| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH COMMUNICATION |
1 From the Vitamins and Carcinogenesis Laboratory (HEG, JWC, S-WC, MKK, HJ, ZL, and JBM), the Vitamin Metabolism Laboratory (HG, MN, and AJ), and the Biostatistics Unit (GED), Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University, Boston, MA, and the Forsythe Dental Research Institute, Harvard School of Dentistry, Boston, MA (DM)
2 Supported by the US Department of Agriculture, under agreement No. 581950-9-001. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the US Department of Agriculture. This work was partially supported by grants from the NCI (KO5 CA100048-01) and from the NIDDK (T32 DK007651-16). 3 Reprints not available. Address correspondence to J Crott, Vitamins and Carcinogenesis Laboratory, Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts University, 711 Washington Street, Boston, MA 02111. E-mail: jimmy.crott{at}tufts.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: Our aim was to compare the presence of folate metabolites and select indicators of genetic damage in the mouths of chronic smokers and nonsmokers.
Design: Dietary, biochemical, and molecular correlates of folate status were measured in healthy smoker (n = 35) and nonsmoker (n = 21) groups of comparable age, sex, and body mass indexes.
Results: After correction for dietary intake, the smokers displayed lower plasma, erythrocyte, and buccal mucosal cell (BMC) folate (20%, 32%, and 50% lower, respectively; P < 0.05) and lower plasma vitamin B-12 and pyridoxal 5-phosphate (P < 0.05) than did nonsmokers. Folate in the BMCs of smokers comprised significantly greater proportions of pteroylmonoglutamate, formyltetrahydrofolate, and 5,10-methenyltetrahyrofolate than did folate in the BMCs of nonsmokers. Although the degree of genomic methylation and uracil incorporation in the buccal cells of the 2 groups were not significantly different, the BMC micronucleus index, a cytologic indicator of genetic damage, in the smokers was 2-fold that of the nonsmokers (9.57 compared with 4.44 micronuclei/1000 cells; P < 0.0001). Neither systemic nor oral folate status was an independent predictor of micronuclei.
Conclusions: Chronic smoking is associated with a lower systemic status of several B vitamins, reduced oral folate, and changes in folate form distribution in the mouth. However, the cytologic damage that is evident in the mouths of smokers does not correlate with oral folate status.
Key Words: Folate • smoking • micronuclei • buccal mucosal cell
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
275000 new cases per year. The incidence of this cancer in the developing world is 2–3-fold that in the developed world, and it is in the former regions where deficiencies of various micronutrients remain prevalent (2). Tobacco smoking (3-5) and alcohol consumption (3-6) are among the strongest risk factors for oral and pharyngeal cancers, whereas a high intake of fruit (4, 7, 8) and vegetables (4, 8) seems to be protective. The carcinogenic effect of cigarette smoking is driven largely by the mutagenicity of various chemicals in the smoke. Tobacco smoke induces an array of genetic aberrations, including gene mutations, chromosome aberrations, micronuclei, sister chromatin exchanges, DNA strand breaks, and oxidative DNA adducts in various models (9). In particular, the frequency of micronuclei is consistently elevated in exfoliated buccal mucosal cells (BMC) of tobacco smokers (10-15). Buccal cell micronuclei are a putative biomarker for oral cancer risk; evidence suggests that micronuclei are elevated in BMCs of persons who harbor precancerous lesions and in cancer patients (10, 12, 16).
High folate intake probably protects against colorectal cancer (17) and possibly against breast (18-20), uterine cervix (21-23), and lung (24, 25) cancers as well. Because folate is critical for DNA synthesis, methylation, and repair in all cells, folate status may also modulate the risk of oral cancers, particularly because the oral epithelium is a continually proliferating and regenerating tissue. Only a few epidemiologic studies have examined this link, and the results have not been consistent: high folate intake was reported to afford significant protection in one study (26), but no such protection was observed in 2 others (7, 27). In support for a protective role of high folate intake, folate repletion (111–516 µg/d) for 7 wk was shown to halve buccal micronuclei frequency in a small sample of women (28). Furthermore, recent studies have observed an association between specific polymorphisms in folate-dependent enzymes and the risk of head and neck cancers (29-31).
Smokers have lower plasma (13, 32-34), red blood cell (RBC; 32, 34), and BMC (13, 33, 34) folate concentrations than do nonsmokers, a trend which persists after correction for folate intake (32, 35). This suggests that cigarette smoking alters the systemic uptake or metabolism of folate either directly or by modifying other components of one-carbon metabolism. The latter point is exemplified by depletion of vitamin B-6, which serves as a cofactor for the interconversion of different forms of folate, in smokers (32).
In addition to the direct genotoxic effect of tobacco smoke on BMCs, a second procarcinogenic pathway initiated by smoking may involve the depletion of folate and thereby the reduction of cofactors necessary to support efficient DNA synthesis, repair, and methylation. Folate depletion causes various genetic and epigenetic aberrations in mammalian cells, such as uracil misincorporation (36, 37), genomic (38) and p53-specific (39) DNA breaks, and genomic (40) and p53-specific (39) hypomethylation.
Therefore, we sought to investigate the relations between chronic cigarette use, total and specific folate pools within the oral mucosa, and putative biomarkers for the risk of oral carcinogenesis. Because folate is only one of several dietary cofactors involved in one-carbon metabolism, we examined these relations in the context of the systemic status of several other one-carbon nutrients.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To eliminate any confounding effects of excessive alcohol intake, partcipants were excluded if they reported habitual consumption of >2 alcoholic beverages/d. Smokers were required to have been smoking 
10 cigarettes/d for 1 y before the start of the study. Nonsmokers were required to have abstained from all forms of smoking and chewing tobacco for a period of 1 y before the start of the study. Fifty-six healthy participants aged between 30 and 80 y and of varying ethnic backgrounds were recruited. Thirty-five of the participants were current smokers and 21 were current nonsmokers. We intentionally enrolled a greater number of participants who were smokers than nonsmokers because we wished to additionally examine the effects of heavy compared with light smoking. Before the study, all participants were required to abstain from the use of any vitamin or nutritional supplements for a 3-mo period. Because we aimed to study the effect of chronic, and not acute, cigarette exposure, the participants must have had 2 cigarettes the day before the study, but were instructed to abstain from smoking for 4 h before blood and buccal cell collections.
On entry into the study, fasting blood samples and cheek scrapings were obtained from the participants. Plasma was collected and analyzed for pyridoxal 5-phosphate (PLP) by radioimmunoassay (41), for vitamin B-12 by radioimmunoassay (BioRad Quantiphase II, Hercules, CA), and for total homocysteine by HPLC with fluorescence detection (42). RBC and plasma total folate concentrations were measured by using the Lactobacillus casei method of Horne and Patterson (43) with modifications (44). Peripheral blood mononuclear cells were isolated from blood that was collected in Cell Preparation Tubes (Becton Dickinson, Franklin Lakes, NJ) according to the manufacturer's instructions.
BMCs were collected by gently rubbing the inside of both cheeks with an extra-soft toothbrush for 1 min each. The participants rinsed their mouths with 20 mL 0.9% saline and expectorated into a 50-mL conical-based tube. The toothbrush was then rinsed into the tube with an additional 30 mL saline before the cells were pelleted and washed once with phosphate buffered saline (pH 7.4). BMCs were resuspended in 1 mL phosphate buffered saline and counted with a hemocytometer. Approximately 4 x 105 BMCs were allocated for total and folate form distribution and 2.5 x 105 cells for DNA isolation.
Ten µL of BMC suspension was smeared on a microscope slide and scored by a single blinded observer for the presence of micronuclei according to the method of Titenko-Holland (28), except slides were stained with May-Grunwald Giemsa (Sigma, St Louis, MO). BMC folate content and conenzymatic distribution was measured by using affinity chromatography followed by reversed-phase chromatography with electrochemical detection (45). BMC total folate was verified by using the microbiologic method, including conjugase pretreatment, according to the method by Horne and Patterson (43) with modifications (44). A strong correlation was observed between the BMC total folate values measured by HPLC and those measured by L. casei (r = 0.85, P < 0.0001). Because the outcome of comparisons between the groups was the same for both methods, only the HPLC results for BMC folate are reported here.
DNA was extracted from peripheral blood mononuclear cells and BMCs with the Easy DNA Kit (Invitrogen, Carlsbad, CA) and quantified by measuring absorbance at 260 nM on a standard UV-Vis plate reader (Biotek, Winooski, VT). Genomic DNA methylation was measured in 1 µg DNA with liquid chromatography–mass spectrometry by using the method of Friso et al (46). DNA uracil content was measured in 7.5 µg DNA with gas chromatography–mass spectrometry by using the method of Blount and Ames (47).
The participants also completed a self-administered oral and systemic health history form, a lifestyle form, and the Willett semiquantitative Food-Frequency Questionnaire (FFQ), version III (48). Habitual folate intake assessed by this questionnaire correlates well with erythrocyte folate concentrations (r = 0.55, P < 0.01) (49).
Statistical analyses
Differences between smokers and nonsmokers were tested with unpaired t tests, analysis of covariance, and multiple regression. Hotelling's T2 statistic (50) was used to test whether the distribution of the different coenaymatic forms of folate was different. Associations between variables were analyzed with Pearson correlation coefficients and multiple regression. Statistical calculations were performed with SAS version 8.02 (SAS Institute, Cary, NC), and results were judged statistically significant when the 2-sided significance level (P value) was < 0.05. Data are reported as means ± SEMs.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Dietary data were incomplete on 2 of the 35 smokers; therefore, the dietary analyses were performed with data from 33 of 35 smokers. Alcohol, B-vitamin, methionine, and caloric intakes were not significantly different between the 2 groups (P > 0.05; Table 1).
|
). Smokers not only had lower total folate content in their BMCs, but the proportions of the different folate coenzymes that comprised the total folate content were also significantly different from those of nonsmokers (Hotelling's T2 P < 0.0001). Specifically, smokers had higher proportions of folic acid (pteroylglutamate), formylated tetrahydrofolate, and 5,10-methenyltetrahydrofolate, and lower proportions of tetrahydrofolate and 5,10-methylenetetrahydrofolate than did nonsmokers. 5-methyltetrahydrofolate was not significantly different between the groups (Table 2).
|
).
The frequency of micronuclei in the buccal cells of smokers was approximately twice that of nonsmokers (P < 0.0001; Table 3). In this regard, we could not detect a dose effect, because there was no significant difference in the micronucleus frequency between those who smoked <20 cigarettes/d and those who smoked >20 cigarettes/d (9.3 ± 0.4 and 10.1 ± 0.8, respectively; P = 0.396). The concentrations of BMC uracil, BMC DNA methylation, or lymphocyte DNA methylation were not statistically different between the smokers and nonsmokers (Table 3).
|
2 alcoholic beverages/d, habitual alcohol intake still constituted a significant determinant of the micronucleus index (r = 0.30, P = 0.029). A dose effect of alcohol was not observed (data not shown), but, as mentioned above, we restricted ourselves to participants with a relatively narrow range of alcohol intake. BMC folate (r = –0.48, P = 0.0002), RBC folate (r = –0.33, P = 0.011), and plasma folate (r = –0.29, P = 0.042) each appeared to be negatively correlated with the micronucleus index in univariate analyses, but this was due to confounding with smoking status. Each of these correlations became nonsignificant after correction for smoking history. We were also interested in establishing what the strongest determinants of folate status were in the plasma, RBC, and BMC compartments. By using multiple regression, all possible combinations of age, sex, no. of cigarettes/d, alcohol intake, and folate equivalent intake were tested to identify the strongest models for the prediction of folate concentrations in each of the 3 compartments. In all 3 compartments, the overwhelming predominant predictor of folate concentration was smoking. The number of cigarettes smoked per day constituted the only significant independent predictor among the abovementioned factors, with the exception of sex for BMC folate concentration, which increased the adjusted R2 in that compartment from 0.21 to 0.25 (P < 0.03). Adding the intake of dietary folate equivalents to these models did not strengthen the adjusted R2 value (data not shown). This may be a consequence of using FFQs to assess mean dietary intakes in a relatively small population of subjects, which is a known limitation of FFQs, or may reflect the fact that the effect of smoking overwhelmed the predictive value of dietary folate.
In contrast, smoking was not a strong predictor for plasma homocysteine, a metabolic marker which reflects folate, vitamin B-12, and vitamin B-6 status. We tested all possible combinations of age, sex, number of cigarettes, alcohol intake, RBC and plasma folate, vitamin B-12, and PLP concentrations. All regression models that strongly predicted plasma homocysteine included plasma B-12 as a predictor (adjusted R2 for B-12 alone = 0.1774; adjusted R2 for the model with all predictors = 0.1814).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
20% lower in smokers than in nonsmokers (P = 0.037). This particular result differs somewhat from previous studies that show that although BMC vitamin B-12 is mildly reduced in smokers, plasma vitamin B-12 is not affected by smoking status (13, 34). This discrepancy may be due to the greater cigarette use in our smoking group, who smoked a mean (±SD) of 21.12 ± 1.33 (range: 10 –36) cigarettes/d, and to a 10 y higher mean age than those subjects in at least one of the other studies (34). Chronic tobacco exposure appeared to have quite a robust effect on folate status in the present study. Although one should remain cautious about over-interpreting dietary data drawn from FFQs in populations of limited size, our multiple regression analyses indicated that the effect of smoking was such an overwhelmingly strong determinant of folate status that the predictive value of dietary folate intake was nonsignificant in each of the 3 body compartments that we analyzed (plasma, RBC, and BMC).
To the best of our knowledge, this is the first study of the effect of smoking on the distribution of folate coenzymes in any tissue. We showed that smokers possess a relative depletion of tetrahydrofolate and 5,10-methylenetetrahydrofolate and a relative accumulation of folic acid (pteroylmonoglutamate), formyltetrahydrofolate, and 5,10-methenyltetrahydrofolate in buccal cells. Concomitant with this altered folate profile, smokers exhibited lower concentrations of both plasma vitamin B-6 and B-12 than did nonsmokers. This change in the folate coenzyme profile may be a secondary consequence of changes in other B-vitamins; it was previously shown that vitamin B-6 depletion significantly inhibited the transfer of 2H from serine to methionine in rats, indicating a reduction in serine hydroxymethyltransferase activity, which is a pathway for the interconversion of different forms of folate (51). In addition, the oxidation and inhibition of the methionine synthase–vitamin B-12 complex by exposure to nitrous oxide (which is present in cigarette smoke) induces the accumulation of 5-methyltetrahydrofolate and the depletion of tetrahydrofolate in rats (52). Thus, it is plausible that the disruption we observed in folate form distribution in smokers is partly due to a reduced activity of vitamin B-6–and B-12–dependent enzymes. Reduced serine hydroxymethyltransferase activity may explain the lower percentage of 5,10-methylenetetrahydrofolate observed, whereas reduced methionine synthase activity might increase the fraction of 5,10-methylenetetrahydrofolate that is metabolized to 5,10-methenyltetrahydrofolate and then to 10-formyltetrahydrofolate because of an inability to process 5-methyltetrahydrofolate. Additional studies are clearly needed to elucidate the mechanisms responsible for the observed shifts in folate form distribution due to smoking. The ramifications of altered folate form distribution, as it applies to carcinogenesis, are not known. However, they are likely to be important, because the enhanced risk of cancer produced by low folate availability is thought to be partly due to an imbalance between the supply of folates that support biological methylation and those that support nucleotide synthesis (53).
Plasma homocysteine concentrations were not significantly different between the smokers and nonsmokers despite significant differences in blood folate, vitamin B-6, and vitamin B-12 concentrations. Previously, smoking was found to be a significant determinant of plasma homocysteine in the Framingham Offspring cohort (54); however, only the persons who smoked >26 cigarettes/d had significantly higher homocysteine concentrations than did nonsmokers. It is likely that the small number of people who smoked this many cigarettes in our study (n = 8) precluded us from detecting a significant smoking effect on homocysteine concentrations. A multiple regression analysis uncovered that, in this population, the strongest (and only demonstrable) determinant of plasma homocysteine concentrations was the plasma vitamin B-12 concentration. This observation is consistent with findings from much larger studies, which showed that both intake and plasma concentrations of vitamin B-12 are important determinants of plasma homocysteine concentrations (54).
Because the molecular pathogenesis of tobacco-induced oral cancer is poorly understood, we examined 3 molecular and cytogenetic anomalies that are affected by folate status and which could affect carcinogenesis of the mouth, although only in the instance of micronuclei is there reasonable evidence for folate as an intermediary biomarker of oral cancer (10, 12, 16). Smoking was associated with a 2-fold micronucleus frequency in BMCs, as has been previously reported (10-15), but was not a predictor of genomic DNA methylation or uracil content in the BMCs. A multivariate regression analysis indicated that, of all the factors examined in the present study, the predominant determinant of micronucleus formation was cigarette smoking.
Although oral micronuclei and low oral folate are each associated with smoking, we did not detect an association between oral concentrations of folate and oral micronuclei, nor did we observe a correlation with systemic indicators of folate status. The absence of association between systemic concentrations of folate and oral micronuclei was also reported by Titenko-Holland et al (28), although they only studied 9 subjects, did not measure oral concentrations of folate, and did not study smokers, and, therefore, their observations are much more limited. Even when we eliminated the overwhelming effect of smoking, ie, when we considered nonsmokers alone, no regression models consisting of BMC total folate, folate distribution, DNA uracil, and DNA methylation could significantly explain the variance in micronucleus frequency (data not shown). Although one must be careful in inferring causality in a cross-sectional design, the absence of any correlation between BMC uracil and micronuclei suggests that folate depletion–induced uracil misincorporation is not a major route of micronucleus formation in the healthy buccal mucosa. Similarly, despite a relative depletion of 5,10-methylenetetrahydrofolate in the BMCs of smokers, it seems that the remaining pool of this folate species is sufficient to maintain an adequate deoxythymidylate pool for DNA synthesis, because the DNA uracil content was not significantly different between smokers and nonsmokers.
In conclusion, our data confirm the observations that cigarette smoking induces a systemic depletion of folate and vitamin B-6 and a localized depletion of folate in the mouth. We also showed that smokers had reduced plasma vitamin B-12 and a redistribution of folate forms in the buccal mucosa. We also showed that smoking induced an accumulation of micronuclei in the buccal mucosa, which is apparently unrelated to smoking-induced changes in folate metabolites. The implications of our observations remain open to interpretation, because the molecular carcinogenesis of oral cancer is still poorly understood. Oral folate concentrations did not correlate with the micronuclei index in the present study, and changes in genomic methylation or uracil were not apparent in the mouths of smokers despite the low concentrations of total folate and the altered folate form distribution in this group. Although such observations do not support a mechanistic role for folate in oral carcinogenesis, they do not exclude a potential chemopreventive effect of adequate or supplemental folate. Although folate concentrations were markedly lower in the mouths of smokers than in the mouths of nonsmokers, we cannot exclude the possibility that, in the present era of folate fortification, folate concentrations were not low enough in the smokers to alter the molecular and cytologic endpoints of our study. Furthermore, altered folate status in the mouth might affect molecular pathways that we did not examine; the micronucleus index, similar to any intermediary biomarker of cancer (55), merely reflects a limited portion of the entire process by which cancer evolves.
|
ACKNOWLEDGMENTS |
|---|
JWC and JBM were responsible for drafting the manuscript. JBM, HEG, S-WC, JWC, and GED were responsible for the experimental design. JWC, GED, and ZL were responsible for the statistical analyses. HEG, HG, AJ, and MN were responsible for the folate analyses. MKK, HJ, and S-WC were responsible for the DNA analyses. HEG, JWC, and DM were responsible for the buccal cell collection and micronuclei analyses. None of the authors had any conflicts of interest.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. C. Larsson, S. Mannisto, M. J. Virtanen, J. Kontto, D. Albanes, and J. Virtamo Folate, Vitamin B6, Vitamin B12, and Methionine Intakes and Risk of Stroke Subtypes in Male Smokers Am. J. Epidemiol., April 15, 2008; 167(8): 954 - 961. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gissot, S.-W. Choi, R. F. Thompson, J. M. Greally, and K. Kim Toxoplasma gondii and Cryptosporidium parvum Lack Detectable DNA Cytosine Methylation Eukaryot. Cell, March 1, 2008; 7(3): 537 - 540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Li, L.-E Wang, P. Chang, A. K. El-Naggar, E. M. Sturgis, and Q. Wei In vitro Benzo[a]pyrene Diol Epoxide-Induced DNA Adducts and Risk of Squamous Cell Carcinoma of Head and Neck Cancer Res., June 15, 2007; 67(12): 5628 - 5634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fenech, C. Bolognesi, M. Kirsch-Volders, S. Bonassi, E. Zeiger, S. Knasmuller, and N. Holland Harmonisation of the micronucleus assay in human buccal cells--a Human Micronucleus (HUMN) project (www.humn.org) initiative commencing in 2007 Mutagenesis, January 1, 2007; 22(1): 3 - 4. [Full Text] [PDF]
|
|
|
|
 |
|
 S. J Weinstein, R. Stolzenberg-Solomon, P. Pietinen, P. R Taylor, J. Virtamo, and D. Albanes Dietary factors of one-carbon metabolism and prostate cancer risk. Am. J. Clinical Nutrition, October 1, 2006; 84(4): 929 - 935. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K Nersesyan Does cigarette smoking induce micronuclei in buccal cells? Am. J. Clinical Nutrition, October 1, 2006; 84(4): 946 - 947. [Full Text] [PDF]
|
|
|
|
|
|
J. W Crott Reply to AK Nersesyan Am. J. Clinical Nutrition, October 1, 2006; 84(4): 947 - 948. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Nersesyan, M. Kundi, K. Atefie, R. Schulte-Hermann, and S. Knasmuller Effect of staining procedures on the results of micronucleus assays with exfoliated oral mucosa cells. Cancer Epidemiol. Biomarkers Prev., October 1, 2006; 15(10): 1835 - 1840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. E. Gabriel, Z. Liu, J. W. Crott, S.-W. Choi, B. C. Song, J. B. Mason, and E. J. Johnson A Comparison of Carotenoids, Retinoids, and Tocopherols in the Serum and Buccal Mucosa of Chronic Cigarette Smokers versus Nonsmokers. Cancer Epidemiol. Biomarkers Prev., May 1, 2006; 15(5): 993 - 999. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
