Micronuclei in exfoliated epithelial cells are widely used as biomarkers of cancer risk in humans. To elucidate the effect of different staining procedures on the outcome of such investigation, we conducted a study in which the micronuclei frequencies in oral mucosa cells of heavy smokers (n = 20) and nonsmokers (n = 10) were evaluated with nonspecific (Giemsa, May-Grünwald-Giemsa) and DNA-specific (4′,6-diamidino-2-phenylindole, Feulgen, acridine orange) stains, whereas with Giemsa-based stains, the frequencies of micronuclei in smokers were significantly (4- to 5-fold) higher in the smokers group, no significant increase was observed with any of the DNA-specific stains. Furthermore, the evaluation of cells of the two study groups with Feulgen stain showed that oral mucosa cells from smokers had significantly increased levels of nuclear anomalies other than micronuclei. These anomalies are consequences of cell injury found in epithelial cells and are paralleled by formation of keratin bodies in the cytoplasm that resemble micronuclei. Correlation analyses showed that micronuclei frequencies scored in Giemsa-stained slides correlated significantly with karyorrhexis, karyolysis, condensed chromatin, and binucleates, whereas no such correlations were found with DNA-specific stains. These findings indicate that nuclear anomalies (and possibly keratin bodies) may be misinterpreted as micronuclei with nonspecific DNA stains and lead to false-positive results in studies with cells of epithelial origin. Furthermore, our results show that exposure of oral mucosa cells to genotoxic carcinogens contained in tobacco smoke does not lead to induction of micronuclei in these cells. (Cancer Epidemiol Biomarkers Prev 2006;15(10):1835–40)

About 25 years ago, Stich et al. (1) developed a protocol for micronucleus assays with exfoliated buccal mucosa cells, which was widely used in occupational and lifestyle studies (for reviews, see refs. 2 and 3). It was repeatedly emphasized that this noninvasive method might be a suitable biomonitoring approach for the detection of increased cancer risk in man because >90% of all human cancers are of epithelial origin (4). Approximately 200 studies with epithelial cells have been published in the last 25 years and apart from mouth cells, cells from other organs, such as nasal mucosa, cervix, bladder, esophagus, and bronchi, were also used (2, 3).

It was soon realized that strong interlaboratory and intralaboratory variations exist in micronuclei studies with exfoliated cells and efforts were made to standardize the assay. For example, Tolbert et al. (5) and Sarto et al. (6) defined morphologic scoring criteria for micronuclei and other confounding factors, such as age and gender of the study groups. However, only little attention has been, until now, given to the effect of different staining procedures on the results of micronuclei assays. An evaluation of the literature shows that a variety of different stains is used in micronuclei studies. Among the DNA-specific stains, the ones most widely used are Feulgen and acridine orange; in some experiments, 4′,6-diamidino-2-phenylindole (DAPI) and propidium iodide were also used. About 30% of the studies in epithelial cells were conducted with nonspecific stains (Giemsa, May-Grünwald-Giemsa, and less frequently orcein). According to our knowledge, only one study (7) has been published in which the effect of different stains, namely Giemsa, Feulgen, Hoechst 33342, and propidium iodide, on the results of micronuclei experiments, was compared in cells from patients with different pathologies of the oral mucosa. Casartelli et al.'s findings indicate that the micronuclei frequencies are higher with Giemsa; however, no firm conclusions can be drawn from this study as no control group was included, the samples were collected from heterogeneous groups, and Hoechst 33342 and propidium iodide have rarely been used in micronuclei assays.

The aim of the present study was to further investigate if, and to what extent, different stains have an effect on the results of micronuclei studies in exfoliated cells. Therefore, buccal mucosa cells were collected from nonsmokers and heavy smokers, and the micronuclei frequencies were comparatively evaluated with two of the most commonly used DNA nonspecific stains (May-Grünwal-Giemsa and Giemsa) and with three specific stains (Feulgen, acridin orange, and DAPI). Smokers were included as it is well established that they have increased risks for oral and other forms of epithelial cancer (8-10) and a number of studies on micronuclei induction in buccal cells of smokers has been published, which gave highly inconsistent results (refs. 11-31; for review, see ref. 28). Damage of epithelial cells leads to nuclear abnormalities other than micronuclei (metanucleated cells; ref. 32), such as karyorrhexis, karyolysis, pyknosis, condensed chromatin, broken eggs, binucleates, and it has been stressed that these anomalies are difficult to distinguish from classic micronuclei and may led to misinterpretations (5, 32). Therefore, we additionally evaluated these variables in both study groups and analyzed if the micronuclei frequencies scored with DNA-specific and nonspecific stains correlate with these anomalies.

Participants

In total, 10 nonsmokers and 20 heavy smokers participated in the study. All subjects were healthy and none of them was exposed to DNA-damaging cytostatic drugs and ionizing radiation. All of them consumed a mixed diet and had moderate alcohol consumption (<1,000 mL beer or 400 mL wine/d).

Other characteristics of the two study groups are listed in Table 1. The number of cigarettes indicated is the average amount consumed per day during 3 months before the experiment was started. Informed consent was obtained from all participants concerning their participation.

Table 1.

General characteristics of the participants

VariableSmokersNonsmokers
No. participants 20 10 
Age (mean ± SD, y) 36.1 ± 8.1 30.3 ± 8.7 
Gender   
    Females 
    Males 16 
No. consumed cigarettes/d (mean ± SD) 34 ± 11.9 — 
Duration of smoking (mean ± SD, y) 12.2 ± 4.0 — 
Body mass index (kg/m2, mean ± SD)   
    Males 24.5 ± 2.0 26.0 ± 2.3 
    Females 22.5 ± 2.5 21.5 ± 3.0 
VariableSmokersNonsmokers
No. participants 20 10 
Age (mean ± SD, y) 36.1 ± 8.1 30.3 ± 8.7 
Gender   
    Females 
    Males 16 
No. consumed cigarettes/d (mean ± SD) 34 ± 11.9 — 
Duration of smoking (mean ± SD, y) 12.2 ± 4.0 — 
Body mass index (kg/m2, mean ± SD)   
    Males 24.5 ± 2.0 26.0 ± 2.3 
    Females 22.5 ± 2.5 21.5 ± 3.0 

Sample size was determined based on power considerations. Assuming a Poisson distribution of the number of micronucleated cells (33) and introducing a square-root transformation, a SD of ∼0.55 resulted that was independent of the average number of micronuclei for the expected range of values. The power to detect a doubling of the average number of micronucleated cells at the 5% level of significance exceeds 90% for a 2:1 ratio of the number of subjects in the smokers and nonsmokers groups, respectively, if, overall, 30 subjects are included.

Sample Collection

Immediately before cell collection, the participants rinsed their mouths twice with tap water. Subsequently, the cells were scraped with wooden spatulas from each cheek, washed twice in plastic tubes (Sarstedt, Germany) containing 10 mL buffer solution [0.1 mol/L EDTA, 0.01 mol/L Tris-HCl, and 0.02 mol/L NaCl (pH 7.0)], and fixed in 80% cold methanol overnight (for details, see ref. 34).

Staining Procedures

Fifty microliters of the cell suspensions were dropped onto wet cold glass slides and dried overnight in the dark at room temperature. From each sample, five to seven slides were prepared and stained. All stains were purchased from Sigma (Munich, Germany).

Giemsa stain (10%) was prepared in Sorensen's buffer (pH 6.8) as described by Vives Corrons et al. (35); the staining time was 20 minutes. For May-Grünwald, the cells were first stained in May-Grünwald solution for 3 to 5 minutes and subsequently with Giemsa (10%; see above) for 10 minutes (35). Acridine orange was dissolved in bidistilled water (0.01%); the staining time was 15 minutes (36). For DAPI, the stock solution was diluted (1 μg/mL) with bidistilled water; the staining time was 30 minutes (15). For Feulgen, the cells were placed in beakers with 5.0 mol/L HCl at room temperature for 15 minutes, rinsed with distilled water (15 min), and subsequently stained with Schiff's reagent for 90 minutes (37).

For direct comparisons of the results obtained with the different procedures, some of the Giemsa and May-Grünwald-Giemsa–stained slides were destained in methanol/glacial acetic acid (3:1) overnight (7) and subsequently restained with either acridine orange or Feulgen.

Evaluation of the Slides

From each participant, at least 1,500 cells were evaluated with each staining procedure. Micronuclei were scored in normal cells according to the criteria defined by Tolbert et al. (5). In addition, the frequencies of nuclear anomalies, namely karyolysis, pyknosis, karyorrhexis, broken eggs, and condensed chromatin, were recorded in Feulgen-stained slides as described by Tolbert et al. (ref. 5; for details, see also ref. 32). The analysis of the slides was carried out by two experienced scorers; micronuclei and metanucleated cells (i.e., cells with karyorrhexis, karyolysis, pyknosis, condensed chromatin, broken eggs, binucleates) were only registered after consensus.

Giemsa-, May-Grünwald-Giemsa–, and Feulgen-stained slides were evaluated under a light microscope (Nikon Eclipse E200) with 1,000-fold magnification using oil immersion. DAPI- and acridine orange–stained slides were evaluated under a fluorescence microscope (Nikon Microphot FXA, filters UV-2a, 330-380 nm and B-2A, 450-499 nm) with 1,000-fold magnification (oil immersion); according to the suggestion of M. Fenech (Commonwealth Scientific and Industrial Research Organization, Adelaide, Australia), Feulgen-stained slides were additionally scored under a fluorescence microscope (filter G-2A, 510-560 nm).

Statistics

A square-root transformation was chosen and homogeneity of variance of transformed values was tested by Bartlett's test. Differences between heavy smokers and nonsmokers as well as staining methods were tested by two-factor ANOVA with one between-subjects and one within-subjects factor. For each staining method, the difference between smokers and nonsmokers was tested by linear contrasts. The same methods were applied for comparison of Giemsa-stained and destained preparations. Age and alcohol consumption were included as covariables, and gender was added as a between-subject factor to analyze whether these variables contribute to differences between micronuclei frequencies. For comparison of cell anomalies between heavy smokers and nonsmokers, Wilcoxon's rank-sum test was applied. Results are expressed as prevalence ratios (ratio of the mean number of deviating cells per 1,000 cells) and their confidence intervals were computed based on Fieller's theorem.

Spearman rank correlation coefficients were computed to assess correlations between cell anomalies and average number of micronucleated cells from DNA specific and nonspecific staining.

Figure 1 summarizes the micronuclei frequencies found in heavy smokers and nonsmokers with the different staining procedures. It can be seen that the micronuclei frequencies scored with Giemsa were significantly (∼5-fold) higher in smokers than in nonsmokers; also, with May-Grünwald-Giemsa, the number of micronuclei was substantially (∼4.5-fold) higher. On the contrary, no significant effects were observed with the DNA-specific stains. With acridine orange, the micronuclei frequencies were ∼90% higher in smokers; with DAPI and Feulgen, the differences were 30% and 120%, respectively. When the Feulgen-stained slides were evaluated under fluorescence, an 89% increase was observed. It is also notable that the micronuclei frequencies recorded in the nonsmokers varied when different staining procedures were used; with nonspecific stains, the numbers were almost twice as high as with DNA-specific stains, but these differences failed to reach statistical significance. Age, gender, and alcohol consumption did not significantly affect the micronuclei frequencies with any of the staining methods and these variables had also no effect on the differences between smokers and nonsmokers.

Figure 1.

Results obtained with different staining procedures in micronuclei (MN) assays with exfoliated oral mucosa cells from smokers (n = 20) and nonsmokers (n = 10). From each participant, at least 1,500 cells were scored for micronuclei formation. Columns, mean; bars, 95% CI. Shaded columns, smokers; open columns, nonsmokers. F-fluorescence microscopy, L-light microscopy.

Figure 1.

Results obtained with different staining procedures in micronuclei (MN) assays with exfoliated oral mucosa cells from smokers (n = 20) and nonsmokers (n = 10). From each participant, at least 1,500 cells were scored for micronuclei formation. Columns, mean; bars, 95% CI. Shaded columns, smokers; open columns, nonsmokers. F-fluorescence microscopy, L-light microscopy.

Close modal

To further substantiate the differences between results obtained with DNA-specific and nonspecific stains, we destained May-Grünwald slides from six smokers and an identical number of nonsmokers and evaluated the frequencies after DAPI staining. The results are depicted in Fig. 2, and it is evident that, again, no significant effect was detectable with the DNA-specific stain, whereas a pronounced difference between smokers and nonsmokers was observed with May-Grünwald.

Figure 2.

Results obtained with micronucleus assays with exfoliated oral mucosa cells from heavy smokers (n = 6) and nonsmokers (n = 6) in identical slides after staining with Giemsa and after destaining and subsequent restaining with acridine orange (Acr.O). From each participant, at least 1,500 cells were evaluated. Columns, means; bars, 95% CI. Shaded columns, smokers; open columns, nonsmokers.

Figure 2.

Results obtained with micronucleus assays with exfoliated oral mucosa cells from heavy smokers (n = 6) and nonsmokers (n = 6) in identical slides after staining with Giemsa and after destaining and subsequent restaining with acridine orange (Acr.O). From each participant, at least 1,500 cells were evaluated. Columns, means; bars, 95% CI. Shaded columns, smokers; open columns, nonsmokers.

Close modal

The photographs depict cells with micronuclei after May-Grünwald-Giemsa (Fig. 3A and C), which were restained either with Feulgen (Fig. 3B) or with DAPI (Fig. 3D). It can be seen that intracellular structures that are classified as micronuclei with the non-DNA-specific stain are not detected with the DNA-specific stain.

Figure 3.

Photographs of exfoliated oral mucosa cells with micronuclei (magnification, ×500). A and C, May-Grünwald-Giemsa–stained cells. B and D, cells stained with Feulgen and DAPI. Arrows, cellular structures that are classified as micronuclei according to established scoring criteria.

Figure 3.

Photographs of exfoliated oral mucosa cells with micronuclei (magnification, ×500). A and C, May-Grünwald-Giemsa–stained cells. B and D, cells stained with Feulgen and DAPI. Arrows, cellular structures that are classified as micronuclei according to established scoring criteria.

Close modal

To find out if smoking induces nuclear anomalies other than micronuclei, we evaluated comparatively the frequencies of different types of metanucleated cells in the two study groups. The results are summarized in Table 2. Four types of anomalies, namely karyorrhexis, karyolysis, binucleates, and condensed chromatin, were 146%, 350%, 117%, and 54% higher in smokers than in nonsmokers. In addition, higher numbers of cells with pyknosis were also found in smokers, although this latter effect failed to reach statistical significance.

Table 2.

Frequencies of anomalies in buccal cells per 1,000 cells of smokers (n = 15) and nonsmokers (n = 10)

VariableGroupMean ± SEPrevalence ratioTwo-tailed P
Condensed chromatin Smokers 9.52 ± 0.89 1.54 (1.08-2.18) 0.019820 
 Nonsmokers 6.19 ± 0.94   
Binucleation Smokers 7.02 ± 0.59 2.18 (1.56-3.03) 0.000385 
 Nonsmokers 3.23 ± 0.47   
Karyorrhexis Smokers 7.59 ± 0.68 2.47 (1.67-3.64) 0.000428 
 Nonsmokers 3.08 ± 0.55   
Karyolysis Smokers 9.90 ± 1.16 4.50 (2.64-7.67) 0.000040 
 Nonsmokers 2.20 ± 0.54   
Pyknosis Smokers 2.23 ± 0.41 1.45 (0.83-2.53) 0.244072 
 Nonsmokers 1.54 ± 0.34   
Broken eggs Smokers 0.15 ± 0.06 0.74 (0.25-2.23) 0.617618 
 Nonsmokers 0.20 ± 0.08   
Degenerative cells Smokers 36.42 ± 1.49 2.22 (1.81-2.73) 0.000032 
 Nonsmokers 16.39 ± 1.59   
VariableGroupMean ± SEPrevalence ratioTwo-tailed P
Condensed chromatin Smokers 9.52 ± 0.89 1.54 (1.08-2.18) 0.019820 
 Nonsmokers 6.19 ± 0.94   
Binucleation Smokers 7.02 ± 0.59 2.18 (1.56-3.03) 0.000385 
 Nonsmokers 3.23 ± 0.47   
Karyorrhexis Smokers 7.59 ± 0.68 2.47 (1.67-3.64) 0.000428 
 Nonsmokers 3.08 ± 0.55   
Karyolysis Smokers 9.90 ± 1.16 4.50 (2.64-7.67) 0.000040 
 Nonsmokers 2.20 ± 0.54   
Pyknosis Smokers 2.23 ± 0.41 1.45 (0.83-2.53) 0.244072 
 Nonsmokers 1.54 ± 0.34   
Broken eggs Smokers 0.15 ± 0.06 0.74 (0.25-2.23) 0.617618 
 Nonsmokers 0.20 ± 0.08   
Degenerative cells Smokers 36.42 ± 1.49 2.22 (1.81-2.73) 0.000032 
 Nonsmokers 16.39 ± 1.59   

NOTE: Average number of inspected cells per participant: 2,127 (smokers) and 2,140 (nonsmokers).

Table 3 depicts the results of Spearman correlation analyses; it can be seen that increased micronuclei frequencies were significantly associated with certain other nuclear anomalies, such as karyorrhexis, karyolysis, binucleates, and condensed chromatin, when nonspecific staining procedures were used, whereas no such associations were found with the DNA-specific stains.

Table 3.

Spearman rank correlation coefficients of correlations between frequency of cellular anomalies and average number of micronuclei after non-DNA-specific staining and DNA-specific staining

Nuclear anomalyAverage no. micronuclei, non-DNA-specific stainingAverage no. micronuclei, DNA-specific staining
Condensed chromatin 0.447* 0.131 
Binucleation 0.776 0.122 
Karyorrhexis 0.741 0.222 
Karyolysis 0.734 0.147 
Pyknosis 0.228 0.123 
Broken eggs 0.014 −0.150 
Degenerative cells 0.844 0.299 
Nuclear anomalyAverage no. micronuclei, non-DNA-specific stainingAverage no. micronuclei, DNA-specific staining
Condensed chromatin 0.447* 0.131 
Binucleation 0.776 0.122 
Karyorrhexis 0.741 0.222 
Karyolysis 0.734 0.147 
Pyknosis 0.228 0.123 
Broken eggs 0.014 −0.150 
Degenerative cells 0.844 0.299 
*

P < 0.05.

P < 0.01.

Gender and alcohol consumption did not correlate with the induction of metanucleated cells, but a borderline significant correlation was seen with age (P = 0.05); however, this did not affect the differences between smokers and nonsmokers.

Our findings show that the results of micronuclei assays in exfoliated oral mucosa cells of smokers and nonsmokers depend strongly on the staining methods. Although much higher micronuclei frequencies were found in smokers with the non-DNA-specific stains Giemsa and May-Grünwald, no significant differences were seen with acridine orange, DAPI, and Feulgen. These comparisons indicate that micronuclei formation in epithelial cells may be overestimated when non-DNA-specific stains are used. This is an important observation as numerous studies have been conducted in the last two decades with exfoliated cells in which Giemsa stains were used.

Our observation that micronuclei formation in oral cells is overestimated with Giemsa-based stains is also supported by the results reported by Casartelli et al. (7), who compared micronuclei numbers after Giemsa and propidium iodide staining in cells from patients with different pathologies of the oral cavity (leucoplakias, squamous cell carcinomas). They found a 3-fold higher frequency of micronuclei with Giemsa than with Hoechst 33258. Comparisons between Feulgen, Hoechst 33259, and propidium iodide showed that 3- to 4-fold higher frequencies are obtained with the latter stain.

Also, the evaluation of earlier studies concerning the formation of micronucleated cells in mucosa of smokers confirms our findings (the evaluation is based on computer-aided searches in medical databases, studies in which the number of participants and/or the number of cigarettes were not specified were not included). Although consistently positive results were obtained with Giemsa (20, 21, 29, 31, 38), negative results were obtained in most studies with DNA-specific stains. We found in total 24 articles in which negative results were obtained in smokers with Feulgen, whereas an increase in micronuclei frequencies was reported in six articles only (6, 22, 23, 39-41). One group reported a positive result (6) but failed to confirm it in a subsequent study (42). Also, with other DNA-specific stains (DAPI, acridine orange), mainly negative results were obtained (15, 18, 24, 30, 37).

The effect of alcohol consumption, age, and gender on micronuclei formation has been studied in several earlier investigations, and the results are strongly controversial. Although some investigators (6, 13, 26, 27, 30) found no effect of alcohol on micronuclei formation, Gabriel et al. (31) showed a significant effect. Also, Stich and Rosin (26) reported on a pronounced synergistic effect of alcohol in combination with smoking, whereas in another study (25) even an inverse relation was found. Also, the effect of age on micronuclei formation in buccal cells is unclear. In five investigations, no effect was detected (6, 13, 27, 29, 30); in two, an increase with age (39, 40) and in another a decrease with age (25) were detected. Piyathilake et al. (40) reported that the frequencies of micronucleated cells are 2.8-fold higher in females after adjusting for race, age, smoking habit, and levels of selected vitamin, whereas in a Brazilian study (27) the number of micronuclei was significantly (i.e., >3-fold) higher in males. In the present investigation, we found neither in smokers nor in nonsmokers a statistically significant effect of all these variables on micronuclei formation, but it is notable that all participants consumed only small amounts of alcohol if any and that the age range was quite narrow; that is, 36.1 ± 8.1 years in the smokers group and 30.3 ± 8.7 years in the controls.

Table 2 shows that metanucleated cells with anomalies other than micronuclei, such as karyorrhexis, karyolysis, binucleates, and condensed chromatin, are significantly increased in smokers. These anomalies reflect the consequences of cell injury, cell death, and mitotic errors (5, 32). Some of them [pyknosis (PN), condensed chromatin (CC), and karyolysis (KL)] are accompanied by apoptosis (5, 32), which may result from DNA damage. However, the same phenomena are also seen in cells undergoing necrosis (5, 32) and cannot be regarded as reliable markers for increased DNA damage and cancer risk. Only few studies have been published in which such anomalies have been recorded. Significantly increased frequencies were seen, for example, in snuff users, petrol-exposed workers, and also in alcoholic patients with oral carcinomas (5, 22, 25, 32). In all studies including ours, karyolysis was strongly increased compared with the control groups, whereas the pattern of induction of the other anomalies differed strongly. For example, binucleates were increased in smokers but not in snuff users, whereas pyknosis was only elevated in the latter group (5, 32).

The effect of gender, age, and alcohol consumption on the formation of metanucleated cells has not been investigated in detail according to our knowledge. Sarto et al. (6) analyzed the effect of alcohol consumption in combination with smoking on the formation of binucleated cells and failed to detect an effect. In the present study, we observed a weak association between the age of the participants and the overall frequency of metanucleated cell, but as stressed above, the design of our study does not allow drawing of firm conclusions.

It has been stressed by Tolbert et al. (5, 32) that these anomalies are sometimes difficult to interpret and may be misclassified as micronuclei. This assumption is supported by the fact that, in the present study, significant correlations were observed between micronuclei frequencies and the formation of cells with karyorrhexis, karyolysis, binucleates, and condensed chromatin with nonspecific stains (Table 3). Another possible confounding factor in micronuclei studies is the formation of keratin granules that are found in degenerated cells with nuclear anomalies (43). These round cytoplasmic bodies, which are formed as a consequence of cell injury (43), do not contain DNA and may be classified as micronuclei with nonspecific stains.

Degenerated cells with abnormal nuclei (other than micronuclei) are not only found in individuals exposed to toxic substances, but to a lesser extent also in nonexposed subjects (Table 2), and it is possible that the positive results obtained in some studies with smokers and nonsmokers (23, 39) in which significant differences were observed between these groups with Feulgen stain are due to misinterpretation of these structures as the micronuclei frequencies in the controls were unusually high in these investigations.

As in most (>99%) earlier studies, the exfoliated cells were collected from both cheeks of each participant in the present investigation. Because our findings indicate that micronuclei studies with nonspecific stains reflect the levels of keratinization of the cells, it can be expected that strong site-specific effects are found when cells are collected from different areas as it is known the levels of karyorrhexis and keratinization in the oral cavity differ strongly. They are highest in the lips, whereas only low levels are found in the bottom of the mouth (13). Currently, Giemsa stains are also widely used in micronuclei studies with nonepithelial cells originating from a number of organs, e.g., in experiments with polychromatic erythrocytes in bone marrow of rodents (44), lymphocytes (45), splenocytes (46), and liver cells (47), and in in vitro studies with stable cell lines (48). Because the formation of keratin bodies and of the nuclear anomalies described above are an adaptive response to cell damage, which is only seen in (normally nonkeratinizing) epithelial cells (49, 50), it is likely that the overestimation of micronuclei formation with nonspecific stains is only relevant for experiments with epithelial cells, e.g., from the oral cavity, cervix, bladder, and esophagus. This assumption is supported by the findings of Surrales et al. (51), who compared micronuclei frequencies in human lymphocytes stained with DAPI and Giemsa and found even higher frequencies with the former stain.

Our investigation indicates that the results of earlier micronuclei studies in cells of epithelial origin should be interpreted with caution when Giemsa-based stains were used. Because our study and the majority of investigations concerning induction of micronuclei in smokers with DNA-specific stains yielded negative results and positive results may be due to misinterpretation, it is very likely that exposure of oral mucosa cells to genotoxic carcinogens, such as nitrosamines, polycyclic aromatic hydrocarbons, and many other compounds that are contained in tobacco smoke, does not lead to formation of micronuclei in oral mucosa cells (8, 9). The increased risks of oral cancer in smokers may be due to acute toxic effects and inflammation that are not associated with micronuclei formation. Therefore, the use of micronuclei in exfoliated epithelial cells as biomarkers of exposure to lifestyle- and occupation-related genotoxic carcinogens warrants further evaluation in general.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank P. Breit for his help in the preparation of the photographic images.

1
Stich HF, Stich W, Parida BB. Elevated frequency of micronucleated cells in the buccal mucosa of individuals at high risk for oral cancer: betel quid chewers.
Cancer Lett
1982
;
17
:
125
–34.
2
Majer BJ, Laky B, Knasmuller S, Kassie F. Use of the micronucleus assay with exfoliated epithelial cells as a biomarker for monitoring individuals at elevated risk of genetic damage and in chemoprevention trials.
Mutat Res
2001
;
489
:
147
–72.
3
Rosin MP, Dunn BP, Stich HF. Use of intermediate endpoints in quantitating the response of precancerous lesions to chemopreventive agents.
Can J Physiol Pharmacol
1987
;
65
:
483
–7.
4
Cairns J. Mutation selection and the natural history of cancer.
Nature
1975
;
255
:
197
–200.
5
Tolbert PE, Shy CM, Allen JW. Micronuclei and other nuclear anomalies in buccal smears: methods development.
Mutat Res
1992
;
271
:
69
–77.
6
Sarto F, Finotto S, Giacomelli L, Mazzotti D, Tomanin R, Levis AG. The micronucleus assay in exfoliated cells of the human buccal mucosa.
Mutagenesis
1987
;
2
:
11
–7.
7
Casartelli G, Monteghirfo S, De Ferrari M, et al. Staining of micronuclei in squamous epithelial cells of human oral mucosa.
Anal Quant Cytol Histol
1997
;
19
:
475
–81.
8
IARC Monogr Eval Carcinog Risk Chem Hum. Vol. 38. Tobacco smoking. Lyon: IARC; 1986.
9
IARC Monogr Eval Carcinog Risks Hum. Vol. 83. Tobacco smoke and involuntary smoking. Lyon: IARC; 2004.
10
Taybos G. Oral changes associated with tobacco use.
Am J Med Sci
2003
;
326
:
179
–82.
11
Martino-Roth MG, Viegas J, Roth DM. Occupational genotoxicity risk evaluation through the comet assay and the micronucleus test.
Genet Mol Res
2003
;
2
:
410
–7.
12
Martino-Roth MG, Viegas J, Amaral M, Oliveira L, Ferreira FLS, Erdtmann B. Evaluation of genotoxicity through micronuclei test in workers of car and battery repair garages.
Genet Mol Biol
2002
;
25
:
493
–500.
13
Bohrer PL, Filho MS, Paiva RL, da Silva IL, Rados PV. Assessment of micronucleus frequency in normal oral mucosa of patients exposed to carcinogens.
Acta Cytol
2005
;
49
:
265
–72.
14
Belowska J, Fraczek A, Ral M, et al. Preliminary study of the frequency of micronuclei in buccal mucosa cells sampled from patients with laryngeal tumors.
Przegl Lek
2004
;
61
:
248
–50.
15
Pastor S, Creus A, Parron T, et al. Biomonitoring of four European populations occupationally exposed to pesticides: use of micronuclei as biomarkers.
Mutagenesis
2003
;
18
:
249
–58.
16
Burgaz S, Erdem O, Cakmak G, Erdem N, Karakaya A, Karakaya AE. Cytogenetic analysis of buccal cells from shoe-workers and pathology and anatomy laboratory workers exposed to n-hexane, toluene, methyl ethyl ketone and formaldehyde.
Biomarkers
2002
;
7
:
151
–61.
17
Ramos-Remus C, Dorazco-Barragan G, Aceves-Avila FJ, et al. Genotoxicity assessment using micronuclei assay in rheumatoid arthritis patients.
Clin Exp Rheumatol
2002
;
20
:
208
–12.
18
Kiilunen M, Utela J, Rantanen T, et al. Exposure to soluble nickel in electrolytic nickel refining.
Ann Occup Hyg
1997
;
4
:
167
–88.
19
Gattas GJ, Cardoso Lde A, Medrado-Faria Mde A, Saldanha PH. Frequency of oral mucosa micronuclei in gas station operators after introducing methanol.
Occup Med (Lond)
2001
;
51
:
107
–13.
20
Goud KI, Hasan Q, Balakrishna N, Rao KP, Ahuja YR. Genotoxicity evaluation of individuals working with photocopying machines.
Mutat Res
2004
;
563
:
151
–8.
21
Danadevi K, Rozati R, Banu BS, Grover P. Genotoxic evaluation of welders occupationally exposed to chromium and nickel using the Comet and micronucleus assays.
Mutagenesis
2004
;
19
:
35
–41.
22
Celik A, Cavas T, Ergene-Gozukara S. Cytogenetic biomonitoring in petrol station attendants: micronucleus test in exfoliated buccal cells.
Mutagenesis
2003
;
18
:
417
–21.
23
Wu PA, Loh CH, Hsieh LL, Liu TY, Chen CJ, Liou SH. Clastogenic effect for cigarette smoking but not areca quid chewing as measured by micronuclei in exfoliated buccal mucosal cells.
Mutat Res
2004
;
562
:
27
–38.
24
Lucero L, Pastor S, Suarez S, et al. Cytogenetic biomonitoring of Spanish greenhouse workers exposed to pesticides: micronuclei analysis in peripheral blood lymphocytes and buccal epithelial cells.
Mutat Res
2000
;
464
:
255
–62.
25
Ramirez A, Saldanha PH. Micronucleus investigation of alcoholic patients with oral carcinomas.
Genet Mol Res
2002
;
1
:
246
–60.
26
Stich HF, Rosin MP. Quantitating the synergistic effect of smoking and alcohol consumption with the micronucleus test on human buccal mucosa cells.
Int J Cancer
1983
;
31
:
305
–8.
27
Benites CI, Amado LL, Vianna RA, Martino-Roth Mda G. Micronucleus test on gas station attendants.
Genet Mol Res
2006
;
5
:
45
–54.
28
Proia NK, Paszkiewicz GM, Nasca MA, Franke GE, Pauly JL. Smoking and smokeless tobacco-associated human buccal cell mutations and their association with oral cancer-a review.
Cancer Epidemiol Biomarkers Prev
2006
;
15
:
1061
–77.
29
Vuyyuri SB, Ishaq M, Kuppala D, Grover P, Ahuja YR. Evaluation of micronucleus frequencies and DNA damage in glass workers exposed to arsenic.
Environ Mol Mutagen
2006
;
47
:
562
–70.
30
Bolognesi C, Nucci MC, Colacci AM, et al. Biomonitoring of nurses occupationally exposed to antineoplastic drugs: the IMEPA Project.
Epidemiol Prev
2005
;
29
:
91
–5.
31
Gabriel HE, Crott JW, Ghandour H, et al. Chronic cigarette smoking is associated with diminished folate status, altered folate form distribution, and increased genetic damage in the buccal mucosa of healthy adults.
Am J Clin Nutr
2006
;
83
:
835
–41.
32
Tolbert PE, Shy CM, Allen JW. Micronuclei and other nuclear anomalies in buccal smears: a field test in snuff users.
Am J Epidemiol
1991
;
134
:
840
–50.
33
Kirkland DJ, editor. Statistical evaluation of mutagenicity test data. Cambridge (United Kingdom): Cambridge University Press; 1989.
34
Rosin MP. The use of the micronucleus test on exfoliated cells to identify anti-clastogenic action in humans: a biological marker for the efficacy of chemopreventive agents.
Mutat Res
1992
;
267
:
265
–76.
35
Vives Corrons JL, Albarede S, Flandrin G, et al. Guidelines for blood smear preparation and staining procedure for setting up an external quality assessment scheme for blood smear interpretation. Part I: Control material.
Clin Chem Lab Med
2004
;
42
:
922
–6.
36
Cao J, Liu Y, Sun H, Cheng G, Pang X, Zhou Z. Chromosomal aberrations, DNA strand breaks and gene mutations in nasopharyngeal cancer patients undergoing radiation therapy.
Mutat Res
2002
;
504
:
85
–90.
37
Tian D, Ma H, Feng Z, et al. Analyses of micronuclei in exfoliated epithelial cells from individuals chronically exposed to arsenic via drinking water in inner Mongolia China.
J Toxicol Environ Health A
2001
;
64
:
473
–84.
38
Bloching M, Hofmann A, Lautenschlager C, Berghaus A, Grummt T. Exfoliative cytology of normal buccal mucosa to predict the relative risk of cancer in the upper aerodigestive tract using the MN-assay.
Oral Oncol
2000
;
36
:
550
–5.
39
Ozkul Y, Donmez H, Erenmemisoglu A, Demirtas H, Imamoglu N. Induction of micronuclei by smokeless tobacco on buccal mucosa cells of habitual users.
Mutagenesis
1997
;
12
:
285
–7.
40
Piyathilake CJ, Macaluso M, Hine RJ, Vinter DW, Richards EW, Krumdieck CL. Cigarette smoking, intracellular vitamin deficiency, and occurrence of micronuclei in epithelial cells of the buccal mucosa.
Cancer Epidemiol Biomarkers Prev
1995
;
4
:
751
–8.
41
Fontham E, Correa P, Rodriguez E, Lin Y. Validation of smoking history with the micronucleus test. In: Hoffmann D, Harris CC, editors. Mechanisms in Tobacco Carcinogenesis. New York: Cold Spring Harbor; p. 113–9.
42
Sarto F, Tomanin R, Giacomelli L, Iannini G, Cupiraggi AR. The micronucleus assay in human exfoliated cells of the nose and mouth: application to occupational exposures to chromic acid and ethylene oxide.
Mutat Res
1990
;
244
:
345
–51.
43
Frost JK. The cells in health and disease: An evaluation of cellular morphologic expression of biologic behaviour. Vol. 2. In: Orell SR, editor. Monographs in clinical cytology. New York: S Karger; 1986.
44
Guzman A, Garcia C, Marin AP, Ruiz MT, Tortajada A, Fernandez de Henestrosa AR. Induction of micronuclei in mouse bone-marrow erythrocytes in association with hypothermia after administration of the σ receptor ligand E-5842.
Mutat Res
2004
;
565
:
11
–22.
45
Bonassi S, Neri M, Lando C, et al. HUMN collaborative group. Effect of smoking habit on the frequency of micronuclei in human lymphocytes: results from the Human MicroNucleus project.
Mutat Res
2003
;
543
:
155
–66.
46
Villani P, Andreoli C, Crebelli R, Pacchierotti F, Zijno A, Carere A. Analysis of micronuclei and DNA single-strand breaks in mouse splenocytes and peripheral lymphocytes after oral administration of tetramethylthiuram disulfide (thiram).
Food Chem Toxicol
1998
;
36
:
155
–64.
47
Slamenova D, Chalupa I, Robichova S, et al. Effect of dietary intake of vitamin A or E on the level of DNA damage, chromosomal aberrations, and micronuclei induced in freshly isolated rat hepatocytes by different carcinogens.
Nutr Cancer
2002
;
42
:
117
–24.
48
Garriott ML, Phelps JB, Hoffman WP. A protocol for the in vitro micronucleus test. I. Contributions to the development of a protocol suitable for regulatory submissions from an examination of 16 chemicals with different mechanisms of action and different levels of activity.
Mutat Res
2002
;
517
:
123
–34.
49
Pindborg JJ, Reibel J, Roed-Peterson B, Mehta FS. Tobacco-induced changes in oral leukoplakic epithelium.
Cancer
1980
;
45
:
2330
–6.
50
Shklar G. Patterns of keratinization in oral leukoplakia.
Arch Otolaryngol
1965
;
87
:
400
–4.
51
Surralles J, Catalan J, Creus A, Norppa H, Xamena N, Marcos R. Micronuclei induced by alachlor, mitomycin-C and vinblastine in human lymphocytes: presence of centromeres and kinetochores and influence of staining technique.
Mutagenesis
1995
;
10
:
417
–23.