Micronuclei in Humans Induced by Exposure to Low Level of Ionizing Radiation Influence of Polymorphisms in DNA Repair Genes

Mutation Research 570 (2005) 105–117

Micronuclei in humans induced by exposure to low levelof ionizing radiation: influence of polymorphisms

in DNA repair genes

Sabrina Angelinia,b,∗, Rajiv Kumarb,c, Fabio Carbonea, Francesca Maffeia,Giorgio Cantelli Fortia, Francesco Saverio Violanteb,d, Vittorio Lodib,d,

Stefania Curtib,d, Kari Hemminkib,c, Patrizia Hreliaa

a Department of Pharmacology, University of Bologna, Via Irnerio 48, Bologna 40126, Italyb Department of Biosciences, Karolinska Institute, Novum, Huddinge 141 57, Sweden

c Division of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germanyd Occupational Medicine Unit, S. Orsola-Malpighi Hospital, Via Pelagi 9, Bologna 40100, Italy

Received 4 June 2004; received in revised form 27 October 2004; accepted 29 October 2004Available online 10 December 2004

Abstract

Understanding the risks deriving from protracted exposure to low doses of ionizing radiation has remarkable societal importancei tudied thef diationa orphismsi ffects ofc tly higheri nta ntlyh either ine combinedp n though

ir; SNPs,ition; FFQ,; PCR-., standard

0d

n view of the large number of work settings in which sources of IR are encountered. To address this question, we srequency of micronuclei (MN), which is an indicator of DNA damage, in a population exposed to low levels of ionizing rand in matched controls. In both exposed population and controls, the possible influence of single nucleotide polym

n XRCC1, XRCC3andXPD genes on the frequency of micronuclei was also evaluated. We also considered the eonfounding factors, like smoking status, age and gender. The results indicated that MN frequency was significann the exposed workers than in the controls [8.62± 2.80 versus 6.86± 2.65;P= 0.019]. Radiological workers with varialleles forXRCC1orXRCC3polymorphisms or wild-type alleles forXPDexon 23 or 10 polymorphisms showed a significaigher MN frequency than controls with the same genotypes. Smoking status did not affect micronuclei frequencyxposed workers or controls, while age was associated with increased MN frequency in the exposed only. In theopulation, gender but not age exerted an influence on the yield of MN, being higher in females than in males. Eve

0

Abbreviations:CA, chromosomal aberrations; MN, micronuclei; BER, base excision repair; HRR, homologous recombination repasingle nucleotide polymorphisms; NER, nucleotide excision repair; EPIC, European prospective investigation into cancer and nutrfood frequency questionnaire;Hwb, dose equivalent of ionizing radiation to the whole body; BN, binucleated; NDI, nuclear division indexRFLP, polymerase chain reaction-restriction fragment length polymorphism; SSCP, single strand conformation polymorphism; S.Ddeviation; FISH, fluorescence in situ hybridization

∗ Corresponding author. Tel.: +39 051 2091782; fax: +39 051 248862.E-mail address:[email protected] (S. Angelini).

027-5107/$ – see front matter © 2004 Elsevier B.V. All rights reserved.oi:10.1016/j.mrfmmm.2004.10.007

106 S. Angelini et al. / Mutation Research 570 (2005) 105–117

there is a limitation in this study due to the small number of subjects, these results suggest that even exposures to low levelof ionizing radiation could have genotoxic effects and thatXRCC3, XRCC1andXPD polymorphisms might contribute to theincreased genetic damage in susceptible individuals occupationally exposed to chronic low levels of ionizing radiation. For aclear conclusion on the induction of DNA damage caused by protracted exposure to low doses of ionizing radiation and thepossible influence of genetic polymorphism in DNA repair genes larger studies are needed.© 2004 Elsevier B.V. All rights reserved.

Keywords: Ionizing radiation; Occupational exposure; Micronuclei; Genetic polymorphisms; DNA repair genes

1. Introduction

Over the years ionizing radiation has become auniversal diagnostic and therapeutic tool, making thelargest man-made contribution to the population dose[1]. Thus, medical personnel represent the group mostconsistently exposed to low dose of ionizing radiation.The high doses of ionizing radiation are clearly knownto produce deleterious consequences in humans, in-cluding, but not exclusively, cancer induction. How-ever, the effect of such radiations at lower doses, as inoccupational work settings, is less clear. The scientificcommunity is primarily concerned with the biologicalconsequences from protracted exposure at low doses[2]. In particular, in spite of the deepening scientificknowledge about radiation adverse health effects, thereis a particular need to introduce new epidemiologicalapproach in radioprotection programs for the health

posure to both clastogenic and aneugenic agents, suchas ionizing radiation[8]. The method has already beensuccessfully employed to assess cytogenetic damagein groups occupationally exposed to ionizing radiation[9,10] or population living in areas with a high back-ground of radioactivity[11].

Ionizing radiation interacts with mammalian cellsby inducing a wide range of detrimental effects, withthe most important damage occurring to the cellu-lar DNA. Ionizing radiation damages cellular DNA inmany ways requiring the concerted action of a num-ber of DNA repair enzymes for the maintenance of itsstructural integrity[12]. Thus, DNA repair plays a vitalrole in faithful maintenance of genomic integrity, anddeficiencies in repair function, as recently highlightedin several reviews, are known to promote cancer devel-opment[13,14]. The deleterious clinical consequencesof inherited defects in DNA repair systems are apparent

mes,niz-n-oc-x-

eenus-antlar

surveillance of hospital workers chronically exposedto ionizing radiation[3–5].

In the last decade, several biomarkers, such as cy-togenetic analysis, have been developed to performbiomonitoring of populations occupationally exposedto ionizing radiation. Chromosomal aberrations (CA)have been widely accepted as reliable biomarkers forevaluating damage induced by ionizing radiation in hu-mans. Increased frequencies of CA are well known

from several human cancer predisposition syndrosome of which shows a hypersensitivity towards ioing radiation[15]. A significant variation in radiosesitivity between healthy individuals has also been dumented[16,17] and the genetic polymorphisms eisting in a number of DNA repair enzymes have bproposed as a source of this individual variability (sceptibility). These polymorphisms may be importin determining an individual’s ability to repair cellu

among radiological workers, as compared to controls.However, since the technique is time consuming andd rgeg )a asinB e-s edurt -m e notd nse-q x-

DNA after ionizing radiation exposure, and therefore,to modulate the toxicological outcome. Oxidative basedamage and strand breaks induced by ionizing radi-

pairR)

ncendely,earchible

emands skilled personnel, the biomonitoring of laroups of workers is difficult. Thus, micronuclei (MNnalysis in human lymphocytes using the cytochaltechnique[6] has found its place in performing larg

cale studies as a valuable easier and faster prochan the CA assay[7]. MN consists of acentric chroosome fragments or whole chromosomes that aristributed to the main nuclei during anaphase. Couently, MN formation is a reliable biomarker of e

e

ation are repaired mainly by the base excision re(BER) and homologous recombination repair (HRpathways[18,19]. XRCC1plays an important role iBER, andXRCC3functions in the HRR pathway. SinXRCC1andXRCC3mutant cells show increased amoderate sensitivity to ionizing radiation, respectivSNPs in these two genes have been attracting resinterest[20–22]. Several studies also suggest a poss

S. Angelini et al. / Mutation Research 570 (2005) 105–117 107

role for NER proteins in the repair of some types of ox-idative damage induced by ionizing radiation[23,24].In particular, since oxidative damage induced by ion-izing radiation is preferentially repaired on the tran-scribed strand, there is a plausible role for its repair fortheXPD protein, which participates in transcription-coupled repair[25,26].

A multi-biomarker approach can provide a valuabletool for human biological monitoring of occupationalexposure to ionizing radiation. In particular, the inte-grated application of biomarkers of effect and suscep-tibility can be a valid approach to ascertain adversehealth effects in the exposed, which is an important steptowards a timely establishment of effective programsfor cancer prevention. Establishing predictive markersof radiosensitivity could be used to identify individ-uals who are at increased risk of ionizing radiation-induced damage and prevent adverse outcomes in hu-mans routinely exposed to low level of ionizing radia-tion. Herein, we report a preliminary study about MNanalysis on peripheral blood lymphocytes in a group ofhospital workers occupationally exposed to low levelof ionizing radiation and a group of unexposed sub-jects as controls. We investigate whether MN forma-tion, indicator of DNA damage, is associated with oc-cupational exposure to low level of ionizing radiation.Moreover, we evaluate the possible influence on theMN frequency of genetic polymorphism in three DNArepair genes:XRCC1(exon 10, G to A; Arg399Gln),XRCC3(exon 7, C to T; Thr241Met), andXPD (exon1 lna avea e inl3t be-t oi

2

2

heh heS ar-t ject

was voluntary and could be cancelled by the subjectat any time during the study, according to the Helsinkideclaration and later amendments. Twenty-one werephysicians and technicians in the Unit of Radiol-ogy, Radiotherapy and Cardiology at the Sant’Orsola-Malpighi Hospital, occupationally exposed to ionizingradiation. None of them were professionally exposedto any carcinogenic agent other than ionizing radiation.The control group consisted of 21 individuals workingin the administrative staff of the same hospital whohad never been occupationally exposed to ionizing ra-diation or other known carcinogenic agents. Each par-ticipant was briefed about the study protocol, with spe-cific written information about cytogenetic and genet-ics tests, the aims of the study and signed an informedconsent. Moreover, the study was conducted in accor-dance to the Italian Law no. 675/96[30].

2.2. Demographic characteristic, life-style factorsand sample collection

In connection with a routine occupational health ex-amination, each subject was extensively interviewed byan occupational health physician, who filled in a struc-tured questionnaire to gather information about age,gender, smoking status, dietary habits, alcohol con-sumption, medical history, included previous exposureto diagnostic or therapeutic X-ray and medical use oftherapeutic drugs, and, for the radiological workers,duration of occupational exposure to ionizing radiation( rac-tA twog g thee werec eref dif-f ermso ion.

ag-n dia-t foret

siono pedaq mealc nts

0, G to A, Asp312Asn; exon 23, A to C, Lys751Gnd intron 10, IVS10 + 17C-ins). Several studies hlso reported that 312Asp and 751Lys alleles ar

inkage disequilibrium, that is, carriers with theXPD12Asp allele tend to have the 751Lys allele[27–29],

he possible effect of the recombinant haplotypeween codons 312 and 751 of theXPD gene is alsnvestigated in this study.

. Materials and methods

.1. Study population

Forty-two healthy individuals, enrolled with telp of the Occupational Medicine Unit of tant’Orsola-Malpighi Hospital in Bologna, Italy, p

icipated in the study. The participation of each sub

years of employment). Detailed demographic chaeristics of the two study groups are listed inTable 1.ge and gender distributions were similar in theroups. Regarding smoking status, seven amonxposed workers and seven among the controlsurrent smokers at sampling time (volunteers who wormer smokers were not included in the study). Noerence was observed between the two groups in tf years of smoking or of daily cigarette consumpt

None of the forty-two individuals had received diostic X-ray examinations or any therapeutic irra

ion or chemotherapeutic drugs in the 12 months behe study took place.

Participants were presented with a modified verf the food frequency questionnaire (FFQ) develond validated for the Italian EPIC cohorts[31]. Theuestionnaire was structured by courses within aharacteristic of Italian dietary habits, and participa


Table 1Demographic characteristics of the study population

Radiological workers Controls

Sample size 21 21

Age (years)Mean± S.D. 45.8± 8.6 45.29± 8.0Range 28–58 28–56

GenderFemales (%) 7 (33.3) 7 (33.3)Males (%) 14 (66.7) 14 (66.7)

Smoking statusNever smokers (%) 14 (66.7) 14 (66.7)Current smokers (%) 7 (33.3) 7 (33.3)

Smoking yearsMean± S.D. 27.14± 12.28 25.00± 5.51

Cigarette/dayMean± S.D. 13.86± 8.91 21.67± 12.34

Years of employmentMean± S.D. 18.2± 9.10 –Range 4–34 –

Equivalent dose (mSv)Mean± S.D. 40.61± 37.70 –Range 0.90–116.74 –

were asked how often, on average, they consume eachfood. Nutrient intakes were calculated by multiplyingthe consumption frequency of each food by the nu-trient content of the portion specified. Values for thenutrient amounts in food were obtained from exist-ing nutrient tables for Italy, derived from the NationalResearch Institute for Food and Nutrition[32]. Noneof the participants had a deficient diet, or peculiar di-etary habits (e.g. vegetarian and vegans) or consumedmulti-Vitamins supplements. Dietary habits were ho-mogeneous among the participants and the mean dailyintakes of folate and Vitamin B12 among radiologi-cal workers and controls were similar (300�g versus290�g and 5.4�g versus 6�g, respectively).

Fifty-seven percent of the exposed workers and 66%of controls consumed wine and/or beer, and they hadan alcohol intake range of 1–2 l/week. The remainingmembers of the two groups did not consume alcohol.The routine medical examinations revealed no clini-cal or biochemical abnormality in any of the membersof either group, and all the subjects were defined ashealthy.

Heparinized blood samples were taken at the occa-sion of the periodical medical examination. Each bloodsample was divided into two aliquots of 5 and 15 ml,respectively. The former were frozen (−80◦C) for sub-sequent DNA isolation and genotyping analysis, andthe latter were freshly processed for the micronucleustest.

2.3. Assessment of ionizing radiation exposure

Subjects in the exposed group are routinely exposedto X- and�-rays. Their occupational exposure to ion-izing radiation is routinely monitored by personal ex-posure measurements devices (film badges) that areread every 40 days. The personal film badge is hungon the hospital uniform and all radiological workersare trained to use it correctly. The dose equivalent ofionizing radiation to the whole body (Hwb), accumu-lated over the entire working-life period and adoptedfrom the personal dosimetry records, ranged from 0.90to 116.74 mSv (mean 40.61± 37.70). No differencein meanHwb and qualities of radiation was recordedeither between the never smokers and current smok-ers (42.00± 34.95 versus 37.83± 45.57; P= 0.818)or between females and males (49.63± 39.74 versus22.56± 27.30;P= 0.123). None of the workers hadrecorded doses above the annual limit of 20 mSv or the5 yearly limit of 100 mSv. Routine activity quality con-trol and exposure assessment assurance activities areperformed in accordance with Italian Law no. 230/95[

2

sis-b o-l cul-t1 um,1 -b Bw t2 eret toss inedb Ina

33] based on Euratom Directive no. 84/466[34].

.4. Micronuclei assay

The MN assay was performed using the cytokinelock technique[6]. Peripheral lymphocytes were is

ated by Histopaque gradient centrifugation andured at a concentration of 2× 106 in 5 ml RPMI640 medium supplemented with 15% fetal calf ser% phytohemaglutinin, 1 mMl-glutamine and incuated at 37◦C in 5% CO2 for 72 h. Cytochalasinas added (final concentration 6�g/ml) for the las8 h. At the end of the incubation time, cells w

reated with a mild hypotonic and fixed accordingtandard procedures in our laboratory[35,36]. Thelides were prepared by cytocentrifugation and stay conventional May Gr̈unwald–Giemsa staining.ccordance with standard criteria[37], MN analysis


was performed on coded slides by scoring 2000 bin-ucleated (BN) lymphocytes for each subject. Clas-sifying 1000 cells according to the number of nu-clei evaluated cell cycle parameters. The nuclear divi-sion index (NDI) was calculated following the formulaNDI = (M1 + 2M2 + 3M3 +4M4)/n, whereM1–M4 in-dicates the number of cells with 1–4 nuclei andn thetotal number of cells scored[38]. All the reagents werepurchased from Sigma, St. Louis, MO.

2.5. DNA extraction

DNA was isolated from peripheral lymphocytes us-ing a modified protocol from Daly et al.[39]. The DNAwas purified by adding 500�l 70% ethanol, twice, fol-lowed by centrifugation at 4◦C. The purified DNA wasspooled, allowed to air dried at room temperature for anhour, and resuspended in 100�l of TE buffer [10 mMTris–HCl, pH 7.4; 1 mM EDTA, pH 8.0]. For PCR am-plification, DNA samples were diluted and stored as10 ng/�l aliquots at−20◦C. All the reagents were pur-chased from Sigma.

2.6. Genotype analysis by PCR-RFLP

Single nucleotide polymorphisms inXRCC1,XRCC3and exon 23 of theXPD genes were deter-mined by PCR-RFLP based method as described ear-lier [40,41]. The amplified fragments were digested at37◦C for at least 3 h to overnight with the appropriater d-u ands erer m-p

2

ler ly-m plesa uslyd1 andi 5’-G er5 onw g Ta

ble

2R

estr

ictio

nen

zym

esus

edfo

rge

noty

ping

SN

Ps

inD

NA

repa

irge

nes

and

alle

lefr

eque

ncie

sin

the

stud

ypo

pula

tion

Gen

e(b

ase

chan

ge)

Res

tric

tion

enzy

me

Gen

otyp

edi

strib

utio

nin

the

stud

ypo

pula

tion

Rad

iolo

gica

lw

orke

rs(n)

Fre

quen

cyva

riant

alle

leC

ontr

ols

(n)F

requ

ency

varia

ntal

lele

Pa

XRCC1,

exon

10(G

toA

)Msp

IG

G(1

2)G

A(5

)A

A(4

)qA

=0.

310

GG

(11)

GA

(8)

AA

(2)

qA

=0.

286

0.20

0XRCC3,

exon

7(C

toT

)Hsp

92II

CC

(5)

CT

(10)

TT

(6)

qT

=0.

524

CC

(8)

CT

(7)

TT

(5)

qT

=0.

425

0.27

1XPD

,exo

n10

(Gto

A)

Non

eG

G(7

)G

A(8

)A

A(6

)qA

=0.

476

GG

(9)

GA

(7)

AA

(4)

qA

=0.

375

0.45

2XPD

,int

ron

10(I

VS

10+

17C

)N

one

C−/−

(16)

C−/+

(5)

C+

/+(0

)qC+/+

=0.

119

C−/−

(16)

C−/+

(4)

C+

/+(0

)qC+/+

=0.

100

0.88

2XPD

,exo

n23

(Ato

C)

PstI

AA

(8)

AC

(8)

CC

(5)

qC

=0.

429

AA

(8)

AC

(8)

CC

(4)

qC

=0.

400

0.88

2

Num

ber

ofin

divi

dual

sw

ithpa

rtic

ular

geno

type

sar

egi

ven

(n)

.a

χ2

test

was

used

for

com

paris

onof

geno

type

dist

ribut

ions

betw

een

radi

olog

ical

wor

kers

and

cont

rols

.
estriction enzyme (Table 2). The digested PCR procts were analyzed on 10% polyacrylamide gelstained with ethidium bromide. Genotype results wegularly confirmed by random repetition of the sales.
.7. Genotype analysis by SSCP

XPD exon 10 polymorphism, with no utilizabestriction site, and the novel intron 10 C-ins poorphism were analyzed by SSCP. The princind details of the method have been previoescribed by us in Kumar et al.[42]. Briefly a03 bp fragment containing parts of exon 10

ntron 10 was amplified using forward primerGAGACGGACGCCCACCTG and reverse prim’-GACGGGGAGGCGGGAAAG. The PCR reactias carried out in 10�l volume containing 10 n


template DNA, 50 mM KCl, 1 mM MgCl2, 0.11 mMdNTPs, 1�Ci [�-32P]dCTP, 0.15�M each primer and0.30 UTaqDNA polymerase. DMSO (10%) was in-cluded as co-solvent. Temperature conditions for PCRwere initial denaturation at 94◦C for 1 min, followedby three cycles at 95◦C (denaturation) for 45 s, 68◦C(annealing) for 45 s, and 72◦C (extension) for 45 s. Thiswas followed by another 32 cycles with an annealingtemperature of 67◦C and each segment time reducedto 30 s. The final extension was for 5 min at 72◦C. Theamplified products, denatured at 95◦C for 3 min, wereanalysed by SSCP on a 10% polyacrylamide gel withelectrophoresis carried out overnight at 4◦C using 5 Wpower. Electrophoresis under these conditions gave op-timal resolution of both polymorphisms and the sepa-rate strands containing the G23591A and 23623C-inspolymorphisms migrated differently. The genotype re-sults were confirmed by direct DNA sequencing usingRhodamine dye terminator cycle sequencing kit (Bigdye, Applied Biosystem, Foster City, CA).

2.8. Statistical analysis

The Wilcoxon rank-sum test was used to test thedifference in MN frequency between exposed workersand controls and to test significant associations betweenMN and the various genotypes. This non-parametrictest allows problems related to equal variance, normal-ity assumption or those regarding the use of frequenciesto be obviated[43]. For data even not-normally dis-t ed tos es-s n be-t rs.P eval-u andd ncyi tely.A evers eno-t witht iuma posedg aly-s n ofn sisf ter-m ex

is also discussed. The level of significance was set atP< 0.05. All statistical analyses were performed usingthe STATA-6 program.

3. Results

3.1. MN analysis

The results of the MN analysis, reported as to-tal number of MN per 1000 BN cells, are shown inTable 3. MN frequency was significantly higher in ra-diological workers than in controls [8.62± 2.80 versus6.86± 2.65;P= 0.019].Hwbof ionizing radiation didnot influence the MN frequency observed in exposedworkers [β = 0.004,P= 0.941, 95% CI =−0.12–0.13;Fig. 1]. When smoking habits were taken into accountfor each group, MN frequencies were higher in cur-rent smokers than in never smokers, but the differenceswere not statistically significant. Among current smok-ers, a higher but non-significant MN frequency wasfound in the exposed workers group than in controls[9.50± 3.46 versus 7.00± 1.85]. The range of expo-sure duration (years of employment) to ionizing radia-tion for radiological workers was 4–34 years (Table 1).No significant association emerged between years ofemployment and MN frequency. As regards NDI, nosignificant overall difference was found between theexposed workers and the controls [1.58± 0.18 ver-sus 1.52± 0.12;P= 0.158], even when smoking habitsw

t in-c t MNf der-l roup[ didn oup.P tudyp ef-f e ont gh(

3

udyw ilib-

ributed standard deviation (S.D.) has been reporthow variability in the study population. Linear regrion analysis was applied to assess the correlatioween theHwband MN frequency in exposed workeoisson regression analysis was also applied toate the influence of age, gender, smoking statusose equivalent of ionizing radiation on MN freque

n the overall population and in both groups separas regards smoking, only current smokers and nmoker were considered in the analysis. The gype frequencies were checked for consistencyhose expected from the Hardy–Weinberg equilibrnd compared between the exposed and the unexroups by�2 test. Single and Poisson regression anis were performed after logarithmic transformatioon-normally distributed variables. A power analy

or the design of future population studies to deine the effect of DNA repair genotype on MN ind

ere taken into account (data not shown).The results of Poisson regression analysis tha

luded age, gender and smoking status show tharequency tended to rise with age, although a borine association emerged only in the exposed gP= 0.079] (Table 4). Smoking status and genderot influence genetic damage in any studied groisson regression analysis applied to the overall sopulation revealed no smoking- and age-related

ects whereas gender exerted a marginal influenche yield of MN, with micronuclei frequency beinigher in females as compared to males [P= 0.053]Table 4).

.2. Genotype analysis

The genotype distribution found in the present stas in accordance with the Hardy–Weinberg equ


Table 3MN frequencies in the study population subdivided by exposure status and genotype distribution

Radiological workers Controls

Subjectsn MN/1000 BN mean± S.D. CI Subjectsn MN/1000 BN mean± S.D. CI

All 21 8.62 ± 2.80a 7.34–9.89 21 6.86± 2.65 5.65–8.06Never smokers 13 8.08± 2.29 6.69–9.46 13 6.77± 3.11 4.89–8.65Current smokers 8 9.50± 3.46 6.60–12.40 8 7.00± 1.85 5.45–8.55

XRCC1-399b

GG 12 7.92± 2.47 6.35–9.48 11 7.09± 1.38 6.17–8.01GA + AA 9 9.56 ± 3.09c 7.18–11.93 10 6.60± 3.66 3.98–9.22

XRCC3-241CC 5 8.00± 4.47 2.45–13.55 8 6.75± 1.04 5.88–7.61CT 10 8.20± 1.23d 7.32–9.08 7 6.00± 1.63 4.49–7.51TT 6 9.83± 3.19 6.49–13.18 5 6.40± 2.61 3.16–9.64

XPD-312b

GG + GA 15 9.00± 2.48e 6.69–9.60 16 6.50± 1.48 5.33–7.80AA 6 7.67 ± 3.56 3.93–11.40 4 6.00± 1.63 3.40–8.60

XPD-751AA 8 9.87 ± 3.27f 7.14–12.6 8 6.50± 1.41 5.32–7.68AC 8 8.50± 2.39 6.50–10.50 8 6.50± 2.07 4.77–8.23CC 5 6.80± 1.79 4.58–9.02 4 6.00± 2.22 3.40–8.60

Haplotypeg

1 10 9.60± 2.95h 7.49–11.71 11 6.73± 1.62 5.64–7.812 11 7.73± 2.45 6.08–9.37 9 6.00± 1.73 4.67–7.33

BN, binucleated cells.a Significantly different from total controls (Wilcoxon test,P= 0.01).b Heterozygous and homozygous allele carriers were pooled to increase sample size.c Significantly different from controls with the same genotype (Wilcoxon test,P= 0.026).d Significantly different from controls with the same genotype (Wilcoxon test,P= 0.013).e Significantly different from controls with the same genotype (Wilcoxon test,P= 0.026).f Significantly different from controls with the same genotype (Wilcoxon test,P= 0.024).g Recombinant haplotype between exon 10 and 23 of theXPDgene.h Significantly different from controls with the same haplotype (Wilcoxon test,P= 0.011).

rium for all the polymorphisms studied, both in ra-diological workers and controls. The genotype fre-quencies were in agreement with those reported pre-viously [19,40–42]. The polymorphisms in codon 312and 751 of theXPDgene appeared to be in linkage dis-equilibrium in this study, consistent with other reports[27–29]. The recombinant haplotypes were found inboth studied groups; there was no difference in inferredhaplotype distribution between radiological workersand controls.

3.3. Links between genotypes and MN frequency

Distribution of MN frequencies by DNA repairgenotypes and exposure status are shown inTable 3.

Individuals homozygotes for theXRCC1variant al-lele were rare and to increase statistical power weregrouped together with heterozygotes [Arg-Gln/Gln-Gln]. Among radiological workers genetic polymor-phism inXRCC1gene was associated with marginallyelevated DNA damage, MN frequency being higherin the Arg-Gln/Gln-Gln genotype group compared tosubjects homozygous wild-type [9.56± 3.09 versus7.92± 2.47; P= 0.160]. On the contrary, a decreasein MN frequency was observed in controls with Arg-Gln/Gln-Gln genotype as compared to homozygouswild-type individuals, however, the two value wheresimilar [7.09± 1.38 versus 6.60± 3.66]. When wecompared radiological workers with controls, in theexposed both genotype groups had a higher MN fre-


Fig. 1. Relationship between DNA damage, assessed as MN frequency in peripheral lymphocytes, and dose equivalent of ionizing radiation(Hwb). Hwb represent the personal exposure to ionizing radiation accumulated over the working-life period. The thick line is the result of linearregression analysis of the data, after their logarithmic transformation.

quency compared to controls. A significantly higherMN frequency was only found in radiological work-ers in the Arg-Gln/Gln-Gln genotype group com-pared to controls in the same pooled genotype group[9.56± 3.09 versus 6.60± 3.66;P= 0.026].

No detectable influence of theXRCC3polymor-phism on the MN frequency was observed in the

Table 4Poisson regression analysis of confounding factors on MN frequen-cies in peripheral lymphocytes of the study groups

Confounding factorsa IRR P 95% CI

AllAge (in years) 1.008 0.290 0.993–1.023Gender (1, 2) 0.780 0.053 0.606–1.003Smoking status (0, 1) 1.018 0.882 0.802–1.292Hwb (mSv) 1.252 0.044 1.006–1.559

ControlsAge (in years) 0.999 0.913 0.980–1.018Gender (1, 2) 0.824 0.272 0.584–1.164Smoking status (0, 1) 1.118 0.495 0.812–1.540

Exposed workersAge (in years) 1.021 0.079 0.998–1.044Gender (1, 2) 0.737 0.107 0.508–1.068Smoking status (0, 1) 0.910 0.603 0.637–1.299

IRR: incidence rate ratio,Hwb: dose equivalent to the whole bodyof ionizing.

a Gender: 1, females; 2, males; smoking status: 0, never; 1, cur-rent.

control group. Regarding hospital workers, subjectshomozygous for the variant allele showed a ten-dency to a higher MN frequency when comparedto the ones heterozygous and homozygous wild-type[9.83± 3.19, 8.20± 1.23 and 8.00± 4.473, respec-tively]. However, these differences were not statisti-cally significant. In the radiological workers, all thethree genotype groups had an higher MN frequencywhen compared to controls, although a significanceemerged only in the exposed heterozygous as com-pared to controls with the same genotype [8.20± 1.23versus 6.00± 1.63;P= 0.013], while only a marginalinfluence was observed in the exposed homozygouspolymorphic when compared to controls with the samegenotype [9.83± 3.19 versus 6.40± 2.61;P= 0.098].

ConcerningXPD exon 23, in the exposed work-ers MN frequencies were decreased in individualshomozygous for the variant allele when comparedwith those homozygous wild-type and heterozygous[6.80± 1.79, 9.87± 3.27 and 8.50± 2.39, respec-tively]. In particular, the difference was of borderlinesignificance among homozygous polymorphic as com-pared to homozygous wild-type [P= 0.075]. Amongcontrols only subjects homozygous for the variant al-lele had a slightly lower frequency of genetic damage ascompared to heterozygous and homozygous wild-type.When comparing radiological workers with controls, astatistically significant higher MN frequency was found


in exposed workers homozygous for the wild-type al-lele as compared to controls with the same genotype[9.87± 3.27 versus 6.50± 1.41;P= 0.024].

RegardingXPD exon 10 polymorphism, individ-uals homozygote wild-type and heterozygote [Asp-Asp/Asp-Asn] were grouped together both in controlsand radiological workers. In controls MN frequencywas not affected by genetic polymorphism inXPDexon 10. Among radiological workers, MN frequencywas higher in the Asp-Asp/Asp-Asn group than thewild-type one. In radiological workers, both genotypegroups showed higher MN as compared to controls; asignificant difference was observed only in the exposedAsp-Asp/Asp-Asn genotype group as compared to con-trols in the same Asp-Asp/Asp-Asn genotype group[9.00± 2.48 versus 6.50± 1.71;P= 0.023]. Intron 10polymorphisms did not show any effect on the modu-lation of the genetic damage (data not shown).

When the individual haplotype for the exon 10 and23 polymorphisms of theXPD gene was considered,we categorized both radiological workers and controlsinto two subgroups, namely haplotype 1 and haplotype2. Haplotype 1 was defined by grouping individualshomozygous wild-type in exon 23 and homozygouswild-type or polymorphic or heterozygous in exon 10,or vice versa homozygous wild-type in exon 10 and ho-mozygous wild-type or polymorphic or heterozygousin exon 23. Haplotype 2 was defined by grouping in-dividuals with at least one variant allele in both theexons. This include individuals heterozygous or ho-m et-e Wef di-o redts e-ma ervedi icalw riousg am-p fora on-s allya er ofc e of0 onf

Poisson regression analysis, including genotypingdata, smoking status and duration of employment wasdone revealing no significant associations with MN fre-quency.

4. Discussion

Knowledge about the carcinogenic effects of ion-izing radiations has advanced substantially since theassociation between skin cancer and radiations was ob-served among radiologists in 1902. Although the prin-ciple that exposure to high doses of ionizing radiationcauses cancer is well-established, the real cancer riskof chronic low dose exposures has not yet been eluci-dated and studies on individuals professionally exposedto ionizing radiation have the potential to address thisquestion.

The strengths and the novelty of the present studyis the investigation of biological markers of effectand susceptibility on the same population exposed tochronic low level of ionizing radiation. To our bestknowledge, the present population study is the first invivo study that combines genotype analysis in DNA re-pair genes with both exposure and MN frequency in so-matic cells of hospital workers, occupationally exposedto low levels of ionizing radiation, and controls. Thisapproach offers the opportunity to complete cancer pre-vention programs for health surveillance of radiologi-cal workers, still largely based on physical dosimetry.

bleb op-u tion[ en-h cals rols.T tionw as-s ituh onM ig-i neg-a cen-t nicaT fer-ec in-

ozygous polymorphic in exon 23 and individuals hrozygous or homozygous polymorphic in exon 10.

ound significantly higher MN frequency in the ralogical workers having the haplotype 1 as compa

o controls with the same haplotype [9.60± 2.95 ver-us 6.73± 1.62;P= 0.011]. This finding is in agreent with the results observed for theXPD exon 10nd 23 separately. Based on the differences obs

n the frequency of MN between exposed radiologorkers and controls and differences between vaenotypic groups in this study, it is desirable that sle size for future studies should be large enoughttaining necessary statistical power. Taking into cideration the differences observed in this study, idesample size of 250 cases and a matched numb

ontrols should give 90% power at an alpha valu.001, which will also include provision for correcti

or multiple comparisons.

MN frequency is already shown to be a reliaiomarker in earlier cytogenetic studies among plations occupationally exposed to ionizing radia

9,10,36,44–54]. Most of these studies reportedanced yields of MN, some of them with statistiignificance, in exposed populations than in conto date only a few biomonitoring studies on radiaorkers have used the micronucleus-centromereay[53,55–58]. The application of fluorescence in sybridization (FISH) with pan-centromeric probesN preparation allows distinction between MN or

nating from chromosome breakage (centromeritive) and those formed from chromosome loss (

romeri positive), reflecting, respectively, clastogend aneugenic actions of ionizing radiation[53,54].he results ranged from a lack of statistical difnces between exposed and controls[56] to a signifi-antly higher centromeri negative MN in uranium m


ers [55,58] and industrial radiographers[53], or in-versely, a significantly higher centromeri positive MNin radiological workers[57]. In our study, MN werenot scored for the presence of centromeri, thus, wecannot state exactly the nature of radiation-induced cy-togenetic damage after protracted low dose exposures.Anyway for a clear-cut conclusion concerning the radi-ation damage in low dose worker groups future largerstudies applying the centromeri micronucleus assay arewarranted.

In our study, as reported in others, no associationbetween chromosome damage andHwb was found[9,36,44,45]. The lack of a dose–effect relationship be-tween chromosome damage and chronic exposure tolow level of ionizing radiation could be attributed tovarious factors, which may, in combination, prevent thedetection of exposure effects by cytogenetic biomark-ers. It has been hypothesized that chronic low levelexposure could lead to the induction of repair enzymesuntil saturation is reached[59]. Another contributorymechanism could be the elimination by apoptosis ofdamaged cells during cell division in vivo[53].

Poisson regression analysis showed that smokingstatus did not affect genetic damage in any of the stud-ied groups. Recently, it has been reported that onlyheavy smokers (daily consumption >30 cigarettes) ex-hibited a significant increase in MN frequency[60]. Inour study, there were few smokers and among themheavy smokers were under-represented. These consid-erations, besides the small sample size could explaint MNf ok-i ansb ex-p fre-qI -q suret ub-j thect by ac -rayi oseo ffecto eticm onicl

MN frequency tended to rise with age, althougha borderline association emerged only in the exposedgroup. This effect could be explained by genetic insta-bility (GI) induced by ionizing radiation, which couldhave the effect of enhancing the hypothesised progres-sive chromosome instability related to the aging pro-cess. The GI, observed in a variety of mammalian sys-tems and mammalian cells in culture, may representthe first critical step in the genesis of certain radiation-induced cancers[61,63] even if the molecular, bio-chemical, and cellular events that initiate this instabilityin irradiated cells remain unknown[64].

Poisson regression analysis on the overall study pop-ulation revealed that gender exerted a marginal influ-ence on the yield of MN, being higher in females ascompared to males. Current knowledge on the effectof gender on genetic damage shows that MN frequen-cies in females are generally 1.2–1.6 times greater thanthose in males[8,65]. In our study, the absence of astatistically observable effect is probably due to therelatively small sample size in which females are un-derrepresented. On the other hand, we were unable tofind a correlation with the age. This is inconsistent withour current knowledge that has come from biomonitor-ing studies, in which MN frequency increases signifi-cantly and steadily with age[66,67]; this age-effect hasrecently been confirmed by data on nearly 7000 sub-jects of the Human MicroNucleus Project[65]. Theinconsistency of our data is probably due to the rela-tively small size of the study population that could haved

n be-ta NPsh ionm ba-s hatca ub-o eser n bei n-i o ac-c rk-e di-a ana zingr

he apparent null effect of smoking on the observedrequencies. Regarding the possible influence of smng habits on chromosomal damage induced in humy ionizing radiation, in a previous study we foundosed smokers exhibited significantly higher MNuencies as compared with exposed non-smokers[36].

nterestingly, when Wang et al.[61] measured the freuency of chromosome damage after in vitro expo

o �-rays in peripheral lymphocytes of healthy sects, they showed that cigarette smoking affectedell radiosensitivity. Moreover, Au et al.[62] foundhat chromosome aberration frequency, evaluatedhallenge assay, was consistently higher, after Xrradiation, in lymphocytes from smokers than in thf non-smokers. These findings suggest that the ef cigarette smoking should be considered in genonitoring for assessing risk associated with chr

ow level radiation exposure.

ecreased the power of statistical analysis.Genotype analysis revealed a clear associatio

ween theXRCC1, XRCC3andXPD polymorphismsnd MN frequencies. Our observation that these Save a clear effect at low level of ionizing radiatight imply a gene-exposure interaction. Even the

is for this low level effect is not clear, is possible tommonXPDalleles andXRCC1andXRCC3variantlleles may alter the protein products resulting in sptimal repair of X-ray-induced DNA damage. Thesults suggest that SNPs in DNA repair genes camportant biomarkers of susceptibility for radiationduced genetic damage, and should be taken intount in future biomonitoring studies of hospital wors chronically exposed to low level of ionizing ration. To date few in vitro studies have reportedssociation between increased sensitivity to ioniadiation and SNPs in DNA repair genes[19,26]. Since


some of the amino acid variants are very common, ourfindings have biological and public health significance.

The small number of individual in our study pop-ulation precluded further statistical analysis of possi-ble relationship between the studied polymorphismsand other genetic and life-style factors. Moreover otherevents, such as effects of life-style or genetic factorscould contribute to the increase of MN frequency ob-served by us in workers exposed to ionizing radia-tion. In particular, dietary may be as well a key fac-tor in determining chromosome damage in humans. Itis known that many micronutrients, mineral and Vi-tamins act as substrates and/or co-factor in key DNAmaintenance reactions, and their exact concentrationsin the cells may be critical[68]. Folate and VitaminB12 deficiency, for example, have been shown to causeMN formation[69,70]. In our study, plasma concentra-tions of folic acid and Vitamin B12 were not measured,however, the daily intakes of these micronutrients, esti-mated by a FFQ, resulted for all the participants abovethe recommended daily intake. Moreover, all the sub-jects enrolled in the study were considered as healthy,not being alcohol addicted and not having dietary defi-ciencies, suggesting that the increased MN frequencyobserved may not be associated with dietary factors.On the other hand, one must be mindful that the useof FFQ may over- or underestimate actual nutrient in-take, thus is not possible to rule out the influence ofthese micronutrients on MN frequencies and future in-vestigation should be made to clarify the relationshipb 12a erall sedt

fini-t cedb ce ofg the-l eful-n arkerf osedt hatX -t ue tor vi-d con-s l ex-p di-

tional epidemiological studies of these and other DNArepair-polymorphisms will provide essential informa-tion about the in vivo relationships between the DNArepair mechanisms and radiation-induced chromosomedamage and will complement in vitro analysis. In thefuture larger studies could be targeted to healthy indi-viduals under controlled life-style conditions, known tointerfere with MN formation (e.g. nutrient dietary in-take and smoking habits). This approach will allow us abetter chance to investigate the role of this polymorphicDNA repair genes on MN frequency. If an unequivocalassociation between MN frequency and DNA repairgenotypes could be developed, the identification of in-dividuals/groups at increased risk of radiation-induceddamages would then be helpful to the field of occupa-tional medicine. In particular, larger and well-designedepidemiological studies, combined with advances inour understanding of molecular mechanisms, may behelpful to further illuminate the complex landscape ofDNA repair and cancer risk associated with protractedexposure to low level of ionizing radiation.

Acknowledgment

We are grateful to Dr. Stefano Mattioli, Occupa-tional Medicine Unit, S. Orsola-Malpighi Hospital, forstatistical evaluation of data.

R

Ef-mic

nd,skin,ncersing

ncil,ion,DC,

n on60,

allianaly,

etween plasma levels of folic acid and Vitamin Bnd the increased MN frequency observed in periph

ymphocytes from individuals occupationally expoo ionizing radiation.

Even though the small sample size prevents a deive conclusion on the enhanced MN frequency induy occupational exposure, and the possible influenenetic polymorphisms in DNA repair genes, never

ess, our experimental evidences highlight the usess of the micronucleus assay, as a biological m

or assessing genetic damage in populations expo low levels of ionizing radiation, and suggest tRCC3, XRCC1andXPDpolymorphisms might con

ribute to increase the genetic damage, possibly deduced DNA repair function. This effect might be eent in individuals with accumulated damage as aequence of low repair capacity and occupationaosure to chronic low level of ionizing radiation. Ad

eferences

[1] UNSCEAR, Ionizing Radiation: Sources and Biologicalfects, United Nations Scientific Committee on Effects of AtoRadiation, New York, NY, 1982.

[2] D.J. Brenner, R. Doll, D.T. Goodhead, E.J. Hall, C.E. LaJ.B. Little, J.H. Lubin, D.L. Preston, R.J. Preston, J.S. PuE. Ron, R.K. Sachs, J.M. Samet, R.B. Setlow, M. Zaider, Carisks attributable to low doses of ionizing radiation: asseswhat we really know, PNAS 100 (2003) 13761–13766.

[3] National Academy of Sciences–National Research CouHealth effects of exposure to low levels of ionizing radiatBEIR V Report, National Academy Press, Washington,1989.

[4] IRCP, Recommendations of the International CommissioRadiological Protection, in: Annals of the ICRP, PublicationPergamon Press, Oxford, UK, 1991.

[5] AIRM, Guidelines for Diagnostic Examination in MedicSurveillance of Workers Exposed to Ionizing Radiation, ItaAssociation of Medical Radiological Protection, Rome, It1995.


[6] M. Fenech, A.A. Morley, Measurement of micronuclei in lym-phocytes, Mutat. Res. 147 (1985) 29–36.

[7] M. Fenech, N. Holland, W.P. Chang, E. Zeiger, S. Bonassi,The human micronucleus project. An international collabora-tive study on the use of the micronucleus technique for measur-ing DNA damage in humans, Mutat. Res. 428 (1999) 271–283.

[8] M. Fenech, Important variables that influence base-line mi-cronucleus frequency in cytokinesis-blocked lymphocytes asbiomarker for DNA damage in human populations, Mutat. Res.404 (1998) 155–165.

[9] R.S. Cardoso, S. Takahashi-Hyodo, P. Peitl Jr., T. Ghilardi-Neto,E.T. Sakamoto-Hojo, Evaluation of chromosomal aberrations,micronuclei, and sister chromatid exchanges in hospital workerschronically exposed to ionizing radiation, Teratog. Carcinog.Mutagen. 21 (2001) 431–439.

[10] V.B. Hadjidekova, M. Bulanova, S. Bonassi, M. Neri, Micronu-cleus frequency is increased in peripheral blood lymphocytesof nuclear power plant workers, Radiat. Res. 160 (2003) 684–690.

[11] W.P. Chang, B.F. Hwang, D. Wang, J.D. Wang, Cytogeneticeffect of chronic low-dose, low-dose-rate gamma-radiation inresidents of irradiated buildings, Lancet 350 (1997) 330–333.

[12] J.H. Hoeijmakers, Genome maintenance mechanisms for pre-venting cancer, Nature 411 (2001) 366–374.

[13] M. Berwick, P. Vineis, Markers of DNA repair and susceptibilityto cancer in humans: an epidemiologic review, J. Natl. CancerInst. 92 (2000) 874–897.

[14] E.C. Friedberg, DNA damage and repair, Nature 421 (2003)436–440.

[15] M. Digweed, Response to environmental carcinogens in DNArepair-deficient disorders, Toxicology 193 (2003) 111–124.

[16] J. Hall, S. Ang̀ele, Radiation, DNA damage and cancer, Mol.Med. Today 5 (1999) 157–164.

[17] R.A. Gatti, The inherited basis of human radiosensitivity, ActaOncol. 40 (2001) 702–711.

[ uman35.

[ A.

vity,

[ hy-pli-

[ en,aks–97.

[ g

[ neticma

[ ion-val ofNatl.

[25] S.A. Leadon, P.K. Cooper, Preferential repair of ionizingradiation-induced damage in the transcribed strand of an activehuman gene is defective in Cockayne syndrome, Proc. Natl.Acad. Sci. U.S.A. 90 (1993) 10499–10503.

[26] R.M. Lunn, K.J. Helzlsouer, R. Parshad, D.M. Umbach, E.L.Harris, K.K. Sanford, D.A. Bell, XPD polymorphisms: effectson DNA repair proficiency, Carcinogenesis 21 (2000) 551–555.

[27] D. Butkiewicz, M. Rusin, L. Enewold, P.G. Shields, M. Chorazy,C.C. Harris, Genetic polymorphisms in DNA repair genes andrisk of lung cancer, Carcinogenesis 22 (2001) 593–597.

[28] M.R. Spitz, X. Wu, Y. Wang, L.E. Wang, S. Shete, C.I. Amos,Z. Guo, L. Lei, H. Mohrenweiser, Q. Wei, Modulation of nu-cleotide excision repair capacity by XPD polymorphisms inlung cancer patients, Cancer Res. 61 (2001) 1354–1357.

[29] D. Tomescu, G. Kavanagh, T. Ha, H. Campbell, D.W. Melton,Nucleotide excision repair gene XPD polymorphisms and ge-netic predisposition to melanoma, Carcinogenesis 2 (2001)403–408.

[30] Law no. 675, Tutela delle persone e di altri soggetti rispetto altrattamento dei dati personali. In: Gazzetta ufficiale 1997, No.5, S.O., 1996.

[31] P. Pisani, F. Faggiano, V. Krogh, D. Palli, P. Vineis, F. Berrino,Relative validity and reproducibility of a food frequency di-etary questionnaire for use in the Italian EPIC centres, Int. J.Epidemiol. 26 (1997) S152–S160.

[32] E. Carnevale, L. Marletta, Tabelle di Composizione degli ali-menti, Istituto Nazionale di Ricerca per gli Alimenti e la Nu-trizione, Roma, 2000.

[33] Law no. 230. Attuazione delle direttive Euratom 80/835,84/466, 84/667, 86/618, 90/641, 92/3 in material di radiazioniionizzanti. In: Gazzetta ufficiale 1995, No. 136, S.O., 1995.

[34] Council Directive 84/466/Euratom, Laying down basic mea-sures for the radiation protection of persons undergoing medicalexamination or treatment, 3 September 1984.

[35] F. Maffei, C. Fimognari, E. Castelli, G.F. Stefanini, G. Can-d by

rom

[ e,rk-ion:is 17

[ e: aeno-93)

[ dy-cytes. 13

[ D6

[ orm-ly-can-

18] T. Lindahl, Suppression of spontaneous mutagenesis in hcells by DNA excision-repair, Mutat. Res. 462 (2000) 129–1

19] J.J. Hu, T.R. Smith, M.S. Miller, H.W. Mohrenweiser,Golden, L.D. Case, Amino acid substitution variants ofAPE1andXRCC1 genes associated with ionizing radiation sensitiCarcinogenesis 22 (2001) 917–922.

20] L.F. Fuller, R.B. Painter, A Chinese hamster ovary cell linepersensitive to ionizing radiation and deficient in repair recation, Mutat. Res. 193 (1988) 109–121.

21] M.A. Brenneman, A.E. Weiss, J.A. Nickoloff, D.J. ChXRCC3 is required for efficient repair of chromosome breby homologous recombination, Mutat. Res. 459 (2000) 89

22] L.H. Thompson, M.G. West,XRCC1 keeps DNA from gettinstranded, Mutat. Res. 459 (2000) 1–18.

23] R. Parshad, R.E. Tarone, F.M. Price, K.K. Sanford, Cytogeevidence for differences in DNA incision activity in Xeroderpigmentosum group A, C and D cells afterx-irradiation duringG2 phase, Mutat. Res. 294 (1993) 149–155.

24] M.S. Satoh, C.J. Jones, R.D. Wood, T. Lindahl, DNA excisrepair defects of xeroderma pigmentosum prevents remoa class of oxygen free radical-induced base lesion, Proc.Acad. Sci. U.S.A. 90 (1993) 6335–6339.

telli Forti, P. Hrelia, Increased cytogenetic damage detecteFISH analysis on micronuclei in peripheral lymphocytes falcoholics, Mutagenesis 15 (2000) 517–523.

36] F. Maffei, S. Angelini, G. Cantelli Forti, V. Lodi, F.S. ViolantS. Mattioli, P. Hrelia, Micronuclei frequencies in hospital woers occupationally exposed to low levels of ionizing radiatinfluence of smoking status and other factors, Mutagenes(2002) 405–409.

37] M. Fenech, The cytokinesis-block micronucleus techniqudetailed description of the method and its application to gtoxicity studies in human population, Mutat. Res. 285 (1935–44.

38] D.A. Eastmond, J.D. Tucker, Identification of aneuploiinducing agents using cytokinesis-blocked human lymphoand an anti-kinetochore antibody, Environ. Mol. Mutagen(1989) 34–43.

39] A.K. Daly, V.M. Steen, K.S. Fairbrother, J.R. Idle, CYP2multiallelism, Meth. Enzymol. 272 (1996) 199–210.

40] S. Sanyal, F. Festa, S. Sakano, Z. Zhang, G. Steineck, U. Ning, H. Wijkstrom, P. Larsson, R. Kumar, K. Hemminki, Pomorphisms in DNA repair and metabolic genes in bladdercer, Carcinogenesis 25 (2004) 729–734.


[41] A. Forsti, S. Angelini, F. Festa, S. Sanyal, Z. Zhang, E. Grzy-bowska, J. Pamula, W. Pekala, H. Zientek, K. Hemminki, R.Kumar, Single nucleotide polymorphisms in breast cancer, On-col. Rep. 11 (2004) 917–922.

[42] R. Kumar, S. Angelini, K. Hemminki, Simultaneous detectionof the exon 10 polymorphism and a novel intronic single baseinsertion polymorphism in theXPD gene using single strandconformation polymorphism, Mutagenesis 18 (2003) 207–209.

[43] R.B. D’Agostino, A. Balanger, R.B. D’Agostino Jr., A sugges-tion for using powerful and informative test of normality, Am.Statist. 44 (1990) 316–321.

[44] J.F. Barquinero, L. Barrios, M.R. Caballin, R. Miro, M. Ribas,A. Subias, J. Egozcue, Cytogenetic analysis of lymphocytesfrom hospital workers occupationally exposed to low levels ofionizing radiation, Mutat. Res. 286 (1993) 275–279.

[45] A.D. Da Cruz, A.G. McArthur, C.C. Silva, M.P. Curado, B.W.Glickman, Human micronucleus counts are correlated with age,smoking and cesium-137 dose in the Goiana (Brazil) radiolog-ical accident, Mutat. Res. 313 (1994) 57–68.

[46] H. Thierens, A. Vral, L. De Ridder, A cytogenetic study of ra-diological workers: effect of age, smoking and radiation burdenon the micronucleus frequency, Mutat. Res. 360 (1996) 75–82.

[47] V. Garaj-vrhovac, A. Fucic, D. Kubelka, A. Hebrang, Assess-ment of genome damage in occupational exposure to ionizingradiation and ultrasound, Mutat. Res. 395 (1997) 101–105.

[48] G. Joksic, V. Spasojevic-Tisma, Chromosome analysis of lym-phocytes from radiation workers in tritium-applying, Int. Arch.Occup. Environ. Health 71 (1998) 213–220.

[49] I.H. Zwingmann, I.J. Welle, M. Van Herwijnen, J.J.M. Engelen,P.A.E.L. Schilderman, T. Smid, J.C.S. Kleijans, Oxidative DNAdamage and cytogenetic effects in flight engineers exposed tocosmic radiation, Environ. Mol. Mutagen. 32 (1998) 121–129.

[50] A. Heimers, Chromosome aberration analysis in concorde pi-lots, Mutat. Res. 467 (2000) 169–176.

[ ofd ul-00)

[ , As-osedssay,

[ A.sing

[ rs,l ex-age

[ i-ure

[ -us-

1999)

[57] H. Thierens, A. Vral, R. Morthier, B. Aousalah, L. De Ridder,Cytogenetic monitoring of hospital workers occupationally ex-posed to ionizing radiation using the micronucleus centromereassay, Mutagenesis 15 (2000) 245–249.

[58] A. Kryscio, W.U. Muller, A. Wojcik, N. Kotschy, S. Gro-belny, C. Streffer, A cytogenetic analysis of the long-term ef-fect of uranium mining on peripheral lymphocytes using themicronucleus-centromere assay, Int. J. Radiat. Biol. 77 (2001)1087–1093.

[59] M. Bauchinger, Quantification of low-level radiation exposureby conventional chromosome aberration analysis, Mutat. Res.339 (1995) 177–189.

[60] S. Bonassi, M. Neri, C. Lando, M. Ceppi, Y.P. Lin, W.P.Chang, N. Holland, M. Kirsch-Volders, E. Zeiger, M. Fenech,HUMN collaborative group. Effect of smoking habit on the fre-quency of micronuclei in human lymphocytes: results from theHuman MicroNucleus project, Mutat. Res. 543 (2003) 155–166.

[61] L.E. Wang, M.L. Bondy, M. DeAndrade, S.S. Strom, X. Wang,A. Sigurdson, M.R. Spitz, Q. Wie, Gender difference in smok-ing effect on chromosome sensitivity to gamma radiation in ahealthy population, Radiat. Res. 154 (2000) 20–27.

[62] W.W. Au, D.M. Walker, J.B. Ward Jr., E. Whorton, M.S. Lega-tor, V. Singh, Factors contributing to chromosome damagein lymphocytes of cigarette smokers, Mutat. Res. 260 (1991)137–144.

[63] J.B. Little, Radiation carcinogenesis, Carcinogenesis 21 (2000)397–404.

[64] L.E. Smith, S. Nagar, G.J. Kim, W.F. Morgan, Radiation-induced genomic instability: radiation quality and dose re-sponse, Health Phys. 85 (2003) 23–29.

[65] S. Bonassi, M. Fenech, C. Lando, Y.P. Lin, M. Ceppi, W.P.Chang, N. Holland, M. Kirsch-Volders, E. Zeiger, S. Ban, R.Barale, M.P. Bigatti, C. Bolognesi, C. Jia, M. Di Giorgio, L.R.Ferguson, et al., Human micronucleus project: international

k mi-tory

cy of

[ si,F.

onu-in anyle,

[ , L.ffect

Age

[ mic

[ ho-ocytes

[ ys-ralian

51] A. Fucic, D. Markucic, A. Mijic, A.M. Jazbec, Estimationgenome damage after exposure to ionizing radiation antrasound used in industry, Environ. Mol. Mutagen. 36 (2047–51.

52] S.W. Maluf, D.F. Passos, A. Bacelar, G. Speit, B. Erdtmannsessment of DNA damage in lymphocytes of workers expto �-radiation using the micronucleus test and the comet aEnviron. Mol. Mutagen. 38 (2001) 311–315.

53] I. Sari-Minodier, T. Orsiere, L. Bellon, J. Pompili, C. Sapin,Botta, Cytogenetic monitoring of industrial radiographers uthe micronucleus assay, Mutat. Res. 521 (2002) 37–46.

54] N. Touil, P.V. Aka, J.P. Buchet, H. Thierens, M. Kirsch-VoldeAssessment of genotoxic effects related to chronic low leveposure to ionizing radiation using biomarkers for DNA damand repair, Mutagenesis 17 (2002) 223–232.

55] C. Streffer, W.U. Muller, A. Kryscio, W. Bocker, Micronuclebiological indicator for retrospective dosimetry after exposto ionizing radiation, Mutat. Res. 404 (1998) 101–105.

56] H. Thierens, A. Vral, M. Barb̀e, B. Aousalah, L. De Ridder, A cytogenetic study of nuclear power plant workersing the micronucleus-centromere assay, Mutat. Res. 445 (105–111.

database comparison for results with the cytokinesis-bloccronucleus assay in human lymphocytes: I. Effect of laboraprotocol, scoring criteria, and host factors on the frequenmicronuclei, Environ. Mol. Mutagen. 37 (2001) 31–45.

66] R. Barale, L. Chelotti, T. Davini, S. Del Ry, M.G. AndreasM. Ballardin, M. Bulleri, J. He, S. Baldacci, F. DiPede,Gemignani, S. Landi, Sister chromatid exchange and micrcleus frequency in human lymphocytes of 1650 subjectsItalian population: II. Contribution of sex, age, and lifestEnviron. Mol. Mutagen. 31 (1998) 228–242.

67] C. Bolognesi, C. Lando, A. Forni, E. Landini, R. ScarpatoMigliore, S. Bonassi, Chromosomal damage and ageing: eon micronuclei frequency in peripheral blood lymphocytes,Ageing 28 (1999) 393–397.

68] M. Fenech, The role of folic acid and Vitamin B12 in genostability of human cells, Mutat. Res. 475 (2001) 57–67.

69] M. Fenech, I.E. Dreosti, J.R. Rinaldi, Folate Vitamin B12,mocysteine status and chromosome damage rate in lymphof older men, Carcinogenesis 18 (1997) 1329–1336.

70] M. Fenech, C. Aitken, J. Rinaldi, Folate Vitamin B12, homocteine status and chromosome damage rate in young Austadults, Carcinogenesis 19 (1998) 1163–1171.

Documents

Micronuclei in Humans Induced by Exposure to Low Level of Ionizing Radiation Influence of Polymorphisms in DNA Repair Genes