**2.1.2 Micronucleus in exfoliated buccal cells**

Regeneration is dependent on the number and division rate of the proliferating (basal) cells, their genomic stability and their propensity for cell death. These events can be studied in the buccal mucosa (BM), which is an easily accessible tissue for sampling cells in a minimally invasive manner and does not cause undue stress to study subjects. This method is increasingly used in molecular epidemiology studies for investigating the impact of nutrition, lifestyle factors, genotoxin exposure and genotype on DNA damage, chromosome malsegregation and cell death (Thomas et al., 2009).

The assay has been successfully to study DNA damage as measured by MN or by the use of fluorescent probes to detect in BM is an indication of the regenerative capacity of this tissue. The BM provides a barrier to potential carcinogens that can be metabolized to generate potential reactive products. As up to 90% of all cancers appear to be epithelial in origin, the BM could be used to monitor early genotoxic events as a result of potential carcinogens entering the body through ingestion or inhalation. Exfoliated buccal cells have been used non-invasively to successfully show the genotoxic effects of lifestyle factors such as tobacco smoking, chewing of betel nuts and/or quids, medical treatments, such as radiotherapy as well as occupational exposure, exposure to potentially mutagenic and/or carcinogenic chemicals, and for studies of chemoprevention of cancer.

In this assay cells derived from the BM are harvested from the inside of a patient's mouth using a small-headed toothbrush. The cells are washed to remove the debris and bacteria, and a single-cell suspension is prepared and applied to a clean slide using a cytocentrifuge. The cells are stained with Feulgen and Light Green stain allowing both bright field and permanent fluorescent analysis that can be undertaken microscopically (Thomas et al., 2009).

The Buccal Mucosa Cytome (BMCyt) assay has been used to measure biomarkers of DNA damage (MN and/or nuclear buds), cytokinetic defects (binucleated cells) and proliferative potential (basal cell frequency) and/or cell death (condensed chromatin, karyorrhexis, pyknotic and karyolitic cells). The protocol can also make use of molecular probes for DNA adduct, aneuploidy and chromosome break measures within the nuclei of buccal cells. Furthermore, chromosome-specific centromeric probes have been used to measure aneuploidy by determining the frequency of nuclei with abnormal chromosome number. Tandem probes have been successfully applied to measure chromosome breaks in specific important regions of the genome (Thomas et al., 2009).

The methodology and concepts described in this protocol may be applied to other types of exfoliated cells such as those of the bladder, nose and cervix but the morphological characteristics, sampling and scoring methods are neither properly described nor standardized for cells from these tissues (Thomas et al., 2009).

The time of sampling is also an important variable to consider. As the buccal cells turn over every 7-21 days, it is theoretically possible to observe the genotoxic effects of an acute exposure approximately 7-21 days later.

complexes (NBUD); chromosomal instability phenotype and breakage-fusion-bridge cycles (simultaneous expression of MN, NPB and NBUD); altered mitotic activity and/or cytostasis (NDI) and cell death by necrosis or apoptosis (ratios of necrotic and apoptotic

Regeneration is dependent on the number and division rate of the proliferating (basal) cells, their genomic stability and their propensity for cell death. These events can be studied in the buccal mucosa (BM), which is an easily accessible tissue for sampling cells in a minimally invasive manner and does not cause undue stress to study subjects. This method is increasingly used in molecular epidemiology studies for investigating the impact of nutrition, lifestyle factors, genotoxin exposure and genotype on DNA damage, chromosome

The assay has been successfully to study DNA damage as measured by MN or by the use of fluorescent probes to detect in BM is an indication of the regenerative capacity of this tissue. The BM provides a barrier to potential carcinogens that can be metabolized to generate potential reactive products. As up to 90% of all cancers appear to be epithelial in origin, the BM could be used to monitor early genotoxic events as a result of potential carcinogens entering the body through ingestion or inhalation. Exfoliated buccal cells have been used non-invasively to successfully show the genotoxic effects of lifestyle factors such as tobacco smoking, chewing of betel nuts and/or quids, medical treatments, such as radiotherapy as well as occupational exposure, exposure to potentially mutagenic and/or carcinogenic

In this assay cells derived from the BM are harvested from the inside of a patient's mouth using a small-headed toothbrush. The cells are washed to remove the debris and bacteria, and a single-cell suspension is prepared and applied to a clean slide using a cytocentrifuge. The cells are stained with Feulgen and Light Green stain allowing both bright field and permanent

The Buccal Mucosa Cytome (BMCyt) assay has been used to measure biomarkers of DNA damage (MN and/or nuclear buds), cytokinetic defects (binucleated cells) and proliferative potential (basal cell frequency) and/or cell death (condensed chromatin, karyorrhexis, pyknotic and karyolitic cells). The protocol can also make use of molecular probes for DNA adduct, aneuploidy and chromosome break measures within the nuclei of buccal cells. Furthermore, chromosome-specific centromeric probes have been used to measure aneuploidy by determining the frequency of nuclei with abnormal chromosome number. Tandem probes have been successfully applied to measure chromosome breaks in specific

The methodology and concepts described in this protocol may be applied to other types of exfoliated cells such as those of the bladder, nose and cervix but the morphological characteristics, sampling and scoring methods are neither properly described nor

The time of sampling is also an important variable to consider. As the buccal cells turn over every 7-21 days, it is theoretically possible to observe the genotoxic effects of an acute

fluorescent analysis that can be undertaken microscopically (Thomas et al., 2009).

cells) (Fenech, 2007).

**2.1.2 Micronucleus in exfoliated buccal cells** 

malsegregation and cell death (Thomas et al., 2009).

chemicals, and for studies of chemoprevention of cancer.

important regions of the genome (Thomas et al., 2009).

exposure approximately 7-21 days later.

standardized for cells from these tissues (Thomas et al., 2009).

Ideally, repeat sampling, at least once every 7 days after acute exposure, should be performed for 28 days or more so that the kinetics and extent of biomarker induction can be thoroughly investigated. In the case of chronic exposure due to habitual diet or alcohol consumption or smoking it is recommend that multiple samples are taken at least once every 3 months to take into account seasonal variation (Thomas et al., 2009).

The uniformity of sampling is one of the many aspects to consider; therefore a circular expanding motion is used with toothbrush sampling to enhance sampling over a greater area and to avoid continual erosion in a single region of the BM. This is performed on the inside of both cheeks using a different brush for sampling left and right areas of the mouth to maximize cell sampling and to eliminate any unknown biases that may be caused by sampling one cheek only. It is important to note that repeated vigorous brushing of the same area can lead to increased collection of cells from the less differentiated basal layer. About transportation, in some investigations buccal cells may have to be collected from a distant site which may cause sample deterioration. About cell fixation, there are many possible alternatives of fixatives such as methanol: glacial acetic acid (3:1), 80% methanol or ethanol: glacial acetic (3:1). The staining technique recommend is Feulgen because is a DNAspecific stain and because permanent slides can be obtained that can be viewed under both transmitted and/or fluorescent light conditions. There are many false-positive results in MN frequency as a result of using Romanowsky-type stains such as Giemsa, May-Grunwald Giemsa and/or Leishmann's which leads to inaccurate assessment of DNA damage. Romanowsky stains have been shown to increase the number of false positives as they positively stain keratin bodies that are often mistaken for MN and are therefore not appropriate for this type of analysis. For these reasons, it is advisable to avoid Romanowsky stains in favour of DNA-specific fluorescent-based stains such as propidium iodide, DAPI, Feulgen, Hoechst 33258 or Acridine Orange (Thomas et al., 2009).

The criterion of scoring is originally based in the described by Tolbert et al. that are intended for classifying buccal cells into categories that distinguish between "normal" cells and cells that are considered "abnormal" on the basis of cytological and nuclear features, which are indicative of DNA damage, cytokinetic failure or cell death. Therefore, some definitions of the cytological findings are (Thomas et al., 2009):

Normal "differentiated" cells have a uniformly stained nucleus, which is oval or round in shape. They are distinguished from basal cells by their larger size and by their smaller nucleus-to-cytoplasm ratio. No other DNA-containing structures apart from the nucleus are observed in these cells. These cells are considered to be terminally differentiated relative to basal cells, as no mitotic cells are observed in this population.

Cells with MN are characterized by the presence of both a main nucleus and one more smaller nuclear structures called MN. The MN are round or oval in shape and their diameter should range between 1/3 and 1/16 of the main nucleus. MN has the same staining intensity and texture as the main nucleus. Most cells with MN will ontain only one MN but it is possible to find cells with two or more MN. Baseline frequencies for micronucleated cells in the BM are usually within the 0.5-2.5 MN/1000 cells range. Cells with multiple MN are rare in healthy subjects but become more common in individuals exposed to radiation or other genotoxic events.

Cells with nuclear buds contain nuclei with an apparent sharp constriction at one end of the nucleus suggestive of a budding process, i.e. elimination of nuclear material by budding.

Genotoxicity Biomarkers: Application in Histopathology Laboratories 141

endogenous metabolic product of N-, O- and S-demethylation reactions in most living systems. It is used mainly in the production of resins and their applications, such as adhesives and binders in wood product, pulp and paper, synthetic vitreous fibre industries, production of plastics, coatings, textile finishing and also as an intermediate in the synthesis of other industrial chemical compounds. Common non-occupational sources of exposure to FA include vehicle emissions, particle boards and similar building materials, carpets, paints and varnishes, food and cooking, tobacco smoke and its use as a disinfectant (Conaway et

Commercially, FA is manufactured as an aqueous solution called formalin, usually containing 37 to 40% by weight of dissolved FA (Zhang et al., 2009), which is commonly used in histopathology laboratories as a cytological fixative to preserve the integrity of

Exogenous FA can be absorbed following inhalation, dermal or oral exposure, being the level of absorption dependent on the route of exposure. The International Agency for Research on Cancer (IARC) reclassified FA as a human carcinogen (group 1) in June 2004 based on *"sufficient epidemiological evidence that FA causes nasopharyngeal cancer in humans"* (IARC, 2006; Zhang et al., 2009). In their review, IARC also concluded that there was *''strong but not sufficient evidence for a causal association between leukaemia and occupational exposure to FA'*' (Zhang et al., 2009, 2010). However, some studies have also led to mixed results and

The inhalation of vapours can produce irritation to eyes, nose and the upper respiratory system. Whilst occupational exposure to high FA concentrations may result in respiratory irritation and asthmatic reactions, it may also aggravate a pre-existing asthma condition. Skin reactions, following exposure to FA are very common, because the chemical is both irritating and allergenic (Pala et al., 2008). FA induces genotoxic and cytotoxic effects in bacteria and mammals cells (Ye et al., 2005) and its genotoxicity and carcinogenicity has been proved in experimental and epidemiological studies that used proliferating cultured mammalian cell lines and human lymphocytes (Pala et al., 2008; Speit et al., 2007) by DNAprotein cross-links, chromosome aberrations, sister exchange chromatides, and MN (Zhang

The goal of this study was to compare the frequency of genotoxicity biomarkers, provided by CBMN assay in peripheral lymphocytes and MN test in buccal cells between workers of histopathology laboratories exposed to FA and individuals non-exposed to FA and other

The study population consisted of 56 workers occupationally exposed to FA from 7 hospital histopathology laboratories located in Portugal (Lisbon and Tagus Valley region), and 85 administrative staff without occupational exposure to FA. The characteristics of both groups

Ethical approval for this study was obtained from the institutional Ethical Board and Director of the participating hospitals, and all subjects gave informed consent to participate in this study. Every person filled a questionnaire aimed at identifying exclusion criteria like history of cancer, radio or chemotherapy, use of therapeutic drugs, exposure to diagnostic X-rays in the past six months, intake of vitamins or other supplements like folic acid as well

al., 1996; Franks, 2005; IARC, 2006; Pala et al., 2008; Viegas & Prista, 2007).

cellular architecture for diagnosis.

et al., 2009).

are described in Table 1.

inconclusive evidence (Franks, 2005; Speit et al., 2010).

environmental factors, namely tobacco and alcohol consumption.

The NBUD and the nucleus are usually in very close proximity and appear to be attached to each other. The NBUD has the same morphology and staining properties as the nucleus; however, its diameter may range from a half to a quarter of that of the main nucleus. The mechanism leading to NBUD formation is not known but it may be related to the elimination of amplified DNA or DNA repair (Thomas et al., 2009).

The scoring method should include coded slides by a person not involved in the study in order to be a blind study. The best magnification to the observation is 1000X. An automated procedure of scoring, by image cytometry have to be developed and validated. The authors suggested first determine the frequency of all the various cell types in a minimum of 1000 cells, following this step, the frequency of DNA damage biomarkers (MN and NBUD) is scored in a minimum of 2000 differentiated cells (Thomas et al., 2009).

At the end the results with the BMCyt are dependent on the level of exposure and potency of genotoxic or cytotoxic agents, genetic background and the age and gender of the donor cells being tested (Thomas et al., 2009).

Is important to define the role of BMCyt in human biomonitoring as a new tool, less invasive in comparison with the CBMN assay, and with many potentialities in molecular epidemiology (Thomas et al., 2009).

Genotoxicity biomonitoring endpoints such as micronucleus, chromosome aberrations and 8-OHdG and DNA repair measured by comet assay are the most commonly used biomarkers in studies evaluating environmental or occupational risks associated with exposure to potential genotoxins. A review made by Knudsen and Hansen (2007) about the application of biomarkers of intermediate end points in environmental and occupational health concluded that MN in lymphocytes provided a promising approach with regard to assessing health risks but concluded that the use of chromosome aberrations in future studies was likely to be limited by the laborious and sensitive procedure of the test and lack of trained cytogeneticists. Methodologies like comet assay in lymphocytes, urine and tissues are increasingly being used as markers of oxidative DNA damage (Battershill et al., 2008).

Studies investigating correlations between endpoints used in genotoxicity biomonitoring studies have yielded inconsistent results, where we can find studies that correlate cytogenetic and comet and studies there do not achieve a correlation between micronucleus, chromosome aberrations and comet. The relative sensitivities of the different endpoints discussed, together with the importance of other factors which influence the persistence of the biomarkers such as DNA repair, may plausibly impact on background levels in the studies considered and would need to be considered before the relationship regarding increases in genotoxicity endpoints with exposure to environmental chemicals or endogenous factors is explored (Battershill et al., 2008).

#### **2.2 Application of genotoxicity biomarkers in an occupational setting – Histopathology laboratories**

A biomonitoring study was conducted in 7 histopathology laboratories in Portugal in order to assess the genotoxicity effects in occupational exposure to formaldehyde (FA).

FA is a reactive, flammable and colourless gas with a strong and very characteristic pungent odour that, when combined with air, can lead to explosive mixtures. FA occurs as an

The NBUD and the nucleus are usually in very close proximity and appear to be attached to each other. The NBUD has the same morphology and staining properties as the nucleus; however, its diameter may range from a half to a quarter of that of the main nucleus. The mechanism leading to NBUD formation is not known but it may be related to the

The scoring method should include coded slides by a person not involved in the study in order to be a blind study. The best magnification to the observation is 1000X. An automated procedure of scoring, by image cytometry have to be developed and validated. The authors suggested first determine the frequency of all the various cell types in a minimum of 1000 cells, following this step, the frequency of DNA damage biomarkers (MN and NBUD) is

At the end the results with the BMCyt are dependent on the level of exposure and potency of genotoxic or cytotoxic agents, genetic background and the age and gender of the donor

Is important to define the role of BMCyt in human biomonitoring as a new tool, less invasive in comparison with the CBMN assay, and with many potentialities in molecular

Genotoxicity biomonitoring endpoints such as micronucleus, chromosome aberrations and 8-OHdG and DNA repair measured by comet assay are the most commonly used biomarkers in studies evaluating environmental or occupational risks associated with exposure to potential genotoxins. A review made by Knudsen and Hansen (2007) about the application of biomarkers of intermediate end points in environmental and occupational health concluded that MN in lymphocytes provided a promising approach with regard to assessing health risks but concluded that the use of chromosome aberrations in future studies was likely to be limited by the laborious and sensitive procedure of the test and lack of trained cytogeneticists. Methodologies like comet assay in lymphocytes, urine and tissues are increasingly being used as markers of oxidative DNA damage (Battershill et al., 2008). Studies investigating correlations between endpoints used in genotoxicity biomonitoring studies have yielded inconsistent results, where we can find studies that correlate cytogenetic and comet and studies there do not achieve a correlation between micronucleus, chromosome aberrations and comet. The relative sensitivities of the different endpoints discussed, together with the importance of other factors which influence the persistence of the biomarkers such as DNA repair, may plausibly impact on background levels in the studies considered and would need to be considered before the relationship regarding increases in genotoxicity endpoints with exposure to environmental chemicals or

elimination of amplified DNA or DNA repair (Thomas et al., 2009).

scored in a minimum of 2000 differentiated cells (Thomas et al., 2009).

cells being tested (Thomas et al., 2009).

epidemiology (Thomas et al., 2009).

endogenous factors is explored (Battershill et al., 2008).

**Histopathology laboratories** 

**2.2 Application of genotoxicity biomarkers in an occupational setting –** 

to assess the genotoxicity effects in occupational exposure to formaldehyde (FA).

A biomonitoring study was conducted in 7 histopathology laboratories in Portugal in order

FA is a reactive, flammable and colourless gas with a strong and very characteristic pungent odour that, when combined with air, can lead to explosive mixtures. FA occurs as an endogenous metabolic product of N-, O- and S-demethylation reactions in most living systems. It is used mainly in the production of resins and their applications, such as adhesives and binders in wood product, pulp and paper, synthetic vitreous fibre industries, production of plastics, coatings, textile finishing and also as an intermediate in the synthesis of other industrial chemical compounds. Common non-occupational sources of exposure to FA include vehicle emissions, particle boards and similar building materials, carpets, paints and varnishes, food and cooking, tobacco smoke and its use as a disinfectant (Conaway et al., 1996; Franks, 2005; IARC, 2006; Pala et al., 2008; Viegas & Prista, 2007).

Commercially, FA is manufactured as an aqueous solution called formalin, usually containing 37 to 40% by weight of dissolved FA (Zhang et al., 2009), which is commonly used in histopathology laboratories as a cytological fixative to preserve the integrity of cellular architecture for diagnosis.

Exogenous FA can be absorbed following inhalation, dermal or oral exposure, being the level of absorption dependent on the route of exposure. The International Agency for Research on Cancer (IARC) reclassified FA as a human carcinogen (group 1) in June 2004 based on *"sufficient epidemiological evidence that FA causes nasopharyngeal cancer in humans"* (IARC, 2006; Zhang et al., 2009). In their review, IARC also concluded that there was *''strong but not sufficient evidence for a causal association between leukaemia and occupational exposure to FA'*' (Zhang et al., 2009, 2010). However, some studies have also led to mixed results and inconclusive evidence (Franks, 2005; Speit et al., 2010).

The inhalation of vapours can produce irritation to eyes, nose and the upper respiratory system. Whilst occupational exposure to high FA concentrations may result in respiratory irritation and asthmatic reactions, it may also aggravate a pre-existing asthma condition. Skin reactions, following exposure to FA are very common, because the chemical is both irritating and allergenic (Pala et al., 2008). FA induces genotoxic and cytotoxic effects in bacteria and mammals cells (Ye et al., 2005) and its genotoxicity and carcinogenicity has been proved in experimental and epidemiological studies that used proliferating cultured mammalian cell lines and human lymphocytes (Pala et al., 2008; Speit et al., 2007) by DNAprotein cross-links, chromosome aberrations, sister exchange chromatides, and MN (Zhang et al., 2009).

The goal of this study was to compare the frequency of genotoxicity biomarkers, provided by CBMN assay in peripheral lymphocytes and MN test in buccal cells between workers of histopathology laboratories exposed to FA and individuals non-exposed to FA and other environmental factors, namely tobacco and alcohol consumption.

The study population consisted of 56 workers occupationally exposed to FA from 7 hospital histopathology laboratories located in Portugal (Lisbon and Tagus Valley region), and 85 administrative staff without occupational exposure to FA. The characteristics of both groups are described in Table 1.

Ethical approval for this study was obtained from the institutional Ethical Board and Director of the participating hospitals, and all subjects gave informed consent to participate in this study. Every person filled a questionnaire aimed at identifying exclusion criteria like history of cancer, radio or chemotherapy, use of therapeutic drugs, exposure to diagnostic X-rays in the past six months, intake of vitamins or other supplements like folic acid as well

Genotoxicity Biomarkers: Application in Histopathology Laboratories 143

Whole blood and exfoliated cells from the buccal mucosa were collected between 10 a.m. and 12 p.m., from every subject and were processed for testing. All samples were coded and analyzed under blind conditions. The criteria for scoring the nuclear abnormalities in lymphocytes and MN in the buccal cells were the ones described by, respectively, Fenech et

Heparinized blood samples were obtained by venipuncture from all subjects and freshly collected blood was directly used for the micronucleus test. Lymphocytes were isolated using Ficoll-Paque gradient and placed in RPMI 1640 culture medium with L-glutamine and red phenol added with 10% inactivated fetal calf serum, 50 ug/ml streptomycin + 50U/mL penicillin, and 10 ug/mL phytohaemagglutinin. Duplicate cultures from each subject were incubated at 37ºC in a humidified 5% CO2 incubator for 44h, and cytochalasin-b 6 ug/mL was added to the cultures in order to prevent cytokinesis. After 28h incubation, cells were spun onto microscope slides using a cytocentrifuge. Smears were air-dried and double stained with May-Grünwald-Giemsa and mounted with Entellan®. One thousand cells were scored from each individual by two independent observers in a total of two slides. Each observer visualized 500 cells per individual. Cells from the buccal mucosa were sampled by endobrushing. Exfoliated cells were smeared onto the slides and fixed with Mercofix®. The standard protocol used was Feulgen staining technique without counterstain. Two thousand cells were scored from each individual by two independent observers in a total of two slides. Each observer visualized 1000 cells per individual. Only cells containing intact nuclei

The deviation of variables from the normal distribution was evaluated by the Shapiro-Wilk goodness-of-fit test. The association between each of the genotoxicity biomarkers and occupational exposure to FA was evaluated by binary logistic regression. The biomarkers were dichotomized (absent/present) and considered the dependent variable in regression models where exposure was an independent variable. Odds ratios were computed to evaluate the risk of biomarkers presence and their significance was assessed. The nonparametric Kuskal-Wallis and Mann-Whitney U-tests, were also used to evaluate interactions involving confounding factors. All statistical analysis was performed using the

Results of FA exposure values were determined using the two methods described – the NIOSH 2541 method for average concentrations (TWA8h) and the PID method for ceiling concentrations. For the first exposure metric, FA mean level of the 56 individuals studied was 0.16 ppm (0.04 – 0.51 ppm), a value that lies below the OSHA reference value of 0.75 ppm. The mean ceiling concentration found in the laboratories was 1.14 ppm (0.18 – 2.93 ppm), a value well above the reference of the American Conference of Governmental Industrial Hygienists (ACGIH) for ceiling concentrations (0.3 ppm). As for the different tasks developed in histopathology laboratories, the highest FA concentration was identified during macroscopic specimens' exam. This task involves a careful observation and grossing of the specimen

preserved in FA, therefore has direct and prolonged contact with its vapors (Table 2).

that were neither clumped nor overlapped were included in the analysis.

al. (1999) and Tolbert et al. (1991).

**2.2.3 Statistical analysis** 

**2.2.4 Results** 

**FA exposure levels** 

SPSS package for windows, version 15.0.


as information related to working practices (such as years of employment and the use of protective measures). In this study, none of the participants were excluded.

Table 1. Characteristics of the studied sample.

#### **2.2.1 Environmental monitoring of FA exposure**

Exposure assessment was based on two techniques of air monitoring conducted simultaneously. First, environmental samples were obtained by air sampling with low flow pumps for 6 to 8 hours, during a typical working day. FA levels were measured by Gas Chromatography analysis and time-weighted average (TWA8h) was estimated according to the National Institute of Occupational Safety and Health method NIOSH 2541 (NIOSH, 1994).

The second method was aimed at measuring ceiling values of FA using Photo Ionization Detection (PID) equipment (11.7 eV lamps) with simultaneous video recording. Instantaneous values for FA concentration were obtained on a per second basis. This method allows establishing a relation between workers activities and FA concentration values, as well to reveal the main exposure sources (McGlothlin et al., 2005; Viegas et al., 2010).

Measurements and sampling were performed in a macroscopic room, provided with fume hoods, always near workers breath.

#### **2.2.2 Biological monitoring**

Evaluation of genotoxic effects was performed by applying the CBMN assay in peripheral blood lymphocytes and exfoliated cells from the buccal mucosa.

Whole blood and exfoliated cells from the buccal mucosa were collected between 10 a.m. and 12 p.m., from every subject and were processed for testing. All samples were coded and analyzed under blind conditions. The criteria for scoring the nuclear abnormalities in lymphocytes and MN in the buccal cells were the ones described by, respectively, Fenech et al. (1999) and Tolbert et al. (1991).

Heparinized blood samples were obtained by venipuncture from all subjects and freshly collected blood was directly used for the micronucleus test. Lymphocytes were isolated using Ficoll-Paque gradient and placed in RPMI 1640 culture medium with L-glutamine and red phenol added with 10% inactivated fetal calf serum, 50 ug/ml streptomycin + 50U/mL penicillin, and 10 ug/mL phytohaemagglutinin. Duplicate cultures from each subject were incubated at 37ºC in a humidified 5% CO2 incubator for 44h, and cytochalasin-b 6 ug/mL was added to the cultures in order to prevent cytokinesis. After 28h incubation, cells were spun onto microscope slides using a cytocentrifuge. Smears were air-dried and double stained with May-Grünwald-Giemsa and mounted with Entellan®. One thousand cells were scored from each individual by two independent observers in a total of two slides. Each observer visualized 500 cells per individual. Cells from the buccal mucosa were sampled by endobrushing. Exfoliated cells were smeared onto the slides and fixed with Mercofix®. The standard protocol used was Feulgen staining technique without counterstain. Two thousand cells were scored from each individual by two independent observers in a total of two slides. Each observer visualized 1000 cells per individual. Only cells containing intact nuclei that were neither clumped nor overlapped were included in the analysis.
