Markers for Cytogenetic Damage in Smokers: Associations with … · phenotype, and indices for tobacco exposure with SCEs and micronuclei in a group of 156 male cigarette smokers

Vol. 2, 44 1 -447, September/October 1993 Cancer Epidemiology, Biomarkers & Prevention 441

Received 11/25/92; revised 4/23/93; accepted 4/23/93.1 To whom requests for reprints should be addressed.

2 The abbreviations used are: SCE. sister chromatid exchange; GST-p, glu-

tathione S-transferase p.

Markers for Cytogenetic Damage in Smokers: Associations withPlasma Antioxidants and Glutathione S-Transferase p

Geert van Poppel,’ Hans Verhagen, Pieter van ‘t Veer,and Peter J. van BladerenTNO Toxicology and Nutrition Institute, Post Office Box 360, 3700Al Zeist, The Netherlands

Abstract

Biomarkers for increased cytogenetic damage in smokersinclude sister chromatid exchanges (SCE) in peripherallymphocytes and micronuclei in sputum cells. Thesemarkers may reflect increased cancer risk. Increasedcancer risk has also been associated with lower bloodlevels of the antioxidants �3-carotene and vitamin C andwith genetic deficiency of the detoxification enzymeglutathione 5-transferase p (GST-p). We thereforeevaluated the associations of plasma antioxidants, GST-pphenotype, and indices for tobacco exposure with SCEsand micronuclei in a group of 1 56 male cigarettesmokers and 38 nonsmokers.

As expected, smokers as compared with nonsmokershad higher SCE levels (5.08 versus 4.71 SCE/lymphocyte)and lower levels of plasma f3-carotene (0.31 versus 0.48pmol/liter) and blood vitamin C (36.6 versus 33.8pmol/liter). In smokers, SCEs were weakly correlatedwith plasma cotinine (r = 0.1 86) but not with plasmaantioxidants (all r < 0.04). Micronuclei in smokers werenot correlated with either cotinine or antioxidants (allr < 0.1 4). As reported previously, SCEs were higher(5.24 versus 4.97 SCE/lymphocyte) in GST-p-deficientsmokers than in nondeficient smokers. Micronuclei,however, were similar in both GST-p phenotypes (4.3versus 4.9 micronuclei/3000 cells). No correlation wasobserved between micronuclei and SCEs (r = -0.025).

Large random variations in both SCEs andmicronuclei make it difficult to interpret the absence ofrelations unambiguously. The results indicate that SCEsand micronuclei have only limited sensitivity tovariations in cigarette smoke exposure. The associationbetween GST-p and cancer risk may be mediatedthrough increases in certain forms of smoking-inducedDNA damage in GST-p deficiency.

Introduction

Biomankers are increasingly used in cancer epidemiology toestimate exposure to carcinogens or putative anticancino-gens, preclinical biological effects, as well as genetic factorsthat may determine individual susceptibility (1). Markers forDNA damage are of special interest since DNA damage is a

crucial step in carcinogenesis (2). Markers for DNA damagesuch as SCE2 (3) and micronuclei (4) are increased in smok-ems, who have a known increased risk of cancer (5).

Despite a large risk, a majority of smokers do not de-velop cancer. Smokers may be partially protected by dietaryantioxidants such as j3-carotene and vitamin C (6, 7). Also,genetic differences in detoxification of tobacco constituentsmay determine individual cancer risk (8-10). The possibleprotection against cancer of antioxidants has been hypoth-

esized to involve decreased DNA damage (1 1 ). Likewise, amore efficient detoxification of tobacco smoke could offerprotection by limiting DNA damage (8-1 0). It can thereforebe hypothesized that increased antioxidant status and moresuccessful detoxification will correspond with less DNAdamage in smokers. We have previously used the SCE meas-ure to demonstrate an association between deficiency in thedetoxification enzyme GST-p and increased cytogeneticdamage in smokers (1 2). We now have new data on bloodlevels of antioxidants and micronuclei in expectorated spu-turn from the same cross-sectional study. This allowed us toevaluate whether the two biomarkers for cytogenetic dam-age are inversely related with markers for antioxidant statusand positively associated with markers for cigarette smokeexposure. Also, the association of GST-p phenotype withmicronuclei was studied, and we evaluated whether SCEand micronuclei are associated.

Subjects and MethodsSubjects. We studied healthy male volunteers, employed atthe AMEV Insurance Company, the taxation office, and thepower company at Utrecht. The study was approved by anexternal medical ethical committee and all participants gavetheir informed consent. Smokers were studied if they me-ported consumption of more than 1 5 cigarettes/day for morethan 2 years. Nonsmokers were included in this study if theyreported never to have smoked and, in addition, did notwork or live with smokers. None of the participants usedvitamin preparations containing retinol or carotenoids ormedications known to influence SCE levels. Moreover, theyreported not to be exposed to xenobiotic chemicals throughtheir occupation or hobbies. Initially, 163 smokers and 38nonsmokers volunteered and were eligible for the study. Thesmokers were studied as part ofan intervention trial (1 3, 14).The present analysis is limited to the 156 smokers and 38nonsmokers for whom SCE data are available. The GST-passay was ambiguous for one smoker and missing for anothersmoker. Sputum samples were only collected by smokerssince nonsmokers do not spontaneously produce sputum. In

on March 30, 2020. © 1993 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

442 Markers for Cytogenetic Damage in Smokers

the smokers group, 29 participants failed to produce sputumsamples, whereas insufficient sputum cells could be evalu-ated in another 7 smokers.

Blood Parameters. Nonfasting blood samples were col-lected between 8.00 a.m. and 1 2.00 am. Directly after ye-

nipunctume blood samples were stored in the dark at 0 to 4#{176}C.After 20 to 23 h overnight dank storage at 4#{176}C,a separateevacuated tube containing hepanin as anticoagulant wasopened to determine the sum of ascorbic acid + dehydro-ascorbic acid (vitamin C) in whole blood by high perfon-mance liquid chromatography with fluonornetnic detection(1 5) (variation coefficient = 7.3%). All-trans-retinol,

ct-tocopherol, f3-canotene, and total carotenoids were as-sayed in EDTA plasma (stoned at -80#{176}C)by high perfon-mance liquid chromatography with colonimetnic detection(16). For these assays, coefficients of variation for samplesfrom a plasma pool were 4.3, 4.1 , 7.0, and 5.7%, respec-tively. Plasma cotinine levels were determined by gas chro-matography (1 7) (variation coefficient = 3%). Presence orabsence of GST-p was established in heparinized wholeblood using an enzyme-linked immunosonbent assay (MU-KIT; Medlabs, Dublin, Ireland).

Sister Chromatid Exchanges. Blood cultures for determina-tion of SCEs in lymphocytes (1 8) were set up within 2 to 6h after venipunctune, after the blood had reached room tern-penatune for 30 mm. For the total study group, blood cultureswere set up on 18 separate days; nonsmokers were studiedon 1 0 of these 1 8 days. Hepaninized whole blood (0.5 ml)was added to 4.4 ml prewarmed RPMI 1 640 (Flow) con-taming 20% fetal calf serum (inactivated for 30 mm at 56#{176}C),2.5% phytohaemaglutinin (HA-i 5; Welcome), 1 00 intemna-tional units/mi penicillin, 100 pg/mI streptomycin, 2 mtvtL-glutamine, and 1 0 pg/mI 5-bnomo-2-deoxyunidine. Theblood was cultured in the dark at 37#{176}Cin T-25 culture flasks(Costar) in 5% CO2 for 68 ± 1 h. Colcemid was added to afinal concentration of 0.2 pg/mI for the last 2 hours of in-cubation. The cells were collected by centnifugation, treatedwith hypotonic KCI (0.075 M) for 8 mm to spread the chro-mosomes and to hemolyse the RBC, and fixed 3 times withmethanol-acetic acid (3:1). After overnight storage in thedark at 4#{176}C,cells were transferred to microscopic slides andair-dried. Preparations aged for 3 days and were stained bythe Fluorescence plus Giemsa Technique (1 8) to obtain ham-lequin chromosomes. For each subject SCEs in 50 second-division metaphases were scored as color changes in thelongitudinal direction of the chrornatid, excluding the cen-tromere. Only metaphases with 46 chromosomes werescored. Individual data are the mean counts of 50 met-aphases. SCEs were scored by a single observer in the non-smoker group, whereas an additional observer assisted inscoring SCEs in the smoker group.

Micronuclei in Sputum. Sputum was collected and pro-cessed as described in detail by Saccomanno eta!. (1 9). Eachparticipant received a careful individual instruction on how

to produce a specimen from “deep in the lungs.” Sputum wascollected at home on three consecutive mornings, directlyafter rising and after carefully rinsing the mouth. The threeon, minimally, two samples collected in preservative (50 ml

50% ethanol with 2% polyethylene glycol) (Cambowax1 540; Merck) were mixed, homogenized, centrifuged, andsmeared onto slides. Sputum was collected on the days fol-lowing the venipunctume. The slides were stained with Feul-gen and fast green, which is specific for DNA and stronglyhighlights micnonuclei (20). For each subject, 3000 cells

Table 1 Characteristics and blood parameters in smokers and

nonsmokers’

. -Characteristic

Smokers)n=156(

Nonsmokers

(n=38)

Age 39.0 ± 9.6 37.0 ± 10.5

Body mass index (kg/m2( 24.5 ± 3.0 23.6 ± 2.4

Alcohol (�day)b,c 1 2.1 ± 1 3.1 2.8 ± 4.2

Cigarettes/day 21.1 ±6.5 0

Duration of smoking (yrs( 20.9 ± 10.0 0

Plasma cotinine” 327.7 ± 117.1 0.6 ± 0.5

Blood vitamin C” 36.6 ± 1 7.9 54.8 ± 1 1 .5

Plasma retinol 2.31 ±0.54 2.16±0.38

Plasma a-tocopherol” 30.9 ± 6.5 27.3 ± 7.0

Plasma f3-carotene” 0.31 ± 0.18 0.48 ± 0.30

Plasma total carotenoids” 1 .56 ± 0.59 1 .91 ± 0.72

SCE/lymphocyte” 5.08 ± 0.98 4.71 ± 0.35

Micronuclei/3000 sputum cells 4.6 ± 3.7 d

., A part of these data have been previously published (1 2, 1 3).

b Smokers different from nonsmokers: P < 0.01.

C Data for 1 43 smokers and 35 nonsmokers.C’ Not sampled.

were examined and evaluated on the basis of the followingcriteria: shape and size typical of epithelial cells; a well de-fined nucleus; and a clearly defined cytoplasm. The criteriain defining a micronucleus were chromatin structure andcolon intensity similar to those of the main nucleus; on fo-cusing, the micronucleus must be on the same level as thenucleus and must be moundish and clearly included in thecytoplasm. The dimensions should be less than one-fifth thatof the main nucleus, and the micronucleus should not beconnected to it. Slides were screened at X400 magnifica-tion, and micronucleated cells were examined at x1000magnification. Slides were read coded/blinded by a singleobserver. Repeated blinded scoring of 9 samples yielded agood correlation (Pearson r = 0.86), with 2 ofthe 9 samplesshowing a difference of more than 1 micronucleus uponrescoring.

Data Analysis. Smoking and nonsmoking groups, as well asGST-p-deficient and nondeficient groups were compared us-ing the Student t test and the x2 test. Associations betweenvariables were evaluated using simple and multiple linearregression. For micronuclei counts, square roots of obsem-vations were taken to stabilize variances before data analy-sis. In addition to the linear regression techniques, the un-transformed rnicronuclei data were also analyzed usingPoisson regression. Data analysis were performed using theBMDP package (21). For the Poisson regression, the GEN-STAT program was used (22).

ResultsTable 1 shows the data for the smoking and nonsmokinggroup. Plasma cotinine levels reflect the large contrast in

cigarette consumption between both groups. Age, bodymass index, and plasma metinol are similar in both groups,although slightly higher in the smoking group. In the smok-ing group, theme are more alcohol users (78 versus 63%) andthe mean alcohol consumption is substantially higher. Allplasma antioxidants are higher in nonsmokers than in smok-

ems, except for plasma a-tocopherol which is slightly thoughsignificantly higher in smokers. The SCE measure is aboutlO% increased in smokers. Adjustment for alcohol con-sumption did not alter this difference (z� SCE = 0.37 before



Fit’. 1. Distribution of sisterchromatid exchanges insmokers (n = 156) and non-

smokers )n = 38). Numberson abscissa, upper bound-aries of categories for SCEs.

n 1O�-�

8-�

20

18H�

16-h

II‘:‘�

� �

�;; [!-

. Non-smok�l

� Smoking

14-

12- �!

6

4..,

,. ‘

.,, I

2 r I’ -�

0 �- Li2.8 3.0

Li Y Li

3.2 3.4 3.6

),i

3.8 4.0 4.2

�.1; tI.; �

4.4 4.6 4.8

�.

5.0 5.2 5.4 5.6 5.8 6.0 6.2 6.4 6.6 6.8 7.0

Cancer Epidemiology, Biomarkers & Prevention 44.3

SCE per lymphocyte

adjustment, 0.33 after adjustment). The distribution of SCEsis given in Fig. 1 , whereas the distribution of micronucleicounts is given in Fig. 2.

The correlations for SCEs, micronuclei, and plasma co-tinine for the smokers group are given in Table 2. Plasmacotinine is clearly positively associated with the reportedcigarette consumption and weakly though significantly withage, smoking years, and, unexpectedly, plasma �-canotene.SCE levels show a weak significant association with bothplasma cotinine and reported cigarette consumption but noassociation whatsoever with plasma antioxidants on alcoholconsumption. No association was observed between SCEsand micronuclei in the smokers group (r = -0.025; Fig. 3).For the nonsmokers group, there were no significant asso-ciations of SCEs with any of the parameters in Table 2 (all r

< 0.14). Micmonuclei counts were not associated with anyofthe indices for tobacco consumption or antioxidant status.Poisson regression models for the micronuclei countsyielded results similar to those in Table 2: none of the as-sociations with the listed parameters was significant.

Of all subjects, 45% were deficient in GST-p. We ob-served no association between GST-p and any of the blood

parameters or characteristics given in Table 2. The associa-tion of GST-p deficiency with SCEs and micnonuclei is givenin Table 3. In nonsmokers, there is no relation betweenGST-p deficiency and SCE levels. In GST-p deficient smok-ems, SCE levels are higher than in nondeficient smokers. Thisp-related difference is more pronounced in “heavy” smokers(plasma cotinine above the median), whereas it is absent in“light” smokers. In the heavy smokers, the p-related differ-ence was similar in multiple regression analysis controllingfor age, body mass index, duration and quantity of smoking,and cotinine levels (5.47 versus 4.99; P = 0.02). In the lightsmokers, the multivariate adjustment also yielded similar me-

suIts (4.94 versus 4.98; P = 0.87). Similar analyses for ml-cronuclei in smokers did not reveal any associations withGST-p phenotype, either in light or heavy smokers (Table 3).

For the SCE count in smokers, we observed an effect ofculturing day variation (multiple partial r = 0.53; P< 0.001)and variation between observers(partial r= 0.51 ; P< 0.001)in a model that included i8 runs and 2 observers (multipler = 0.7i for the total model). Adjustment for culturing dayand observer variation did not materially alter the SCE dataand correlations for SCEs given in Tables 1-3. Only the dif-ference in SCEs between smokers and nonsmokers was morepronounced after this adjustment(4.85 versus 5.66 SCE/lym-phocyte; P < 0.OOi ). For the determination of micronuclei,we did not observe a significant effect ofthe 1 8 different runs

on the micronuclei counts (multiple r = 0.42; P = 0.20 in

linear regression; P = 0.20 in multiple Poisson regression).

DiscussionThe decreased blood levels of carotenoids and vitamin Cthat we observe in smokers are in line with previous studies(23, 24). These differences may reflect a decreased dietaryintake of vitamin C and carotenoids in smokers or a meta-bolic consequence of cigarette smoking (23, 24). A meta-bolic effect is biologically plausible, since cigarette smokeis a major source offree radicals and oxidant stress (25). Ourstudy did not measure dietary intake, but a metabolic effectof smoking is not clearly supported by the low correlationsthat we observe between plasma levels of cotinine, vitaminC, and carotenoids. For plasma cotinine and plasma�3-canotene, we even observe an unexpected weak positiveassociation. The antioxidant vitamin F is slightly, althoughsignificantly, higher in smokers than in nonsmokers, but thismay be explained by higher plasma lipid levels, which are



Fij,’. 2. Distribution o sputum micro-nuclei counts in smokers In = 143).

n

ii �

�1 ‘� � �I I (1 _______� L.._L L__ � ‘-r-’�� C.�..! ��__ � � � ,._.._ �


18

1 6

14t

12 H

�

4�

2H

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Micronuclel per 3000 cells

Table 2 Univariatecorrelations (Pearson coefficients) of severalcharacteristics and blood parameters with SCE, micronuclei, and plasma

cotinine in male cigarette smokers )n = 156)

. . . Plasma MicronucleiCharacteristic . . SCE

cotinine )n= 120)

Age 0.193’ 0.152 0.054

Body mass index (kglm2( -0.163” 0.083 0.006

Alcohol (gJday) -0.085 0.052 -0.057

Cigarettes/day 0. 368 “ 0.248 “ -0.006

Duration of smoking (yrs) 0.239-’ 0.109 0.006

Plasma cotinine 1.000 0.186” -0.013

Blood vitanlin C 0.042 0.039 -0.115

Plasma retinol 0.058 0.063 0.069

Plasma a-tocopherol 0.059 -0.022 0.007

Plasma (3-carotene 0.177” -0.038 0.109

Plasma total carotenoids -0.090 -0.021 0.136

.,P<0.05.

common in smokers (26). We had no lipid measurements tostandardize the vitamin E levels (27).

An increase in SCE levels in smokers is also a wellknown phenomenon, with levels being 1 0-88% higher thanin nonsmokers (5). It should be noted that SCE levels ob-served in several studies cannot readily be compared, sincelaboratory protocols are not standardized (i 8). The sensi-tivity of the SCE determination to laboratory variations isillustrated by the significant variation oven the 18 culturingdays, which we observe despite adherence to a strict pro-tocol. Nonlabomatory-nelated differences between the cul-turing days may also have attributed to this variation. In theanalyses within the group of smokers (Fig. 3; Tables 2 and3), the scorer and run variations did not bias our results, sincethe smokers were randomly divided over the scoring runs,and the observers were equally represented in every scoringnun. The analyses adjusted for run and observer variation

Table 3 SCEs and micronuclei in GST-p-positive and GST-p-negativenonsmokers, smokers, “light” smokers (plasma cotinine < 31 S ng/m)(, and

“heavy” smokers (plasma cotinine � 31 5 ng/ml(”

Smoking status Biomarker GST-p-negative GST-p-positive

Nonsmokers SCEs 4.69 ± 0.35 4.74 ± 0.35

(n = 38) Micronuclei “

Smokers SCEs5 5.24 ± 0.95 4.97 ± 0.98

(0 = 1 54) Micronuclei 4.3 ± 4.3 4.9 ± 2.9

Light smokers SCFs 4.95 ± 0.99 4.97 ± 1 .01

)n = 76) Micronuclei 4.1 ± 3.0 4.9 ± 3.0

Heavy smokers SCEs’ 5.50 ± 0.84 4.97 ± 0.97

(n = 78) Micronuclei 4.5 ± 5.0 5.0 ± 2.9

., A part of these data have been previously published (1 2).

,, P value for comparison, p versus p� = 0.09.‘ IJ_ differentIrom � P= 0.01.(I Not sampled.

therefore yielded results similar to the unadjusted analyses.The potential for bias, however, is illustrated by the contrastbetween nonsmokers and smokers. This contrast increasedafter adjustment for run and scorer variation, reflecting theunequal distribution of nonsmokers over the runs and scor-ems. This is also exemplified by Fig. 1 , showing that the over-all variation in SCE counts is fan less in the nonsmoker groupthan in the smoker group.

The correlations within the group of smokers betweenreported cigarette consumption or plasma cotinine are rathersmall. Reported cigarette consumption may not be a goodmeasure of genotoxic exposure due to differences in brandsand smoking and inhaling practices (28). Also, the relationbetween nicotine content and content of genotoxic agentsin cigarettes is uncertain. However, since we also observe anonly i 0-20% increase in smokers as compared with non-smokers, the conclusion that the SCE measure in lympho-



7-

6-

5-

8

B8

B

Fig. 3. Relationship between spu-

tum micronuclei counts and lympho-

cyte SCE in smokers In = 143).

a)

C)0

-C

E>�

w0C/)

0 8

B

3- 0

2-

1-

0- �-

0I I ( I I I I I I I I I ( I I � I I I

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Cancer Epidemiology, Biomarkers & Prevention 445

Micronuclel counts (per 3000 cells)

cytes is not very sensitive to variations in tobacco smokeexposure seems warranted. In this respect, it is noteworthythat lymphocytes also do not seem to be very sensitive toother types of genetic damage. DNA adducts or HPRT mu-tant frequency are also only moderately elevated in heavysmokers (29, 30).

Micronuclei in sputum (31) on bronchial brushings (32)have been reported to be 3-fold higher in smokers than non-smokers. Micnonuclei in exfoliated epithelial cells reflect theextent ofchmomosome breakage due to mutagenic exposure,when the cells were dividing a few days or weeks earlier, inthe basal layer ofthe epithelium ofthe tracheobronchial tree(33). The mean micmonuclei counts in our study in healthyvolunteers are lower than in studies using hospitalizedpatients (3i , 32). Bennem et a!. (34) also observed lowerrnicronuclei counts in healthy smokers undergoing broncho-scopy than in patients. The previous studies (31 , 32) also didnot describe their scoring criteria, and counts for “high cem-tainty micronuclei” may be much lower than counts includ-ing “medium certainty micronuclei” (35). Within our groupof smokers, the micmonuclei parameter is not associated withsmoking intensity, as was reported previously using countsin only 500 sputum cells in patients (31). The lack of sen-sitivity for micronuclei in sputum in our study may be at-tnibuted to a large random sampling site variation, since ex-pectorated cells may originate from all locations in thetracheobmonchial tree. Studies using buccal micnonuclei (33,35) and bronchial brushings (34) have indeed demonstratedmajor sampling site variations.

Both micmonuclei and SCEs reflect DNA damage, butthe exact molecular mechanisms are not known. Micmonu-clei are considered to reflect chromosome breakage (33),whereas SCEs are considered to be due to perturbations inthe DNA that persist through DNA replication (36). !n vitrostudies indicate that the mechanisms may be mediated byfree radicals and oxidants but can also involve other path-

ways (37-40). The absence of associations of the vitamins C

and E or camotenoids with SCEs and rnicmonuclei does notsupport an involvementoffree radicals and oxidants in vivo.However, theme are numerous other nonenzymatic and en-

zymatic antioxidants that we did not measure (41 ), bloodlevels may not reflect long-term antioxidant levels in lym-phocytes or lung tissue, and the range of antioxidant levelsin our study may not have been sufficiently large to dem-

onstrate associations. Also, large random variations both inmicronuclei and SCE counts may introduce bias towards the

null and thus obscure a possible weak association.The results for the GST-p phenotype indicate that a part

of the variation in SCE counts in smokers is genetically de-

termined, since the GST-p isozyme is inherited in an auto-somal dominant fashion (42). Glutathione-S-tnansferases de-toxify reactive electrophiles, in particular epoxides (43), and

GST-p deficiency may imply a more limited capacity for de-toxification and more carcinogen-mediated DNA damage.

Our results for the SCE measure support this hypothesis andsuggest that increased DNA damage in GST-p-deficientheavy smokers may be involved the association betweenGST-p and lung cancer that is observed in case-control stud-ies (8, 44=46), as we have previously discussed in more de-

tail (1 2). One of the studies showed a clean inverse relationbetween GST-p deficiency and lung cancer in heavy smok-

ems but not in light smokers (1 8), which seems to correspondwith our data. Another study (46) was more equivocal but

also observed an inverse relation (although not statisticallysignificant) only in heavy smokers. Zhong (45) observed aninverse relation for squamous carcinoma but not for adeno-

carcinoma ofthe lung. The results for the micronuclei countsdo not support the hypothesis since micronuclei counts wereeven somewhat lower in GST-p-deficient subjects. The dif-ferent results for the GST-p analysis suggest that micnonucleiand SCE may be different biological phenomena; although




epoxides may contribute to in vivo SCE induction, this maynot be the case for micronuclei.

The concept that SCEs and micnonuclei reflect differentbiological phenomena is also supported by the lack of anassociation between micmonuclei and SCEs, as depicted inFig. 3. Both micnonuclei and SCEs are sensitive to carcino-gens in experimental in vivo and in vitro models (36, 47, 48),but there is little information on correlations between SCEsand micmonuclei in these models to compare with our epi-demiological observations. Here again, caution in intenpre-tation of our data is required since unexplained randomvariation in both parameters may have obscured a weakassociation. Also, the differences in tissues and time framethat the two cytogenetic parameters reflect may have con-tnibuted to the absence of a relation.

This study has evaluated the application of a number ofbiomarkems in a cross-sectional study in smokers and non-smokers. SCEs and micmonuclei have been successfully usedin previous studies to demonstrate differences in DNA dam-age between smokers and nonsmokers. For the SCE measure,our study confirms these previous studies. In addition, theSCE measure in our study could be used to demonstratedifferences in cytogenetic damage between smokers with orwithout a genetically determined detoxification enzyme.Our results do, however, demonstrate that both SCE and mi-cronuclei have only limited or no sensitivity to variations incigarette smoke exposure within a group of smokers. Thislimited sensitivity may be partly attributed to large variationsof a yet unknown origin that we observe between persons inboth SCE and micnonuclei. This presumably random vamia-tion makes it difficult to unambiguously interpret the ab-sence of relations (e.g., between antioxidants and cytoge-netic damage) that we observe in this study. Moreinformation on biological, laboratory, and design factors thatdetermine variations is necessary to be conclusive about theabsence of associations.

References

1 . Hulka, B. S., Wilcosky, T. C., and Griffith, J. D. (eds). Biological Markers

in Epidemiology New York: Oxford University Press, 1990.

2. Weinstein, I. B. The origins ofhuman cancer: molecular mechanisms andtheir implications for cancer prevention and treatment. Cancer Res., 48:

4135-4143, 1988.

3. Wilcosky, T. C., and Rynard, S. M. Sister chromatid exchange. In: B. S.

Hulka, 1. C. Wilcosky, and J. D. Griffith (eds.), Biological Markers in Epide-miology. New York: Oxford University Press, 1990.

4. Vine, M. F. Micronuclei. In: B. S. Hulka, T. C. Wilcosky, and J. D. Griffith

(eds.), Biological Markers in Epidemiology. New York: Oxford UniversityPress, 1990.

S. IARC Monograph on the evaluation ofthe carcinogenic risk of chemicalsto humans, Vol. 38: Tobacco Smoking. WHO/International Agency for Re-

search on Cancer, Lyon, 1986.

6. Byers, 1., and Perry, C. Dietary carotenes, vitamin C, and vitamin E as

protective antioxidants in human cancers. Annu. Rev. Nutr., 12: 1 39-1 59,

1992.

7. Steinmetz, K. A., and Potter, J. D. Vegetables, fruits and cancer II. Mecha-nisms. Cancer Causes Control, 2: 427-442, 1 991.

8. Seideg�rd, J., Pero, R. W., Miller, D. C., and Beanie, E. J. A glutathionetransferase in human leukocytes as a marker for the susceptibility to lung

cancer. Carcinogenesis (Lond.), 7: 751-753, 1986.

9. Gelboin, H. V. Carcinogens, drugs, and cytochromes P-4S0. N. EngI. J.Med., 309:105-107, 1983.

10. Cough, A. C., Miles, J. S., Spurr, N. K., Moss, J. F., Caedigk, A., Eich-

elbaum, M., and Wolf, C. R. Identification of the primary gene defect at thecytochrome P450 CYP2D locus. Nature (Lond.), 347: 773-776, 1990.

1 1 . Sun, Y. Free radicals, antioxidant enzymes, and carcinogenesis. Free

Radical Biol. Med., 8: 583-599, 1990.

1 2. Van Poppel, C., De Vogel, N., Van Bladeren, P., and Kok, F. J. Increasedcytogenetic damage in smokers deficient in glutathione S-transferase isozyme

Ii. Carcinogenesis (Lond.), 13: 303-305, 1992.

1 3. Van Poppel, C., Kok, F. J., Duijzings, P., and De Vogel, N. No influence

of beta-carotene on smoking induced DNA damage as reflected by sister

chromatid exchanges. Int. I. Cancer, 51: 355-358, 1992.

14. Van Poppel, C., Kok, F. J., and Hermus, R. J. J. Beta-carotene supple-

mentation in smokers reduces the frequency of micronuclei in sputum. Br. J.Cancer, 66: 1164-1168, 1992.

1 S. Speek, A. I., Schrijver, J., and Schreurs, W. H. P. Fluorimetric determi-nation of total vitamin C in whole blood by high performance liquid chro-matography with pre-column derivatization. J. Chromatogr., 305: 53-hO,

1984.

16. Van Vliet, T., Van Schaik, F., Van Schoonhoven, J., and Schrijver, I.

Determination of several retinoids, carotenoids, and E vitamers by HPLC.Application to plasma and tissues of rats fed a diet rich in either f3-carotene

or canthaxanthin. I. Chromatogr., 553: 1 79-1 86, 1 991.

1 7. Feyerabend, C., Bryant, A., Jarvis, M. J., and Russell, M. A. H. Deter-

mination of cotinine in biological fluids of non-smokers by packed column

gas-liquid chromatography. J. Pharm. Pharmacol., 38: 1075-1078, 1984.

1 8. Perry, P. E., and Thomson, E. J. The methodology of sister chromatidexchanges. In: B. J. Kilbey, M. Legator, W. Nichols, and C. Ramel (eds.),Handbook of Mutagenicity Test Procedures, ed. 2 Amsterdam: Elseviers Sci-

ences Publishers, 1984.

1 9. Saccomanno, C. Diagnostic Pulmonary Cytology. Ch cago: AmericanSociety of Clinical Pathologists, 1978.

20. Bruck, H. C. Histologische Technik. Leidfaden f#{252}rder Herstellung mi-

croskopischer Praparate in Unterricht und Praxis. Thieme Verlag, Stuttgart:

Ceorg, 1981.

21 . Dixon, W. J. led). BMDP Statistical Software Manual. Berkeley, CA: Uni-versity of California Press, 1988.

22. CENSTAT S Committee. CENSTAT S Reference Manual. Oxford: Oxford

University Press, 1987.

23. Schectman, C., Byrd, I. C., and Cruchow, H. W. The influence of smok-ing on vitamin C status in adults. Am. J. Public Health, 79: 1 58-1 62, 1989.

24. Cerster, H. )3-carotene and smoking. J. Nutr. Growth Cancer, 4:4S-49,1987.

25. Pryor, W. A., Hales, B. J., Premovic, P. I., and Church, D. F. The radicalsin cigarette tar: their nature and suggested physiological complications. Sci-ence (Washington, DC), 220: 425-427, 1983.

26. Craig, W. Y., Palomaki, C. E., and Haddon, J. F. Cigarette smoking andserum lipid and lipoprotein concentrations: an analysis of published data. Br.Med. J., 298: 784-788, 1989.

27. Hunter, D. Biochemical indicators of dietary intake. In: W. C. Willett,Nutritional Epidemiology. Monographs in Epidemiology and Biostatistics,Vol. 15. New York: Oxford University Press, 1990.

28. Pojer, R., Whitfield, J. B., Poulos, V., Eckhard, I. F., Richmond, R., andHensley, W. I. Carboxyhemoglobin, cotinine and thiocyanate assay com-

pared for distinguishing smokers from non-smokers. Clin. Chem., 30: 1377-1380, 1984.

29. Perera, F. P., Santella, R. M., Brenner, D., Poirier, M. C , Munshi, A. A.,

Fischman, H. K., and Ryzin, I. V. DNA adducts, protein adducts, and sisterchromatid exchange in cigarette smokers and nonsmokers. J. NatI. CancerInst., 79:449-456, 1987.

30. Vrieling, H., Thijssen, J. C. P., Rossi, A. M., Van Dam, F. J., Natarajan,A. T., Tates, A. D., and Van Zeeland, A. A. Enhanced hprt mutant frequency

but no significant difference in mutation spectrum between a smoking and anon-smoking human population. Carcinogenesis (Lond.), 13: 1625-1631,

1992.

31. Fontham, E., Correa, P., Rodriguez, E., and Lin, Y. Validation of smokinghistory with the micronuclei test. In: D. Hoffmann and C. C. Harris (eds.),Mechanisms in Tobacco Carcinogenesis. Banbury Report 23. New York: Cold

Spring Harbor Laboratory, 1986.

32. Lippman, S. M., Peters, E. J., Wargovich, M. J., et al. Bronchial micro-

nuclei as a marker of an early stage of carcinogenesis in the human tracheo-bronchial epithelium. Int. J. Cancer, 45: 811-815, 1990.

33. Stich, H. F., and Rosin, M. P. Micronuclei in exfoliated human cells asa tool for studies in cancer risk and cancer intervention. Cancer Lett., 22:

241-253, 1984.

34. Benner, S. E., Lippman, S. M., Wargovich, M. 1., Velasco, M., Peters, E.1., Morice, R. C., and Hong, W. K. Micronuclei in bronchial biopsy specimensfrom heavy smokers: characterization of an intermediate marker of lung car-cinogenesis. Int. I. Cancer, 52: 44-47, 1992.

35. Tolbert, P. E., Shy, C. M., and Allen, W. I. Micronuclei and other nuclearanomalies in buccal smears: a field test in snuff users. Am. J. Epidemiol., 134:840-850, 1991.



Cancer Epidemiology, Biomarkers & Prevention 447

36. Wolff, S. Sister chromatid exchange: the most sensitive mammalian sys-tem for determining the effect of mutagenic carcinogens. In: K. Berg led.),

Genetic Damage in Man Caused by Environmental Agents, pp. 229-246.New York: Academic Press, 1979.

37. Lee, C. K., Brown, B. C., Rice, W. Y., and Doolittle D. J. Role of oxygenfree radicals in the induction of sister chromatid exchanges by cigarettesmoke. Environ. Mol. Mutagen., 13: 54-59, 1989.

38. Weitberg, A. B., Weitzman, S. A., Clarck, F. P., and Stossel, T. P. Effects

of antioxidants on oxidant induced sister chromatid exchange formation. J.Clin. Invest., 75: 1835-1841, 1985.

39. Lasne, C., Cu, Z. W., Venegas, W., and Chouroulinkov, I. The in vitromicronucleus assay for detection of cytogenetic effects induced by mutagen-carcinogens: comparison with the in vitro sister-chromatid exchange assay.Mutat. Res., 130: 273-282, 1984.

40. Vanparys, P., Vermeiren, F., Sysmans, M., and Temmerman, R. The mi-

cronucleus assay as a test for the detection of aneugenic activity. Mutat. Res.,

244:95-103, 1990.

41 . Halliwell, B., and Cutteridge, J. M. C. Free radicals in biology and medi-

cine. New York: Oxford University Press, 1985.

42. Seideg#{226}rd, J., and Pero, P. W. The hereditary transmission of high glu-tathione transferase activity towards trans-stilbene oxide in human mono-nuclear leukocytes. Hum. Cenet., 69: 66-68, 1985.

43. Mannervik, B., and Danielson, U. H. Glutathione-transferases: structure

and catalytic activity. CRC Crit. Rev. Biochem., 23: 283-337, 1988.

44. Seideg�rd, J., Pero, R. W., Markowitz, M. M., Roush, C., Miller, D. C.,

and Beattie, F. J. Isoenzyme(s) of glutathione transferase (class p1 as a marker

for the susceptibility to lung cancer: a follow up study. Carcinogenesis (Lond.),

11:33-36, 1990.

45. Zhong, S., Howie, A. F., Ketterer, B., Taylor, J., Hayes, I. D., Beckett, C.

J., Wathen, C. C., Wolf, C. R., and Spurr, N. K. Glutathione S-transferase p

locus: use of genotyping and phenotyping assays to assess association with

lung cancer susceptibility. Carcinogenesis (Lond.(, 12: 1 533-1 537, 1 991.

46. Heckbert, S. R., Weiss, N. S., Hornung, S. K., Eaton, D. L., and Motulsky,

A. C. Clutathione S-transferase and epoxide hydrolase activity in human leu-kocytes in relation to risk of lung cancer and other smoking-related cancers.

J. NatI. Cancer Inst., 84:414-422, 1992.

47. Jensen, D., and Ramel, C. The micronuclei test as part of short termmutagenicity test program for the prediction of carcinogenity evaluated by

1 43 agents tested. Mutat. Res., 75: 1 79-1 91 , 1980.

48. Stich, H. F., Stich, W., and Panda, B. B. Elevated frequency of micro-nucleated cells in the buccal mucosa ofindividuals at high risk fororal cancer:

betel quid chewers. Cancer Lett., 17: 125-134, 1982.



1993;2:441-447. Cancer Epidemiol Biomarkers Prev G van Poppel, H Verhagen, P van 't Veer, et al. plasma antioxidants and glutathione S-transferase mu.Markers for cytogenetic damage in smokers: associations with

Updated version

http://cebp.aacrjournals.org/content/2/5/441

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cebp.aacrjournals.org/content/2/5/441To request permission to re-use all or part of this article, use this link



http://cebp.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Markers for Cytogenetic Damage in Smokers: Associations with … · phenotype, and indices for tobacco exposure with SCEs and micronuclei in a group of 156 male cigarette smokers