 underground and 42 surface workers (128). Similarly, no association between DNA strand breaks and metabolic enzyme polymorphisms was found in 38 subjects exposed to PAH in a coke-oven and graphite-electrode-producing factory (129). Van Delft et al. (130) assessed whether the current exposure to PAH of coke-oven workers in a Dutch plant induced biological effects and whether these effects are influenced by tobacco smoking and by genetic polymorphisms in GSTM1 and GSTT1; no effects of genetic variation on exposure were found. All these studies were relatively small. On the other hand, a study in northern Bohemia on women exposed to air pollution showed that comet assay % tail DNA correlated with exposure to respirable particles and that GSTM1 genotype had a significant effect on urinary PAH metabolites, urine mutagenicity and % tail DNA when the GSTM1 genotype was considered as a single factor affecting these biomarkers (131). However, a study from the same group (125) did not find any effect of GSTM1 polymorphism on the level of strand breaks in 322 pregnant women from northern Teplice and 220 from southern (Prachatice) regions. Wilms et al. (109) found an interesting association in 168 healthy volunteers consuming blueberry/apple juice between DNA damage and GSTT1 polymorphism. Wild types benefit more from its protecting effects against oxidative DNA damage. The authors conclude that genotyping for relevant polymorphisms enables the selection of subgroups among the general population that will benefit more from the DNA damage-modulating effects of micronutrients. Marcon et al. (101) in a pilot study of 31 healthy subjects found a modulator effect of smoking habits and GSTM1 genotype on individual DNA repair capacity measured in lymphocytes, possibly related to the higher expression of enzymes involved in the repair of oxidative DNA damage in heavy smokers and GSTM1-null subjects. Fresh blood samples were irradiated with γ-rays (2 Gy) and the kinetics of DNA repair in leukocytes assessed by measuring breaks at 0, 15 and 30 min after irradiation. In order to elucidate the health effects of occupational exposure to traffic fumes, a few biomarkers of early genetic effect were investigated in 190 Rome traffic policemen and in 57 office workers (132). DNA damage measured by the comet assay did not reveal any statistically significant difference between the exposed and control workers and no association was seen with polymorphisms in the genes for metabolizing enzymes. On the other hand, Novotna et al. (133) found an association between DNA damage, vitamin C levels and GSTM1+ polymorphism in 65 non-smoking city policemen. In our biomonitoring studies, we carry out genotype analysis for polymorphisms in metabolic and DNA repair genes. In several studies we found an interesting association between GST and DNA damage and repair. In a study of 150 middle-aged men, we found that the homozygous GSTP1 a/a genotype was associated with significantly higher levels of GST activity measured in lymphocytes, in comparison with the b/b genotype. Multifactorial statistical analysis revealed significant associations between smoking, GSTP1 genotype, plasma vitamin C and purine base damage in lymphocyte DNA. The GSTP1 b/b group accounted for most of the increased purine oxidation in smokers. The link between smoking and oxidized pyrimidines was seen only with GSTT1 null (96). Thus GST polymorphism influences DNA stability and repair. In the FIBRETOX project, we investigated 139 asbestos-exposed subjects. Those with GSTT1-null genotype had lower repair capacity for oxidative damage than those with GSTT1 wild type. The same association was found in controls and in women (P = 0.035) (M. Dusinska and A. Horska, in preparation). A study of 141 exposed and control subjects in a mineral wool factory showed an association of DNA damage (endonuclease III-sensitive sites) with GST activity (measured in erythrocytes) in all subjects, all women, all men and all exposed (120). The GSTP1aa group had the highest repair rate for oxidative damage. This association was found in women and in the control group. Moreover, an interaction of repair with GSTT1, exposure and smoking was also found in this study population (Dusinska and Horska, in preparation). In another monitoring study, where exposure to styrene was monitored, a general linear model revealed associations between GSTM1 deletion, sex, smoking habit and exposure status and DNA repair capacity (134). The nature of the relationship between DNA damage, DNA repair and genetic polymorphism in GST genes is still unclear due to the limitations in the design of studies as well as in the range of DNA lesions measured. In the FIBRETOX study (120), we found a negative correlation of FPG- and endonuclease III-sensitive sites with GST levels which implies that antioxidant enzymes may play an important role in protection against oxidative DNA damage. GST has an important role in detoxification of xenobiotics, drugs and carcinogens and thus protects the cells against redox cycling and oxidative stress. Xenobiotic-metabolizing enzymes play a major role in regulating the toxic, oxidative-damaging, mutagenic and neoplastic effects of chemical carcinogens. Mounting evidence has indicated that the induction of phase 2 detoxification enzymes such as GSTs results in protection against toxicity and chemical carcinogenesis, especially during the initiation phase (135). The GSTs are a family of enzymes that catalyze the nucleophilic addition of the thiol of GSH to a variety of electrophiles. It seems that GST may have also some other role in protection of DNA against oxidative DNA damage. Recently, it was found that α class GSTs bind with the dinitrosyl–diglutathionyl–iron complex in rat hepatocytes and that a significant part of the bound complex is also associated with the nuclear fraction (136,137). Stella et al. (137) using confocal and electron microscopy reported nuclear localization of GSTs in these cells. Surprisingly, they found that a considerable amount of GST corresponding to 10% of the cytosolic pool is electrostatically associated with the outer nuclear membrane, and a similar quantity is compartmentalized inside the nucleus. Mainly α class GSTs are involved in this double modality of interaction. A quantitative analysis of the membrane-bound αGSTs suggests the existence of a multilayer assembly of these enzymes at the outer nuclear envelope that could represent a considerable novelty in cell physiology. The authors conclude that the interception of potentially noxious compounds to prevent DNA damage could be the possible physiological role of the perinuclear and intranuclear localization of αGSTs. In four independent biomonitoring studies, we have found that GST polymorphisms and levels of GST enzymes can influence DNA stability and repair of oxidative DNA damage. GST enzymes are important in the control of oxidative stress but it seems that they also can influence DNA stability and genotoxic effects of exposure by other mechanisms in addition to antioxidant protection. A pooled analysis of the several thousand subjects for which comet assay and GST polymorphism data or biological material are available could bring more light to bear on mechanisms of cellular and nuclear defence against (oxidative) DNA damage. DNA damage and repair and polymorphisms in repair genes The recent review of Collins and Gaivao (51) summarized reported associations between DNA repair and polymorphisms in repair genes. Studies are usually too small to make definite conclusions, even though significant results may be obtained (138). However, the study of 143 coke-oven workers and 50 non-coke-oven workers (139) showed that the A-allele of G27466A polymorphism of XRCC1 may be associated with decreased DNA repair capacity towards PAH-induced base damage and strand breaks. Vodicka et al. (140) analysed polymorphisms in the DNA repair genes XPD, XPG, XPC, XRCC1 and XRCC3 in a population of 337 healthy subjects. DNA strand break rejoining rates were measured in lymphocytes, and their capacity to repair 8-oxoGua was estimated with the in vitro comet repair assay. Strand break rejoining was significantly influenced by XPC and XRCC1 polymorphisms. The importance of the Ser326Cys polymorphism in hOGG1 has been examined by several groups. Vodicka et al. (104), with 244 subjects and the comet in vitro repair assay, found a relatively lower repair activity associated with the variant. The protein product of XPA is a zinc finger DNA-binding protein that is essential in NER (141), but it seems that NER proteins may also be involved in repair of oxidized bases (62,142). In a monitoring study of several hundred control subjects, it was found that the A-allele of the common polymorphism 23A → G, in the 5′ non-coding region, was associated with less efficient NER (143). Several studies suggested that sequence variation in XPA can modulate susceptibility to lung cancer. The XPA variant homozygous for G at position 23 in the non-coding region (4 nt before the start codon) is supposed to have a protective role, reducing cancer risk. We found that the frequency of this polymorphic allele in the Slovak population is 62%. Interestingly, subjects with highest levels of FPG- and endonuclease III-sensitive sites were found preferentially to be XPA AA (wild type) in all subjects and in the following groups: all controls, all women, all control women, all non-smokers and all controls. Multifactorial analysis showed that the level of FPG-sensitive sites depends on age, exposure and XPA polymorphism (62). Moreover, capacity for repair of 8-oxoGua was higher in subjects from the asbestos biomonitoring study with AA genotype—who have the higher levels of damage. We hypothesized that, if XPA protein is normally involved in repair of oxidized bases and is less active when derived from the A-allele, a higher steady-state level of 8-oxoGua might be expected. This in turn might lead to a compensatory increase in the activity of OGG1 in the BER pathway. We also found a correlation between repair of oxidized guanine and age in several groups—all subjects, all exposed, all controls, all women and all non-smokers—and specifically in subjects with AA genotype. It seems very important to include genetic susceptibility markers in biomonitoring studies and to identify possible interactions between DNA damage and repair and genetic polymorphisms. Conclusions In humans, biomonitoring the comet assay plays an increasingly important role in its various modified forms. Although not all types of carcinogenic exposures should be expected to be revealed as damage to DNA in leukocytes, the comet assay has proved to be a valuable general method for detection of genotoxic exposure in humans (4–7,17,64–66). It has been used as a tool for identifying markers of biologically effective dose, effect and individual susceptibility. In addition to providing data on effects of genotoxic exposure in human populations, the comet assay has yielded a great deal of fundamental information on mechanisms of genotoxicity, and cellular responses, that is crucial to interpretation of biomonitoring data in terms of cancer risk. The use of lesion-specific endonucleases allows the measurement of different kinds of DNA base damage, which can be seen as biomarkers of biologically relevant dose. Oxidative DNA damage, measured with the comet assay, may have relevance not only for cancer risk but also for other diseases associated with oxidative stress. Various modifications of the comet assay for measurement of DNA repair capacity allow the study of this important phenotypic marker of susceptibility, and antioxidant status can be assessed by challenging lymphocytes with H2O2 and measuring the DNA breaks induced. The inclusion of markers of oxidative stress and antioxidant defence in human biomonitoring studies especially where oxidative damage is expected is essential (144). Detailed information on lifestyle, possible additional exposure from other sources, smoking, alcohol consumption and nutrition and other covariate information should be obtained from questionnaires. These are important to identify the possible modulation of DNA response, to identify confounders and to explain the results obtained. Sizes of study groups are often relatively small, and there is a need for larger studies. However, meta-analysis (i.e. pooling the results of individual studies and assigning weight to the results according to aspects of the study design) can lead to statistically much stronger and more credible conclusions. The combination of data from these various phenotypic biomarkers with polymorphism analysis promises to yield much valuable information. However, genotyping requires very large numbers of subjects. To match this demand, high capacity versions of the comet assay (and other phenotype assays) are required. The EC-funded COMICS project (‘comet assay and cell array for fast and efficient genotoxicity testing’) aims to increase by orders of magnitude the throughput of this assay, using multi-well format and ‘cell arrays’ to increase the number of samples per experiment and developing an alternative, automated method of scoring based on differential fluorescence. Funding EC-funded projects, in particular the Centre of Excellence in Environmental Health (HEAR NAS, QLK6-2002-90445); FIBRETOX (QLK4-1999-01629); Intarese (GOCE 018385); NewGeneris (FOOD-2005-016320). We thank all colleagues at Department of Experimental and Applied Genetics, Slovak Medical University, for technical help with comet assay and for providing historical data. We also thank Bozena Smolkova for help with the analysis of results. References 1. Angerer J,  Ewers U,  Wilhelm M.  Human biomonitoring: state of the art,  Int. J. Hyg. Environ. Health ,  2007, vol.  210 (pg.  201- 228) Google Scholar CrossRef Search ADS PubMed  2. Bennett DA,  Waters MD.  Applying biomarker research,  Environ. Health Perspect. ,  2005, vol.  108 (pg.  907- 910) Google Scholar CrossRef Search ADS   3. Collins A,  Dušinská M,  Franklin M, et al.  Comet assay in human biomonitoring studies: reliability, validation, and applications,  Environ. Mol. Mutagen. ,  1997, vol.  30 (pg.  139- 146) Google Scholar CrossRef Search ADS PubMed  4. Møller P,  Knudsen LE,  Loft S,  Wallin H.  The comet assay as a rapid test in biomonitoring occupational exposure to DNA-damaging agents and effect of confounding factors,  Cancer Epidemiol. Biomarkers Prev. ,  2000, vol.  9 (pg.  1005- 1015) Google Scholar PubMed  5. Faust F,  Kassie F,  Knasmüller S,  Boedecker RH,  Mann M,  Mersch-Sundermann V.  The use of the alkaline comet assay with lymphocytes in human biomonitoring studies,  Mutat. Res. ,  2004, vol.  566 (pg.  209- 229) Google Scholar CrossRef Search ADS PubMed  6. Møller P.  The alkaline comet assay: towards validation in biomonitoring of DNA damaging exposures,  Basic Clin. Pharmacol. Toxicol. ,  2006, vol.  98 (pg.  336- 345) Google Scholar CrossRef Search ADS PubMed  7. Tice RR,  Agurell E,  Anderson A, et al.  Single cell gel/comet assay: guidelines for in vitro and in vivo genetic toxicology testing,  Environ. Mol. Mutagen. ,  2000, vol.  35 (pg.  206- 221) Google Scholar CrossRef Search ADS PubMed  8. Dušinská M,  Fábry R,  Szabová E,  Somorovská M,  Petrovská H,  Collins AR.  Occupational exposure to mutagens; monitoring of rubber factory workers,  Mutat. Res. ,  1997, vol.  379  Suppl. S-XVI(pg.  161- 162) Google Scholar CrossRef Search ADS   9. ESCODD Measurement of DNA oxidation in human cells by chromatographic and enzymic methods,  Free Radic. Biol. Med. ,  2003, vol.  34 (pg.  1089- 1099) CrossRef Search ADS PubMed  10. Lovell DP,  Omori T.  Statistical issues in the use of the comet assay. Mutagenesis, (in press) 11. Betti C,  Davini T,  Giannessi L,  Loprieno N,  Barale R.  Comparative studies by comet test and SCE analysis in human lymphocytes from 200 healthy subjects,  Mutat. Res. ,  1995, vol.  343 (pg.  201- 207) Google Scholar CrossRef Search ADS PubMed  12. Anderson D,  Yu TW,  Dobrzyńska MM,  Ribas G,  Marcos R.  Effects in the Comet assay of storage conditions on human blood,  Teratog. Carcinog. Mutagen. ,  1997, vol.  17 (pg.  115- 125) Google Scholar CrossRef Search ADS PubMed  13. Collins AR.  The comet assay for DNA damage and repair: principles, applications, and limitations,  Mol. Biotechnol. ,  2004, vol.  26 (pg.  249- 261) Google Scholar CrossRef Search ADS PubMed  14. Collins AR.  Cutter RG,  Rodriguez H.  Oxidative DNA damage biomarkers: a need for quality control,  Critical Reviews of Oxidative Stress and Ageing ,  2003 World Scientific(pg.  469- 484) 15. Speit G,  Hartmann A.  The comet assay: a sensitive genotoxicity test for the detection of DNA damage and repair,  Methods Mol. Biol. ,  2006, vol.  2314 (pg.  275- 286) 16. Møller P.  Assessment of reference values for DNA damage detected by the comet assay in human blood cell DNA,  Mutat. Res. ,  2006, vol.  612 (pg.  84- 104) Google Scholar CrossRef Search ADS PubMed  17. Møller P.  Genotoxicity of environmental agents assessed by the alkaline comet assay,  Basic Clin. Pharmacol. Toxicol. ,  2005, vol.  96  Suppl. 1(pg.  1- 42) Google Scholar CrossRef Search ADS PubMed  18. Hellman B,  Vaghef H,  Friis L,  Edling C.  Alkaline single cell gel electrophoresis of DNA fragments in biomonitoring for genotoxicity: an introductory study on healthy human volunteers,  Int. Arch. Occup. Environ. Health ,  1997, vol.  69 (pg.  185- 192) Google Scholar CrossRef Search ADS PubMed  19. De Boeck M,  Touil N,  De Visscher G,  Vande PA,  Kirsch-Volders M.  Development and validation of an internal standard in comet assay analysis,  Mutat. Res. ,  2000, vol.  469 (pg.  181- 197) Google Scholar CrossRef Search ADS PubMed  20. Speit G,  Witton-Davies T,  Heepchantree W,  Trenz K,  Hoffmann H.  Investigations on the effect of cigarette smoking in the comet assay,  Mutat. Res. ,  2003, vol.  542 (pg.  33- 42) Google Scholar CrossRef Search ADS PubMed  21. Rojas E,  Valverde M,  Sordo M,  Ostrosky-Wegman P.  DNA damage in exfoliated buccal cells of smokers assessed by the single cell gel electrophoresis assay,  Mutat. Res. ,  1996, vol.  370 (pg.  115- 120) Google Scholar CrossRef Search ADS PubMed  22. Pinhal D,  Gontijo AM,  Reyes VA,  Salvadori DM.  Viable human buccal mucosa cells do not yield typical nucleoids: impacts on the single-cell gel electrophoresis/Comet assay,  Environ. Mol. Mutagen. ,  2006, vol.  47 (pg.  117- 126) Google Scholar CrossRef Search ADS PubMed  23. Eren K,  Ozmeriç N,  Sardaş S.  Monitoring of buccal epithelial cells by alkaline comet assay (single cell gel electrophoresis technique) in cytogenetic evaluation of chlorhexidine,  Clin. Oral Investig. ,  2002, vol.  6 (pg.  150- 154) Google Scholar CrossRef Search ADS PubMed  24. Fortoul TI,  Valverde M,  López MC,  Avila-Costa MR,  Avila-Casado MC,  Mussali-Galante P,  Gonzalez-Villalva A,  Rojas E,  Ostrosky-Shejet P.  Genotoxic differences by sex in nasal epithelium and blood leukocytes in subjects residing in a highly polluted area,  Environ. Res. , vol.  94 (pg.  243- 248) CrossRef Search ADS PubMed  25. Szeto YT,  Benzie IF,  Collins AR,  Choi SW,  Cheng CY,  Yow CM,  Tse MM.  A buccal cell model comet assay: development and evaluation for human biomonitoring and nutritional studies,  Mutat. Res. ,  2005, vol.  578 (pg.  371- 381) Google Scholar CrossRef Search ADS PubMed  26. Rojas E,  Valverde M,  Lopez MC,  Naufal I,  Sanchez I,  Bizarro P,  Lopez I,  Fortoul TI,  Ostrosky-Wegman P.  Evaluation of DNA damage in exfoliated tear duct epithelial cells from individuals exposed to air pollution assessed by single cell gel electrophoresis assay,  Mutat. Res. ,  2000, vol.  468 (pg.  11- 17) Google Scholar CrossRef Search ADS PubMed  27. Pool-Zobel BL,  Dornacher I,  Lambertz R,  Knoll M,  Seitz HK.  Genetic damage and repair in human rectal cells for biomonitoring: sex differences, effects of alcohol exposure, and susceptibilities in comparison to peripheral blood lymphocytes,  Mutat. Res. ,  2004, vol.  551 (pg.  127- 134) Google Scholar CrossRef Search ADS PubMed  28. Hughes CM,  Lewis SE,  McKelvey-Martin VJ,  Thompson W.  A comparison of baseline and induced DNA damage in human spermatozoa from fertile and infertile men, using a modified comet assay,  Mol. Hum. Reprod. ,  1996, vol.  2 (pg.  613- 619) Google Scholar CrossRef Search ADS PubMed  29. Sergerie M,  Bleau G,  Teulé R,  Daudin M,  Bujan L.  Sperm DNA integrity as diagnosis and prognosis element of male fertility,  Gynecol. Obstet. Fertil. ,  2005, vol.  33 (pg.  89- 101) Google Scholar CrossRef Search ADS PubMed  30. Migliore L,  Colognato R,  Naccarati A,  Bergamaschi E.  Relationship between genotoxicity biomarkers in somatic and germ cells: findings from a biomonitoring study,  Mutagenesis ,  2006, vol.  21 (pg.  149- 152) Google Scholar CrossRef Search ADS PubMed  31. Duthie SJ,  Pirie L,  Jenkinson AM,  Narayanan S.  Cryopreserved versus freshly isolated lymphocytes in human biomonitoring: endogenous and induced DNA damage, antioxidant status and repair capability,  Mutagenesis ,  2002, vol.  17 (pg.  211- 214) Google Scholar CrossRef Search ADS PubMed  32. Hininger I,  Chollat-Namy A,  Sauvaigo S,  Osman M,  Faure H,  Cadet J,  Favier A,  Roussel AM.  Assessment of DNA damage by comet assay on frozen total blood: method and evaluation in smokers and non-smokers,  Mutat. Res. ,  2004, vol.  558 (pg.  75- 80) Google Scholar CrossRef Search ADS PubMed  33. Collins AR,  Dobson VL,  Dušinská M,  Kennedy G,  Štětina R.  The comet assay: what can it really tell us?,  Mutat. Res. ,  1997, vol.  375 (pg.  183- 193) Google Scholar CrossRef Search ADS PubMed  34. Lee OE,  Lee J,  Sul D,  Lee J.  Use of the tail moment of the lymphocytes to evaluate DNA damage in human biomonitoring studies,  Toxicol. Sci. ,  2004, vol.  81 (pg.  121- 132) Google Scholar CrossRef Search ADS PubMed  35. Kumaravel TS,  Jha AN.  Reliable Comet assay measurements for detecting DNA damage induced by ionizing radiation and chemicals,  Mutat. Res. ,  2006, vol.  605 (pg.  7- 16) Google Scholar CrossRef Search ADS PubMed  36. Collins AR,  Oscoz AA,  Brunborg G,  Gaivao I,  Kruszewski M,  Stetina R,  Smith C.  The comet assay: topical issues,  Mutagenesis ,  2008  Advance Access published on February 17, 2008. doi:10.1093/mutage/gem051 37. Collins AR,  Duthie SJ,  Dobson VL.  Direct enzymic detection of endogenous oxidative base damage in human lymphocyte DNA,  Carcinogenesis ,  1993, vol.  14 (pg.  1733- 1735) Google Scholar CrossRef Search ADS PubMed  38. Doetsch PW,  Henner WD,  Cunningham RP,  Toney JH,  Helland DE.  A highly conserved endonuclease activity present in Escherichia coli, bovine, and human cells recognizes oxidative DNA damage at sites of pyrimidines,  Mol. Cell. Biol. ,  1987, vol.  7 (pg.  26- 32) Google Scholar CrossRef Search ADS PubMed  39. Collins AR,  Dušinská M,  Gedik CM,  Štetina R.  Oxidative damage to DNA: do we have a reliable biomarker?,  Environ. Health Perspect. ,  1996, vol.  104  Suppl. 3(pg.  465- 469) Google Scholar CrossRef Search ADS PubMed  40. Dušinská M,  Collins A.  Detection of oxidised purines and UV-induced photoproducts in DNA of single cells, by inclusion of lesion-specific enzymes in the comet assay,  Altern. Lab. Anim. ,  1996, vol.  24 (pg.  405- 411) 41. Collins AR,  Dušinská M,  Horská A.  Detection of alkylation damage in human lymphocyte DNA with the comet assay,  Acta Biochim. Pol. ,  2001, vol.  48 (pg.  611- 614) Google Scholar PubMed  42. Angelis KJ,  Dušinská M,  Collins AR.  Single cell gel electrophoresis; detection of DNA damage at different levels of sensitivity,  Electrophoresis ,  1999, vol.  20 (pg.  2133- 2138) Google Scholar CrossRef Search ADS PubMed  43. Speit G,  Schütz P,  Bonzheim I,  Trenz K,  Hoffmann H.  Sensitivity of the FPG protein towards alkylation damage in the comet assay,  Toxicol. Lett. ,  2004, vol.  15 (pg.  151- 158) Google Scholar CrossRef Search ADS   44. Smith CC,  O'Donovan MR,  Martin EA.  hOGG1 recognizes oxidative damage using the comet assay with greater specificity than FPG or ENDOIII,  Mutagenesis ,  2006, vol.  21 (pg.  185- 190) Google Scholar CrossRef Search ADS PubMed  45. David SS,  Williams SD.  Chemistry of glycosylases and endonucleases involved in base-excision repair,  Chem. Rev. ,  1998, vol.  98 (pg.  1221- 1261) Google Scholar CrossRef Search ADS PubMed  46. Gedik CM,  Boyle SP,  Wood SG,  Vaughan NJ,  Collins AR.  Oxidative stress in humans: validation of biomarkers of DNA damage,  Carcinogenesis ,  2002, vol.  23 (pg.  1441- 1446) Google Scholar CrossRef Search ADS PubMed  47. Dušinská M,  Vallová B,  Ursínyová M,  Hladíková V,  Smolková B,  Wsólová L,  Rašlová K,  Collins AR.  DNA damage and antioxidants: fluctuation through the year in a central European population group,  Food Chem. Toxicol. ,  2002, vol.  40 (pg.  1119- 1123) Google Scholar CrossRef Search ADS PubMed  48. Smolková B,  Dušinská M,  Rašlová K,  McNeill G,  Spustová V,  Blažíček P,  Horská A,  Collins A.  Effects of seasonal variation in diet on markers of oxidative damage to lipids and DNA in defined population groups in Slovakia,  Mutat. Res. ,  2004, vol.  551 (pg.  135- 144) Google Scholar CrossRef Search ADS PubMed  49. Møller P,  Wallin H,  Holst E,  Knudsen LE.  Sunlight induced DNA damage in human mononuclear cells,  FASEB J. ,  2002, vol.  16 (pg.  45- 53) Google Scholar CrossRef Search ADS PubMed  50. Giovannelli L,  Saieva C,  Masala G,  Testa G,  Salvini S,  Pitozzi V,  Riboli E,  Dolara P,  Palli D.  Nutritional and lifestyle determinants of DNA oxidative damage: a study in a Mediterranean population,  Carcinogenesis ,  2002, vol.  23 (pg.  1483- 1489) Google Scholar CrossRef Search ADS PubMed  51. Collins AR,  Gaivao I.  DNA base excision repair as a biomarker in molecular epidemiology studies,  Mol. Aspects Med. ,  2007, vol.  28 (pg.  307- 322) Google Scholar CrossRef Search ADS PubMed  52. Gaivao I,  Piasek A,  Brevik A,  Shaposhnikov S,  Collins AR.  Comet assay-based methods for measuring DNA repair in vitro; estimates of inter- and intra-individual variation,  Cell Biol. Toxicol.   (in press) 53. Cebulska-Wasilewska A,  Wiecheć A,  Panek A,  Binková B,  Srám RJ,  Farmer PB.  Influence of environmental exposure to PAHs on the susceptibility of lymphocytes to DNA-damage induction and on their repair capacity,  Mutat. Res. ,  2005, vol.  30 (pg.  73- 81) Google Scholar CrossRef Search ADS   54. Cebulska-Wasilewska A,  Pawłyk I,  Panek A,  Wiecheć A,  Kalina I,  Popov T,  Georgieva T,  Farmer PB.  Exposure to environmental polycyclic aromatic hydrocarbons: influences on cellular susceptibility to DNA damage (sampling Kosice and Sofia),  Mutat. Res. ,  2007, vol.  620 (pg.  145- 154) Google Scholar CrossRef Search ADS PubMed  55. Vodicka P,  Kumar R,  Stetina R, et al.  Markers of individual susceptibility and DNA repair rate in workers exposed to xenobiotics in a tire plant,  Environ. Mol. Mutagen. ,  2004, vol.  44 (pg.  283- 292) Google Scholar CrossRef Search ADS PubMed  56. Vodicka P,  Tuimala J,  Stetina R, et al.  Cytogenetic markers, DNA single-strand breaks, urinary metabolites, and DNA repair rates in styrene-exposed lamination workers,  Environ. Health Perspect. ,  2004, vol.  112 (pg.  867- 871) Google Scholar CrossRef Search ADS PubMed  57. Aka P,  Mateuca R,  Buchet J-P,  Thierens H,  Kirsch-Volders M.  Are genetic polymorphisms in Ogg1, XRCC1 and XRCC3 genes predictive for the DNA strand break repair phenotype and genotoxicity in workers exposed to low dose ionising radiation?,  Mutat. Res. ,  2004, vol.  556 (pg.  169- 181) Google Scholar CrossRef Search ADS PubMed  58. Torbergsen AC,  Collins AR.  Recovery of human lymphocytes from oxidative DNA damage; the apparent enhancement of DNA repair by carotenoids is probably simply an antioxidant effect,  Eur. J. Nutr. ,  2000, vol.  39 (pg.  80- 85) Google Scholar CrossRef Search ADS PubMed  59. Collins AR,  Dušinská M,  Horváthová E,  Munro M,  Savio M,  Štetina R.  Inter-individual differences in repair of DNA base oxidation, measured in vitro with the comet assay,  Mutagenesis ,  2001, vol.  16 (pg.  297- 301) Google Scholar CrossRef Search ADS PubMed  60. Dušinská M,  Collins AR,  Kažimírová A, et al.  Genotoxic effects of asbestos in humans,  Mutat. Res. ,  2004, vol.  553 (pg.  91- 102) Google Scholar CrossRef Search ADS PubMed  61. Dušinská M,  Kažimírová A,  Barančoková M,  Horská A,  Burghardtová K,  Volkovová K,  Staruchová M,  Wsólová L,  Collins AR.  Does occupational exposure to mineral fibres cause DNA or chromosome damage?,  Mutat. Res. ,  2004, vol.  553 (pg.  103- 110) Google Scholar CrossRef Search ADS PubMed  62. Dušinská M,  Džupinková Z,  Wsólová Z,  Harrington V,  Collins AR.  Possible involvement of XPA in repair of oxidative DNA damage, deduced from analysis of damage, repair and genotype in a human population study,  Mutagenesis ,  2006, vol.  21 (pg.  205- 211) Google Scholar CrossRef Search ADS PubMed  63. Langie SAS,  Knaapen AM,  Brauers KJJ,  van Berlo D,  van Schooten F-J,  Godschalk RWL.  Development and validation of a modified comet assay to phenotypically assess nucleotide excision repair,  Mutagenesis ,  2006, vol.  21 (pg.  153- 158) Google Scholar CrossRef Search ADS PubMed  64. Kassie F,  Parzefall W,  Knasmüller S.  Single cell gel electrophoresis assay: a new technique for human biomonitoring studies,  Mutat. Res. ,  2000, vol.  463 (pg.  13- 31) Google Scholar CrossRef Search ADS PubMed  65. Faust F,  Kassie F,  Knasmüller S,  Kevekordes S,  Mersch-Sundermann V.  Use of primary blood cells for the assessment of exposure to occupational genotoxicants in human biomonitoring studies,  Toxicology ,  2004, vol.  198 (pg.  341- 350) Google Scholar CrossRef Search ADS PubMed  66. Plappert U,  Raddatz K,  Roth S,  Fliedner TM.  DNA-damage detection in man after radiation exposure—the comet assay—its possible application for human biomonitoring,  Stem Cells ,  1995, vol.  13  Suppl 1(pg.  215- 222) Google Scholar CrossRef Search ADS PubMed  67. Albertini RJ,  Anderson D,  Douglas GR, et al.  IPCS guidelines for the monitoring of genotoxic effects of carcinogens in humans,  Mutat. Res. ,  2000, vol.  463 (pg.  111- 172) Google Scholar CrossRef Search ADS PubMed  68. Bates MN,  Hamilton JW,  LaKind JS,  Langenberg P,  O'Malley M,  Snodgrass W.  Workgroup report: biomonitoring study design, interpretation, and communication—lessons learned and path forward,  Environ. Health Perspect. ,  2005, vol.  113 (pg.  1615- 1621) Google Scholar CrossRef Search ADS PubMed  69. Au WW.  Usefulness of biomarkers in population studies: from exposure to susceptibility and to prediction of cancer,  Int. J. Hyg. Environ. Health ,  2007, vol.  210 (pg.  239- 246) Google Scholar CrossRef Search ADS PubMed  70. ESCODD Comparative analysis of baseline 8-oxo-7,8-dihydroguanine in mammalian cell DNA, by different methods in different laboratories: an approach to consensus,  Carcinogenesis ,  2002, vol.  23 (pg.  2129- 2133) CrossRef Search ADS PubMed  71. Dušinská M,  Lietava J,  Olmedilla B,  Rašlová K,  Southon S,  Collins AR.  Basu TK,  Temple NJ,  Garg ML.  Indicators of oxidative stress, antioxidants and human health,  Antioxidants in Human Health ,  1999 Oxford CABI Publishing 72. Collins AR,  Dušinská M.  Armstrong D.  Oxidation of cellular DNA, measured with the comet assay,  Oxidative Stress Biomarkers and Antioxidant Protocols ,  2002 Totowa, NJ Humana Press(pg.  147- 159) 73. Rison L,  Møller P,  Loft S.  Oxidative stress-induced DNA damage by particulate air pollution,  Mutat. Res. ,  2005, vol.  592 (pg.  119- 137) Google Scholar CrossRef Search ADS PubMed  74. Collins A,  Cadet J,  Epe B,  Gedik C.  Problems in the measurement of 8-oxoguanine in human DNA. Report of a workshop, DNA oxidation, held in Aberdeen, UK, 19–21 January, 1997,  Carcinogenesis ,  1997, vol.  18 (pg.  1833- 1836) Google Scholar CrossRef Search ADS PubMed  75. Ravanat,J-LTuresky RJ,  Gremaud E,  Trudel LJ,  Stadler RH.  Determination of 8-oxoguanine in DNA by gas chromatography-mass spectrometry and HPLC-electrochemical detection: overestimation of the background level of the oxidised base by the gas chromatography-mass spectrometry assay,  Chem. Res. Toxicol. ,  1995, vol.  8 (pg.  1039- 1045) Google Scholar CrossRef Search ADS PubMed  76. Gedik CM,  Collins A.  ESCODD Establishing the background level of base oxidation in human lymphocyte DNA: results of an interlaboratory validation study,  FASEB J. ,  2005, vol.  19 (pg.  82- 84) Google Scholar PubMed  77. Collins A.  Oxidative DNA damage, antioxidants, and cancer,  Bioessays ,  1999, vol.  21 (pg.  238- 246) Google Scholar CrossRef Search ADS PubMed  78. Hagmar L,  Bonassi S,  Strömberg U,  Brøgger A,  Knudsen LE,  Norppa H,  Reuterwall C.  the European Study Group on Cytogenetic Biomarkers and Health Chromosome aberrations in lymphocytes predict human cancer: a report from the European study group on cytogenetic biomarkers and health (ESCH),  Cancer Res. ,  1998, vol.  58 (pg.  4117- 4121) Google Scholar PubMed  79. Bonassi S,  Znaor A,  Ceppi M,  Lando C,  Chang WP,  Holland N, et al.  An increased micronucleus frequency in peripheral blood lymphocytes predicts the risk of cancer in humans,  Carcinogenesis ,  2007, vol.  28 (pg.  625- 631) Google Scholar CrossRef Search ADS PubMed  80. Hoffmann H,  Hogel J,  Speit G.  The effect of smoking on DNA effects in the comet assay: a metaanalysis,  Mutagenesis ,  2005, vol.  20 (pg.  455- 466) Google Scholar CrossRef Search ADS PubMed  81. Wojewódska M,  Kruszewski M,  Iwaneñko T,  Collins AR,  Szumiel I.  Application of the comet assay for monitoring DNA damage in workers exposed to chronic low-dose irradiation I. Strand breakage,  Mutat. Res. ,  1998, vol.  416 (pg.  21- 35) Google Scholar CrossRef Search ADS PubMed  82. Kruszewski M,  Wojewódska M,  Iwaneñko T,  Collins AR,  Szumiel I.  Application of the comet assay for monitoring DNA damage in workers exposed to chronic low-dose irradiation II. Base damage,  Mutat. Res. ,  1998, vol.  416 (pg.  37- 57) Google Scholar CrossRef Search ADS PubMed  83. Cavallo D,  Iavicoli I,  Setini A,  Marinaccio A,  Perniconi B,  Carelli G,  Iavicoli S.  Genotoxic risk and oxidative DNA damage in workers exposed to antimony trioxide,  Environ. Mol. Mutagen. ,  2002, vol.  40 (pg.  184- 189) Google Scholar CrossRef Search ADS PubMed  84. Somorovská M,  Szabová E,  Vodička P, et al.  Biomonitoring of genotoxic risk in workers in a rubber factory: comparison of the comet assay with cytogenetic methods and immunology,  Mutat. Res. ,  1999, vol.  445 (pg.  181- 192) Google Scholar CrossRef Search ADS PubMed  85. Somorovská M,  Jahnová E,  Tulinská J, et al.  Biomonitoring of occupational exposure to styrene in a plastics lamination plant,  Mutat. Res. ,  1999, vol.  428 (pg.  255- 269) Google Scholar CrossRef Search ADS PubMed  86. Staruchova M,  Volkovova K,  Lajdova A,  Mislanova C,  Collins A,  Wsolova L,  Staruch L,  Dusinska M.  Importance of diet in protection against oxidative damage,  Neuro. Endocrinol. Lett. ,  2006, vol.  27  Suppl. 2(pg.  112- 115) Google Scholar PubMed  87. Kopjar N,  Zeljezić D,  Garaj-Vrhovac V.  Evaluation of DNA damage in white blood cells of healthy human volunteers using the alkaline comet assay and the chromosome aberration test,  Acta Biochim. Pol. ,  2006, vol.  53 (pg.  321- 336) Google Scholar PubMed  88. Decordier I,  De Bont K,  De Bock K,  Mateuca R,  Roelants M,  Ciardelli R,  Haumont D,  Knudsen LE,  Kirsch-Volders M.  Genetic susceptibility of newborn daughters to oxidative stress,  Toxicol. Lett. ,  2007, vol.  30 (pg.  68- 84) Google Scholar CrossRef Search ADS   89. Knudsen LE,  Hansen AM.  Biomarkers of intermediate endpoints in environmental and occupational health,  Int. J. Hyg. Environ. Health ,  2007, vol.  210 (pg.  461- 470) Google Scholar CrossRef Search ADS PubMed  90. Jasso-Pineda Y,  Espinosa-Reyes G,  González-Mille D, et al.  An integrated health risk assessment approach to the study of mining sites contaminated with arsenic and lead,  Integr. Environ. Assess. Manag. ,  2007, vol.  3 (pg.  344- 350) Google Scholar CrossRef Search ADS PubMed  91. Pérez-Maldonado IN,  Athanasiadou M,  Yáñez L,  González-Amaro R,  Bergman A,  Díaz-Barriga F.  DDE-induced apoptosis in children exposed to the DDT metabolite,  Sci. Total. Environ. ,  2006, vol.  370 (pg.  343- 351) Google Scholar CrossRef Search ADS PubMed  92. González C,  Nájera O,  Cortés E,  Toledo G,  López L,  Betancourt M,  Ortiz R.  Susceptibility to DNA damage induced by antibiotics in lymphocytes from malnourished children,  Teratog. Carcinog. Mutagen. ,  2002, vol.  22 (pg.  147- 158) Google Scholar CrossRef Search ADS PubMed  93. Ortiz-Pérez MD,  Torres-Dosal A,  Batres LE, et al.  Environmental health assessment of deltamethrin in a malarious area of Mexico: environmental persistence, toxicokinetics, and genotoxicity in exposed children,  Environ. Health. Perspect. ,  2005, vol.  113 (pg.  782- 786) Google Scholar CrossRef Search ADS PubMed  94. Wilhelm M,  Wittsiepe J,  Furst P, et al.  HBM in North Rhine Westphalia, Germany,  Int. J. Hyg. Environ. Health ,  2007, vol.  210 (pg.  307- 318) Google Scholar CrossRef Search ADS PubMed  95. Bajpayee M,  Dhawan A,  Parmar D,  Pandey AK,  Mathur N,  Seth PK.  Gender-related differences in basal DNA damage in lymphocytes of a healthy Indian population using the alkaline Comet assay,  Mutat. Res. ,  2002, vol.  520 (pg.  83- 91) Google Scholar CrossRef Search ADS PubMed  96. Dušinská M,  Ficek A,  Horská A, et al.  Glutathione S-transferase polymorphisms influence the level of oxidative DNA damage and antioxidant protection in humans,  Mutat. Res. ,  2001, vol.  482 (pg.  47- 55) Google Scholar CrossRef Search ADS PubMed  97. Humphreys V,  Martin RM,  Ratcliffe B,  Duthie S,  Wood SJ,  Gunnell D,  Collins AR.  Age-related increases in DNA repair and antioxidant production: a comparison of the Boyd Orr cohort of elderly subjects with a younger cohort,  Age Ageing ,  2007, vol.  36 (pg.  521- 526) Google Scholar CrossRef Search ADS PubMed  98. Cebulska-Wasilewska A,  Panek A,  Zabiński Z,  Moszczyński P,  Au WW.  Occupational exposure to mercury vapour on genotoxicity and DNA repair,  Mutat. Res. ,  2005, vol.  586 (pg.  102- 114) Google Scholar CrossRef Search ADS PubMed  99. Mohankumar MN,  Janani S,  Prabhu BK,  Kumar PR,  Jeevanram RK.  DNA damage and integrity of UV-induced DNA repair in lymphocytes of smokers analysed by the comet assay,  Mutat. Res. ,  2002, vol.  520 (pg.  179- 187) Google Scholar CrossRef Search ADS PubMed  100. Piperakis SM,  Petrakou E,  Tsilimigaki S.  Effects of air pollution and smoking on DNA damage of human lymphocytes,  Environ. Mol. Mutagen. ,  2000, vol.  36 (pg.  243- 249) Google Scholar CrossRef Search ADS PubMed  101. Marcon F,  Andreoli C,  Rossi S,  Verdina A,  Galati R,  Crebelli R.  Assessment of individual sensitivity to ionizing radiation and DNA repair efficiency in a healthy population,  Mutat. Res. ,  2003, vol.  541 (pg.  1- 8) Google Scholar CrossRef Search ADS PubMed  102. Palyvoda O,  Polańska J,  Wygoda A,  Rzeszowska-Wolny J.  DNA damage and repair in lymphocytes of normal individuals and cancer patients: studies by the comet assay and micronucleus tests,  Acta Biochim. Pol. ,  2003, vol.  50 (pg.  181- 190) Google Scholar PubMed  103. Piperakis SM,  Petrakou E,  Tsilimigaki S,  Sagnou M,  Monogiudis E,  Haniotakis G,  Karkaseli H,  Sarikaki E.  Biomonitoring with the comet assay of Greek greenhouse workers exposed to Pesticides,  Environ. Mol. Mutagen. ,  2003, vol.  41 (pg.  104- 110) Google Scholar CrossRef Search ADS PubMed  104. Vodicka P,  Stetina R,  Polakova V, et al.  Association of DNA repair polymorphisms with DNA repair functional outcomes in healthy human subjects,  Carcinogenesis ,  2007, vol.  28 (pg.  657- 664) Google Scholar CrossRef Search ADS PubMed  105. Panayiotidis M,  Collins AR.  Ex vivo assessment of lymphocyte antioxidant status using the comet assay,  Free Rad. Res. ,  1997, vol.  27 (pg.  533- 537) Google Scholar CrossRef Search ADS   106. Boyle SP,  Dobson VL,  Duthie SJ,  Hinselwood DC,  Kyle JAM,  Collins AR.  Bioavailability and efficiency of rutin as an antioxidant: a human supplementation study,  Eur. J. Nutr. ,  2000, vol.  54 (pg.  774- 782) Google Scholar CrossRef Search ADS   107. Collins AR.  Assays for oxidative stress and antioxidant status: applications to research into the biological effectiveness of polyphenols,  Am. J. Clin. Nutr. ,  2005, vol.  81 (pg.  261S- 267S) Google Scholar PubMed  108. Porrini M,  Riso P,  Brusamolino A,  Berti C,  Guarnieri S,  Visioli F.  Daily intake of a formulated tomato drink affects carotenoid plasma and lymphocyte concentrations and improves cellular antioxidant protection,  Br. J. Nutr. ,  2005, vol.  93 (pg.  93- 99) Google Scholar CrossRef Search ADS PubMed  109. Wilms LC,  Boots AW,  de Boer VC, et al.  Impact of multiple genetic polymorphisms on effects of a 4-week blueberry juice intervention on ex vivo induced lymphocytic DNA damage in human volunteers,  Carcinogenesis ,  2007, vol.  28 (pg.  1800- 1806) Google Scholar CrossRef Search ADS PubMed  110. Duthie SJ,  Ma A,  Ross MA,  Collins AR.  Antioxidant supplementation decreases oxidative DNA damage in human lymphocytes,  Cancer Res. ,  1996, vol.  56 (pg.  1291- 1295) Google Scholar PubMed  111. Pool-Zobel BL,  Bub A,  Muller H,  Wollowski I,  Rechkemmer G.  Consumption of vegetables reduces genetic damage in humans: first results of a human intervention trial with carotenoid-rich foods,  Carcinogenesis ,  1997, vol.  18 (pg.  1847- 1850) Google Scholar CrossRef Search ADS PubMed  112. Mitchell JH,  Collins AR.  Effects of a soy milk supplement on plasma cholesterol levels and oxidative DNA damage in men—a pilot study,  Eur. J. Nutr. ,  1999, vol.  38 (pg.  143- 148) Google Scholar CrossRef Search ADS PubMed  113. Collins AR,  Harrington V,  Drew J,  Melvin R.  Nutritional modulation of DNA repair in a human intervention study,  Carcinogenesis ,  2003, vol.  24 (pg.  511- 515) Google Scholar CrossRef Search ADS PubMed  114. Park YK,  Park E,  Kim JS,  Kang MH.  Daily grape juice consumption reduces oxidative DNA damage and plasma free radical levels in healthy Koreans,  Mutat. Res. ,  2003, vol.  28 (pg.  77- 86) Google Scholar CrossRef Search ADS   115. Møller P,  Loft S.  Interventions with antioxidants and nutrients in relation to oxidative DNA damage and repair,  Mutat. Res. ,  2004, vol.  551 (pg.  79- 89) Google Scholar CrossRef Search ADS PubMed  116. Møller P,  Loft S.  Dietary antioxidants and beneficial effect on oxidatively damaged DNA,  Free Radic. Biol. Med. ,  2006, vol.  41 (pg.  388- 415) Google Scholar CrossRef Search ADS PubMed  117. Kažimírová A,  Barančoková M,  Volkovová K,  Staruchová M,  Kudláčková M,  Wsólová L,  Dušinská M.  Does a vegetarian diet influence genomic stability?,  Eur. J. Nutr. ,  2004, vol.  43 (pg.  32- 38) Google Scholar CrossRef Search ADS PubMed  118. Smolkova B,  Dusinska M,  Raslova K,  Barancokova M,  Kazimirova A,  Horska A,  Spustova V,  Collins A.  Folate levels determine effect of antioxidant supplementation on micronuclei in subjects with cardiovascular risk,  Mutagenesis ,  2004, vol.  19 (pg.  469- 476) Google Scholar CrossRef Search ADS PubMed  119. Dušinská M,  Kažimírová A,  Barančoková M,  Beňo M,  Rašlová K,  Smolková B,  Horská A,  Wsólová L,  Collins AR.  Nutritional supplementation with antioxidants decreases chromosomal damage in humans,  Mutagenesis ,  2003, vol.  18 (pg.  371- 376) Google Scholar CrossRef Search ADS PubMed  120. Staruchova M,  Collins AR,  Volkovova K, et al.  Occupational exposure to mineral fibres Biomarkers of oxidative damage and antioxidant defence and associations with DNA damage and repair,  Mutagenesis ,  2008  Advance Access published on February 14, 2008. doi:10.1093/mutage/gen004 121. Watters JL,  Satia JA,  Kupper LL,  Swenberg JA,  Schroeder JC,  Switzer BR.  Associations of antioxidant nutrients and oxidative DNA damage in healthy African-American and White adults,  Cancer Epidemiol. Biomarkers Prev. ,  2007, vol.  16 (pg.  1428- 1436) Google Scholar CrossRef Search ADS PubMed  122. Vodička P,  Souček P,  Tates AD, et al.  Association between genetic polymorphism and biomarkers in styrene-exposed workers,  Muta. Res. ,  2001, vol.  482 (pg.  89- 103) Google Scholar CrossRef Search ADS   123. Naccarati A,  Soucek P,  Stetina R, et al.  Genetic polymorphisms and possible gene-gene interactions in metabolic and DNA repair genes: effects on DNA damage,  Mutat. Res. ,  2006, vol.  593 (pg.  22- 31) Google Scholar CrossRef Search ADS PubMed  124. Srám RJ.  Effect of glutathione S-transferase M1 polymorphisms on biomarkers of exposure and effects,  Environ. Health Perspect. ,  1998, vol.  106 (pg.  231- 239) Google Scholar CrossRef Search ADS PubMed  125. Srám RJ,  Binková B,  Rössner P,  Rubes J,  Topinka J,  Dejmek J.  Adverse reproductive outcomes from exposure to environmental mutagens,  Mutat. Res. ,  1999, vol.  428 (pg.  203- 215) Google Scholar CrossRef Search ADS PubMed  126 Pavanello S,  Clonfero E.  Biological indicators of genotoxic risk and metabolic polymorphisms,  Mutat Res. ,  2000, vol.  463 (pg.  285- 308) Google Scholar CrossRef Search ADS PubMed  127. Laffon B,  Teixeira JP,  Silva S,  Roma-Torres J,  Pérez-Cadahía B,  Méndez J,  Pásaro E,  Mayan O.  Assessment of occupational genotoxic risk in the production of rubber tyres,  Ann. Occup. Hyg. ,  2006, vol.  50 (pg.  583- 592) Google Scholar PubMed  128. Knudsen LE,  Gaskell M,  Martin EA,  Poole J,  Scheepers PT,  Jensen A,  Autrup H,  Farmer PB.  Genotoxic damage in mine workers exposed to diesel exhaust, and the effects of glutathione transferase genotypes,  Mutat. Res. ,  2005, vol.  583 (pg.  120- 132) Google Scholar CrossRef Search ADS PubMed  129. Marczynski B,  Rihs HP,  Rossbach B,  Hölzer J,  Angerer J,  Scherenberg M,  Hoffmann G,  Brüning T,  Wilhelm M.  Analysis of 8-oxo-7,8-dihydro-2′-deoxyguanosine and DNA strand breaks in white blood cells of occupationally exposed workers: comparison with ambient monitoring, urinary metabolites and enzyme polymorphisms,  Carcinogenesis ,  2002, vol.  23 (pg.  273- 281) Google Scholar CrossRef Search ADS PubMed  130. van Delft JH,  Steenwinkel MS,  van Asten JG, et al.  Biological monitoring the exposure to polycyclic aromatic hydrocarbons of coke oven workers in relation to smoking and genetic polymorphisms for GSTM1 and GSTT1,  Ann. Occup. Hyg. ,  2001, vol.  45 (pg.  395- 408) Google Scholar CrossRef Search ADS PubMed  131. Binková B,  Lewtas J,  Mísková I, et al.  Biomarker studies in northern Bohemia,  Environ. Health Perspect. ,  1996, vol.  104  Suppl. 3(pg.  591- 597) Google Scholar CrossRef Search ADS PubMed  132. Carere A,  Andreoli C,  Galati R, et al.  Biomonitoring of exposure to urban air pollutants: analysis of sister chromatid exchanges and DNA lesions in peripheral lymphocytes of traffic policemen,  Mutat. Res. ,  2002, vol.  518 (pg.  215- 224) Google Scholar CrossRef Search ADS PubMed  133. Novotna B,  Topinka J,  Solansky I,  Chvatalova I,  Lnenickova Z,  Sram RJ.  Impact of air pollution and genotype variability on DNA damage in Prague policemen,  Toxicol. Lett. ,  2007, vol.  30 (pg.  37- 47) Google Scholar CrossRef Search ADS   134. Slyskova J,  Dusinska M,  Kuricova M,  Soucek P,  Vodickova L,  Susova S,  Naccarati A,  Tulupova E,  Vodicka P.  Relationship between the capacity to repair 8-oxoguanine, biomarkers of genotoxicity and individual susceptibility in styrene-exposed workers,  Mutat. Res. ,  2007, vol.  634 (pg.  101- 111) Google Scholar CrossRef Search ADS PubMed  135. Matés M.  Effects of antioxidant enzymes in the molecular control of reactive oxygen species toxicology,  Toxicology ,  2000, vol.  153 (pg.  83- 104) Google Scholar CrossRef Search ADS PubMed  136. Pedersen JZ,  De Maria F,  Turella P, et al.  Glutathione transferases sequester toxic dinitrosyl-iron complexes in cells. A protection mechanism against excess nitric oxide,  J. Biol. Chem. ,  2007, vol.  282 (pg.  6364- 6371) Google Scholar CrossRef Search ADS PubMed  137. Stella L,  Pallottini V,  Moreno S, et al.  Electrostatic association of glutathione transferase to the nuclear membrane. Evidence of an enzyme defense barrier at the nuclear envelope,  J. Biol. Chem. ,  2007, vol.  282 (pg.  6372- 6379) Google Scholar CrossRef Search ADS PubMed  138. Mateuca R,  Aka PV,  De Boeck M,  Hauspie R,  Kirsch-Volders M,  Lison D.  Influence of hOGG1, XRCC1 and XRCC3 genotypes on biomarkers of genotoxicity in workers exposed to cobalt or hard metal dusts,  Toxicol. Lett. ,  2005, vol.  156 (pg.  277- 288) Google Scholar CrossRef Search ADS PubMed  139. Leng S,  Cheng J,  Pan Z,  Huang C,  Niu Y,  Dai Y,  Li B,  He F,  Zheng Y.  Associations between XRCC1 and ERCC2 polymorphisms and DNA damage in peripheral blood lymphocyte among coke oven workers,  Biomarkers ,  2004, vol.  9 (pg.  395- 406) Google Scholar CrossRef Search ADS PubMed  140. Vodicka P,  Kumar R,  Stetina R, et al.  Genetic polymorphisms in DNA repair genes and possible links with DNA repair rates, chromosomal aberrations and single-strand breaks in DNA,  Carcinogenesis ,  2004, vol.  25 (pg.  757- 763) Google Scholar CrossRef Search ADS PubMed  141. Sugasawa K,  Ng JMY,  Masutani C,  Iwai S,  van der Spek PJ,  Eker APM,  Hanaoka F,  Bootsma D,  Hoeijmakers JHJ.  Xeroderma pigmentosum group C protein complex is the initiator of global genome nucleotide excision repair,  Mol. Cell ,  1998, vol.  2 (pg.  223- 232) Google Scholar CrossRef Search ADS PubMed  142. Reardon JT,  Bessho T,  Kung HC,  Bolton PH,  Sancar A.  In vitro repair of oxidative DNA damage by human nucleotide excision repair system: possible explanation for neurodegeneration in xeroderma pigmentosum patients,  Proc. Natl Acad. Sci. USA ,  1997, vol.  94 (pg.  9463- 9468) Google Scholar CrossRef Search ADS   143. Wu X,  Zhao H,  Wei Q,  Amos CI,  Zhang K,  Guo Z,  Qiao Y,  Hong WK,  Spitz MR.  XPA polymorphism associated with reduced lung cancer risk and a modulating effect on nucleotide excision repair capacity,  Carcinogenesis ,  2003, vol.  24 (pg.  505- 509) Google Scholar CrossRef Search ADS PubMed  144. Mayne ST.  Antioxidant nutrients and chronic disease: use of biomarkers of exposure and oxidative stress status in epidemiologic research,  J. Nutr. ,  2003, vol.  133 (pg.  933- 940) © The Author 2008. Published by Oxford University Press on behalf of the UK Environmental Mutagen Society. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.
TI - The comet assay in human biomonitoring: gene–environment interactions
JO - Mutagenesis
DO - 10.1093/mutage/gen007
DA - 2008-03-07
UR - https://www.deepdyve.com/lp/oxford-university-press/the-comet-assay-in-human-biomonitoring-gene-environment-interactions-MDJ1kg6T8s
SP - 191
EP - 205
VL - 23
IS - 3
DP - DeepDyve
ER -