Ann. Occup. Hyg., Vol. 50, No. 6, pp. 583–592, 2006# The Author 2006. Published by Oxford University Press

on behalf of the British Occupational Hygiene Societydoi:10.1093/annhyg/mel036

Assessment of Occupational Genotoxic Risk in theProduction of Rubber TyresBLANCA LAFFON1*†, JOAO PAULO TEIXEIRA2†, SUSANA SILVA2,JOANA ROMA-TORRES2, BEATRIZ PEREZ-CADAHIA1,JOSEFINA MENDEZ3, EDUARDO PASARO1 and OLGA MAYAN2

1Toxicology Unit, University of A Coruna, Edificio de Servicios Centrales de Investigacion,

Campus Elvina s/n, 15071-A Coruna, Spain; 2National Institute of Health, Environmental Health

and Toxicology Department, Largo 1 Dezembro, 4000-Porto, Portugal; 3Department of Cell and

Molecular Biology, University of A Coruna, Faculty of Sciences, Campus A Zapateira s/n,

15071-A Coruna, Spain

Received 9 January 2006; in final form 12 April 2006; published online 23 June 2006

A broad spectrum of substances is used in the rubber industry, many of them being genotoxicand/or carcinogenic. Convincing evidence of an excess of certain forms of cancer amongrubber workers has been provided. The objective of this study was to determine the genotoxiceffects in a group of individuals engaged in the production of rubber tyres from a Portuguesefactory. Peripheral blood samples were collected from 32 exposed workers and 32 controls, andmicronucleus (MN) test, sister chromatid exchanges (SCE) and comet assay were performed.Urinary thioethers were measured as a general biomarker of exposure to electrophilic com-pounds, and genetic polymorphisms in metabolizing enzymes (CYP2E1 Dra I, EPHX1 codons113 and 139, GSTP1 codon 105, and GSTM1 and GSTT1 deletion polymorphisms) wereanalysed as susceptibility biomarkers. Excretion of thioethers was found significantly higherin rubber workers. Also, a non-significant increase inMN frequency related to time of exposureand no effect in SCE were observed in the exposed. Comet assay data showed decreasedTL values in the exposed population with respect to the control group, this might indicatethe induction of crosslinks by the substances present in the workplace environment. Significantincrease in MN frequency was obtained for GSTT1 null exposed individuals with respect topositive ones, and interaction with GSTP1 polymorphism was found. Higher levels of cytoge-netic test frequencies were observed in epoxide hydrolase expected low activity donors withrespect to medium and high activity individuals. No effect of CYP2E1 or GSTM1 variants wasobtained in the biomarkers analysed.

Keywords: comet assay; metabolic polymorphisms; micronucleus; rubber tyres; sister chromatid exchanges

INTRODUCTION

Rubber industry uses a broad spectrum of substances

(vulcanization agents, accelerators, activators, col-

orants, solvents, etc.), belonging to many chemical

categories (polycyclic aromatic hydrocarbons [PAH],

N-nitrosamines, 1,3-butadiene, acetonitrile, styrene,

vinyl chloride, ethylene oxide, mineral oils, other

volatile organic compounds, etc.) (Fishbein, 1991;

Sturaro et al., 1993; Oury et al., 1997), some of

which have been shown to be genotoxic and/or

carcinogenic. In the environment of tyre producing

plants, presence of several organic solvent vapours

and airborne particulate matter has been character-

ized (Kromhout et al., 1994; Dost et al., 2000). In

addition, elevated mutagenicity and genotoxicity

have been observed in air and particulate samples

from rubber manufacturing plants (Baranski et al.,

1989; Fracasso et al., 1999; Vermeulen et al.,

2000; Monarca et al., 2001), although the substantial

differences found between companies have been

attributed to differences in rubber chemicals used

and the overall level of control measures

(Vermeulen et al., 2000).

Convincing evidence of an excess of certain forms

of cancer among rubber industry workers has been

*Author to whom correspondence should be addressed. Tel:34 981167000; fax: 34 981167172; e-mail: blaffon@udc.es†The first two authors contributed equally to this work.

583

provided by a large number of industry-based

epidemiological studies (IARC, 1982, 1987). A sub-

sequent updating of those data confirmed increased

risk of bladder, laryngeal and lung cancers and

leukaemia (Kogevinas et al., 1998). Later, exposures

to 1,3-butadiene and dimethyldithiocarbamate in the

synthetic rubber industry were positively associated

with leukaemia in multivariable analyses (Delzell

et al., 2001), and elevated risk of bladder cancer

was reported to be associated with exposure to

b-naphthylamine that reversed when the substance

was removed from the rubber manufacturing process

(Veys, 2004),

Assays measuring micronucleus (MN) and sister

chromatid exchanges (SCE) in peripheral blood

lymphocytes are well-established cytogenetic techni-

ques that have been used extensively for assessing

DNA damage at the chromosomal level in

human biomonitoring (Carrano and Natarajan,

1988; Fenech, 1993; Lando et al., 1998). These

cytogenetic biomarkers constituted valuable tool

for studying the most important occupational and

environmental hazards to public health occurring in

the past few decades (Bonassi et al., 2005) and allow

a reasonable epidemiological evaluation of cancer

predicitivity (Tucker and Preston, 1996). In recent

years, single cell gel electrophoresis or comet

assay has been proven to be a very sensitive method

to investigate the level of DNA damage, and an useful

tool for the detection of genetic damage at the

individual cell level, and in human biomonitoring

(Kassie et al., 2000; Moller et al., 2000). This

assay has been widely used in order to detect strand

breaks, alkali–labile sites, DNA crosslinking and

incomplete excision repair sites (Fairbairn et al.,

1995; Collins et al., 1997).

Biomonitoring studies commonly describe varia-

tion in the level of genotoxicity biomarkers among

healthy individuals exposed to similar concen-

trations of contaminants. Polymorphic genes of low

penetrance but high allele frequency involved in

the metabolism of xenobiotics may modulate the

levels of biomarkers arising from environmental

and/or occupational exposure to genotoxicants

(Pavanello and Clonfero, 2000). Many studies have

shown an elevated cancer proneness for individuals

carrying the potential at-risk alleles of metabolic

genes, but a number of controversial results have

also been obtained (reviewed in Hirvonen, 1999).

Reasons for disagreeing data and recommen-

dations to design studies aimed to measure the effect

of polymorphic metabolic genes on cancer suscep-

tibility have been suggested (Garte, 2001; Imyanitov

et al., 2004). Knowledge of the real impact of

genetic polymorphisms as biomarkers of suscepti-

bility is of key significance in understanding the

processes of genetic damage involved in muta-

genesis and carcinogenesis (Sram and Binkova,

2000) and could help to minimize risks for

susceptible subjects.

The objective of this study was to determine the

genotoxic risk in workers engaged in the production

of rubber tyres from a Portuguese factory. Urinary

thioethers were measured as a general biomarker of

exposure to electrophilic compounds. Effect

biomarkers applied were MN test, SCE and comet

assay. In addition several genetic polymorphisms in

xenobiotic metabolizing enzymes from phase I

(CYP2E1 Dra I, and EPHX1 codons 113 and 139)

and phase II (GSTP1 codon 105, and GSTM1 and

GSTT1 deletion polymorphisms) were analysed as

susceptibility biomarkers.

METHODS

Studied population

The exposed group consisted of 32 Caucasian male

individuals engaged in the production of rubber tyres

in a Portuguese factory located in Oporto. Also,

32 Caucasian males were included in the control

population, matched by age and smoking habits.

A complete questionnaire on lifestyle, consumption

habits such as smoking, alcohol intake, medication,

recent viral infections, vaccinations, diagnostic tests

or previous occupational exposures to chemicals was

filled out by all donors. Written informed consent was

obtained from each subject. The study was approved

by the National Institute of Health Research Ethics

Committee.

Sample collection

Urine samples were collected post-shift. Blood

samples were collected by venipuncture in hepari-

nized sterile tubes for genotoxicity tests, and in sterile

tubes containing EDTA for DNA extraction and

genotyping. Tubes were coded and sent immediately

to the laboratory, where they were processed.

Analysis of thioethers in urine samples

Urinary thioethers were determined in urine col-

lected after the workshift and stored at �20�C until

required (not longer than 15 days). Thioethers were

quantified by the method described by Vainio et al.

(1978).

Cytokinesis-blocked MN assay

Aliquots of 0.5 ml of heparinized whole blood were

used to establish duplicate lymphocyte cultures for

cytokinesis-blocked MN test, as described in Laffon

et al. (2005). A total of 1000 binucleated lymphocytes

with well-preserved cytoplasm were scored ‘blindly’

for each subject (500 from each duplicate culture)

by the same reader to determine the number of

MN. Criteria from Kirsch-Volders et al. (2000) for

584 B. Laffon et al.

identifying binucleated cytokinesis-blocked cells and

MN were followed.

Sister chromatid exchanges

Lymphocyte cultures for SCE were established

in duplicate as described previously (Teixeira et al.,

2004) from 0.5 ml of heparinized whole blood. Dif-

ferential chromatid staining was performed with

the fluorescence-plus-Giemsa procedure (Perry and

Wolff, 1974). A single observer scored 50 second

division metaphases for each donor (25 from each

duplicate culture) on coded slides to determine the

number of SCE/cell.

Comet assay

BD Vacutainer� CPT� Cell Preparation Tubes

with sodium heparin (Becton Dickinson) were

used for the isolation of mononuclear leukocytes,

following manufacturer’s instructions. Cells were

suspended in freezing medium (50% foetal calf

serum, 40% RPMI 1640 and 10% DMSO) to

obtain 107 cells ml�1, and frozen at �80�C in a

Nalgene� Cryo 1�C Freezing Container (Nalgene

Nunc International). At the time of analysis

(<2 weeks), cells were quickly thawed at 37�C and

trypan blue exclusion technique was used to check

viability, being >85% in all cases.

The alkaline version of the comet assay, basically

as described by Singh et al. (1988), was applied

with minor modifications (Laffon et al., 2002).

Two slides were prepared for each donor and a

‘blind’ scorer examined 50 randomly selected cells

from each slide (100 cells/donor) using a magnifi-

cation of ·400. QWIN Comet software (Leica

Imaging Systems, Cambridge, UK) was used for

image capture and analysis. Comet tail length

(TL), measured from the estimated centre of the

cell, was evaluated as DNA damage parameter.

Genotyping

Genomic DNA was isolated by means of

Puregene� DNA isolation kit (Gentra Systems,

Minneapolis, MI, USA). All genotype analyses

were performed at least in duplicate to confirm the

study results.

CYP2E1 Dra I site polymorphism (T7632A, intron

6) was analysed by means a polymerase chain

reaction–restriction fragment length polymorphism

(PCR–RFLP) assay following the method of Lin

et al. (1998).

For EPHX1, two PCR–RFLP assays were used to

detect the T to C mutation in exon 3 (Tyr113His) and

the A to G transition in exon 4 (His139Arg). Codon

113 polymorphism was analysed as described in

Laffon et al. (2003). For determination of codon

139 polymorphism, a 540 bp fragment was amplified

using 0.75 U Taq polymerase, 0.2 mM of each primer

(50-CCA GCT GTC AGG GGG CAC C-30 and

50-TGG CGA GGA CGG GGC AGT-30), 0.2 mM

deoxynucleoside triphosphates, 1.5 mM MgCl2 and

30 ng genomic DNA in a total volume of 30 ml. PCR

conditions were 35 cycles of 30 s at 94�C, 30 s at 68�C

and 1 min at 72�C, preceded by an initial denaturation

of 90 s at 94�C and followed by a final extension of

10 min at 72�C. The PCR product was digested

with Rsa I, generating fragments of 499 and 41 bp

in the case of wild-type allele and 322, 177 and 41 bp

in the case of variant allele. Once the individuals were

genotyped for codons 113 and 139, they were clas-

sified according to the expected epoxide hydrolase

enzyme activity (Sarmanova et al., 2000).

A multiplex PCR method was used to detect the

presence or absence of the GSTM1 and GSTT1 genes,

using b-globin as internal control, as described in

Laffon et al. (2003). Genotypes were classified as

positive (at least one undeleted allele) or null (both

alleles deleted). GSTP1 exon 5 (Ile105Val) polymor-

phism analysis was performed using a PCR–RFLP

technique, following Saarikoski et al. (1998).

Statistical analysis

All statistical analyses were conducted using the

SPSS for Windows statistical package, version 11.5

(IL, USA). Distribution of every variable obtained in

this study did not depart significantly from normality

(Kolmogorov–Smirnov goodness of fit test), and

therefore parametric tests were considered adequate

for the statistical analysis of these data. Analysis

of variance (ANOVA), followed by Bonferroni’s

correction for multiple comparisons among groups

when the overall F-test was significant, was used

to assess the contribution of exposure, age, smoking

habits and genotypes to the variability of genotoxicity

variables studied. The associations between two

variables were analysed by Pearson’s correlation.

For all the polymorphisms considered (excepting

GSTM1 and GSTT1), the homozygous and hetero-

zygous carriers of the variant alleles were combined

in the statistical analyses, owing to the low number of

variant homozygotes.

RESULTS

The present article illustrates the results of a

genotoxicity study (MN test, SCE and comet

assay) on peripheral blood leukocytes obtained

from 32 workers of a Portuguese rubber tyre factory

and 32 control subjects. Characteristics of the study

populations are gathered in Table 1. Age was very

similar in both groups, as well as smoking habits, 39%

of the individuals being smokers. Exposure time

covered a wide range, with 19 individuals (59.4%)

exposed for <7 years and 13 subjects (40.6%) exposed

for >14 years. To assess the possible influence of

585Genotoxicity in the production of rubber tyres

metabolic polymorphisms on the genetic biomarkers

investigated, the study subjects were genotyped for

the following polymorphisms: CYP2E1 Dra I,

EPHX1 codons 113 and 139, GSTM1 and GSTT1

deletion polymorphisms and GSTP1 codon 105.

The distribution of CYP2E1, EPHX1 and GSTP1

analysed genotypes were in Hardy–Weinberg equi-

librium (c2-test). The frequencies of variant alleles or

genotypes (in case of GSTM1 and GSTT1) were the

expected for the Caucasian population and similar to

those reported previously (Pemble et al., 1994;

Sarmanova et al., 2000; Laffon et al., 2003;

Teixeira et al., 2004).

Comet assay results were evaluated by means of

the three most frequently used parameters to describe

DNA damage: TL, percentage of DNA in the comet

tail and tail moment (which results from multiplying

the other two variables). Statistical analysis obtained

the same results for all of them (in terms of statistical

significances), thus we only show TL data. Table 2

shows results obtained in the measurement of

internal exposure dose and in genotoxicity assays,

both in the control and exposed groups and stratified

by tobacco consumption. Exposed subjects had

levels of urinary thioethers significantly higher

than controls. When smokers were separated from

non-smokers, both groups presented elevated values

for urinary thioethers in the exposed subjects as

compared to controls, but statistical significance

was only maintained in non-smokers. No significant

increase was observed in MN or SCE frequencies in

the exposed population as compared with control

individuals, although MN frequencies were higher

in the exposed group. Nevertheless, a significant

decrease in TL was obtained for the exposed with

respect to control individuals. This decrease was

also significant when comparing DNA damage in

non-smoker exposed with non-smoker controls, but

not when the comparison was made between

smokers. Smoking did not significantly affect results

of exposure or effect biomarkers analysed.

The influence of duration of exposure on MN, SCE

and comet tests is reflected in Fig. 1. MN frequencies

increased with time of exposure, although signifi-

cance was not reached. SCE results were not affected

by the duration of the employment, and TL decreased

significantly in both groups as compared to controls,

but they did not differ from each other. Correlation

between MN frequency and TL with exposure time

was found to be significant at the level of 0.05 (r =0.274 and r = �0.267, respectively).

Table 1. General characteristics of study subjects

Exposed Controls

Total no. of individuals 32 32

Age (years)a 37.41 – 12.71 38.16 – 12.62

Range 20–57 20–58

Smoking habits

Non-smokers 23 23

Smokers 9 9

No. of cigarettes/day 15.55 – 9.88 20.00 – 6.61

Range 8–40 10–35

Time of exposure (years)a 12.44 – 11.17

Range 1–34

CYP2E1 intron 6 genotype

*1A/*1A 28 21

*1A/*6 4 11

EPHX1 codon 113

Tyr/Tyr 24 20

Tyr/His 7 8

His/His 1 4

EPHX1 codon 139

His/His 19 20

His/Arg 12 11

Arg/Arg 1 1

GSTM1 genotype

Positive 19 18

Null 13 14

GSTT1 genotype

Positive 20 28

Null 12 4

GSTP1 codon 105 genotype

Ile/Ile 17 14

Ile/Val 12 13

Val/Val 3 5

aMean – SD.

Table 2. Exposure and effect biomarkers in the populations examined (mean – SE)

Parameter Controls Exposed

Total(N = 32)

Non-smokers(N = 23)

Smokers(N = 9)

Total(N = 32)

Non-smokers(N = 23)

Smokers(N = 9)

Urinarythioethers (mM)

0.24 – 0.02 0.23 – 0.03 0.29 – 0.03 0.41 – 0.05a 0.44 – 0.06b 0.34 – 0.05

MN frequency 1.84 – 0.29 1.91 – 0.38 1.67 – 0.37 2.34 – 0.33 2.30 – 0.39 2.44 – 0.67

SCE frequency 4.38 – 0.17 4.26 – 0.21 4.72 – 0.32 4.35 – 0.20 4.35 – 0.25 4.34 – 0.30

TL (mm) 48.25 – 0.71 48.65 – 0.74 47.22 – 1.73 44.72 – 0.66a 44.87 – 0.85b 44.34 – 0.91

aP < 0.01, significant difference with respect to controls.bP < 0.01, significant difference with respect to non-smoker controls.

586 B. Laffon et al.

Figure 2 shows the effect of age on genotoxicity

assays performed. Higher MN and SCE frequencies

were observed in the older group, although significant

differences were only obtained between the two esta-

blished groups (<35 and >35 years) among controls

in MN frequencies and among exposed subjects in

SCE frequencies. Significant associations between

age and cytogenetic parameters were found (r =0.387, P < 0.01 for MN and r = 0.225, P < 0.05

for SCE). TL values were not affected by this factor.

Possible modulation of results obtained in the effect

biomarkers by metabolic polymorphisms is reflected

in Table 3 for CYP2E1 and GST genes and in

Fig. 3 for EPHX1 gene (expressed as expected acti-

vity of epoxide hydrolase). Significant increase in

MN frequency was observed for GSTT1 null exposed

individuals with respect to positive ones. No effect of

CYP2E1, GSTM1 or GSTP1 variants was obtained in

SCE or MN test. Nevertheless, when individuals were

stratified by their GSTT1 genotype, among the

exposed population those GSTT1 positive carriers

of GSTP1 105Val variant allele showed significantly

higher MN frequencies than GSTP1 105Ile homo-

zygotes (2.10 – 0.43 versus 1.00 – 0.30, P < 0.05,

N = 10 for both groups). Combining these two poly-

morphisms (Fig. 4) exposed subjects GSTT1 positive

and GSTP1 105Ile homozygous carriers have shown

significantly lower MN frequencies than both GSTT1

null and GSTP1 105Ile homozygotes or carriers of

105Val allele.

Fig. 1. Effect of exposure time on (a) MN frequency, (b) SCE frequency and (c) TL. The number of individuals included in eachgroup is indicated inside each bar. **P < 0.01, *P < 0.05, significant difference with respect to the control group.

Fig. 2. Effect of age on (a) MN frequency, (b) SCE frequency and (c) TL. The number of individuals included in each group isindicated inside each bar. **P<0.01, *P<0.05, significant difference with respect to the minor age group.

Table 3. Effect of genetic polymorphisms on genotoxicity tests

Polymorphism MN frequency SCE frequency TL (mm)

N Controls N Exposed Controls Exposed Controls Exposed

CYP2E1

*1A/*1A 21 1.67 – 0.39 28 2.36 – 0.34 4.33 – 0.21 4.46 – 0.21 48.04 – 0.83 44.91 – 0.73

*1A/*6 11 2.18 – 0.42 4 2.25 – 1.44 4.50 – 0.34 3.62 – 0.30 48.66 – 1.39 43.42 – 1.08

GSTM1

positive 18 1.78 – 0.33 19 2.58 – 0.45 4.49 – 0.22 4.19 – 0.22 48.41 – 0.98 45.30 – 1.05

null 14 1.93 – 0.53 13 2.00 – 0.49 4.25 – 0.28 4.58 – 0.35 48.05 – 1.07 43.87 – 0.45

GSTT1

positive 28 1.86 – 0.32 20 1.55 – 0.29 4.51 – 0.18 4.44 – 0.27 48.61 – 0.78 45.27 – 0.89

null 4 1.75 – 0.63 12 3.67 – 0.59a 3.53 – 0.28 4.20 – 0.27 45.74 – 1.05 43.80 – 0.93

GSTP1

Ile/Ile 14 1.71 – 0.41 17 2.00 – 0.51 4.21 – 0.27 4.39 – 0.26 48.92 – 1.04 45.02 – 0.84

Ile/Val+Val/Val 18 1.94 – 0.42 15 2.73 – 0.42 4.53 – 0.23 4.30 – 0.30 47.74 – 0.99 44.38 – 1.05

Number of subjects included in each group in SCE and comet assay is the same as in MN test.aP<0.01, significant difference with respect to the positive genotype.

587Genotoxicity in the production of rubber tyres

As for EPHX1, higher MN frequency among

controls and SCE frequency among exposed were

detected for the epoxide hydrolase expected low

activity donors with respect to medium and high

activity individuals, reaching statistical significance

only in the later case. TL values appeared not to be

affected by any of the genetic polymorphisms

analysed.

DISCUSSION

The present report aims to evaluate the genotoxic

effects associated with the exposure to xenobiotic

substances in the production of rubber tyres, and to

determine the relationships between transient mark-

ers of genotoxic/carcinogenic effect and several gen-

etic polymorphisms in genes coding for xenobiotic

metabolizing enzymes.

Owing to the great variety of chemical compounds

present in the ambient air of rubber tyre factories,

evaluation of the internal exposure dose was per-

formed by analyzing urinary thioethers, since this

measure constitutes a general biomarker of exposure

to electrophilic compounds. Significantly increased

levels of this biomarker were observed in the exposed

population as compared to controls, but when the

population was divided by tobacco consumption, sig-

nificance for the difference between exposed and

controls was only maintained among non-smokers.

This was caused by the elevated excretion of

thioethers compounds by smokers, as has been

described previously (Sinues et al., 1990), that

decreased the importance of urinary thioethers com-

ing from the exposure.

A range of potentially carcinogenic compounds

occur as workplace contaminants during rubber

manufacturing, such as PAH, N-nitrosamines, 1,3-

butadiene, styrene and various alkenes. The geno-

toxic effects that these agents may simultaneously

induce are largely unknown, but synergistic interac-

tions can be expected. In this study a non-significant

increase in MN frequency has been observed in

exposed individuals, but no effect of exposure was

detected in SCE. Our results agree with those from

Moretti et al. (1996), who reported elevated MN freq-

uencies but not SCE frequencies in a rubber-exposed

population, although thioethers excretion described

was clearly higher than in our study. Nevertheless,

other authors found increases in chromosome aber-

rations and SCE in workers from the rubber industry

(Sorsa et al., 1983; Sasiadek et al., 1992), but unfor-

tunately no indication on the quantitative level of

exposure was given in these studies.

Comet assay data have shown decreased TL values

in the exposed population with respect to the control

group that might indicate the induction of cross-

links by the substances present in the workplace

environment. In contrast to other DNA alterations,

crosslinks may stabilize chromosomal DNA and

inhibit DNA migration in the comet assay (Pfuhler

and Wolf, 1996; Merk and Speit, 1999). As some

of the active metabolites from rubber production

compounds can react with nucleophilic regions of

both protein and DNA, similar to those from alkenes

and styrene (Philips and Farmer, 1994), the occur-

rence of DNA–protein crosslinks would be expected

(Zhu et al., 2000). In fact, it has been reported the

induction of DNA–protein crosslinks by several PAH

(Perin-Rousel et al., 1984; Park et al., 2002). On the

other hand, the comet assay also detects DNA breaks

coming from incomplete excision repair processes, so

data obtained may reflect a lower DNA repair capac-

ity of the exposed subjects. Moretti et al. (1996)

Fig. 3. Effect of epoxide hydrolase expected activity on (a) MN frequency, (b) SCE frequency and (c) TL. The number ofindividuals included in each group is indicated inside each bar. *P<0.05, significant difference with respect to low activity

genotype.

Fig. 4. Effect of combined GSTT1 and GSTP1 codon 105polymorphisms on MN frequency in the exposed population.The number of individuals included in each group is indi-cated inside each bar. **P<0.01, *P<0.05, significant dif-ference with respect to GSTT1 positive and GSTP1 Ala/Ala

individuals.

588 B. Laffon et al.

and Vodicka et al. (2004) did not find significant

differences in comet results in rubber-exposed

individuals, and Zhu et al. (2000) described increase

in DNA damage in rubber workers but they used

proteinase K in the comet assay to break down pos-

sible DNA–protein crosslinks, which would retard the

migration of DNA fragments. However, Somorovska

et al. (1999) reported higher level of DNA breaks,

evaluated by the comet assay, and also of chromo-

some aberration and MN frequencies in workers from

a rubber factory. Differences found between studies

could be attributed to the different spectra of chemical

substances used (Vermeulen et al., 2000).

Tobacco consumption did not affect results

obtained in cytogenetic tests, except that a non-

significant increase in SCE frequency was observed

in smokers in comparison to non-smokers among

control group. Smoking is known to increase SCE

(Lazutka et al., 1994; Barale et al., 1998), and there-

fore smoking effect could be masked by exposure in

rubber workers, although the low number of smoker

subjects included in this study may also affect our

results. On the other hand, tobacco smoking did not

appear to have a clear effect on the frequency of MN

(Thierens et al., 1996; Barale et al., 1998). According

to this, results from the HUman MicroNucleus Project

have shown a significant increase of MN frequency

only in heavy smokers (30 cigarettes or more per day)

(Bonassi et al., 2003). Smoking habit did not either

significantly increase the levels of DNA damage in

control or exposed subjects, in agreement with find-

ings by Hellman et al. (1997, 1999) and Wojewodzka

et al. (1999).

The present investigation shows increasing MN

frequencies with duration of exposure, confirming

the fact that this biomarker reflects accumulated

chromosomal damage (Kirsch-Volders and Fenech,

2001). However, maybe owing to the small sample

size the difference does not reach statistical signi-

ficance. In this study, significant associations were

found between age and cytogenetic parameters.

Age is known to affect the frequency of MN and

SCE (Thierens et al., 1996; Bolognesi et al., 1997;

Barale et al., 1998; Bonassi et al., 2003), its effect

being particularly clear on MN, apparently mostly

because of the age-dependent micronucleation of

sex chromosomes (Catalan et al., 1998).

Cytochrome P450 (CYP) monooxygenases are

enzymes which catalyse the insertion of one atom

of molecular oxygen into a substrate in a typical reac-

tion of activation (phase I). Of particular interest for

the industrial field is the isozyme CYP2E1, since

substrate spectrum of this enzyme includes many

compounds of classical relevance for industrial tox-

icology (Thier et al., 2003). Genetic polymorphisms

in CYP2E1 appear to act on the transcription levels of

the enzyme, increasing its activity (Pavanello and

Clonfero, 2000). We did not observe any association

between biomarkers measured in the present study

and Dra I polymorphism in CYP2E1.

Microsomal epoxide hydrolase, coded by the

EPHX1 gene, catalyses the addition of a molecule

of H2O to an epoxide to form a dihydrodiol, genera-

ting more soluble and less reactive metabolites that

can be readily conjugated and excreted. Replacement

of Tyr to His at codon 113 reduces enzyme activity,

while substitution of His to Arg at codon 139 of exon

4 is associated with increased activity (Hasset et al.,

1994). It has been suggested that the amino acid

substitutions may result in altered protein stability,

since they do not affect the specific activity of the

enzyme (Raaka et al., 1998). In this work, signifi-

cantly increased SCE frequencies have been obser-

ved in exposed workers with low expected epoxide

hydrolase activity, with respect to those individuals

with medium and high activity. The same effect,

although non-significant, has been also obtained in

MN frequencies in the control population. In a pre-

vious report, individuals working in a tyre plant with

low epoxide hydrolase expected activity exhibited

higher CA frequencies than those with medium and

high activity (Vodicka et al. 2004). In addition,

elevated HPRT lymphocyte mutant frequencies

were observed in workers with low epoxide hydrolase

activity exposed to 1,3-butadiene (Abdel-Rahman

et al., 2003) or to low levels of PAH (Viezzer

et al., 1999). These results confirm the role of epoxide

hydrolase in the detoxification of xenobiotic sub-

stances present in the environment of rubber tyre

factories.

Glutathione S-transferases (GST) are a superfamily

of polymorphic enzymes involved in the conjugation

of reactive chemical intermediates, and play an

important role in the detoxification of endogenous

and exogenous compounds. The polymorphisms of

GSTM1 and GSTT1 owing to gene deletions result

in null alleles, and homozygous individuals for the

deletions lack enzyme activity. As in the present

study, no association between chromosomal aberra-

tions or DNA damage and GSTM1 deletion polymor-

phism was detected in rubber workers (Vodicka et al.,

2004). The lack of GSTM1 appears to be associated

with increased sensitivity to genotoxicity of tobacco

smoking (Norppa 2004). Nevertheless, no significant

increase in cytogenetic or DNA damage among

GSTM1 null smokers has been detected in this

study in comparison with GSTM1 positive smokers,

neither in controls nor in exposed (data not shown),

probably owing to the fact that the effect of smoking

itself on the mean number of SCE per cell is usually

small.

Deletion of GSTT1 gene has been found to yield an

increase in baseline SCE frequency (Norppa, 2004).

Also, increased MN frequency was consistently

shown for GSTT1 null individuals (Bonassi et al.,

2005). The reported effect of GSTT1 genotype on

589Genotoxicity in the production of rubber tyres

baseline SCE level was quite small, which may

explain why it has not been detected in this study,

in addition to the low number of individuals included

in the group of GSTT1 null controls. However, a clear

influence of this polymorphism has been obtained in

MN frequencies among rubber workers. Since this

effect has been obtained only in the exposed

group, it suggests the existence of an exposure–

genotype interaction (Norppa, 2003).

GSTP1 is the most abundant isoform in the lungs;

thus, it has particular importance in the detoxification

of inhaled toxicants (Saarikoski et al., 1998). Poly-

morphism in codon 105 of GSTP1 produces an

enzyme with different thermal stability and substrate

affinity (Sarmanova et al., 2000). In this study, effect

of this polymorphism has been detected in association

with GSTT1 genotype, since the lack of GSTT1 acti-

vity and the presence of GSTP1 105Val variant allele

determined increasing MN frequencies. This obser-

vation suggests an increased risk of genotoxic effects

in individuals with particular genotype combinations.

Data obtained in this study indicated that genotoxic

risk of occupational exposure associated with the

production of rubber tyres cannot be excluded. In

addition, a certain role of genetic polymorphisms

in EPHX1 and GSTT1 in the modulation of cyto-

genetic tests results has been suggested, together

with an interaction between GSTP1 and GSTT1 poly-

morphisms. Nevertheless, these results must be

cautiously interpreted, owing to the relatively low

number of exposed and control individuals included

in this study.

Acknowledgements—This work was partially supported by theComissao de Fomento a Investigacao em Cuidados de Saudeand by a grant from the Xunta de Galicia (PGIDT04P-XIB10602PR). B.L. was supported by a postdoctoral fellowshipfrom the Xunta de Galicia. B.P.-C. was supported by a predoc-toral fellowship from the University of A Coruna.

REFERENCES

Abdel-Rahman SZ, El-Zein RA, Ammenhaeuser MM, Yang Zet al. (2003) Variability in human sensitivity to 1,3-butadiene: influence of the allelic variants of the microsomalepoxide hydrolase gene. Environ Mol Mutagen; 41: 140–6.

Barale R, Chelotti L, Davini T et al. (1998) Sister chromatidexchanges and micronucleus frequency in human lympho-cytes of 1,650 subjects in an Italian population: II. Contribu-tion of sex, age, and lifestyle. Environ Mol Mutagen; 31:228–42.

Baranski B, Indulski J, Janik-Spiechowicz E et al. (1989)Mutagenicity of airborne particulates in the rubber industry.J Appl Toxicol; 9: 389–93.

Bolognesi C, Abbondandolo A, Barale R et al. (1997) Age-related increase of baseline frequencies of sister chromatidexchanges, chromosome aberrations, and micronuclei inhuman lymphocytes. Cancer Epidemiol Biomarkers Prev;6: 249–56.

Bonassi S, Neri M, Lando C et al. (2003) Effect of smokinghabit on the frequency of micronuclei in human lymphocytes:

results from the HUman MicroNucleus project. Mutat Res;543: 155–66.

Bonassi S, Ugolini D, Kisch-Volders M et al. (2005) Humanpopulation studies with cytogenetic biomarkers: review ofthe literature and future prospectives. Environ Mol Mutagen;45: 258–70.

Carrano AV, Natarajan AT. (1988) Considerations for popula-tion monitoring using cytogenetic techniques. Mutat Res;204: 379–406.

Catalan J, Autio K, Kuosma E et al. (1998) Age-dependentinclusion of sex chromosomes in lymphocyte micronucleiof man. Am J Hum Genet; 63: 1464–72.

Collins A, Dusinska M, Franklin M et al. (1997) Comet assay inhuman biomonitoring studies: reliability, validation andapplications. Environ Mol Mutagen; 30: 139–46.

Delzell E, Macaluso M, Sathiakumar N et al. (2001) Leukemiaand exposure to 1,3–butadiene, styrene and dimethyldithio-carbamate among workers in the synthetic rubber industry.Chemi-Biol Interact; 135–136: 515–534.

Dost AA, Redman D, Cox G. (2000) Exposure to rubber fumeand rubber process dust in the general rubber goods, tyremanufacturing and retread industries. Ann Occup Hyg; 44:329–342.

Fairbairn DW, Olive PL, O’Neill KL. (1995) The comet assay: acomprehensive review. Mutat Res; 339: 37–59.

Fenech M. (1993) The cytokinesis-block micronucleus tech-nique: a detailed description of the method and its applicationto genotoxicity studies in human populations. Mutat Res;285: 35–44.

Fishbein L. (1991) Chemicals used in the rubber industry. SciTotal Environ; 101: 33–43.

Fracasso ME, Franceschetti P, Mossini E et al. (1999) Exposureto mutagenic airborne particulate in a rubber manufacturingplant. Mutat Res; 441: 43–51.

Garte S. (2001) Metabolic susceptibility genes as cancer riskfactors: time for a reassessment? Cancer Epidemiol Biomark-ers Prev; 10: 1233–7.

Hellman B, Friis L, Vaghef H et al. (1999) Alkaline single cellgel electrophoresis and human biomonitoring for genotoxi-city: a study on subjects with residential exposure to radon.Mutat Res; 442: 121–32.

Hellman B, Vaghef H, Friis L et al. (1997) Alkaline single cellgel electrophoresis of DNA fragments in biomonitoring forgenotoxicity: an introductory study on healthy human vol-unteers. Int Arch Occup Environ Health; 69: 185–92.

Hirvonen A. (1999) Polymorphisms of xenobiotic-metabolizingenzymes and susceptibility to cancer. Environ HealthPerspect; 107: 37–47.

IARC. (1982) IARC monographs on the evaluation of carcino-genic risks to humans, vol. 28. The rubber industry. Lyon:International Agency for Research on Cancer. ISBN 92 8321528 1.

IARC. (1987) IARC monographs on the evaluation of carci-nogenic risks to humans, Suppl. 7. Overall evaluations ofcarcinogenicity: an updating of IARC monographs 1–42.Lyon: International Agency for Research on Cancer.pp. 332–333. ISBN 92 832 1411 0.

Imyanitov EN, Togo AV, Hanson KP. (2004) Searchingfor cancer-associated gene polymorphisms: promises andobstacles. Cancer Lett; 204: 3–14.

Kassie F, Parzefall W, Knasmuller S. (2000) Single cell gelelectrophoresis assay: a new technique for human bio-monitoring studies. Mutat Res; 463: 13–31.

Kirsch-Volders M, Fenech M. (2001) Inclusion of micronucleiin non-divided mononuclear lymphocytes and necrosis/apoptosis may provide a more comprehensive cytokinesisblock micronucleus assay for biomonitoring purposes.Mutagenesis; 16: 51–58.

Kirsch-Volders M, Sofuni T, Aardema M et al. (2000) Reportfrom the in vitro micronucleus assay working group. EnvironMol Mutagen; 35: 167–72.

590 B. Laffon et al.

Kogevinas M, Sala M, Boffetta P et al. (1998) Cancer risk inthe rubber industry: a review of the recent epidemiologicalevidence. Occup Environ Med; 55: 1–12.

Kromhout H, Swuste P, Boleij JS. (1994) Empirical modellingof chemical exposure in the rubber-manufacturing industry.Ann Occup Hyg; 38: 3–22.

Laffon B, Pasaro E, Mendez J. (2002) Evaluation of genotoxiceffects in a group of workers exposed to low levels of styrene.Toxicology; 171: 175–86.

Laffon B, Perez-Cadahıa B, Pasaro E et al. (2003) Effect ofepoxide hydrolase and glutathione S-transferase genotypeson the induction of micronuclei and DNA damage by styrene-7,8–oxide in vitro. Mutat Res; 536: 49–59.

Laffon B, Perez-Cadahıa B, Pasaro E et al. (2003)Individual sensitivity to DNA damage induced by styrenein vitro: influence of cytochrome P450, epoxide hydrolaseand glutathione S-transferase genotypes. Toxicology; 186:131–41.

Laffon B, Teixeira JP, Silva S et al. (2005) Genotoxic effects ina population of nurses handling antineoplastic drugs, andrelationship with genetic polymorphisms in DNA repairenzymes. Am J Ind Med; 48: 128–36.

Lando C, Hagmar L, Bonassi S. (1998) Biomarkers of cytoge-netic damage in human and risk of cancer. The EuropeanStudy Group on Cytogenetic Biomarkers and Health. MedLav; 89: 124–31.

Lazutka JR, Dedonyte V, Krapavickaite D. (1994) Sister-chromatid exchanges and their distribution in humanlymphocytes in relation to age, sex and smoking. MutatRes; 306: 173–80.

Lin D-X, Tang Y-M, Peng Q et al. (1998) Susceptibility toesophageal cancer and genetic polymorphisms in glutathioneS-transferase T1, P1, and M1 and cytochrome P450 2E1.Cancer Epidemiol Biomarkers Prev; 7: 1013–8.

Merk O, Speit G. (1999) Detection of crosslinks with the cometassay in relationship to genotoxicity and cytotoxicity.Environ Mol Mutagen; 33: 167–172.

Moller P, Knudsen LE, Loft S, Wallin H. (2000) The cometassay as a rapid test in biomonitoring occupational exposureto DNA-damaging agents and effect of confounding factors.Cancer Epidemiol. Biomarkers Prev; 9: 1005–15.

Monarca S, Feretti D, Zanardini A et al. (2001) Monitoringairborne genotoxicants in the rubber industry using geno-toxicity tests and chemical analyses. Mutat Res 490: 159–69.

Moretti M, Villarini M, Scassellati-Sforzolini G et al. (1996)Biological monitoring of genotoxic hazard in workers of therubber industry. Environ Health Perspect; 104: 543–5.

Norppa H. (2003) Genetic susceptibility, biomarker responses,and cancer. Mutat Res; 544: 339–48.

Norppa H. (2004) Cytogenetic biomarkers. In Buffler P,Rice J, Baan R, Bird M, Boffetta P, editors. Mechanismsof carcinogenesis: contributions of molecular epidemiology.IARC Scientific Publications No. 157. Lyon: InternationalAgency for Research on Cancer. pp. 179–205. ISBN 92 8322157 5.

Oury B, Limasset JC, Protois JC. (1997) Assessment of expo-sure to carcinogenic N-nitrosamines in the rubber industry.Int Arch Occup Environ Health; 70: 261–71.

Park H-S, Ha E-H, Lee K-H, Hong Y-C. (2002)Benzo[a]pyrene-induced DNA–protein crosslinks in culturedhuman lymphocytes and the role of GSTM1 and GSTT1genotypes. J Korean Med Sci; 17: 316–321.

Pavanello S, Clonfero E. (2000) Biological indicators of geno-toxic risk and metabolic polymorphisms. Mutat Res; 463:285–308.

Pemble L, Schoreder KR, Spencer SR et al. (1994) Humanglutathione S-transferase Theta (GSTT1): cDNA cloningand the characterization of a genetic polymorphism. BiochemJ; 300: 271–6.

Perin-Rousel O, Ekert B, Barat N et al. (1984) DNA–proteincrosslinks induced by exposure of cultured mouse fibroblasts

to dibenzo[a,e]fluorathene and its bay- and pseudo-bayregion dihydrodiols. Carcinogenesis; 5: 379–83.

Pfuhler S, Wolf HU. (1996) Detection of DNA-crosslinkingagents with the alkaline comet assay. Environ Mol Mutagen;27: 196–201.

Philips DH, Farmer PB. (1994) Evidence for DNA and proteinbinding by styrene and styrene oxide. Crit Rev Toxicol; 24:S11–33.

Raaka S, Hassett C, Omiecinski CJ. (1998) Human microsomalepoxide hydrolase: 50-flanking region genetic polymor-phisms. Carcinogenesis; 19: 387–93.

Saarikoski ST, Voho A, Reinkainen M et al. (1998)Combined effect of polymorphic GST genes onindividual susceptibility to lung cancer. Int J Cancer; 77:516–21.

Sarmanova J, Tynkova L, Susova S, Gut I, Soucek P. (2000)Genetic polymorphisms of biotransformation enzymes: allelefrequencies in the population of the Czech Republic.Pharmacogenetics; 10: 781–8.

Sasiadek M. (1992) Cytogenetic studies of workers from therubber industry. Mutat Res; 279: 195–8.

Singh NP, McCoy MT, Tice RR et al. (1988) Asimple technique for quantitation of low levels of DNAdamage in individual cells. Exp Cell Res; 175:184–91.

Sinues B, Izquierdo M, Perez-Viguera J. (1990) Chromosomeaberrations and urinary thioethers in smokers. Mutat Res;240: 289–93.

Somorovska M, Szabova E, Vodicka P et al. (1999)Biomonitoring of genotoxic risk in workers in a rubberfactory: comparison of the Comet assay with cyto-genetic methods and immunology. Mutat Res; 445:181–92.

Sorsa M, Maki-Paakkanen J, Vainio H. (1983) A chromosomestudy among worker groups in the rubber industry. Scand JWork Environ Health; 9: 43–47.

Sram RJ, Binkova B. (2000) Molecular epidemiology studies onoccupational and environmental exposure to mutagens andcarcinogens, 1997–1999. Environ Health Perspect; 108:57–70.

Sturaro A, Parvoli G, Doretti L, Gori G, Bartolucci GB. (1993)GC-MS investigation of polycyclic aromatic compounds inthe manufacture of rubber tubes. Ann Occup Hyg; 37:327–34.

Teixeira JP, Gaspar J, Silva S et al. (2004) Occupational expo-sure to styrene: modulation of cytogenetic damage and levelsof urinary metabolites of styrene by polymorphisms in genesCYP2E1, EPHX1, GSTM1, GSTT1 and GSTP1. Toxicology;195: 231–42.

Thier R, Bruning T, Roos PH et al. (2003) Markers of geneticsusceptibility in human environmental hygiene and toxico-logy: the roles of selected CYP, NAT and GST genes. Int JHyg Environ Health; 206: 149–71.

Thierens H, Vral A, De Ridder L. (1996) A cytogenetic study ofradiobiological workers: effect of age, smoking and radiationburden on the micronucleus frequency. Mutat Res; 360:75–82.

Tucker JD, Moore II DH. (1996) The importance of age andsmoking in evaluating adverse cytogenetic effects of expo-sure to environmental agents. Environ Health Perspect; 104:489–92.

Vainio H, Savolainen H, Kilpikari I. (1978) Urinarythioether of employees of a chemical plant. Br J Ind Med;35: 232–4.

Vermeulen R, Bos RP, de Hartog J et al. (2000) Mutagenicprofile of rubber dust and fume exposure in two rubber tirecompanies. Mutat Res; 468: 165–71.

Veys CA. (2004) Bladder tumours in rubber workers: a factorystudy 1946–1995. Occup Med; 54: 322–9.

Viezzer C, Norppa H, Clonfero E et al. (1999)Influence of GSTM1, GSTT1, GSTP1, and EPHX gene

591Genotoxicity in the production of rubber tyres

polymorphisms on DNA adduct level and HPRTmutant frequency in coke-oven workers. Mutat Res; 431:259–69.

Vodicka P, Kumar R, Stetina R et al. (2004) Markers of indi-vidual susceptibility and DNA repair rate in workers exposedto xenobiotics in a tire plant. Environ Mol Mutagen; 44:283–92.

Wojewodzka M, Kruszewski M, Iwanenko T et al. (1999) Lackof adverse effect of smoking habit on DNA strand breakageand base damage, as revealed by the alkaline comet assay.Mutat Res; 440: 19–25.

Zhu CQ, Lam TH, Jiang CQ et al. (2000) Increased lymphocyteDNA strand breaks in rubber workers. Mutat Res; 470:201–9.

592 B. Laffon et al.